Ateneo de Zamboanga University

College of Science, Information Technology, and Engineering

`

Improving Flood Monitoring Systems: Utilizing

Ultrasonic, Turbidity, and Pressure Sensors for Early
Flood Detection and Response at Pasonanca Diversion
Weir

A Senior Design Project

Presented to the Faculty of the Engineering Department
College of Science, Information Technology, and Engineering
ATENEO DE ZAMBOANGA UNIVERSITY

by:
Joyen Rap Benzer I. Lagmay
Jan Angelo Tan Belda

Abdul Hadi H. Nograles

Adviser

July 2024
 APPROVAL SHEET

This thesis hereto entitled:

Improving Flood Monitoring Systems: Utilizing Ultrasonic, Turbidity, and

Pressure Sensors for Early Flood Detection and Response at Pasonanca
Diversion Weir

Prepared and submitted by Joyen Rap Benzer Lagmay and Jan Angelo Tan
Belda, in partial fulfillment of the requirements for the degree of Bachelor of
Science in Electronics Engineering has been examined and is recommended for
acceptance and approval by the adviser(s)

Engr. Abdul Had H. Nograles

Adviser

This thesis paper is hereby approved and accepted by the College of Science,
Information Technology, and Engineering as fulfillment of the thesis requirement
for the degree of Bachelor of Science in Electronics Engineering

Engr. Janet G. Tan

Chair, Engineering Department

Jocelyn D. Partosa, PhD, LPT

Dean, College of Science, Information Technology, and Engineering

2
 ACKNOWLEDGEMENTS

The success of this study can be attributed to the combined efforts of the team,
which have proven to be more beneficial than the efforts of the individual
researchers. These experts have opened up opportunities for progress and
learning in the study of this topic. The researchers would like to acknowledge and
appreciate the key individuals who played a crucial role in their achievements:

Above all, we express our gratitude to our Almighty God, who has given us
wisdom, strength, guidance, and determination to overcome obstacles and
complete this important task. Your limitless love and amazing grace leave us in
awe. You are the one who truly deserves all honor and praise!

To our parents, whose endless love and unwavering support have motivated us
to always give our best in everything we do. We are grateful for your care of our
financial, emotional, and spiritual needs. Your love and support have always
pushed us to excel!

To our thesis adviser, Engr. Abdul Hadi H. Nograles, for his unwavering
assistance in building prototypes, sharing expertise, and directing
implementation. His encouragement and guidance were crucial to the success of
this research project. We are truly grateful for his significant contribution to our
study.

To the Environmental Management Bureau (EMB) and Zamboanga City Water

District (ZCWD), we extend our heartfelt thanks for their warm welcome, granting
us permission to work on this project, and providing us with the privilege, trust,
and direction during the implementation phase.

To our relatives and friends, we express our gratitude for sharing their wisdom
and continuous prayers. Your steadfast faith has been a driving force in helping
us reach this milestone!

Finally, to each and every person who contributed in their own special way, thank
you for helping us overcome the greatest challenge of our college experience.

3
 TABLE OF CONTENTS

TITLE PAGE............................................................................................................................................

APPROVAL PAGE...................................................................................................................................

ACKNOWLEDGEMENT............................................................................................................................

TABLE OF CONTENTS.............................................................................................................................

LIST OF TABLES......................................................................................................................................

LIST OF FIGURES....................................................................................................................................

ABSTRACT..............................................................................................................................................

CHAPTER 1: INTRODUCTION..................................................................................................................

Background of the Study....................................................................................................

Statement of the Problem...................................................................................................

Objectives of the Study......................................................................................................

Significance of the Study....................................................................................................

Scope and Delimitations......................................................................................................

CHAPTER 2: REVIEW OF RELATED LITERATURE.......................................................................................

Background Context............................................................................................................

Research Questions.............................................................................................................

Synthesis of Literature.......................................................................................................

Methodological Issues........................................................................................................

Critical Evaluation..............................................................................................................

Integration with Research Questions/ Objectives................................................................

Summary and Conclusion...................................................................................................

4
 Implications for Future Research........................................................................................

CHAPTER 3: METHODOLOGY..................................................................................................................

Conceptual Framework......................................................................................................

Theoretical Framework......................................................................................................

Methodology....................................................................................................................

Integration of Sensor Data Modules...............................................................................

Ultrasonic Sensor .....................................................................................................

Sensor height sensitivity test ...............................................................................

Field testing accuracy reading ............................................................................

Comparison of Sensor height sensitivity test through error

checking .............................................................................................................

Comparison of Field testing accuracy reading through error

checking .............................................................................................................

Turbidity Sensor ......................................................................................................

Validation of Turbidity Sensor Accuracy Against Turbidimeter

Readings .............................................................................................................

Turbidity Sensor calibration using a serial dilution of NTU

(formazine) .........................................................................................................

Voltage Reading comparison (TS-300B vs Eutech Instrument

Turbidimeter) ......................................................................................................

Field Testing of Turbidity Sensor Accuracy Against Turbidimeter

Readings..............................................................................................................

5
 Atmospheric Pressure Sensor ..................................................................................

Standardized Atmospheric Pressure and Temperature Versus

Atmospheric Pressure and Temperature Sensor Data ..........................................

Atmospheric Pressure and Temperature Difference of Enclosed

Case Versus Open Case .......................................................................................

Integration of data storage and logging.........................................................................

Data storage and logging test.............................................................................

Gantt Chart........................................................................................................................

CHAPTER IV RESULTS AND DISCUSSION.................................................................................................

Sensor Height Sensitivity Test for A01NYUB ......................................................................

Sensor Height Sensitivity Test for JSN SR04T ......................................................................

Comparison of Sensor Height Sensitivity Test Through Error Checking ...............................

Field Testing Accuracy Reading of A01NYUB ......................................................................

Field Testing Accuracy Reading of JSN SR40T ......................................................................

Comparison of Field Testing Accuracy Reading Through Error Checking .............................

Turbidity Sensor Calibration Using a Serial Dilution of NTU (formazine) ..............................

Voltage Reading Comparison (TS-300B vs Eutech Instrument Turbidimeter)

.....................................................………………………………………………………………………………………

Field Testing of Turbidity Sensor Accuracy Against Turbidimeter Readings) .......................

………………………………………………………………………………………………………………91

Turbidimeter vs TS_300B NTU vs True NTU reading ...........................................................

6
 Standardized Atmospheric Pressure and Temperature versus Atmospheric

Atmospheric Pressure and Temperature Sensor Data .........................................................

Atmospheric Pressure and Temperature Difference of Exposed Case Versus

Enclosed Case .....................................................................................................................

Data Storage and Logging Test.............................................................................................

CHAPTER V CONCLUSION AND RECOMMENDATIONS............................................................................

REFERENCES..........................................................................................................................................

APPENDICES..........................................................................................................................................

7
 LIST OF TABLES
Table 2.0. Water level Ultrasonic Sensors used in Related Studies..............................................
Table 2.1. Turbidity Sensors used in Related Studies...................................................................
Table 2.2. Pressure Sensors used in Related Studies...................................................................
Table 3.0: Sensor height sensitivity test.........................................................................................
Table 3.1: Field testing accuracy reading......................................................................................
Table 3.2: Comparison of Sensor height sensitivity.......................................................................
Table 3.3: Comparison of Field testing accuracy...........................................................................
Table 3.4: Serial Dilution of 800 NTU (formazine) to create different NTU levels..........................
Table 3.5: Voltage reading of Turbidity sensor from different NTU solutions and
NTU readings of Eutech Instrument from the diluted NTU solutions..............................................
Table 3.6: Field testing of turbidity level: Eutech Instrument Turbidimeter vs
TS_300B NTU vs True NTU reading.............................................................................................
Table 3.7: Field testing of turbidity level: Eutech Instrument Turbidimeter vs Hach
2100Q............................................................................................................................................
Table 3.8: Data gathering for Standardized tools and BMP108 module at 15
minutes interval..............................................................................................................................
Table 3.9: Data gathering for the atmospheric pressure of the 3 devices: Sensor
A, Sensor B, and Reference device...............................................................................................
Table 3.10: Data gathering for the temperature of the 3 devices: Sensor A,
Sensor B, and Reference device...................................................................................................
Table 3.11 Data Storage and Logging Test...................................................................................
Table 4.0: Water Level Sensor height sensitivity.........................................................................
Table 4.1: Water Level Sensor height sensitivity for JSN SR04T...............................................
Table 4.2: Comparison of Sensor height sensitivity.......................................................................
Table 4.3: Field testing data (A01NYUB).......................................................................................
Table 4.4: Field testing data (JSN SR04T)....................................................................................
Table 4.5: Comparison of Field testing accuracy...........................................................................
Table 4.6: Serial Dilution of 800 and 100 NTU (formazine) to create different NTU
levels..............................................................................................................................................
Table 4.7 : Voltage reading of Turbidity sensor from different NTU solutions and
NTU readings of Eutech Instrument from the diluted NTU solutions..............................................
Table 4.8: Field testing of turbidity level: Eutech Instrument Turbidimeter vs
TS_300B NTU vs True NTU reading.............................................................................................
Table 4.9: Field testing of turbidity level: Eutech Instrument Turbidimeter vs Hach
2100Q............................................................................................................................................
Table 4.10: Data gathering for Standardized tools and BMP108 module at 15
minutes interval..............................................................................................................................
Table 4.11: F-statistics for temperature readings (standard vs. sensor)........................................
Table 4.12: Data gathering for the temperature of the 3 devices: Sensor A,
Sensor B, and Reference device...................................................................................................
Table 4.13: Comparison of Temperature Readings.......................................................................
Table 4.14: Data gathering for the air pressure of the 3 devices: Sensor A,
Sensor B, and Reference device...................................................................................................
Table 4.15: Comparison of Pressure Readings.............................................................................
Table 4.16: Comparison of Stored data vs Expected data.............................................................

8
 LIST OF FIGURES

Figure 2.0 Types of floods experienced by surveyed Flood Early Warning

Systems.........................................................................................................................................
Figure 2.1 Flood monitoring methods............................................................................................
Figure 2.2 Figure 2.2. Wang et al. (2018) turbidity sensor design.................................................
Figure 2.3 Figure 2.3: Schematic cross-section ............................................................................
Figure 3.0 Conceptual Design........................................................................................................
Figure 3.1: Pin layouts of the ultrasonic sensor.............................................................................
Figure 3.2: Power management of Ultrasonic sensor of the transmitter and
receiver..........................................................................................................................................
Figure 3.2: Power management of Ultrasonic sensor of the transmitter and
receiver .........................................................................................................................................
Figure 3.3: Pin layouts of the Turbidity Sensor connected to the microcontroller..........................
Figure 3.4: Power Management of the Turbidity sensor and microcontroller
device.............................................................................................................................................
Figure 3.5: Pin layouts of the BMP180 in the microcontroller........................................................
Figure 3.6: Power Management of BMP180 and microcontroller...................................................
Figure 3.7: Turbidity vs voltage level relationship, Turbidity graph................................................
Figure 3.8: Error plot of Standardized pressure data vs BMP108 pressure data...........................
Figure 3.9: Error plot of Standardized temperature data vs BMP108 temperature
data................................................................................................................................................
Figure 3.10: Pin layouts of the RTC and SD Card module integrated to the
system............................................................................................................................................
Figure 4.0: Error plot of water level readings (Gauge Reading vs A01NYUB)..............................
Figure 4.1: Turbidity vs voltage level relationship, Turbidity graph................................................
Figure 4.2: Error plot of Atmospheric pressure (Standardized vs BMP180)..................................
Figure 4.3: Error plot of Temperature (Standardized vs BMP180).................................................

9
 ABSTRACT
This study aims to enhance flood monitoring systems by evaluating the accuracy
and reliability of various sensors, including ultrasonic, turbidity, and atmospheric
pressure sensors, integrated with data storage and logging systems. Conducted
at the Pasonanca Diversion Weir, the research involved a series of tests to
measure sensor performance in real-world conditions. The ultrasonic sensor
demonstrated high precision in measuring distances ranging from 1 to 6 meters,
with most errors falling within ±1%. Conversely, the turbidity sensor exhibited
limited accuracy at low NTU values, though it performed reliably in high-turbidity
environments. The atmospheric pressure sensor provided consistent data,
aligning closely with standardized values in both enclosed and open cases. Data
storage and logging were verified through the RTC DS1302 module and microSD
card, which showed reliable long-term performance without data loss. The
findings suggest the ultrasonic sensor is suitable for applications requiring
precise distance measurements, such as water level monitoring, while the
turbidity sensor should be limited to high-turbidity environments. The atmospheric
pressure sensor's consistent performance supports its use in diverse
environmental conditions. The study underscores the importance of selecting
appropriate sensors based on specific use cases and environmental conditions,
and recommends further calibration and advanced techniques to enhance sensor
accuracy, particularly for turbidity measurements. These insights contribute to
more effective and reliable flood monitoring systems, ensuring timely and
accurate environmental data collection for disaster management and response.

10
 CHAPTER I
INTRODUCTION
Background of the Study
Flooding in coastal communities during significant storm events – which is
made worse with sea-level rise – has prompted coastal communities to
more closely monitor water levels. This includes roadway monitoring
systems that combine water level sensors and telemetry for a real-time
alert system. The alerts can be directed at city managers, news stations,
and mobile apps that drivers can access directly from their phones.
In contemporary industrial settings, information technologies play a pivotal
role, particularly in applications involving both simple and complex
automation [49]. Water reservoirs and tanks are integral components for
storing water resources in agricultural and industrial contexts. To ensure
the prevention of adverse consequences and to gather essential
information regarding water levels, continuous monitoring of these storage
facilities is imperative. Remote monitoring and data collection systems
become essential in such scenarios, facilitating the acquisition of data
based on predetermined parameters and the delivery of processed
information to users when necessary or for decision-making in complex
situations.
In various regions of the Philippines, numerous monitoring sensors, such
as those for water level and weather conditions, have been strategically
placed. The implementation of early warning and flood monitoring systems
is crucial to mitigate the impacts of floods. These sensors serve the
purpose of alerting the population about approaching typhoons or heavy
rainfall, as well as monitoring water levels in dams and along riverbanks.
Traditional flood monitoring systems, however, encounter challenges such
as inaccurate water level estimations arising from erroneous readings
from ultrasonic and water level sensors. Additionally, these conventional
systems often exhibit limited coverage and a slow rate of data
transmission, leading to delayed responses to flood events. Moreover,
their effectiveness is compromised in remote areas lacking cellular
network coverage [52].
Flood monitoring systems do not solely concentrate on water levels, but
also on water quality parameters like turbidity, which is a critical indicator
of water clarity and quality. Turbidity, the measurement of water
cloudiness or haziness, can have a significant impact on both human
health and the environment. Elevated turbidity levels can signal the
presence of harmful microorganisms, sediments, and pollutants, posing a
threat to the safety of drinking water and aquatic ecosystems. In
agricultural and industrial settings, maintaining low turbidity levels is
crucial to ensure the efficiency of irrigation systems and industrial

11
 processes that rely on clean water. In times of low flow (base flow),
numerous rivers exhibit a transparent green hue, with turbidity devices
typically measuring below 10 NTU. However, when a rainstorm occurs,
sediment particles from the adjacent land are carried into the river,
resulting in a murky brown appearance that signifies elevated turbidity
levels. The integration of turbidity sensors into flood monitoring systems
can offer a comprehensive understanding of water quality, enabling timely
interventions to prevent contamination and safeguard public health. This
addition enhances the functionality of flood monitoring systems, ensuring
they not only warn about potential flooding but also about deteriorating
water quality conditions that may necessitate immediate attention.
This scenario poses a significant and serious challenge, as delays in
response could result in substantial harm to both individuals and property.
Therefore, there is a pressing need to develop a more efficient and
reliable flood monitoring and early warning system capable of covering
remote areas and facilitating real-time data transmission.
The trajectory of information technology has delineated research avenues,
encompassing the systematic assessment of water properties through the
utilization of sensors that convert mechanical quantities into electrical
ones. This research also delves into the classification of water quality for
human consumption, contributing to human health and ecosystem
preservation [53]–[57]. Professionals across various industries face time
constraints, limiting their ability to engage in continuous water level
monitoring. Considering water as an indispensable resource in our daily
lives [58], the primary objective of this project is to develop a device
capable of measuring water levels at the diversion weir and notifying users
in the monitoring stations near the diversion weir.
Existing water level measurements exhibit several drawbacks. Traditional
water level monitoring gauges require constant human supervision, exhibit
low accuracy levels, and present challenges in terms of automation. The
article addresses these limitations by employing an ultrasonic sensor for
water level measurement. The ultrasonic sensor emits waves towards the
diversion weir, records their propagation and return times, and after
multiple measurements, provides precise information about the water
level. This pertinent information is then relayed to users' mobile phones
through a GSM module, notifying them via SMS. This streamlined process
not only facilitates efficient water storage but also ensures stability for
daily tasks, prevents water wastage, and mitigates unnecessary costs.

Statement of the Problem

The problem of warning communities of impending disasters quickly
becomes complex due to its multifaceted nature. At the most basic level,
the problem breaks down to identifying the event, communicating that

12
event to the proper authorities, warning the communities affected, and
evacuating those communities.
Public and emergency management officials in Zamboanga City often rely
on an hourly report on the water level reading at the water diversion weir,
which is done manually, and relayed to the nearest station a few
kilometers away. Conducting the study will answer few specific problems:
Sensor Data Reliability
Ultrasonic Sensor
To address the issue of sensor data reliability, particularly
with regards to the water level readings at the diversion weir,
our system will incorporate an ultrasonic sensor. This sensor
will provide accurate and reliable data on the water levels,
ensuring that the information relayed to the authorities and
communities is timely and dependable. Additionally, the low-
cost aspect of the ultrasonic sensor aligns with the need for
cost-effective solutions in implementing disaster warning
systems. By utilizing this sensor, we aim to enhance the
accuracy and efficiency of the monitoring process, ultimately
contributing to timely warnings and effective evacuation
measures in the event of disasters. The current system of
manually measuring and relaying water level readings at the
water diversion weir in Zamboanga City poses several
problems, including the lack of real-time data, potential for
human error, and delays in communication. By integrating an
ultrasonic sensor into the monitoring system, we can ensure
the reliability and timeliness of the sensor data. This will
significantly improve the accuracy of water level readings
and enable rapid decision-making for disaster response and
evacuation measures. The proposed system addresses the
problem of sensor data reliability by incorporating an
ultrasonic sensor.
Turbidity Sensor
To further enhance the reliability of data in our disaster
warning system, it is crucial to incorporate a turbidity sensor
that can provide accurate and real-time measurements of
water clarity and suspended solids in the water. The turbidity
sensor will play a key role in monitoring water quality,
especially during events such as heavy rainfall or flooding,
when sediment and pollutants can affect the water clarity
and quality. By integrating a low-cost yet reliable turbidity
sensor into the system, we can ensure that the data on
water quality and turbidity levels is consistently monitored
and communicated to the relevant authorities and

13
 communities. This will enable timely and informed decision-
making for disaster response and evacuation measures, as
well as facilitate the implementation of necessary measures
to address any water quality issues that may arise during
disaster events. The inclusion of a turbidity sensor in our
system will contribute to the overall reliability and
effectiveness of the disaster warning and management
system in Zamboanga City.
Atmospheric pressure
To improve the reliability of our flood prediction system, it's
essential to incorporate sensors for both air pressure and
temperature to provide accurate, real-time atmospheric data.
These sensors will be crucial for monitoring weather
conditions, especially during heavy rainfall or storms, which
can affect water levels and flood risks. By integrating low-
cost yet reliable air pressure and temperature sensors, we
ensure continuous monitoring and communication of critical
data to authorities and communities. This will support timely,
informed decision-making for disaster response and
evacuation, enhancing the overall effectiveness of flood
management in Zamboanga City.

Objectives of the Study

General Objective:
A. To develop an ultrasonic sensor module to be tailored for flood
monitoring integration.
B. To develop a turbidity sensor module to be tailored for flood
monitoring integration.
C. To develop a pressure sensor module to be tailored for flood
monitoring integration.
D. To install data logging capability to be tailored for the flood
monitoring module.

Specific Objectives:
A. The Implementation of the A01NYUB and JSN SR04T Ultrasonic
Sensor Modules

a) To implement the A01NYUB and JSN SR04T ultrasonic

sensor modules and test their efficiency by measuring:
● The calibration of the sensor modules through linear
measurement devices.

14
 ● The height sensitivity of the A01NYUB and JSN
SR04T ultrasonic sensor modules.
b) To verify the ultrasonic sensor readings by comparing it to
the diversion weir’s gauge readings through percent error.

B. Comparison of the A01NYUB and JSN SR04T Ultrasonic Sensor

Modules

a) To compare the A01NYUB and JSN SR04T ultrasonic

sensors’ efficiency in terms of the height sensitivity of the
A01NYUB and JSN SR04T ultrasonic sensor modules using
the percent errors at different heights.
b) To verify the ultrasonic sensor readings by comparing the
A01NYUB and JSN SR04T to the diversion weir’s gauge
readings through the percent error.

C. To implement the TS-300B Turbidity sensor module and test its

efficiency by measuring in terms of

a) The calibration of the sensor module through standardized

turbidimeter

D. To compare the TS-300B Turbidity sensor module and standardized

turbidimeters and its efficiency using error checking

E. To implement and compare the BMP180 Pressure sensor module

and tests its atmospheric pressure and temperature efficiency by
measuring in terms of:

a) The comparison of the BMP180 pressure sensor module to

a standardized barometer through percent error.

b) The efficiency of the BMP180 pressure sensor module in

exposed versus enclosed setups using paired t-test.

F. To compare the data stored and logged in the Module from the
actual data on continuous monitoring.:***

Significance of the Study

This research study significantly contributes to addressing critical
challenges in water level monitoring and early warning systems. The
study's relevance is underscored by its solutions to coastal flooding,
offering an IoT-based system tailored for real-time alerting to enhance the
resilience of coastal communities. It also advances industrial water
storage monitoring by emphasizing remote monitoring and data collection

15
 systems, providing valuable insights for engineers engaged in industrial
automation. Furthermore, the research aims to enhance flood monitoring
in the Philippines by overcoming the limitations of traditional systems,
offering a more efficient and reliable flood monitoring and early warning
system applicable to remote areas with limited cellular network coverage.
The integration of information technology in water quality assessment,
focusing on parameters like turbidity and pH, aligns with contemporary
trends in the field, contributing to broader goals of human health and
ecosystem preservation.
This study's innovative approach overcomes limitations associated with
traditional water level monitoring by proposing the use of ultrasonic
sensors, along with turbidity sensors and pressure sensors, and LoRa
modules. This not only improves accuracy and automation but also
streamlines the communication of water level information to users,
fostering efficient water storage practices and minimizing wastage. By
focusing on the systematic assessment of water properties through sensor
technology, the research aligns with the trajectory of information
technology. The outcomes are expected to influence the development of
more resilient and efficient monitoring frameworks, impacting both
academic research and practical applications in the field of forestry IoT
systems and beyond.

Scope and Delimitations

This project aims to establish a low-power wireless communication system
for monitoring water level and turbidity at the Zamboanga City Water
District diversion weir in Pasonanca. The study seeks to select and
calibrate suitable sensors, design a low-power wireless monitoring system
for real-time data collection, and implement an alert system. The initial
phase involves choosing and calibrating water level and turbidity sensors
that are appropriate for the diversion weir's conditions, and assessing their
range and accuracy. Subsequently, the focus shifts to creating a low-
power wireless monitoring system that prioritizes real-time data collection
and communication, with an emphasis on water level and turbidity
monitoring. Finally, an alert system is established for real-time water level
monitoring. The study concentrates on evaluating the feasibility of an in-
situ monitoring system using sensors for water level and turbidity,
considering durability against significant natural events such as sudden
rises in river water levels. However, it does not address rare cases of
wireless connectivity interference or analyze empirical data for selecting
the most suitable wireless network. This research will be conducted in
Zamboanga City, Philippines, with system development expected to take a
minimum of three months. Testing will occur in two phases: first in the
Engineering Laboratory (C205) of Ateneo de Zamboanga over two
months, focusing on controlled sensor preparation and testing, and then in

16
 the Tumaga River area for a couple of weeks to assess performance in
the actual environment.

CHAPTER II
REVIEW OF RELATED LITERATURE

Background and context

Status Quo on Flood Prediction System in Zamboanga City
In today's rapidly changing world, the significance of accurate weather
forecasts and flood prediction systems cannot be overstated. Floods are
natural disasters that cause significant damage to both human lives and
infrastructure. To mitigate the impact of floods, early detection and real-
time notification systems are crucial. These systems can provide sufficient
time for inhabitants to evacuate and take necessary precautions.

The construction of the Pasonanca River dam dates back to the American
colonization period. Subsequently, a gauge was installed on the diversion
dam to ascertain the water level in relation to the sea level in Zamboanga
City. As of writing, water level measurements were obtained through
manual means, specifically visual observation by human personnel, and
subsequently communicated to the water district at hourly intervals.
Problems with Manual Data Logging
The monitoring team currently relies on manual data logging, with
observations recorded every hour based on painted sea level markings
near the diversion weir intake pipe. However, during periods of intense
flooding, the monitoring team faces significant risks as they measure
these levels. The difficulties arise from poor visibility caused by harsh
weather conditions, including storms and strong winds.
Existing Solutions for Early Flood Monitoring System
This section examines existing solutions and related studies aimed at
addressing the issues outlined in Section 2.1. Notably, the discussion
includes the non-functional flood monitoring device installed at the
diversion weir. This device was initially implemented by the Project NOAH
of DOST and comprised multiple high-quality systems utilizing SMS
modules for wireless communication. However, the device's current
condition renders it non-operational and has long been disregarded.
Possible reasons for discarding the device include challenges related to

17
 weak service signals in the area, posing difficulties for its primary
communication source to the main base. Additionally, other reasons for
the device's decommissioning are classified.
Zamboanga City Water District ’s Solution
Since then, the division weir team has reverted to manual methods of
flood monitoring, relying solely on the sea level markings near the division
weir. This traditional approach involves periodic visual inspections of water
levels, conducted by team members stationed at the diversion weir site.
Despite its simplicity, this method poses several limitations, particularly
during periods of intense flooding or adverse weather conditions. Visibility
challenges, especially during storms or low-light conditions, can hinder
accurate observations, leading to potential delays or inaccuracies in flood
monitoring efforts. Additionally, manual monitoring requires continuous
human presence at the site, placing personnel at risk during hazardous
weather events. As such, there is a pressing need for more robust and
automated flood monitoring solutions to enhance the safety and efficiency
of monitoring operations at the division weir.

Research Questions
1. What are the limitations and challenges of the traditional
water level monitoring system used by Zamboanga City
Water District (ZCWD)?
Traditional water level monitoring systems, such as those used by
the Zamboanga City Water District (ZCWD), often face numerous
limitations and challenges. These systems typically rely on manual
data collection and rudimentary measuring tools, which can lead to
several inefficiencies and inaccuracies. Manual monitoring is labor-
intensive and time-consuming, requiring personnel to be physically
present at monitoring sites. This method not only increases
operational costs but also poses significant risks during adverse
weather conditions, such as heavy rains or floods, when accessing
monitoring sites becomes dangerous [50], [51].
The lack of sensor data in traditional systems hampers the ability to
make real-time decisions and respond promptly to changing water
levels. Without automated sensors, data collection is sporadic and
often delayed, reducing the ability to predict and mitigate flood risks
effectively. The reliance on manual methods also increases the
likelihood of human error, further compromising the accuracy and
reliability of the data collected [59], [60].

18
2. What are the specific advantages of using A01NYUB
ultrasonic sensor, TS-300B turbidity sensor, and BMP180
barometric pressure sensor for flood monitoring in remote
locations of the diversion weir?
To address these limitations, the integration of advanced sensors
such as the A01NYUB ultrasonic sensor, TS-300B turbidity sensor,
and BMP180 atmospheric pressure sensor presents a promising
solution for flood monitoring in remote locations of the diversion
weir.

A01NYUB Ultrasonic Sensor

The A01NYUB ultrasonic sensor is designed for precise

water level measurement. It emits ultrasonic waves and
measures the time taken for the echo to return, which allows
for accurate distance measurement. This sensor is
particularly advantageous due to its wide range, capable of
measuring distances from a few centimeters to several
meters. Its high-speed measurement capability ensures real-
time data acquisition, and its low power consumption makes
it suitable for remote installations where power supply may
be limited. This sensor's durability and resistance to harsh
environmental conditions make it ideal for continuous
monitoring in varying weather scenarios,Cheng [3].

TS-300B Turbidity Sensor

The TS-300B turbidity sensor measures the clarity of water

by detecting the amount of light scattered by suspended
particles. This sensor is essential for assessing water
quality, as high turbidity levels can indicate the presence of
pollutants or sediment. The TS-300B provides rapid and
accurate readings, enabling timely detection of changes in
water quality. Its compact design and low power
requirements make it an excellent choice for remote
monitoring systems, ensuring reliable operation with minimal
maintenance, Huang [5].

BMP180 Barometric Pressure Sensor

The BMP180 sensor is a versatile MEMS-based device

capable of measuring temperature, pressure, and altitude by
Subair [15]. It has been utilized in various applications,
including weather monitoring stations and high-altitude
balloon experiments in Gaikwad [16] studies . The sensor
communicates using the I2C protocol and can be integrated

19
 with microcontrollers like Arduino or Raspberry Pi for data
collection and processing [16]. In weather monitoring
applications, the BMP180 enables real-time tracking of
environmental parameters, which can be stored in
databases and displayed through graphical user interfaces.
For high-altitude balloon experiments, the sensor can trigger
cut-down mechanisms based on specific altitude,
temperature, or pressure conditions. Additionally, the
BMP180 has been incorporated into healthcare monitoring
systems, working alongside other sensors to measure vital
signs and transmit data to mobile applications , Abburu [18].

The BMP280 and similar sensors (BMP085, BMP180) are

widely used for measuring air pressure and altitude. These
sensors offer high accuracy, precision, and low power
consumption ,Wang Cheng [20], M. Vasylenko & V. Dzhus
[17]. Studies have shown that the BMP280 sensor's
performance can be improved through statistical analysis
and calibration techniques. One-way ANOVA and Tukey
tests can identify differences between sensors, while linear
regression can enhance accuracy, H. Kusuma [18]. When
compared to standard meteorological instruments, BMP280-
based devices demonstrate high accuracy and precision,
with a calibration equation yielding a coefficient of
determination (R) of 0.99998, Khaery [19]. These sensors
can be integrated with microcontrollers like Arduino Uno or
C8051F310 to create portable altitude measurement
systems with features such as temperature compensation,
noise processing, and OLED display, Wang Cheng [20] , M.
Vasylenko & V. Dzhus [17].

3. How does the proposed addition of data logging ensure

reliable data transmission in the location of the diversion
weir with limited connectivity?
Incorporating remote sensing and data logging technologies
enhances the reliability of data transmission in locations with limited
connectivity, such as the diversion weir. The proposed system
includes wireless communication modules and data logging devices
that work together to ensure continuous monitoring and data
collection.

Data Logging Devices

20
 Data logging devices store sensor data locally, providing a backup
in case of communication failures. These devices can store large
amounts of data and are equipped with time-stamping features to
ensure data integrity. By using data loggers, the monitoring system
can continue to collect and save data during periods of connectivity
loss, which can be transmitted once the connection is re-
established. This redundancy ensures no data is lost, maintaining
the reliability of the monitoring system, Patel [8].

By integrating these advanced sensors and communication

technologies, the proposed monitoring system offers a significant
improvement over traditional methods. It provides accurate, real-
time data essential for effective flood management and risk
mitigation, ensuring the safety and resilience of the Zamboanga
City Water District's water resources.

Synthesis of Literature
Water Level Monitoring
The implementation of flood monitoring and warning services necessitates
a scientifically rigorous approach . The fundamental components of a flood
forecasting system involve the application of hydrological models using
observed hydrometeorological data to derive river discharges and the
utilization of hydraulic/hydrodynamic models to simulate river stages
based on predicted streamflow. In certain scenarios, georeferenced water
surface elevations are employed to generate flood inundation maps, which
prove invaluable in disaster relief operations and government-led public
awareness initiatives. To ensure the provision of accurate and timely early
warning information, the operation of a Flood Early Warning System
(FEWS) demands robust technical capabilities. This discussion of survey
responses encompasses key aspects such as the levels of hydro-
meteorological observation systems in operation, prevalent types of
floods, the usage of hydrological/hydrodynamic models, and the nature of
forecasts generated.
According to A. Smith [10], Survey findings reveal that most river basins
have both rainfall and water level gauging stations, with approximately half
equipped with telemetric systems; the remainder either rely on manual
data recording, collection, and transfer, or employ semi-automated
methods where data recording is automatic while collection and
transmission are manual. Streamflow gauges are present in only half of all
basins, with 52% operated manually, and the remaining 48% divided into
38% telemetric and 10% semi-automated systems. The survey responses
indicate that the basins commonly experience fluvial and flash floods,
followed by pluvial floods and human-induced floods resulting from

21
 upstream reservoir operations (Figure 2.1). Flash floods, posing a
significant threat to human lives globally, represent a major concern due to
their ability to be forecasted but with lead times too short for appropriate
preventive action .

Figure 2.0. Types of floods experienced by surveyed Flood Early Warning Systems

Numerous researchers have developed a flood alert system for real-time

monitoring and alerting of river surges in flood-prone areas. This review
specifically focuses on the identification and early warning signals that
notify local residents of potential flood events.
Despite significant advancements in aid from satellite technology, ground
observations continue to play a predominant role in operational flood
monitoring and warning systems worldwide (Figure 2.2). Survey
responses indicate that some warning systems in Asia and North America
leverage drones and Closed-Circuit Television (CCTV) cameras for flood
monitoring, while remote sensing data is the primary source in other
regions. However, the situation in South America remains uncertain due to
a limited number of responses.

22
 Figure 2.1. Flood monitoring methods

Approximately 49% of Flood Forecasting Centers (FFCs) utilize models

specifically tailored for their target river basins, reflecting the considerable
variability in hydrological and climatological characteristics across these
basins. Among the remaining FFCs, 31% employ open source models,
and 19% rely on commercial models. The primary output from nearly all
operational Flood Early Warning Systems (FEWS) is water level, followed
by streamflow and inundation information, which are available for almost
half of the systems.
Event-based flood forecasts are generated for about 40% of the surveyed
basins, with seasonal and continuous updates provided for approximately
30% of them. Daily temporal resolution emerges as the most commonly
employed frequency by FFCs for continuous forecasts.
With the existing gauge installed at the diversion channel in Tumaga
River, an established 74 m above sea level tends to be the baseline of the
diversion weir in which sufficient water can be provided to the community.
Real-time gauge height data can indicate when river levels are beginning
to exceed the baseline, providing the local community an early warning of
dangerous flood conditions.
The utilization of water-level sensor technology extends beyond its
common use in elevated water tanks of high-rise buildings, urban sewage

23
pipe networks, underground cisterns, and groundwater wells. It finds
extensive application in large-scale water conservancy projects, ocean
wave and tide observations, water quality monitoring, as well as water
level measurements in industrial and agricultural settings, G. J. Pottie and
W. J. Kaiser [23].

Various water-level sensing techniques based on mechanical, electrical,

magnetic, microwave, and optical methods have been documented.
Electrical water-level sensors, while widely adopted, face limitations in
monitoring erosive or impure water and are not well-suited for remote
control and monitoring due to high electrical signal loss.
Microwave water-level sensors, offering advantages such as high
reflectivity from water, low sensitivity to ambient conditions, and non-
contact operation [61], also struggle with remote control implementation,
similar to electrical sensors, M. Hall [24], [25]T. De Groeve[25]. Some
microwave techniques for flood monitoring using satellites have been
developed, but their temporal and spatial resolutions are not precise
enough, leading to delays in flood observation.
Optical sensors present significant advantages in challenging conditions
due to their high sensitivity, dielectric properties, resistance to erosion,
substantial distance between signal generation and detection, and
immunity to electromagnetic interference , J. E. Antonio-Lopez [26].
However, their sensitivity to water impurities and limitations during heavy
fog and rainy conditions must be considered. The poor temporal and
spatial resolutions of previous sensors for remote water-level monitoring
have hindered accurate flood location in the past.
In the study conducted by J. Band and F. I. Anyasi [27], the paper outlines
the design of an automatic water level controller system that incorporates
a mercury flow switch as a key component. Two contactors within the
system are activated to facilitate a direct online start of the motor.
Additionally, an overload relay is integrated to detect excess current and
disconnect the supply. The mercury flow switch utilizes the Archimedes
principle of flotation, providing electrical contact to switch the supply ON
and OFF when the tank reaches empty or full states, respectively. While
this system is characterized by its cost-effectiveness and durability, the
primary drawback lies in the use of the mercury switch. Mercury switches
exhibit a relatively slow operating rate attributed to the inertia of mercury;
however, they pose significant environmental concerns due to their high
toxicity and potential accumulation in the food chain.
Sabre and Abdullah [28] designed a system intended for determining flood
levels by gathering data on water level, water pressure, and rainfall. The
gathered information is transmitted to a microcontroller for subsequent
processing, and the processed data is then linked to a server via Wi-Fi,

24
facilitating the sharing of information on the internet. The system
demonstrated efficiency through its collaborative role in overseeing,
processing, and communicating various operational aspects related to
background issues.
Getu et.al [29] devised a water level monitoring system with an integrated
automatic water pump controller, utilizing the electrical conductivity
property of water for determining tank water levels. The tank's total height
was divided into nine segments, each equipped with a metallic electrode
probe. These electrodes were linked to a voltage divider arrangement
utilizing series resistors (R1 and R2) and a supply voltage (Vcc). A certain
arrangement involving the floating terminals of the electrodes and the
grounded bottom electroderesulted to the output voltage becoming zero
when an electrode was immersed in water due to the water's conductivity.
A digital logic controller regulated the water pump based on the highest
and lowest voltage sensed by the electrodes.
Susheel [30] designed an automatic water pump controller based on a
float switch sensor, featuring a floating bob that moved in response to
changes in water level. When the water level was low, the bob rested at
the bottom of the tank, floating at the top when the tank was full. The bob's
movement triggered a reed switch, responsible for opening or closing to
activate the water pump through an electrical current. The limitation of this
system was that it only turned on the pump when the tank was completely
empty.

Water Level Monitoring via Ultrasonic Sensor

In this chapter, we explore the selection of ultrasonic sensors for
integration into a flood prediction system designed for river
monitoring. Ultrasonic sensors play a crucial role in accurately
measuring water levels, providing essential data for flood prediction
and mitigation strategies. This chapter evaluates several ultrasonic
sensor options based on their speed of transmissions, distance
reading accuracy, data accuracy, and effectiveness in water level
monitoring for flood prediction systems.
Sensor Options and Evaluation Criteria
These ultrasonic sensors offer various features and capabilities
suited for flood prediction systems, providing essential data for
monitoring water levels in rivers and other water bodies.
JSN SR04T
According to R. Yelekar et al [31] the JSN SR04T is an
ultrasonic sensor commonly used for distance measurement
applications. It operates on the principle of emitting

25
 ultrasonic waves and measuring the time taken for the
waves to bounce back after hitting an object. The sensor's
waterproof design (indicated by "T" in the model name)
makes it suitable for outdoor use, including river monitoring
for flood prediction.

A01NYUB Waterproof Ultrasonic Sensor

The A01NYUB Waterproof Ultrasonic Sensor is an ultrasonic
sensor designed for outdoor applications where water
resistance is essential. It features a waterproof design to
withstand exposure to moisture and is suitable for use in wet
environments such as river monitoring for flood prediction
[32].
Sensor Options and Evaluation Criteria
In evaluating the suitability of each sensor for integration into the
flood prediction system, several critical criteria are considered.
These include the speed of transmissions, ensuring timely updates
on water level changes; distance reading accuracy, crucial for
precise measurement of water levels; data accuracy, ensuring
reliability in flood prediction; effectiveness in water level monitoring,
evaluating overall performance in real-world conditions; and coding,
data gathering, and data communication capabilities, essential for
seamless integration into the system architecture. By assessing
these criteria comprehensively, the most suitable sensor can be
identified for accurate and reliable flood prediction.

JSN SR04T
The JSN SR04T exhibits comparable data transmission
speed to the MB7389 HRXL-MaxSonar-WR, ensuring timely
updates on water level changes essential for real-time
monitoring applications. While generally accurate, the JSN
SR04T may display slightly lower reliability under certain
conditions compared to the MB7389 HRXL-MaxSonar-WR.
Nonetheless, it maintains good accuracy within a few
millimeters to centimeters, sufficient for precise distance
readings in flood prediction systems. Additionally, the JSN
SR04T provides reliable data accuracy, although it may be
more susceptible to environmental factors such as
temperature fluctuations and interference. Despite these
considerations, its accuracy remains effective in contributing
to the overall efficiency of flood prediction models and early
warning systems.ctive flood prediction, but careful calibration

26
 may be necessary for optimal performance. Please note that
the device is waterproof, but it couldn't work underwater [31].

A01NYUB Waterproof Ultrasonic Sensor

The A01NYUB Waterproof Ultrasonic Sensor boasts fast
data transmission speeds, making it well-suited for real-time
monitoring of water levels in flood prediction systems,
ensuring timely updates on changes in river levels. While
capable of providing distance measurements, the accuracy
of the A01NYUB sensor may vary depending on
environmental conditions and calibration efforts.
Nonetheless, it typically maintains accuracy within a few
millimeters to centimeters, facilitating precise distance
readings crucial for flood prediction applications.
Furthermore, the A01NYUB sensor offers reliable data
accuracy, though calibration may be necessary to optimize
its performance. Its waterproof design enhances its
suitability for outdoor use in flood prediction systems,
ensuring robust and reliable operation even in challenging
environmental conditions [32].

Table 2.0: Water level Ultrasonic Sensors used in Related Studies

Sensor Low Commercially Response Range Accuracy Waterpro

Cost use Time of

JSN SR04T Yes Yes <10 ms 4-5 ±1 cm Yes

meters

A01NYUB Yes Yes <20 ms 6-7 ±2 mm to Yes

meters ±5 mm

Turbidity monitoring
It is noteworthy that despite the longstanding focus on turbidity
measurement and monitoring in resarch, there have been limited
endeavors to create and implement low-cost sensors specifically designed
for managing fresh-water quality. The majority of water quality monitoring
products, as discussed in the preceding section, and turbidity sensor
designs [6],[7], tend to be sophisticated and often come with high costs,
thereby deviating from the overarching goal of this study.

27
Numerous online water quality monitoring systems, facilitated by WSN,
have been proposed, utilizing either Commercial Off-The-Shelf (COTS)
sensors [33], [34], [35], or custom-designed sensors [33] to track changes
in water quality in domestic running water, lakes, and rivers. These WSN
systems typically exhibit low power consumption and heavily rely on the
sensors and instruments integrated into their platforms. However, none of
the previously mentioned systems, incorporating low-cost COTS sensors,
have demonstrated the ability to provide turbidity values within the range
of 0 to 1000 NTU with sufficient precision and accuracy.
In Azman et al. [33], the authors proposed a sensor design based on the
nephelometric method. In this system, a photodetector positioned 90
degrees from the light source detects the scattered light intensity. A
microcontroller regulates the LED light source and the orthogonal light
detection receiver. While a calibration experiment validated the reliability
of the sensor for water quality monitoring, the results did not explicitly
confirm the sensor's capacity to measure samples exceeding 100 NTU.
Lambrou [36] developed a low-cost sensor design employing the ratio
method for online monitoring. This design closely resembles existing low-
cost COTS products, initially utilizing transmitted light detection. Notably,
an enhancement involving the addition of an orthogonal detection unit was
introduced to improve accuracy. Through calibration, the designed sensor
demonstrated a precision level of 0.1 NTU with ideal accuracy. However,
the authors did not provide detailed information on how the ±0.5 NTU
accuracy of the showcased sensor was achieved, and the detection range
was confined to 0–100 NTU.
To address the issues on the lack of range, sensitivity, accuracy,
repeatability, consistency, and calibration for fluctuating ambient
temperatures, Wang et al. developed a sensor that utilizes both
transmitted light and orthogonal (90 degrees) scattered light detection
principles. This sensor is 2-3 orders of magnitude lower in cost compared
to existing solutions and provides improved accuracy and robustness
compared to existing low-cost turbidity sensors [62].

28
Figure 2.2. Wang et al. [62] turbidity sensor design. (a) Dimensional layout of hardware.
The 20mm diameter hole is used for acrylic tube adhesion, and water flows through the
tube during measurements. (b) Circuit schematic design. Four pin connections are
required: 5V, 0V ground, and dual sensor output pins (VoutN and VoutD). (c) Overall
structural view of the sensor [69].
Turbidity Monitoring via Turbidity Sensor
Turbidity monitoring plays a critical role in flood prediction systems,
providing insights into changes in water quality and sediment
transport dynamics within river ecosystems. This chapter examines
the application of turbidity sensors in flood prediction systems on
rivers, focusing on three specific models: the SKU:SEN0189, TS
300B, and Atlas Scientific EZO-Turbidity sensors. These sensors
are designed to measure the turbidity of water, typically operating
on the principle of light scattering to quantify the presence of
suspended particles. Turbidity, expressed in Nephelometric
Turbidity Units (NTU), serves as an indicator of water clarity, with

29
 higher turbidity values suggesting increased levels of suspended
solids and potential sediment transport during flood events [36],
[37],[38],[29].

SKU:SEN0189 Turbidity Sensor

The SKU:SEN0189 turbidity sensor is a compact device designed
for measuring water turbidity, operating on the principle of light
scattering, where the amount of light scattered by suspended
particles correlates with turbidity levels. Its durability and resistance
to water ingress make it suitable for long-term deployment in flood
prediction systems, while integration with microcontrollers for data
monitoring and gathering is facilitated through coding libraries or
custom programming. Data on turbidity levels can be gathered and
stored for analysis using appropriate logging techniques, and
communication with external devices or systems is possible via
analog or digital interfaces [40].
The SKU:SEN0189 turbidity sensor provides accurate
measurements within a defined tolerance. While the exact accuracy
specification may vary depending on calibration and environmental
conditions, it typically offers an accuracy of around ±0.1 NTU within
the specified NTU range of 0 to 30 NTU [40].

TS 300B Turbidity Sensor

According to T. P. Lambrou et al [36] the TS 300B turbidity sensor
is a compact and robust device tailored specifically for measuring
water turbidity, functioning on the principle of light scattering, where
the intensity of scattered light correlates directly with the turbidity of
the water sample. Typically, it measures turbidity within a specific
range, often customizable to suit diverse applications, with common
measurement ranges spanning from 0 to 1000 Nephelometric
Turbidity Units (NTU), facilitating a comprehensive assessment of
turbidity levels in various water bodies.

Accuracy in turbidity measurement is paramount for reliable water

quality assessment, and the TS 300B sensor is renowned for its
low precision, typically providing an accuracy of around ±30 NTU
within its specified range. Although calibration and environmental
factors may influence accuracy, the sensor consistently delivers
precise turbidity measurements crucial for flood prediction systems,
as corroborated by studies [63]. Moreover, engineered to withstand
demanding water environments, the TS 300B turbidity sensor

30
 boasts excellent durability and resilience to water ingress. Its
rugged construction enables sustained deployment in flood
prediction systems, ensuring dependable performance even under
adverse conditions.
Atlas Scientific EZO-Turbidity Sensor
The Atlas Scientific EZO-Turbidity Sensor is a compact and robust
device engineered specifically for measuring water turbidity,
functioning on the principle of light scattering. It offers a
customizable measurement range typically spanning from 0 to 1000
Nephelometric Turbidity Units (NTU), allowing for versatile
applications across various water bodies. Renowned for its high
precision, the EZO-Turbidity Sensor provides an accuracy of
approximately ±0.1 NTU within its specified range, ensuring reliable
water quality assessment even in challenging environmental
conditions. Its rugged construction enables sustained deployment
in flood prediction systems and other demanding water
environments, making it a dependable tool for continuous
monitoring applications [39].
Studies have validated the effectiveness of the Atlas Scientific
EZO-Turbidity Sensor in providing dependable turbidity
measurements crucial for environmental monitoring and research
applications [37],[38]. With its compact design, high accuracy, and
durability, this sensor serves as an indispensable component in
water quality assessment systems, facilitating comprehensive
assessments of turbidity levels in diverse water bodies. Its robust
performance underscores its importance in ensuring reliable data
collection for flood prediction systems and other critical
applications, contributing to enhanced understanding and
management of water resources.
Table 2.1. Turbidity Sensors used in Related Studies

Sensor Low Commerciall Response Accuracy Waterpro

Cost y use Time of

SKU: SEN0189 Yes Yes <100 ms ±0.1 NTU Yes

typical

TS 300B No Yes <100 ms ±30 NTU Yes

typical

31
Atlas Scientific EZO- No Yes <100 ms ±0.1 NTU Yes
Turbidity Sensor typical

Atmospheric Pressure Sensor

Piezoresistive pressure transducers use a combination of silicon, as the
construction material for the mechanical stress amplifiers (membranes,
cantilever beams and bridges), and piezoresistive strain gauges. The
mechanical stress amplifiers are used to transform the pressure into
stress and the particular mechanical design is chosen to provide both the
required sensitivity and resolution for a given application. Pressure
transducers are usually formed from a membrane that is hermetically
sealed to a support and also separates the reference pressure from the
pressure to be measured. Silicon-to-Pyrex anodic bonding techniques
used by Wallis & Pomerantz [41] and Cozma and Puersare [42] often
used for absolute pressure sensors. Pressure transducer drift may arise
when either the active
elements or electrical connections are constructed from materials with
different thermal expansion coefficients. This is the case with silicon and
Pyrex and designs, shown in Figure XX, have been used to minimize
these effects. Silicon piezoresistive strain gauges are machined into
supporting membranes with a standard integrated circuit (IC) process.
Silicon-based MEMS can provide pressure transducers with μm
dimensions, which can withstand pressures up to 100 MPa. Mono-
crystalline silicon, owing to the stability of the crystal, reduces the
observed hysteresis. In Figure 2.4 a schematic of a typical pressure
sensor is shown.

32
Figure 2.3: Schematic cross-section through a pressure gauge formed from an Si
membrane and strain gauge doped into specific locations to detect the
mechanical deformation arising from the application of a pressure difference.
This Si element is sealed to the support with Pyrex, which is in turn mounted on
a base to provide support for the electrical connections. Extracted from [70].

Chau and Wise [43] and Spencer [45] have evaluated the performance of
both piezoresistive and capacitive pressure transducers. The theoretical
performances of miniature capacitive and piezoresistive pressure
transducers have been described in a review article on micromachined
pressure sensors presented by Eaton and Smith [44].

Spencer [45]) introduced the concept of a minimum detectable signal, β,

for evaluating pressure transducers. The β value represents the
theoretical detection limit, defined as the noise expressed as an equivalent
pressure fluctuation, indicating the uncertainty from the transduction
process and the transducer's resolution. This definition assumes the
cancellation of all systematic errors and does not consider long-term drift.

There are three types of noise in electrical circuits: Schottky effect (shot
noise), Johnson (thermal) noise, and 1/f noise. Schottky noise arises from
electric potential barriers at p-n junctions, while Johnson noise is due to
energy dissipation processes, and both have flat spectral densities up to
GHz frequencies. 1/f noise results from trapping centers near the device
surface. The change in the resistor bridge output voltage arising from a
pressure change p is given by:

33
where is the pressure sensitivity of the transducer and the applied
voltage. The is obtained by equating with the r.m.s. Johnson noise to
give

where is the frequency bandwidth, R the resistance, k Boltzmanns'

constant, and T temperature. For capacitive pressure sensors the noise,
therefore cannot be generalized and must be defined for that particular
detection circuit.

According to Suski et al. [64], five primary types of pressure transducers

are recommended for high-accuracy gas pressure measurements ranging
from 0.1 to 1000 Pa. These transducers respond to the force exerted by
an applied pressure difference, and as long as there are no chemical
reactions between the gas and the materials used in the construction of
the transducer, the readings remain consistent regardless of the gas used.

Temperature variations can influence the performance of all these

transducers. Consequently, capacitive diaphragm gauges and quartz
Bourdon gauges are typically operated with their temperature regulated to
around 318 K (45 °C). Quartz resonance gauges feature an internal
quartz-crystal temperature sensor for thermal compensation. Both MEMS
resonant Si gauges and MEMS piezoresistive Si gauges are temperature
compensated. Capacitive diaphragm gauges, quartz Bourdon gauges, and
quartz resonance gauges are precision instruments that perform optimally
when protected from mechanical or thermal shock and over-pressure.

Bande & Shete [46] developed an IoT-based flood monitoring system

employing an artificial neural network where temperature, pressure,
humidity, rainfall and water level data were collected for temporal analysis
for flood prediction. The IoT approach is used for data collection from
sensors and ANN is used for data prediction.

BMP180

The BMP180 sensor is a versatile MEMS-based device capable of

measuring temperature, pressure, and altitude [15], [16]. It has
been utilized in various applications, including weather monitoring
stations and high-altitude balloon experiments Gaikwad [25]. The
sensor communicates using the I2C protocol and can be integrated
with microcontrollers like Arduino or Raspberry Pi for data collection
and processing [15]. In weather monitoring applications, the
BMP180 enables real-time tracking of environmental parameters,
which can be stored in databases and displayed through graphical

34
 user interfaces [15]. For high-altitude balloon experiments, the
sensor can trigger cut-down mechanisms based on specific
altitude, temperature, or pressure conditions (Youn, 2020).
Additionally, the BMP180 has been incorporated into healthcare
monitoring systems, working alongside other sensors to measure
vital signs and transmit data to mobile applications[16].

The BMP280 and similar sensors (BMP085, BMP180) are widely

used for measuring air pressure and altitude. These sensors offer
high accuracy, precision, and low power consumption [20], [17].
Studies have shown that the BMP280 sensor's performance can be
improved through statistical analysis and calibration techniques.
One-way ANOVA and Tukey tests can identify differences between
sensors, while linear regression can enhance accuracy reported by
H. Kusuma et al. [18]. When compared to standard meteorological
instruments, BMP280-based devices demonstrate high accuracy
and precision, with a calibration equation yielding a coefficient of
determination (R) of 0.99998. These sensors can be integrated with
microcontrollers like Arduino Uno or C8051F310 to create portable
altitude measurement systems with features such as temperature
compensation, noise processing, and OLED display [15],[17],[19].

MPL3115A2

The MPL3115A2 is a commercially available MEMS pressure

sensor that has been studied for its performance under extreme
conditions. Lall [21] investigated the effects of high-temperature
operating life (HTOL) at 125°C and low-temperature storage (LTS)
at -35°C on this sensor. Their research aimed to quantify damage
progression and identify potential failure sites in harsh
environments. The study focused on measuring incremental shifts
in parameters such as absolute pressure and offset. While not
specifically mentioning the MPL3115A2, Linke [22] provides an
overview of 1-Wire technology and its applications in environmental
sensing, including barometric pressure measurement. This
technology enables the construction of sensors that can measure
various parameters on a single twisted-pair cable. Although the
MPL3115A2 is not explicitly mentioned in Linke's paper, it
demonstrates the broader context of pressure sensing technologies
and their applications in environmental monitoring.

35
Table 2.2. Pressure Sensors used in Related Studies

Sensor Low Commercially Resolution Accuracy (Pressure)

Cost Used (Pressure)

MPL3115A2 No Yes 1.5 Pa typical ± 0.4 kPa typical

BMP180 Yes Yes 0.01 hPa ±0.12 hPa typical

typical

Methodological Issues
The methodological procedures detailed in the reviewed literature provide
a comprehensive foundation for this study. For instance, Wang Cheng [15]
and M. Vasylenko & V. Dzhus [17] highlighted the integration of BMP280
sensors with microcontrollers like Arduino Uno for portable altitude
measurement systems, emphasizing features such as temperature
compensation, noise processing, and OLED display. H. Kusuma et al. [18]
further refined this approach by using statistical analysis and calibration
techniques, including one-way ANOVA and Tukey tests to enhance
sensor accuracy. Similarly, Lall et al. [21] explored the performance of
MPL3115A2 sensors under extreme conditions, providing insights into
their durability and accuracy in harsh environments.

Critical Evaluation
When evaluating the usefulness and reliability of the studies, certain
methodologies stand out. Wang Cheng [15] and M. Vasylenko & V. Dzhus
[17] offer highly reliable methods for integrating BMP280 sensors due to
their detailed focus on calibration and precision. H. Kusuma et al. [18] also
ranks high in reliability due to their comprehensive statistical validation.
Lall et al. [21], while valuable for understanding sensor performance in
extreme conditions, may be less directly applicable to the typical
operational environment of the ZCWD but provides critical insights into
sensor durability. Studies focusing on specific sensors like the A01NYUB
ultrasonic sensor [32] and the TS-300B turbidity sensor [5], [7] are
particularly useful for their practical insights into sensor performance and
integration.

Integration with Research Questions/ Objectives

1. What are the limitations and challenges of the traditional water level
monitoring system used by Zamboanga City Water District (ZCWD)?

36
 Traditional water level monitoring systems at ZCWD face significant
limitations, including reliance on manual data collection and
rudimentary measuring tools. This method is labor-intensive, time-
consuming, and prone to human error, leading to inefficiencies and
inaccuracies[6],[12]. Moreover, manual monitoring requires
personnel to be present at the site, posing risks during adverse
weather conditions. The lack of automated sensors hampers real-
time decision-making, reducing the effectiveness of flood risk
mitigation[6],[12]. This highlights similar challenges in other
contexts, reinforcing the need for automated solutions [47],[48].

2. What are the specific advantages of using the ultrasonic , turbidity and
atmospheric pressure data modules for flood monitoring in remote
locations of the diversion weir?
The A01NYUB ultrasonic sensor offers precise water level
measurements with high-speed data acquisition, making it suitable
for real-time monitoring in remote locations. Its waterproof design
and resistance to harsh environmental conditions ensure reliability
[31]). The TS-300B turbidity sensor provides accurate readings of
water clarity, essential for assessing water quality and detecting
pollutants [6], [12]. The BMP180 atmospheric pressure sensor is
versatile and can measure temperature, pressure, and altitude,
contributing to comprehensive environmental monitoring [15], [16].
These sensors' integration addresses the limitations of traditional
systems by providing real-time, accurate data, enhancing flood
prediction and response capabilities.
3. How does the proposed addition of remote sensing and data logging
ensure reliable data transmission in the location of the diversion weir with
limited connectivity?
The proposed system leverages LoRa technology for reliable data
transmission over long distances, even in areas with limited
connectivity. LoRa modules, like the SX1278, offer low power
consumption and long-range communication, ensuring data can be
transmitted from remote monitoring sites to a central server without
significant loss or delay. Data logging using microcontrollers like
Arduino or Raspberry Pi ensures continuous data collection,
storage, and processing, even when connectivity is intermittent.
This setup enhances the system's robustness, allowing for
continuous monitoring and timely data transmission, which is
crucial for effective flood management in remote locations.

37
Summary and Conclusion
The solution devised to address challenges at the diversion weir of the
Pasonanca River in Zamboanga City focused on developing an efficient system
utilizing remote sensing technology to reduce labor requirements. The sensor
nodes were designed to operate independently and sustainably, optimizing
energy usage, wireless connectivity, and cost-effectiveness. Building on
methodologies outlined and summarized in prior studies of the ways of gathering
data these sensors, data collection relied on specialized sensor technologies
tailored for various parameters. Among all of the models, A01NYUB ultrasonic
sensor and SKU:SEN family turbidity sensor were identified as suitable for
outdoor surveillance applications and integrated into the device systems.

Implications for Future Research

The reviewed studies underscore the significant potential for advanced sensor
technologies and remote communication systems in enhancing environmental
monitoring and flood prediction. However, several areas warrant further
investigation to optimize these systems' efficacy and broaden their application
scope. Future research should explore the integration of multi-sensor networks
with advanced data analytics and machine learning algorithms to improve
predictive accuracy and response times. Additionally, addressing the challenges
of sensor calibration and environmental factors that affect measurement
accuracy remains crucial. Studies should also examine the scalability and cost-
effectiveness of deploying these systems in diverse geographic and climatic
conditions, ensuring their viability for widespread adoption. Finally, investigating
the social and economic impacts of these technologies on local communities can
provide insights into their practical benefits and potential areas for improvement.
Future research should focus on enhancing the integration of multi-sensor
networks with advanced data analytics and machine learning, addressing
calibration challenges, examining scalability and cost-effectiveness, and
evaluating the social and economic impacts of advanced environmental
monitoring technologies.

38
 CHAPTER III
METHODOLOGY
Conceptual Framework
The system integrates LoRa technology for communication, ensuring low power
consumption. It focuses on the transmission aspect, utilizing separate nodes for
data transmission: one for the Bangwang Diversion weir house and one for the
main base. The system relies on two separate power supply sources: for the
Bangwang Diversion weir, the existing power supply available at the site,
leveraging the power line infrastructure for operation, is utilized by the main
transmitter system and the receiver inside the guard house. The remaining nodes
installed in the main base will be utilizing batteries as their power supplies, along
with adequate infrastructure to house the devices.

Figure 3.0: Conceptual Design

The design will consist of modules built for specific functions. The study will
develop microcontroller-based nodes positioned near the diversion weir and
outskirts of the river site. These nodes will be responsible for recording
parameters such as water level, pressure/temperature, and turbidity. While the
sensor node measures these parameters, the microcontroller will record the data,
facilitating wireless communication between the river site transmitter and the
monitoring station receiver. Two receiving stations will be installed on-site: one
inside the monitoring station near the diversion dam and the other at the base.
Although the long-term plan is for the ZCWD old reservoir to receive the data, the

39
system will initially be adjusted so that succeeding nodes can be placed, creating
additional receiving stations apart from those at the diversion weir. Each
receiving station will feature a wireless communication device and basic user
interface for monitoring purposes, alongside a specific module for data logging.
The system will be powered in two ways: the transmitter and receiver near the
dam’s location will be connected to an AC/DC adapter linked to the main line of
the diversion dam’s location, while other nodes will utilize a battery or power
bank module as the power supply. To conserve power, the node's sensing and
transmission activity, considered the active mode in this study, will follow a
predetermined schedule. Additionally, a time control feature will be integrated
into the system, ensuring that it only sends data for a specific amount of time to
minimize power consumption. Figure 3.0 depicts a diagram representing the
conceptual design of the system.

Theoretical Framework
Ultrasonic sensor
The ultrasonic system utilized in early flood monitoring relies on the fundamental
principle of measuring distance using ultrasonic pulses. The standard formula for
calculating distance based on time-of-flight principle is commonly employed:

This formula assumes a constant speed of sound (c) in air, which traditionally
depends on environmental factors such as temperature (T), humidity, and
pressure (P). However, in the context of the early flood monitoring system,
precise environmental corrections are essential to ensure accurate distance
measurements.

To achieve this, it is crucial to understand how temperature, humidity, and

pressure affect the speed of sound. These factors influence the density and
compressibility of the air, thereby impacting the propagation of ultrasonic waves.
For instance, as temperature increases, the speed of sound generally increases,
while humidity and pressure variations can also affect its value.

Rutsch et al. [65], Hassen [66], and A. M. Sabatini [67] discussed environmental
measurements considered to validate simulations based on the speed of sound
and atmospheric attenuation. The speed of sound was determined using a
combination of temperature, pressure, and humidity values. However, the
specific formula used for these calculations was not provided in the excerpts.

40
Despite the potential influence of environmental factors on the speed of sound,
the ultrasonic sensor system employed in the early flood monitoring project is
designed to compensate for these variations. By utilizing calibration techniques
and algorithms, the system effectively accounts for changes in temperature,
humidity, and pressure, ensuring consistent and accurate distance
measurements.

Therefore, in practical implementation, the theoretical considerations of

environmental factors affecting the speed of sound become negligible. Instead,
the standard ultrasonic formula suffices for calculating theoretical distances in the
project. This approach simplifies the system design and ensures reliable
performance under varying environmental conditions, ultimately enhancing the
effectiveness of early flood monitoring efforts.

Methodology
This section summarizes the methodologies used to achieve each specific
objective of the project.
Specific Objectives Methodology
A. Sensor height sensitivity
A. The Implementation of the A01NYUB and test
JSN SR04T Ultrasonic Sensor Modules
B. Field testing accuracy
a) To implement the A01NYUB and reading
JSN SR04T ultrasonic sensor
modules and test their efficiency by
measuring:
● The calibration of the
sensor modules
through linear
measurement devices.
● The height sensitivity of
the A01NYUB and JSN
SR04T ultrasonic
sensor modules.
b) To verify the ultrasonic sensor
readings by comparing them to the
diversion weir’s gauge readings

41
 using percent error checking

C. Comparison of Sensor
B. Comparison of the A01NYUB and JSN height sensitivity test
SR04T Ultrasonic Sensor Modules through error checking

a) To compare the A01NYUB and JSN D. Comparison of Field

SR04T ultrasonic sensors’ efficiency testing accuracy reading
in terms of the height sensitivity of through error checking
the A01NYUB and JSN SR04T
ultrasonic sensor modules using
percent error checking

b) To verify the ultrasonic sensor

readings by comparing the
A01NYUB and JSN SR04T to the
diversion weir’s gauge readings
using percent error checking

E. Turbidity Sensor
C. To implement the TS-300B Turbidity sensor calibration using a serial
module and test its efficiency by the calibration of dilution of NTU
the sensor module through standardized (formazine)
turbidimeter

F. Voltage Reading
D. To compare the TS-300B Turbidity sensor comparison (TS-300B vs
module and standardized turbidimeters and its Eutech Instrument
efficiency using error checking Turbidimeter)

E. Standardized Atmospheric
E. To compare the BMP180 Pressure sensor pressure and temperature
module and tests its atmospheric pressure and versus Atmospheric pressure
temperature efficiency by measuring in terms of: and temperature sensor data

a) Using standardized barometer and F. Atmospheric pressure and

thermometer temperature difference of
enclosed case versus open case

42
 b) THe efficiency of the BMP180 pressure
sensor module in exposed versus enclosed
setups using paired t-test.

G. Data Storage and

To compare the data stored and logged in the Logging Test
Module from the actual data on continuous
monitoring.

Integration of Sensor Data Modules

This section discusses the integration of the Ultrasonic, turbidity and
Atmospheric pressure sensor’s hardware requirements, software
requirements and test validation.
Ultrasonic Sensor
This section discusses the hardware and software requirements for
integrating the A01NYUB Waterproof Ultrasonic Sensor. The A01NYUB
sensor boasts fast data transmission speeds, making it well-suited for
real-time monitoring of water levels in flood prediction systems, ensuring
timely updates on changes in river levels. While capable of providing
distance measurements, the accuracy of the A01NYUB sensor may vary
depending on environmental conditions and calibration efforts.
Nonetheless, it typically maintains accuracy within a few millimeters to
centimeters, facilitating precise distance readings crucial for flood
prediction applications. Furthermore, the A01NYUB sensor offers reliable
data accuracy, though calibration may be necessary to optimize its
performance. Its waterproof design enhances its suitability for outdoor use
in flood prediction systems, ensuring robust and reliable operation even in
challenging environmental conditions [68].

Ultrasonic Sensor Hardware Requirements

The system consists of two modules: the LoRa SX1278 module
and the ANYUB01 module. The LoRa SX1278 module is connected
to seven digital pins, labeled as LoRa_PINS, and the ANYUB01
module is connected to four pins labeled in blue, as shown in
Figure 3.2. The LoRa module uses pins 3.3, 13, 12, 11, 10, and 9,
which are the default pins required by the specific library in use.
Additionally, pins 6 and 7 are designated for the Echo and Trigger
pins of the ultrasonic sensor, respectively.

43
 The primary power source for the transmitting and receiving
devices is the power line available at the diversion weir, eliminating
the need for battery power supplies. This setup explains the
absence of power supply circuits in the transmitter as shown in
Figure 3.3, which is relatively similar to the power supply of the
receiving station as well. The devices heavily rely on AC-DC
adapters during calibration and field testing. An alternative power
supply option is the USB connector available on each Arduino
NANO microcontroller, which is used for debugging or supervising
the device’s performance during various validation tests.

Figure 3.1: Pin layouts of the ultrasonic sensor

Figure 3.2: Power management of Ultrasonic sensor of the transmitter and receiver

Ultrasonic Sensor Software Requirements

44
#define TRIG 7
#define ECHO 6
The provided code snippet initializes the ultrasonic sensor's trigger
pin (TRIG) and echo pin (ECHO) by defining them as digital pins 6
and 5, respectively. These pins are commonly used to control and
receive signals from the ultrasonic sensor.

void setup() {
pinMode(TRIG, OUTPUT);
pinMode(ECHO, INPUT_PULLUP);
}
Within the setup function, the pinMode function is utilized to
configure the TRIG pin as an output and the ECHO pin as an input
with a pull-up resistor enabled. This setup ensures proper
communication between the Arduino board and the ultrasonic
sensor

void loop() {
digitalWrite(TRIG, LOW);
delayMicroseconds(2);

digitalWrite(TRIG, HIGH);
delayMicroseconds(20);
}

In the loop function, the code begins by sending a short LOW pulse
to the TRIG pin to trigger the ultrasonic sensor. After a brief delay,
a HIGH pulse is sent to the TRIG pin for a specified duration to
generate an ultrasonic pulse. Subsequently, the TRIG pin is set
back to LOW to conclude the pulse generation process.

digitalWrite(TRIG, LOW);
int distance, time = pulseIn(ECHO, HIGH,26000);
distance= 3 + time/58;

The pulseIn function is then employed to measure the duration of

the pulse received by the ECHO pin. The duration of the pulse

45
 corresponds to the time taken for the ultrasonic signal to travel to
the object and back. This time value is stored in the 'time' variable.

Using the time value obtained from the pulseIn function, along with
the speed of sound in air (approximately 3 microseconds per
millimeter), the distance to the object is calculated. The distance
calculation considers the time taken for the ultrasonic signal to
travel to the object and back, and it is determined by dividing the
time by a conversion factor of 58, as per the typical calculation
method for ultrasonic sensors. This was obtained through the
calibration of the device to match the compared distance to the
value the ultrasonic sensor is outputting.

Serial.print("Time ");
Serial.print(time);
Serial.print(" | ");
Serial.print("Distance ");
Serial.print(distance);
Serial.println("cm");
delay(1000);
}
Finally, the calculated time and distance values are printed to the
serial monitor for observation and analysis. Each iteration of the
loop function includes a delay of 1000 milliseconds (1 second) to
control the sampling frequency of the ultrasonic sensor.

Turbidity Sensor
Turbidity Sensor Hardware Requirements
The system comprises two modules: the LoRa SX1278 and the TS-
300B turbidity sensor. The LoRa SX1278 module connects to
seven digital pins labeled LoRa_PINS, while the TS-300B module
connects to two pins labeled in blue, as shown in Figure 3.4. The
LoRa module uses pins 3.3, 13, 12, 11, 10, and 9, which are the
default pins specified by the library in use. Additionally, analog pin
A5 serves as the data pin for the turbidity sensor.
The primary power source for both the transmitting and receiving
devices is the power line available at the diversion weir, eliminating
the need for battery power supplies. This explains the absence of
power supply circuits in the transmitter, as illustrated in Figure 3.5,
and is similar to the power setup of the receiving station. The

46
 devices rely on AC-DC adapters during calibration and field testing.
An alternative power source is the USB connector on each Arduino
NANO microcontroller, which is used for debugging or monitoring
the device’s performance during various validation tests.

Figure 3.3: Pin layouts of the Turbidity Sensor connected to the microcontroller

Figure 3.4: Power Management of the Turbidity sensor and microcontroller device

Turbidity Sensor Software Requirements

#include <SPI.h>
#define Turbidity A5

47
The SPI library is used for communication with devices over the
SPI bus. Turbidity is defined as an analog pin (A5) where the
turbidity sensor is connected.

//-----VARIABLES---//
float Turb_reading;
float Turb_turbidity;
float voltage;
String turbidity_text;
uint16_t turbRaw;

The sections above are variables for storing sensor readings,

calculated turbidity, voltage, turbidity text, and raw data.

void setup() {
// INITIALIZATION //
Serial.begin(9600);
// PIN INITIALIZATION //
pinMode(Turbidity, INPUT);
}

The setup() function only initializes the serial monitor at a baud rate
of 9600 and sets the variable “Turbidity” as an INPUT variable.

void loop() {
Turb_reading = analogRead(Turbidity);

voltage = Turb_reading / 1023.0 * 5.0;

// Apply the conditions and calibration formula

if (voltage <= 1.58) {
turbidity_text = "greater than 800 NTU";
Turb_turbidity = 800;
} else if (voltage >= 1.90) {
turbidity_text = "less than 10 NTU";
Turb_turbidity = 10;
} else {
Turb_turbidity = (-4683.8 * pow(voltage, 2) + 13748 * voltage -
9218);
if (Turb_turbidity < 0) {
turbidity_text = "Invalid Reading";
Turb_turbidity = 0;
} else {
turbidity_text = String(Turb_turbidity) + " NTU";

48
 }
}

Within the loop function, the analogRead() function is called to read

the voltage output from the turbidity sensor, which is connected to
pin A5. The analogRead() function returns a value between 0 and
1023, representing the voltage level measured by the analog pin.
This value is stored in the 'reading' variable. The voltage is
calculated by converting the reading to voltage using the formula:
voltage = (reading * 5.00) / 1023. This formula scales the analog
reading to the voltage level, considering that the Arduino's analog
pins operate at a voltage range of 0 to 5 volts.

Finally, the reading and voltage values are printed to the serial
monitor using the Serial.print() function. This allows the user to
monitor the sensor's output voltage in real time for calibration and
analysis purposes.

Atmospheric Pressure Sensor

Atmospheric Pressure Sensor Hardware Requirements
The system consists of two modules: the LoRa SX1278 and the
BMP180 atmospheric pressure sensor. The LoRa SX1278 module
connects to seven digital pins, labeled as LoRa_PINS, while the
BMP180 module connects to two pins labeled in blue, as shown in
Figure 3.6. The LoRa module uses pins 3.3, 13, 12, 11, 10, and 9,
which are the default pins specified by the library in use.
Additionally, analog pins 4 and 5 are designated for the SCL and
SDA pins of the BMP180 sensor, respectively.
The power management for this system is unique due to the
experimental nature of the data collection. Two transmitters were
built to send data to one main receiver, with the transmitters located
far from each other. The BMP180 operates well when connected to
the power supply of the Arduino Nano microcontroller, making
batteries a suitable option to power the devices. As shown in Figure
3.7, a 5V voltage regulator and bypass capacitor are added to the
power supply to regulate voltages higher than the intended supply
voltage. An alternative power source is the USB connector on each
Arduino Nano microcontroller, which is used for debugging or
monitoring the device’s performance during various validation tests.

49
 Figure 3.5: Pin layouts of the BMP180 in the microcontroller

Figure 3.6: Power Management of BMP180 and microcontroller

Atmospheric Pressure Sensor Software Requirements

//-----LIBRARIES-----//
#include <Adafruit_BMP085.h>

The code includes the Adafruit BMP085 library, which is used to

interface with the BMP180 sensor. This library provides functions to
communicate with and read data from the BMP180 sensor.

50
//-----PINOUTS AND PRESETS-----//
Adafruit_BMP085 BMP;

//-----VARIABLES-----//
uint32_t BMP_pres;
float BMP_temp;

The next set of code snippets declare variables to store

temperature, pressure, and raw sensor readings. The 'BMP_pres'
and ‘BMP_temp’ variables are of type float to store the temperature
and pressure values respectively.

void setup() {
//-----PERIPHERAL INITIALIZATION-----//
Serial.begin(9600);
BMP.begin();
} // End of setup

void loop() {
//-----SENSOR READING-----//
BMP_temp = BMP.readTemperature(); // BMP Temperature
BMP_pres = BMP.readPressure(); // BMP Pressure

Moving on to the setup function, the serial monitor is initialized with

a baud rate of 9600 to enable communication between the Arduino
board and the computer. Additionally, the BMP.begin() function is
called to initialize communication with the BMP180 sensor module.

Within the loop function, the temperature and pressure values are
read from the BMP180 sensor using the BMP.readTemperature()
and BMP.readPressure() functions, respectively. These values are
then stored in the 'temp' and 'pressure' variables.

51
 Serial.print("Temp: ");
Serial.print(BMP_temp); Serial.println("degC");
Serial.print("Pres:");
Serial.print(float(BMP_pres/100.000)); Serial.println("hPa");
Serial.println(".................................................");
Finally, the temperature and pressure values are printed to the
serial monitor using the Serial.print() and Serial.println() functions.
This allows the user to monitor the sensor readings in real-time for
calibration and comparison purposes during the study.

Ultrasonic Sensor Height Sensitivity Test

The validation of the water level sensor primarily relies on gauge readings
observed at the diversion weir dam. As the water level sensor is already
calibrated, no additional code adjustments or modifications are necessary,
allowing for a direct approach to field testing.

During the field test, pre-existing water level markers were used for
calibration. These markers were labeled based on their elevation above
sea level, ranging from 72 meters to 90 meters, with intervals of 10
centimeters. This corresponds to a range of 0 to 6 meters from the
positioned ultrasonic station. The correlation between different height
depths (in centimeters) and their respective measurements is shown in
Table 3.0 below. The test involved seven trials for each water level
marking in centimeters to determine the overall mean, standard deviation,
and coefficient of variation for each marking.

Table 3.0: Sensor height sensitivity test

Standar
d Stan
measur Trial Trial Trial Trial Trial Trial Trial dard
ement 1 2 3 4 5 6 7 Mea Devi
(m) (cm) (cm) (cm) (cm) (cm) (cm) (cm) n ation CV
1
1.25
1.5

52
 2
2.25
2.5
2.75
3
3.25
3.5
3.75
4
4.25
4.5
4.75
5
5.25
5.5
5.75
6

Field testing accuracy reading

Field testing will involve a systematic comparison of ultrasonic sensor
readings to gauge readings over multiple days and times. The process
includes recording both sensor and gauge readings at specified intervals,
calculating the average of these readings, and determining the percentage
error between the sensor readings and the gauge readings. This will help
to assess the accuracy and reliability of the sensors in various
environmental conditions.

Table 3.1: Field testing accuracy reading

Date Time Gauge Sensor Average % error

Reading Reading
Jun-14

53
Comparison of Sensor height sensitivity test through percent
error checking
This test is designed to compare the height sensitivity of the A01NYUB
and JSN SR04T ultrasonic sensors by evaluating the percentage error of
each sensor at different height levels. Table 3.3 is structured to list
measurement points (in meters) and the corresponding percentage error
for each sensor. The verdict column provides a comparative assessment
based on the calculated errors.

Table 3.2: Comparison of Sensor height sensitivity

Distance (m) A01NYUB JSN SR04T Verdict

PERCENT ERROR PER DISTANCE

1.25

1.5

2.25

2.5

2.75

54
 3

3.25

3.5

3.75

55
Comparison of Field testing accuracy reading through error
checking
This test aims to compare the real-world performance of the A01NYUB
and JSN SR04T ultrasonic sensors by evaluating the percentage error in
their readings over multiple dates and times. Table 3.4 captures specific
dates and times, the measurement points or distances, and the
corresponding percentage error for each sensor. The verdict column
summarizes the comparative accuracy of the sensors. Tis to assess the
accuracy and reliability of the sensors in field conditions. By recording and
comparing the percentage error over time, researchers can determine
which sensor provides more accurate readings in practical, real-world
scenarios.

Table 3.3: Comparison of Field testing accuracy

DATE & TIME DISTANCE A01NYUB JSN SR04T Verdict

(?)
PERCENT ERROR PER
SAMPLE DATA

1.25

1.5

2.25

2.5

2.75

3.25

3.5

56
Validation of Turbidity Sensor Accuracy Against Turbidimeter
Readings
The validation of the turbidity sensor's accuracy involved two processes:
calibration through serial dilution of formazine (NTU) and field testing of
the sensor with comparisons to two standardized turbidimeters. The initial
calibration was done by creating different NTU (formazine solutions
diluting them in different milliliters of water and submerging the sensor in
different NTU solutions to get its voltage reading. The NTU solutions were
also compared to a standardized turbidimeter reading to determine the
accuracy of the dilution. For the field testing, comparisons were made with
two standardized turbidimeters (the Eutech instrument turbidity meter from
the university laboratory and the Hach 2100Q turbidimeter from the water
quality testing laboratory of the Zamboanga City Water District) and the
NTU reading from TS- 300B turbidity sensor. The sensor was then
deployed for field tests to gather water data samples under specific
weather conditions near the river diversion dam.

Turbidity Sensor calibration using a serial dilution of NTU

(formazine)
Table 3.4 shows the serial dilution for different NTU levels for the
calibration of the turbidity sensor. A turbidity testing kit sourced from the
Physics Laboratory was utilized. The calibration procedure entailed
standard 800 and 100 formazine solutions diluted using distilled water to
create seven (7) different NTU levels. Using the following equation, which
states that C₁V₁ = C₂V₃ where C₁ andV₁ represent the concentration
and volume of the initial concentrated solution and C ₂ and V ₃ represent
the concentration and volume of the final diluted solution, the researchers
may relate the concentrations and volumes before and after a dilution.
Using the formula below, the final volume of the NTU solutions can be
achieved. Subsequently, the correlation between NTU's measured
potential and turbidity is evaluated for linearity and graphically
represented, as shown in Table 3.4.

57
 Given: 20 mL of 800 NTU and 20 mL of 100 NTU

Table 3.4: Serial Dilution of 800 NTU (formazine) to create different NTU levels
NTU Distilled H2O Outcome

20 mL of 800 NTU Given

15 mL of 800 NTU 5 mL

15 mL of 600 NTU 7.5 mL

10 mL of 400 NTU 10 mL

20 mL of 100 NTU Given

10 mL of 100 NTU 10 mL

10 mL of 50 NTU 40 mL

Voltage Reading comparison (TS-300B vs Eutech Instrument

Turbidimeter)
There were seven (7) different NTU solutions used in the study for the
calibration of the turbidity sensor. The solutions were used to determine
the different voltage readings of the turbidity sensor and determine the
relationship using a graph and getting its equation as shown in Figure 3.5.
Additionally, the diluted solutions will also be compared with the NTU
reading of the Eutech Instrument Turbidimeter to determine the accuracy
of the diluted solution.

Table 3.5: Voltage reading of Turbidity sensor from different NTU solutions and NTU
readings of Eutech Instrument from the diluted NTU solutions
NTU SOLUTION Voltage Reading from TS- NTU reading from Eutech
(Formazine) 300B Turbidity Sensor Instrument Turbidity Meter

10

50

100

200

400

600

58
 800

Figure 3.7: Turbidity vs voltage level relationship, Turbidity graph

Figure 3.7 will show the equation of the line between the relationship of the
Turbidity Level and the Voltage levels in millivolts. This is then compared and
adjusted in accordance with the results of the Standardized Eutech turbidimeter
for field testing.

Field Testing of Turbidity Sensor Accuracy Against

Turbidimeter Readings
To ensure the reliability of calibrated voltage levels and actual turbidity
readings, percent errors are calculated based on three readings for each
sample. This test assesses how closely the sensor outputs values within
the minimum and maximum allowable turbidity ranges, evaluating its
sensitivity under varying turbidity conditions which is shown by the tables
3.6 & 3.7 below. table 3.7 only differs with the addition of another
turbidimeter from the laboratories of the Zamboanga CIty Water District.

Table 3.6: Field testing of turbidity level: Eutech Instrument Turbidimeter vs TS_300B
NTU vs True NTU reading

59
 Eutech Turbidity Arduino’s
Date Sampl Instrument Sensor True NTU Percent Error
es Turbidity TS- 300B Reading
Meter NTU NTU

10

Turbidimeter vs TS_300B NTU vs True NTU reading

Table 3.7: Field testing of turbidity level: Eutech Instrument Turbidimeter vs Hach 2100Q

Eutech Hach Turbidity Percent Percent

Instrument 2100Q Sensor Error Error
Date Sampl Turbidimet Turbidimete TBS- 300B (Eutech (Hach
es er (ADZU) r (ZCWD) NTU Instrument 2100Q vs
NTU NTU True vs TS- TS- 300B)
Reading 300B)

60
 9

10

Standardized Atmospheric Pressure and Temperature Versus

Atmospheric Pressure and Temperature Sensor Data
During field testing, the BMP180 module's atmospheric pressure and
temperature readings will be compared with those from standard
barometers and thermometers. These comparisons will occur at fixed 15-
minute intervals to ensure continuous monitoring, as detailed in the table
below. An error plot will visually depict the discrepancies between the
atmospheric pressure and temperature readings from each interval,
illustrated in Figure 3.9.
Table 3.8: Data gathering for Standardized tools and BMP108 module at 15
minutes interval

BMP10
BMP10
Baromet Thermome 8
8 Perce Perce
er ter Sensor
Date Time Sensor nt nt
Reading Reading Readin
Readin Error Error
(hPa) (degC) g
g (hPa)
(degC)
09:00
09:15
09:30
09:45
10:00
10:15
10:30
10:45
11:00
11:15
11:30
11:45

61
12:00
12:15
12:30
12:45
13:00
13:15
13:30
13:45
14:00
14:15
14:30
14:45
15:00
15:15
15:30
15:45
16:00

62
 Error Plot for Pressure
1011.2

1011

Air Pressure (hPa) 1010.8

1010.6

1010.4

1010.2

1010

1009.8
Time of Reading

Barometer Reading (hPa)

Figure 3.8: Error plot of Standardized pressure data vs BMP108 pressure data

Error Plot for Temperature

28
27
26
Temperature (degC)

25
24
23
22
21
20
0 0 0 0 0 0 0 0 0 0 0 0 0 0 0
:0 :3 :0 :3 :0 :3 :0 :3 :0 :3 :0 :3 :0 :3 :0
09 09 10 10 11 11 12 12 13 13 14 14 15 15 16
Time of Reading

Thermometer Reading (degC)

Figure 3.9: Error plot of Standardized temperature data vs BMP108 temperature data

63
Atmospheric Pressure and Temperature Difference of Enclosed
Case Versus Open Case
Studies have indicated that the internal atmospheric pressure and
temperature within the housing of the BMP180 module may impact the
accuracy of its readings. This test aims to assess this effect on the device
casing and determine if necessary modifications are needed for optimal
placement and display. Tables 3.9 and 3.10 below outline the data
collection format for atmospheric pressure and temperature readings,
along with reference data from standardized measurement tools for both
parameters.
Table 3.9: Data gathering for the atmospheric pressure of the 3 devices: Sensor A,
Sensor B, and Reference device

ATMOSPHERIC PRESSURE
SETUP A SETUP B
REFERENCE
EXPOSED ENCLOSED EXPOSED ENCLOSED

Table 3.10: Data gathering for the temperature of the 3 devices: Sensor A, Sensor B,
and Reference device

TEMPERATURE

64
 SETUP A SETUP B
REFERENCE
EXPOSED ENCLOSED EXPOSED ENCLOSED

Integration of Data Storage and Logging

Data Storage and Logging Hardware Requirements

65
Figure 3.10: Pin layouts of the RTC and SD Card module integrated to the
system
Data Storage and Logging Software Requirements

#include <SD.h>
#include <virtuabotixRTC.h>
#define CS_LoRa 10
virtuabotixRTC myRTC(5, 6, 7);
File myFile;

In this section, the SD library is included, enabling communication

with the SD card. Two constants, CS_LoRa and CS_MicroSD, are
defined to represent the chip-select pins for tdhe LoRa module and
the SD card module, respectively. The File object myFile is
declared to manage file operations.

unsigned long previousMillis = 0;

const long interval = 900000;
int previousMonth = 0;
char currentFileName[20];

void storeData(const char* fileName);

void createNewFileForMonth();.

This section sets the integers of various variable needed for the
program. previousMillis and interval is responsible for the systems
operation and sending of data in intervals. In this case, interval is
set to 900000 milliseconds or 15 minutes, which prompts the loop
function to operate to the set interval variable.
currentFileName[20] sets how may bits are used to name the file
were the data is being logged. StoreData() and

66
createNewFileForMonth() are functions used for data storage and
logging.

void setup() {

//-----PERIPHERAL INITIALIZATION-----//
Serial.begin(9600); // *TROUBLESHOOTING
SD.begin(CS_MicroSD);
myRTC.updateTime();
// Store the initial month
previousMonth = myRTC.month;
createNewFileForMonth();
}

Here, within the setup() function, the pins corresponding to the

LoRa and SD card chip selects are configured as output pins.
Additionally, serial communication is initialized at a baud rate of
9600, which will be essential for debugging and monitoring
purposes. Initializes the RTC module to update the current Time
and store the current month under the variable previousmonth. The
function createNewFileFOrMonth() is called upon to create a new
folder for the file.

void loop() {

//-----codes to process the reception, conversion and transmission

of data before storing and logging -----//
etc…

storeData(currentFileName);

if (myRTC.month != previousMonth) {
createNewFileForMonth();
// Create a new file for the new month
previousMonth = myRTC.month;
// Update the previous month
}

After each line of code handling data reception, conversion, and

transmission, the `storeData(currentFileName)` function is invoked
for storing and logging the data. Another function is called to create
a new file for the current month when needed. The system logic
relies on the current month displayed by the RTC module to

67
determine this. The current month is then saved as
`previousMonth` for future reference.

void storeData(const char* fileName) {

myFile = SD.open(fileName, FILE_WRITE);

if (myFile) {
Serial.print("Storing in memory...");
myFile.print( “all the necessary data to print out”);
myFile.close();
Serial.println("Data stored!");
} else {
Serial.println("Error opening file!");
}
}
This function calls out the SD card to open a file under the current
filename variable, and inserts all the necessary data using the
“variable”.print command. The serial prints include are for
debugging purposes.

void createNewFileForMonth() {

snprintf(currentFileName, sizeof(currentFileName), "%02d_

%04d.txt", myRTC.month, myRTC.year);
if (SD.exists(currentFileName)) {
Serial.print("File already exists: ");
Serial.println(currentFileName);
} else {
myFile = SD.open(currentFileName, FILE_WRITE);
if (myFile) {
myFile.println("New month data");
myFile.close();
Serial.print("Created new file: ");
Serial.println(currentFileName);
} else {
Serial.println("Error creating new file");
}
}
}

This function sets the file name format thru the use of snprintf ()
function. It is coded to format the file name as “__month no.__” + _
+ “__year__” via calling out the month and year data from the RTC
module. The function also handles error checking for multiple files

68
 under the same file name. it is coded so that the system does not
make another file with the same file name. If the system detects
there is no initialized SD card module, it prints out “error creating
new file!”. This is helpful for debugging purposes.

Data Storage and Logging Test

This assessment aimed to validate the operational capability of the
module, ensuring it could reliably receive and store data from the
microcontroller. The test involved 10 trials daily over a monitoring period,
totaling 50 trials, with a target success rate of 90%. This ensured that the
MicroSD card effectively stored data transmitted by the microcontroller in
the majority of instances. Additionally, a comparison made with data from
another standard source using a GSM module as a secondary receiver to
track data packet. This comparison confirmed the satisfactory functionality
of the data storage and logging modules.
Table 3.11 Data Storage and Logging Test

Date Test Case No. Data Stored Expected Output

69
Gantt Chart
The Gantt chart below outlines the progress report for the various tasks
involved in the project from January to July. The tasks are categorized into
four main areas: Functional Testing, Hardware and Software Design, Data
Gathering, and Field Testing. Each category includes the integration of
four modules: ANYUB01 ultrasonic sensor, TS-300B turbidity sensor,
BMP108 atmospheric pressure and temperature sensor, and the data
storage and logging modules.

JANUARY FEBRUARY MARCH APRIL

METHODS OR WEEK
TASKS
1 2 3 4 5 6 7 8 9 1 1 1 1 1 1 1
0 1 2 3 4 5 6
FUNCTIONAL TESTING
Integration of
ANYUB01
ultrasonic sensor
Integration of
TS-300B
turbidity sensor
Integration of
BMP103
atmospheric
pressure and
temperature
sensor
Integration of
data storage and
logging module
HARDWARE AND SOFTWARE DESIGN
Integration of
ANYUB01
ultrasonic sensor
Integration of
TS-300B
turbidity sensor

70
Integration of
BMP103
atmospheric
pressure and
temperature
sensor
Integration of
data storage and
logging module
DATA GATHERING
Integration of
ANYUB01
ultrasonic sensor
Integration of
TS-300B
turbidity sensor
Integration of
BMP103
atmospheric
pressure and
temperature
sensor
Integration of
data storage and
logging module
FIELD TESTING
Integration of
ANYUB01
ultrasonic sensor
Integration of
TS-300B
turbidity sensor
Integration of
BMP103
atmospheric
pressure and
temperature

71
sensor
Integration of
data storage and
logging module

MAY JUNE JULY

METHODS OR TASKS WEEK
17 18 19 20 21 22 23 24 25 26 27 2
8
FUNCTIONAL TESTING
Integration of ANYUB01
ultrasonic sensor
Integration of TS-300B
turbidity sensor
Integration of BMP103
atmospheric pressure
and temperature sensor
Integration of data
storage and logging
module
HARDWARE AND SOFTWARE DESIGN
Integration of ANYUB01
ultrasonic sensor
Integration of TS-300B
turbidity sensor
Integration of BMP103
atmospheric pressure
and temperature sensor
Integration of data
storage and logging
module
DATA GATHERING

72
Integration of ANYUB01
ultrasonic sensor
Integration of TS-300B
turbidity sensor
Integration of BMP103
atmospheric pressure
and temperature sensor
Integration of data
storage and logging
module
FIELD TESTING
Integration of ANYUB01
ultrasonic sensor
Integration of TS-300B
turbidity sensor
Integration of BMP103
atmospheric pressure
and temperature sensor
Integration of data
storage and logging
module

73
 CHAPTER IV
RESULTS AND DISCUSSION

Methodology Summary and Results Verdict

A. Ultrasonic Water level sensor height The performance of the
sensor sensitivity test A01NYUB ultrasonic
height sensor module
sensitivity The ultrasonic sensor demonstrated acceptable
test demonstrated high accuracy and ranges of error across
B. Field testing precision in measuring distances. various height sensitivities
accuracy Sensitivity tests showed that the and proved its capability
reading test sensor could measure distances to monitor water levels
from 1 meter to 6 meters with over extended periods in
minimal deviation. For instance, at the field. This reliability
1 meter, the mean reading was makes the A01NYUB
approximately 100.4 cm with a ultrasonic sensor a
standard deviation of 0.34 cm. practical choice for flood
monitoring applications.

Field testing accuracy reading

test
Field testing validates the sensor's
performance, with most errors
falling within ±1% sometimes at
only ±3% error.
C. Comparison Comparison of Sensor height The A01NYUB sensor's
of Sensor sensitivity test through error minor errors and
height checking and Comparison of consistent performance
sensitivity Field testing accuracy reading make it a reliable choice
test through through error checking for precise water level
error measurements, while the

74
 checking Both JSN SR04T and A01NYUB JSN SR04T sensor's high
D. Comparison sensor's performance confirms its errors and limitations
of Field suitability for accurate water level highlight its impracticality
testing measurements, as it maintains for this purpose. These
accuracy consistent accuracy and reliability. results advocate for the
reading On the other hand, the JSN SR04T use of the A01NYUB
through error sensor's distance limitations render sensor in scenarios
checking it ineffective for scenarios requiring requiring accurate and
measurements beyond its effective dependable
range. measurements.
E. Turbidity Validation of Water Level Sensor The performance of the
Sensor Accuracy Against Turbidimeter TS-300B Turbidity sensor
calibration was quite low and poor
using a When the readings from the TS- due to its specification.
300B
serial dilution turbidity sensor were The sensor had high
of NTU compared with the readings from accuracy when NTU is
(formazine) the Eutech Instrument, 92.86% greater than 100 NTU,
was the total average percent error however, its accuracy
among the 34 water samples was poor below 100 NTU
shown in Table 4.4. Among the 10 since its ratio range was 0
F. Voltage water samples in Table 4.3,
Reading - 1000 NTU with ± 30.
113.86% was the percent error of This means it cannot
comparison the turbidity sensor when
(TS-300B vs compared accurately measure 50
with the Eutech NTU and below.
Eutech Instrument.
Instrument
Turbidimeter In addition, when the turbidity
) sensor was also compared with the
readings from the Hach 2100Q
turbidimeter, the result was 57. 80
percent error.

E. Standardized Standardized Atmospheric The BMP180 sensor

Atmospheric pressure and temperature demonstrates excellent
pressure and versus Atmospheric pressure reliability in measuring
temperature versus and temperature sensor data atmospheric pressure,
Atmospheric 28 readings were collected to with percent errors well
pressure and gather the temperature and within an acceptable
temperature sensor pressure at 15-minute intervals. range, making it highly
data Upon observations, the percent suitable for applications
errors for atmospheric pressure requiring precise pressure
readings range from -0.10% to measurements. Although
F. Atmospheric 0.03%, and the percent errors for the sensor shows more
pressure and temperature readings range from - variability in temperature
temperature 3.40% to 2.22%. F-test was readings, the lack of

75
difference of employed to account for the wide significant difference in
enclosed case range percent errors in variances (as determined
versus open case temperature readings, but found no by the F-test) suggests
significant difference between the that the BMP180 sensor
readings. still provides reasonably
accurate temperature
measurements.
Atmospheric pressure and
temperature difference of Setup B shows a
enclosed case versus open case significant difference (p-
T-test was employed to compare value = 0.0160) between
the means of the two setups in exposed and enclosed
different condtions. Comparing conditions, indicating that
Setup A (Exposed) and Setup B temperature readings
(Exposed) to the reference differ significantly
temperature readings yields p- between these setups.
values of 0.0595 and 0.3163,
respectively, both exceed the Setup A shows no
significance level of 0.05 (0.0595 >significant difference in
0.05 and 0.3163 > 0.05). pressure readings
Comparing Setup A (Exposed) and compared to the
Setup B (Exposed) to Setup A reference, while Setup B
(Enclosed) and Setup Bdoes. Both setups exhibit
(Enclosed), respectively, yields p- significant differences
values of 0.7968 and 0.0160. between exposed and
enclosed conditions,
Comparing Setup A (Exposed) and indicating that the
Setup B (Exposed) to the reference enclosure affects
pressure readings yields p-values pressure readings.
of 0.3551 and 4.7945E-07,
respectively. Comparing Setup A
(Exposed) and Setup B (Exposed)
to Setup A (Enclosed) and Setup B
(Enclosed), respectively, yields p-
values of 0.0320 and 0.0303, both
exceeding the p-value (<0.05).
G. Data Storage Data Storage and Logging Test
and Logging Test
Throughout the monitoring period, The RTC DS1302, in
10 randomly selected data sets collaboration with the
comprising 100 data points were microSD card, performs
meticulously reviewed. Upon as expected, ensuring
comparison with the device's reliable data storage and
expected output, all data proved to logging capabilities
be accurate without any errors throughout the monitoring
detected. The RTC DS1302, process.
integrated with the microSD card,

76
 operated seamlessly and reliably,
fulfilling its intended function of
data storage and logging.

Water Level Sensor Height Sensitivity for A01NYUB

The A01NYUB ultrasonic sensor was tested for its sensitivity and
accuracy from 1 meter to 6 meters, as well as through field testing. The
data from the sensitivity test show the sensor's ability to measure
distances accurately across multiple trials, providing the mean, standard
deviation, and coefficient of variation (CV) for each standard
measurement. Table 4.1 shows the height readings of the ultrasonic
sensor in different trials.
Table 4.0: Water Level Sensor height sensitivity

Tria Stand
Standard Trial Trial Trial Trial Trial Trial
l1 ard
measure 2 3 4 5 6 7 Mean CV
(cm Devia
ment (m) (cm) (cm) (cm) (cm) (cm) (cm)
) tion
0.343
100 100. 100. 100. 100. 100.41 0.3422
1 100.5 100 64987
.2 8 1 9 4 42857 %
7

0.106
125 125. 125. 125. 125. 125. 125.51 0.0852
1.25 125.6 90449
.7 5 5 4 5 4 42857 %
7

0.214
150 150. 150. 150. 150. 150. 150.45 0.1428
1.5 150.3 91969
.5 3 4 9 5 3 71429 %
7

0.138
200. 200. 200. 199. 200. 200.02 0.0690
2 200 199.9 01311
1 1 2 8 1 85714 %
2

0.355
224 225. 225. 225. 225. 225. 225.25 0.1577
2.25 225.1 23298
.7 1 7 3 2 7 71429 %
9

0.234
250 250. 250. 250. 250. 250. 250.48 0.0934
2.5 250.5 01261
.3 4 3 9 3 7 57143 %
7

77
 0.177
274 275. 275. 275. 274. 274.98 0.0645
2.75 275 275 28105
.7 1 1 2 8 57143 %
2

0.167
300 300. 300. 300. 300. 300. 300.41 0.0558
3 300.4 61634
.5 7 2 3 5 3 42857 %
2

0.127
324 325. 325. 325. 325.04 0.0391
3.25 325.1 325 325 24180
.8 2 1 1 28571 %
2

0.281
350 350. 350. 350. 350. 350. 350.47 0.0802
3.5 350.2 15408
.4 3 9 5 2 8 14286 %
4

0.292
375 375. 375. 375. 375. 375. 375.52 0.0780
3.75 375.6 77002
.8 7 2 9 3 2 85714 %
2

0.195
400 400. 400. 400. 400. 400. 400.41 0.0487
4 400.4 18001
.5 6 2 3 7 2 42857 %
5

0.242
425 425. 425. 425. 425. 425. 425.37 0.0571
4.25 425.8 99715
.6 4 2 2 2 2 14286 %
9

0.149
450 450. 450. 450. 450. 450. 450.27 0.0332
4.5 450.1 60264
.3 5 2 4 3 1 14286 %
8

0.302
475 475. 475. 475. 475. 475.38 0.0636
4.75 475.7 475 37157
.5 8 4 2 1 57143 %
8

0.179
500 500. 500. 500. 500. 500. 500.27 0.0360
5 500 94708
.2 4 3 1 4 5 14286 %
2

0.256
525 525. 525. 525. 525. 525. 525.57 0.0488
5.25 525.2 34797
.6 8 9 5 7 3 14286 %
8

0.095
550 550. 550. 550. 550. 550. 550.42 0.0173
5.5 550.4 11897
.4 5 6 4 3 4 85714 %
3

78
 0.053
575 575. 575.04 0.0093
5.75 575.1 575 575 575 575 45224
.1 1 28571 %
8

0.281
600 600. 600. 600. 600. 600.34 0.0470
6 600.2 600 99966
.1 8 3 4 6 28571 %
2
As shown in Table 4.0, the means of the readings for each standard
measurement are very close to the actual standard measurements,
indicating the sensor is generally accurate. The mean readings are very
close to the standard measurements, indicating that the sensor does not
have a significant bias or systematic error. Furthermore, the standard
deviations are relatively small, suggesting that the sensor readings are
consistent across multiple trials. The small standard deviations across all
measurements suggest that the sensor's performance is stable and does
not fluctuate significantly from trial to trial.
The CV values across different standard measurements range from
0.0093% to 0.3422%. This indicates that the sensor exhibits a high degree
of precision, with variability in the sensor's readings being very low relative
to the mean values. This also shows that at lower standard measurements
(e.g., 1 m to 2 m), the CV values are slightly higher but still very low (<
0.35%), indicating good accuracy and precision, but as the standard
measurements increase (from 2.25 m to 6 m), the CV values generally
decrease, with the lowest CV being 0.0093% at 5.75 m, suggesting even
greater precision at higher measurements.
The sensor exhibits excellent sensitivity, accuracy, and precision in
measuring the standard values across multiple trials. The low coefficient of
variation across all standard measurements indicates that the sensor is
reliable and consistent. The small standard deviations and means that
closely match the standard measurements further affirm the sensor's
robustness and suitability for precise applications. This performance
suggests that the sensor is highly reliable for practical use where precise
measurements are critical.

Water Level Sensor Height Sensitivity for JSN SR04T

The JSN SR04T ultrasonic sensor was tested for its sensitivity and
accuracy from 1 meter to 4 meters, as well as through field testing. The
data from the sensitivity test show the sensor's ability to measure
distances accurately across multiple trials, providing the mean, standard
deviation, and coefficient of variation (CV) for each standard
measurement. No stable data has been recorded for measurements
beyond 4 meters. Table 4.1 shows the height readings of the ultrasonic
sensor in different trials.
Table 4.1: Water Level Sensor height sensitivity for JSN SR04T

79
 Standard
Trial 1 Trial 2 Trial 3 Standard
measure Mean CV
(cm) (cm) (cm) Deviation
ment (m)
1 103.57 104.3 103.9 103.9233 0.365558933 0.3518%

1.25 127.59 128.5 127.9 127.9967 0.462637367 0.3614%

1.5 136.2 135.3 135.8 135.7667 0.450924975 0.3321%

2 171.74 172.8 172 172.18 0.552449093 0.3209%

2.25 210.95 209.7 210.3 210.3167 0.625166644 0.2973%

2.5 230.12 229.5 230.8 230.14 0.650230728 0.2825%

2.75 278.4 277.6 278.1 278.0333 0.404145188 0.1454%

3 313.2 314 312.6 313.2667 0.702376917 0.2242%

3.25 326.32 330.1 332 329.4733 2.891389516 0.8776%

3.5 243.73 244.6 243.2 243.8433 0.706847461 0.2899%

3.75 376.53 377.8 376.9 377.0767 0.653171749 0.1732%

4 270.74 269.5 271.2 270.48 0.879317917 0.3251%

Comparison of Sensor Height Sensitivity Test Through

Error Checking
The comparison of sensor height sensitivity involved evaluating the
performance of the JSN SR04T and A01NYUB ultrasonic sensors across
different height measurements ranging from 1 to 4 meters. The key metric
for this comparison was the coefficient of variation (CV), which measures
the relative variability of the sensor readings.
Table 4.2: Comparison of Sensor height sensitivity

Distance (m) A01NYUB JSN SR04T Coefficient of

variation
Coefficient of variation difference

1 0.3422% 0.351758% 0.0095%

1.25 0.0852% 0.361445% 0.2763%

1.5 0.1428% 0.332132 0.1893%

80
 2 0.0690% 0.320856% 0.2519%

2.25 0.1577% 0.29725% 0.1395%

2.5 0.0934% 0.282537% 0.1891%

2.75 0.0645% 0.145359% 0.0809%

3 0.0558% 0.224211% 0.1684%

3.25 0.0391% 0.877579% 0.8384%

3.5 0.0802% 0.289878% 0.2097%

3.75 0.0780% 0.17322% 0.0953%

4 0.0487% 0.325095% 0.2764%

The data presented in Table 4.2 compares the height sensitivity of two
sensors, A01NYUB and JSN SR04T, across various distances ranging
from 1 meter to 4 meters. The coefficient of variation (CV) is used as a
measure of the relative variability of the sensor readings.

At a distance of 1 meter, the A01NYUB sensor exhibits a CV of 0.3422%,

while the JSN SR04T sensor has a slightly higher CV of 0.351758%,
indicating a minor difference of 0.0095%. As the distance increases to
1.25 meters, the A01NYUB sensor shows a remarkably low CV of
0.0852%, whereas the JSN SR04T sensor has a CV of 0.361445%,
resulting in a substantial difference of 0.2763%. This trend of higher
variability in the JSN SR04T sensor compared to the A01NYUB sensor
continues across most distances, with the differences in CV fluctuating.

At 1.5 meters, the A01NYUB sensor's CV is 0.1428%, and the JSN

SR04T's CV is 0.332132%, leading to a difference of 0.1893%. At 2
meters, the CVs are 0.0690% and 0.320856% for the A01NYUB and JSN
SR04T sensors, respectively, resulting in a difference of 0.2519%. At 2.25
meters, the CV difference narrows to 0.1395% with the A01NYUB sensor
showing a CV of 0.1577% and the JSN SR04T sensor at 0.29725%.

When the distance is 2.5 meters, the A01NYUB sensor has a CV of

0.0934%, while the JSN SR04T sensor's CV is 0.282537%, resulting in a
difference of 0.1891%. At 2.75 meters, the CVs are 0.0645% for the
A01NYUB sensor and 0.145359% for the JSN SR04T sensor, with a
difference of 0.0809%. The CVs continue to show a greater discrepancy at
3 meters, where the A01NYUB sensor exhibits a CV of 0.0558%
compared to the JSN SR04T sensor's 0.224211%, leading to a difference
of 0.1684%.

81
 Notably, at 3.25 meters, the JSN SR04T sensor shows a significant
increase in variability with a CV of 0.877579%, while the A01NYUB sensor
maintains a much lower CV of 0.0391%, resulting in the largest difference
of 0.8384%. At 3.5 meters, the CVs are 0.0802% for the A01NYUB sensor
and 0.289878% for the JSN SR04T sensor, with a difference of 0.2097%.
At 3.75 meters, the A01NYUB sensor has a CV of 0.0780% and the JSN
SR04T sensor 0.17322%, resulting in a difference of 0.0953%. Finally, at
4 meters, the A01NYUB sensor shows a CV of 0.0487%, while the JSN
SR04T sensor has a CV of 0.325095%, with a significant difference of
0.2764%.

In summary, the A01NYUB sensor consistently demonstrates lower

variability in its readings across all distances compared to the JSN SR04T
sensor. This indicates that the A01NYUB sensor is generally more reliable
and consistent in measuring height sensitivity, especially as the distance
increases. The substantial differences in the coefficient of variation,
particularly at longer distances, highlight the superior performance of the
A01NYUB sensor in maintaining consistent and accurate measurements.

Field Testing Accuracy Reading Test of A01NYUB

Ultrasonic Sensor
Field testing involved comparing the sensor readings to gauge readings
over multiple days and times. The sensor demonstrated an average
percentage error mostly within ±1%, with a few instances reaching up to
±3%. For instance, on June 14th at 8 am, the sensor reading was 75.8 cm
compared to the gauge reading of 74.28 cm, resulting in a -2% error. This
trend of minor errors was consistent throughout the testing period,
showing that the sensor maintained a high level of accuracy and reliability
in practical applications. An error plot will visually depict the discrepancies
between the atmospheric pressure and temperature readings from each
interval, illustrated in Figure 4.0.
Table 4.3: Field testing data (A01NYUB)
Date Time Gauge Sensor Average % error
Reading Reading
Jun-14 8am 74.28 75.8 75.04 -2%
9am 74.28 76.8 75.54 -3%
10am 74.28 75.8 75.04 -2%
11am 74.26 75.2 74.73 -1%
12pm 74.26 75.2 74.73 -1%
1pm 74.24 75.3 74.77 -1%

82
 2pm 74.24 75.1 74.67 -1%
3pm 74.24 74.8 74.52 -1%
4pm 74.22 74.8 74.51 -1%
5pm 74.22 74.1 74.16 0%

Jun-15 8am 74.2 75.1 74.65 -1%

9am 74.18 75.2 74.69 -1%
10am 74.18 75.2 74.69 -1%
11am 74.18 75.1 74.64 -1%
12pm 74.18 75.1 74.64 -1%
1pm 74.18 75.1 74.64 -1%
2pm 74.18 75.1 74.64 -1%
3pm 74.18 75.1 74.64 -1%
4pm 74.18 75.1 74.64 -1%
5pm 74.16 75.1 74.63 -1%

Jun-17 8am 74.18 74.8 74.49 -1%

9am 74.18 75.1 74.64 -1%
10am 74.18 74.8 74.49 -1%
11am 74.16 75.1 74.63 -1%
12pm 74.16 75.1 74.63 -1%
1pm 74.16 75.1 74.63 -1%
2pm 74.16 75.1 74.63 -1%
3pm 74.16 75.1 74.63 -1%
4pm 74.16 75.1 74.63 -1%
5pm 74.16 75.1 74.63 -1%
Jun-18 8am 74.12 73.6 73.86 1%
9am 74.1 73.2 73.65 1%
10am 74.1 72.2 73.15 3%

83
 11am 74.1 72.1 73.1 3%
12pm 74.1 72.9 73.5 2%
1pm 74.1 73.6 73.85 1%
2pm 74.1 73.6 73.85 1%
3pm 74.1 72.3 73.2 2%
4pm 74.1 71.9 73 3%
5pm 74.1 72.1 73.1 3%

Jun-19 7am 74.16 75.1 74.63 -1%

8am 74.16 75.1 74.63 -1%
9am 74.14 75.1 74.62 -1%
10am 74.14 75.1 74.62 -1%
11am 74.12 75.1 74.61 -1%
12pm 74.12 74.6 74.36 -1%
1pm 74.12 75.1 74.61 -1%
2pm 74.16 75.1 74.63 -1%
3pm 74.28 75.2 74.74 -1%
4pm 74.46 75.4 74.93 -1%
5pm 75.36 76.3 75.83 -1%
7am 74.38 75.3 74.84 -1%

Jun-20 8am 74.36 74.3 74.33 0%

9am 74.34 74.3 74.32 0%
10am 74.32 75.3 74.81 -1%
11am 74.3 75.3 74.8 -1%
12pm 74.3 75.2 74.75 -1%
1pm 74.3 75.2 74.75 -1%
2pm 74.34 75.3 74.82 -1%

3pm 74.36 74.3 74.33 0%

84
 4pm 74.36 75.3 74.83 -1%
5pm 74.34 75.3 74.82 -1%

Jun-21 7am 74.26 74.2 74.23 0%

8am 74.26 75.2 74.73 -1%
9am 74.26 75.2 74.73 -1%
10am 74.24 75.2 74.72 -1%
11am 74.24 74.1 74.17 0%
12pm 74.24 75.2 74.72 -1%
1pm 74.24 75.2 74.72 -1%
2pm 74.22 75.2 74.71 -1%
3pm 74.22 75.2 74.71 -1%
4pm 74.22 75.1 74.66 -1%
5pm 74.22 75.1 74.66 -1%

Jun-22 8am 74.2 75.1 74.65 -1%

9am 74.2 75.1 74.65 -1%
10am 74.2 74.8 74.5 -1%
11am 74.18 75.1 74.64 -1%
12pm 74.18 75.2 74.69 -1%
1pm 74.18 74.1 74.14 0%
2pm 74.18 74.5 74.34 0%
3pm 74.18 74.1 74.14 0%
4pm 74.18 75.1 74.64 -1%
5pm 74.18 75.1 74.64 -1%

Jun-24 7am 74.18 75.1 74.64 -1%

8am 74.18 74.2 74.19 0%
9am 74.18 75.1 74.64 -1%
10am 74.18 74.6 74.39 -1%

85
 11am 74.18 75.1 74.64 -1%
12pm 74.18 75.1 74.64 -1%
1pm 74.18 75.1 74.64 -1%
2pm 74.18 73.9 74.04 0%
3pm 74.18 75.1 74.64 -1%
4pm 74.18 75.1 74.64 -1%
5pm 74.18 75.1 74.64 -1%

Jun-25 7am 74.16 75.1 74.63 -1%

8am 74.16 75.1 74.63 -1%
9am 74.16 74.1 74.13 0%
10am 74.16 75.1 74.63 -1%
11am 74.16 75.1 74.63 -1%
12pm 74.16 75.1 74.63 -1%
1pm 74.16 73.9 74.03 0%
2pm 74.16 75.1 74.63 -1%
3pm 74.16 75.1 74.63 -1%
4pm 74.16 75.1 74.63 -1%
5pm 74.16 75.1 74.63 -1%
Jun-26 7am 74.18 75.1 74.64 -1%
8am 74.18 75.1 74.64 -1%
9am 74.18 75.1 74.64 -1%
10am 74.18 75.1 74.64 -1%
11am 74.18 75.1 74.64 -1%
12pm 74.18 75.1 74.64 -1%
1pm 74.18 75.1 74.64 -1%
2pm 74.18 75.1 74.64 -1%
3pm 74.18 75.3 74.74 -2%

86
 4pm 74.18 75.1 74.64 -1%
5pm 74.18 75.2 74.69 -1%

Jun-27 7am 74.76 75.7 75.23 -1%

8am 74.7 75.6 75.15 -1%
9am 74.16 75.1 74.63 -1%
10am 74.6 75.5 75.05 -1%
11am 74.56 75.5 75.03 -1%
12pm 74.52 75.6 75.06 -1%
1pm 74.5 75.4 74.95 -1%
2pm 74.48 75.4 74.94 -1%
3pm 74.46 74.8 74.63 0%
4pm 74.44 75.4 74.92 -1%
5pm 74.42 75.3 74.86 -1%

Jun-28 7am 74.3 75.3 74.8 -1%

8am 74.28 75.2 74.74 -1%
9am 74.28 74.8 74.54 -1%
10am 74.28 74.1 74.19 0%
11am 74.28 75.2 74.74 -1%
12pm 74.28 75.3 74.79 -1%
1pm 74.26 75.3 74.78 -1%
2pm 74.26 75.2 74.73 -1%
3pm 74.26 75.2 74.73 -1%
4pm 74.26 75.2 74.73 -1%
5pm 74.24 75.1 74.67 -1%
Jun-29 7am 74.22 75.2 74.71 -1%
8am 74.22 75.2 74.71 -1%
9am 74.2 73.9 74.05 0%

87
 10am 74.2 74.1 74.15 0%
11am 74.2 75.1 74.65 -1%
12pm 74.2 75.2 74.7 -1%
1pm 74.2 75.2 74.7 -1%
2pm 74.2 75.3 74.75 -1%
3pm 74.2 75.1 74.65 -1%
4pm 74.2 75.1 74.65 -1%
5pm 74.2 75.1 74.65 -1%

Jul-01 7am 74.16 75.1 74.63 -1%

8am 74.16 75.1 74.63 -1%
9am 74.16 75.1 74.63 -1%
10am 74.16 75.2 74.68 -1%
11am 74.16 75.2 74.68 -1%
12pm 74.16 75.1 74.63 -1%
1pm 74.16 75.1 74.63 -1%
2pm 74.96 75.9 75.43 -1%
3pm 74.86 75.8 75.33 -1%
4pm 74.84 75.7 75.27 -1%
5pm 74.7 75.6 75.15 -1%
Jul-02 7am 74.34 75.3 74.82 -1%
8am 74.32 75.3 74.81 -1%
9am 74.3 74.6 74.45 0%
10am 74.3 74.2 74.25 0%
11am 74.3 75.2 74.75 -1%
12pm 74.3 75.2 74.75 -1%
1pm 74.3 75.2 74.75 -1%
2pm 74.28 75.2 74.74 -1%

88
 3pm 74.2 75.1 74.65 -1%
4pm 74.2 75.1 74.65 -1%
5pm 74.2 75.1 74.65 -1%

Figure 4.0: Error plot of water level readings (Gauge Reading vs A01NYUB)

Field Testing Accuracy Reading Test of JSN SR04T

Ultrasonic Sensor
Field testing involved comparing the sensor readings to gauge readings
over multiple days and times. The sensor demonstrated an average
percentage error mostly within ±1%.
Table 4.4: Field testing data (JSN SR04T)

Date Time Gauge Sensor Average % error

Reading Reading

Jun-14 7am 74.18 74.91 74.545 -1%

8am 74.18 74.91 74.545 -1%

9am 74.18 74.9 74.54 -1%

10am 74.18 74.9 74.54 -1%

11am 74.18 74.91 74.545 -1%

12pm 74.18 74.89 74.535 -1%

89
 1pm 74.18 74.91 74.545 -1%

2pm 74.18 74.91 74.545 -1%

3pm 74.18 74.9 74.54 -1%

4pm 74.18 74.9 74.54 -1%

Comparison of Field Testing Accuracy Reading Through

Error Checking
The table provides a detailed analysis of the percent error differences
between the A01NYUB and JSN SR04T sensors at various times
throughout the day. The percent error is determined by comparing the
sensor readings to gauge readings, offering insight into the accuracy and
reliability of each sensor under practical conditions.

Table 4.5: Comparison of Field testing accuracy

TIME A01NYUB JSN SR04T Percent Error

difference
PERCENT ERROR PER SAMPLE
DATA

7am -1% -1% 0%

8am -1% -1% 0%

9am 0% -1% 1%

10am 0% -1% 1%

11am -1% -1% 0%

12pm -1% -1% 0%

1pm -1% -1% 0%

2pm -1% -1% 0%

3pm -1% -1% 0%

4pm -1% -1% 0%

90
 The data presented in Table 4.5 compares the field testing accuracy of
two sensors, A01NYUB and JSN SR04T, over a span of ten hours from 7
am to 4 pm. The percent error for each sensor was recorded hourly, with a
focus on determining the differences in accuracy between the two
sensors.

At 7 am, both sensors exhibited a percent error of -1%, resulting in no

difference between them (0%). This trend of identical performance
continues at 8 am, with both sensors again showing a -1% error and a
difference of 0%. At 9 am, the A01NYUB sensor showed an error of 0%,
while the JSN SR04T sensor had a -1% error, leading to a 1% difference
in favor of the A01NYUB sensor. This pattern repeats at 10 am, with the
same 1% difference observed. From 11 am through 4 pm, both sensors
consistently recorded a -1% error each hour, resulting in no differences in
their percent errors (0%).

Overall, the data indicates that the performance of both sensors is highly
similar, with the only deviations observed at 9 am and 10 am, where the
A01NYUB sensor demonstrated slightly better accuracy. However, for the
majority of the testing period, both sensors performed identically,
suggesting that they are equally reliable for field measurements in these
conditions.

Despite their similarities in performances, one must account to the JSN

SR04T sensor's limited measurement range and its inability to accurately
measure distances beyond 3 meters. Given that the dam height is at least
6 meters, the JSN SR04T sensor proves unsuitable for this application
due to its significant limitations.

These findings emphasize the importance of selecting the appropriate

sensor for specific field conditions. The A01NYUB sensor's performance
confirms its suitability for accurate water level measurements, as it
maintains consistent accuracy and reliability. On the other hand, the JSN
SR04T sensor's significant errors render it ineffective for scenarios
requiring measurements beyond its effective range.

Turbidity Sensor Calibration Using a Serial Dilution of

NTU (formazine)
Using the dilution formula V2 = (C1V1)/C2, results showed (Table 4.6)
different milliliters of NTU solutions ranging from 10 NTU to 800 NTU.
There were seven (7) NTU solutions made: 800 NTU, 600 NTU, 400 NTU,
200 NTU, 100 NTU, 50 NTU, and 10 NTU. These NTU solutions were
used to calibrate the TS-300B turbidity sensor.

91
Table 4.6: Serial Dilution of 800 and 100 NTU (formazine) to create different NTU levels

NTU Distilled H2O Outcome

20 mL of 800 NTU Given 20 mL of 800 NTU
15 mL of 800 NTU 5 mL 20 mL of 600 NTU
15 mL of 600 NTU 7.5 mL 22.5 mL of 400 NTU
10 mL of 400 NTU 10 mL 20 mL of 200 NTU
20 mL of 100 NTU Given 20 mL 0f 100 NTU
10 mL of 100 NTU 10 mL 20 mL of 50 NTU
10 mL of 50 NTU 40 mL 50 mL of 10 NTU

Voltage reading of the Turbidity Sensor from different

NTU solutions
After the dilution process of different NTU solutions, seven (7) NTU
solutions were used to calibrate the TS-300B turbidity sensor by getting its
voltage reading from different NTU solutions. The voltage readings of 10
NTU, 50 NTU, and 100 NTU were notably close. On the other hand, the
voltage readings of 100 NTU, 200 NTU, 400 NTU, 600 NTU, and 800 NTU
were relatively far from each other. In addition, when comparing the
diluted solutions to the readings of the Eutech Instrument Turbidimeter,
there were minor differences in NTU readings. The diluted solution was
10 NTU, however, the readings from the turbidimeter showed 15.28 NTU.
Similarly, 200 NTU showed 230 NTU, 400 NTU showed 371 NTU, and
600 NTU showed 532 NTU in the Eutech Instrument Turbidimeter

Table 4.7 : Voltage reading of Turbidity sensor from different NTU solutions and NTU
readings of Eutech Instrument from the diluted NTU solutions

NTU SOLUTION Voltage Reading from TS- NTU Reading from Eutech
(Formazine) 300B Turbidity Sensor Instrument Turbidity Meter

10 NTU 1.90 15.28 NTU

50 NTU 1.89 50.3 NTU

100 NTU 1.87 102 NTU

200 NTU 1.83 230 NTU

400 NTU 1.77 371 NTU

600 NTU 1.74 532 NTU

800 NTU 1.58 800 NTU

92
 Figure 4.1: Turbidity vs voltage level relationship, Turbidity graph

Figure 4.1 shows the equation of the line between the relationship of the NTU
Level and the Voltage levels in millivolts. The quadratic equation was used for
the calibration of the sensor.

Field Testing of Turbidity Sensor Accuracy Against

Turbidimeter Readings
To ensure the reliability of calibrated voltage levels and actual turbidity
readings, percent errors are calculated based on three readings for each
sample. This test assesses how closely the sensor outputs values within
the minimum and maximum allowable turbidity ranges, evaluating its
sensitivity under varying turbidity conditions which is shown by tables 4.8
& 4.9 below. Table 4.6 only differs with the addition of another turbidimeter
from the laboratories of the Zamboanga CIty Water District.

Table 4.8: Field testing of turbidity level: Eutech Instrument Turbidimeter vs TS_300B
NTU vs True NTU reading

Eutech
Instrumen Turbidity Arduino’s
Date Samples t Sensor True NTU Absolute
Turbidity TS- 300B Reading Percent
Meter NTU Error
(ADZU)
NTU

93
 1 22.5 Less than 7.21 67.96%
10

2 23.6 Less than 35.25 49.36%

10

June 27, 3 140 129.8 129.8 7.29%

2024
4 124 114.63 114.63 7.56%

5 138 139.45 139.45 1.05%

6 145 139.45 139.45 3.83%

June 28, 7 16.15 Less than 6.11 62.17%

2024 10

9 10.67 Less than 21.85 104.78%

10

10 4.18 Less than 12.14 190.43%

10

11 4.79 Less than 15.56 224.84%

10

12 6.87 Less than 13.08 90.39%

10

13 6.47 Less than 12.13 87.48%

10

14 12.39 Less than 25.71 107.51%

10

July 02, 15 4.14 Less than 11.12 168.60%

2024 10

16 5.73 Less than 12.92 125.48%

10

17 5.24 Less than 20.03 282.25%

10

18 5.64 Less than 21.12 274.47%

10

19 6.97 Less than 10.23 46.77%

10

94
 20 5.21 Less than 22.11 324.38%
10

July 03, 21 10.9 Less than 23.35 114.22%

2024 10

22 10.81 Less than 15.56 43.94%

10

23 10.79 Less than 20.17 86.93%

10

24 10.83 Less than 14.44 33.33%

10

25 12.45 Less than 15.42 23.86%

10

26 13.48 Less than 19.12 41.84%

10

27 10.33 Less than 18.05 74.73%

10

July 04, 28 10.81 Less than 22.68 109.81%

2024 10

29 10.79 Less than 18.92 75.35%

10

30 10.83 Less than 19.35 78.67%

10

31 12.45 Less than 17.24 38.47%

10

32 10.44 Less than 15.53 48.75%

10

33 10.05 Less than 13.87 38.01%

10

34 10.67 Less than 13.87 29.99%

10

Mean 92.86%
Absolute
Percent
Error:

95
 The percent errors vary significantly, with some readings having very high
errors (e.g., 224.84% for a true NTU of 15.56) and others having much
lower errors (e.g., 1.05% for a true NTU of 139.45). The average percent
error when NTU is greater than 100 is 4.93%, indicating good accuracy for
high turbidity levels. The total average percent error when NTU is less
than 30 is 104.99%, indicating poor accuracy for low turbidity levels. The
TS-300B sensor shows high accuracy for higher turbidity levels but
struggles with lower turbidity levels, resulting in significant errors. This
suggests the sensor is more reliable for applications where turbidity is
expected to be high.

Turbidimeter vs TS_300B NTU vs True NTU reading

Table 4.9: Field testing of turbidity level: Eutech Instrument Turbidimeter vs Hach 2100Q

Eutech Hach Turbidity Percent Percent

Instrument 2100Q Sensor Error Error
Date Sample Turbidimet Turbidime TBS- (Eutech (Hach
s er (ADZU) ter 300B Instrume 2100Q vs
NTU (ZCWD) NTU nt vs TS- TS-
NTU True 300B) 300B)
Reading
July 05, 1 6.67 6.94 12.81 92.05% 84.58%
2024
2 7.2 9.22 14.01 94.58% 51.95%
3 6.35 10.9 15.02 136.54% 37.80%
4 6.97 10.3 15.1 116.64% 46.60%
5 6.21 9.92 18.31 194.85% 84.58%
6 7.18 9.71 18.31 155.01% 88.57%
7 6.44 7.63 10.77 67.34% 41.15%
8 6.29 7.5 11.12 76.79% 48.27%
9 7.25 11.3 16.43 126.62% 45.40%
10 6.29 7.52 11.21 78.22% 49.07%
Averag 113.864 57.80%
e %
Percent
Error:

96
 The percent errors for the Eutech Instrument vs. TS-300B range from
67.34% to 194.85%, with an average percent error of 113.864%. The
percent errors for the Hach 2100Q vs. TS-300B range from 37.80% to
88.57%, with an average percent error of 57.80%. The Hach 2100Q
shows better agreement with the TS-300B sensor readings than the
Eutech Instrument, as indicated by the lower average percent error. The
TS-300B sensor still shows significant errors, suggesting the need for a
new turbidity sensor since its data sheet showed that the sensitivity of the
sensor ranges from 0-1000 NTU with +- 30 NTU which means that the
sensor's readings can be off by up to ± 30 NTU.

Standardized Atmospheric Pressure and Temperature

versus Atmospheric Atmospheric Pressure and
Temperature Sensor Data
During field testing, the BMP180 module's atmospheric pressure and
temperature readings will be compared with those from standard
barometers and thermometers. These comparisons will occur at fixed 15-
minute intervals to ensure continuous monitoring, as detailed in the table
below.
Table 4.10: Data gathering for Standardized tools and BMP108 module at
15 minutes interval
BMP1
BMP10 08
Baromet
8 Perce Thermomet Senso Perce
er
Date Time Sensor nt er Reading r nt
Reading
Readin Error (degC) Readi Error
(hPa)
g (hPa) ng
(degC)

1011 1011.6 -0.06% 27.5 27.7 -0.73%

09:00 3

1010.75 1011.6 -0.09% 26.5 27.4 -3.40%

09:15 1

09:30
1010.75 1011.7 -0.10% 27 27.8 -2.96%
5

97
09:45 1011 1011.9 -0.09% 27 27.1 -0.37%

1011 1011.9 -0.09% 26.25 26.5 -0.95%

10:00 1

1011 1011.7 -0.08% 26 26.1 -0.38%

10:15 7

1011 1011.7 -0.08% 25.5 25.9 -1.57%

10:30 9

1010.75 1011.7 -0.10% 25.25 25.8 -2.18%

10:45 9

1011 1011.9 -0.10% 25.25 25.6 -1.39%

11:00 8

1011 1011.7 -0.07% 25.25 25.8 -2.18%

11:15 3

1011 1011.6 -0.06% 25 25.7 -2.80%

11:30 4

1011 1011.5 -0.06% 25 25.7 -2.80%

11:45 7

1011 1011.5 -0.05% 25 25.4 -1.60%

12:00 1

12:15
1011 1011.3 -0.03% 25 25.6 -2.40%

98
 2

12:30 - - - - - -

1011 1010.9 0.01% 24.5 24.7 -0.82%

12:45 4

1011 1010.8 0.02% 24 24.3 -1.25%

13:00 4

1011 1010.7 0.02% 24 24 0.00%

13:15 7

1011 1010.8 0.02% 24 24 0.00%

13:30 1

1010.75 1010.4 0.03% 23.5 23.7 -0.85%

13:45 5

1010.75 1010.4 0.03% 23.5 23.7 -0.85%

14:00 3

14:15 1010.5 1010.4 0.01% 23.5 23.7 -0.85%

1010.5 1010.4 0.01% 24 24.3 -1.25%

14:30 4

14:45
1010.25 1010.3 -0.01% 24.75 24.2 2.22%
4

99
 1010.5 1010.4 0.01% 24 24 0.00%
15:00 2

1010.75 1010.4 0.03% 23.5 23.6 -0.43%

15:15 2

1010.5 1010.4 0.00% 23 23.1 -0.43%

15:30 5

1010.25 1010.5 -0.03% 23 22.9 0.43%

15:45 2

1010.75 1011.0 -0.03% 22.75 22.9 -0.66%

16:00 6

Through thorough inspection, the percent errors for atmospheric pressure

readings range from -0.10% to 0.03%, and the percent errors for
temperature readings range from -3.40% to 2.22%. The BMP180 sensor
demonstrates high reliability in measuring atmospheric pressure, as
indicated by the consistently low percent errors close to zero. This
reliability is crucial for applications requiring accurate pressure monitoring,
such as weather forecasting or altitude sensing.
However, while the BMP180 sensor generally provides accurate
temperature measurements, the wider range of percent errors suggests
variability. Negative errors (underestimation) up to -3.40% and positive
errors (overestimation) up to 2.22% indicate potential limitations in
temperature accuracy, especially in scenarios where precise temperature
control is critical. Therefore, an F-test was conducted to quantitatively
assess whether the variances of two samples are significantly different,
providing insights into the variability of the temperature readings. Table
4.12 shows the calculated F-statistics:
Table 4.11: F-statistics for temperature readings (standard vs. sensor)

F-statistics
Degrees of freedom (v1) 27
Degrees of freedom (v2) 27
f Statistic (f) 1.1228
Probability: P(F1.1228) 0.617 100
Probability: P(F1.1228) 0.383
As shown in Table 4.11 the degrees of freedom for the two samples are both 27, the calculated F-

Figures 4.2 and 4.3 visually illustrates the discrepancies between the
atmospheric pressure and temperature readings from each interval.

Figure 4.2: Error plot of Atmospheric pressure (Standardized vs BMP180)

The error plot shows minimal deviation between the two sets of readings,
with no significant outliers. The BMP180 sensor provides consistent and
accurate atmospheric pressure readings, as evidenced by the low error
values in the plot.

101
 Figure 4.3: Error plot of Temperature (Standardized vs BMP180)

Similar to the pressure error plot, the temperature error plot shows minor
discrepancies between the readings.The BMP180 sensor also performs
well in measuring temperature, with errors remaining small and within
acceptable limits.

Atmospheric Pressure and Temperature Difference of

Exposed Case Versus Enclosed Case
Studies have indicated that the internal atmospheric pressure and
temperature within the housing of the BMP180 module may impact the
accuracy of its readings. This test aims to assess this effect on the device
casing and determine if necessary modifications are needed for optimal
placement and display. Tables 4.12 and 4.13 below outline the data
collection format for atmospheric pressure and temperature readings,
along with reference data from standardized measurement tools for both
parameters.

102
Table 4.12: Data gathering for the temperature of the 3 devices: Sensor A, Sensor B,
and Reference device

TEMPERATURE
SETUP A SETUP B
REFERENCE
EXPOSED ENCLOSED EXPOSED ENCLOSED

26.5 23.1 22.8 26.9 23.5 27

25.1 23.4 23.8 26.2 23 25.25

24.3 22.4 22.7 25.4 22.5 24

23.7 21.9 22.4 24.8 22.5 23.5

22.8 23.2 23.3 24.5 23.5 22.75

22.7 22.9 22.6 24 23 22.5

21.1 22.7 22.7 23.7 23 22

20.9 22.6 23 23.1 23.5 22

20.7 23.3 22.8 22.7 22.5 21.75

20.5 22.2 22.8 22 22.75 22

22.1 22.8 22.5

22.3 22.8 23

22.4 22.9 24

103
 Due to the varying and limited number of samples in each setup, a two-
tailed t-test was employed to assess if there is a significant difference
between the means of the temperature readings of the two setups. These
t-scores were then converted into p-values and compared to the
significance level of 0.05. Table 4.12 shows the corresponding p-values of
comparing the temperature readings between setups under different
conditions:

Table 4.13: Comparison of Temperature Readings

Comparison P-value

Setup A (Exposed) vs. Reference 0.0595

Setup B (Exposed) vs. Reference 0.3163

Setup A (Exposed) vs. Setup A 0.7968

(Enclosed)

Setup B (Exposed) vs. Setup B 0.0160

(Enclosed)

As shown in Table 4.13, comparing Setup A (Exposed) and Setup B (Exposed)

to the reference readings yields p-values of 0.0595 and 0.3163, respectively.
Since these p-values (0.0595 > 0.05 and 0.3163 > 0.05), we conclude that there
are no significant differences between the temperature readings of the sensors in
both setups compared to the standardized temperature meter. It is noteworthy
that the p-value of Setup A (0.0595) is relatively close to the significance level of
0.05, indicating a marginal difference that may warrant further investigation for
reliability in generating conclusions.
Also shown in Table 4.14 are the comparisons of both setups in different
conditions (exposed vs. enclosed). comparing Setup A (Exposed) and Setup B
(Exposed) to Setup A (Enclosed) and Setup B (Enclosed), respectively, yields p-
values of 0.7968 and 0.0160, respectively. Setup B shows a significant difference
(p-value = 0.0160) between exposed and enclosed conditions, indicating that
temperature readings differ significantly between these setups.

104
Table 4.14: Data gathering for the air pressure of the 3 devices: Sensor A, Sensor B,
and Reference device

PRESSURE
SETUP A SETUP B
REFERENCE
EXPOSED ENCLOSED EXPOSED ENCLOSED

1009.14 1009.22 1010.45 1010.58 1009.25 1010

1008.84 1009.29 1010.37 1010.07 1009 1009

1008.94 1009.34 1010.55 1010.06 1009.25 1009

1008.94 1009.35 1010.68 1010.04 1009.25 1008.5

1008.83 1009.36 1010.63 1009.93 1009.5 1008.25

1008.68 1009.15 1010.47 1009.73 1009.5 1008.25

1008.37 1009.25 1010.51 1009.66 1009.5 1008.25

1008.49 1009.14 1010.49 1009.76 1009 1008.25

1008.46 1010.18 1010.15 1009.68 1009.5 1008.25

1008.59 1008.59 1009.81 1010.03 1009 1008.25

1008.37 1009.65 1009

1008.48 1009.82 1009.25

1008.52 1009.83 1009.5

Similarly, a two-tailed t-test was employed to assess if there is a significant

difference between the means of the pressure readings of the two setups. Table

105
4.15 shows the corresponding p-values of comparing the temperature readings
between setups under different conditions:

Table 4.15: Comparison of Pressure Readings

Comparison P-value

Setup A (Exposed) vs. Reference 0.3551

Setup B (Exposed) vs. Reference 4.7945E-07

Setup A (Exposed) vs. Setup A 0.0320

(Enclosed)

Setup B (Exposed) vs. Setup B 0.0303

(Enclosed)

As shown in Table 4.15, comparing Setup A (Exposed) and Setup B (Exposed)

to the reference readings yields p-values of 0.3551 and 4.7945E-07, respectively.
With a p-value of 0.3551 (> 0.05), there is no significant difference between the
temperature readings of Setup A and the reference. However, with a very low p-
value of 4.7945E-07 (< 0.05), there is a significant difference between the
temperature readings of Setup B and the reference which must have something
to do with the ambient temperature surrounding the sensor upon testing.
Also shown in Table 4.15 are the comparisons of both setups in different
conditions (exposed vs. enclosed). comparing Setup A (Exposed) and Setup B
(Exposed) to Setup A (Enclosed) and Setup B (Enclosed), respectively, yields p-
values of 0.0320 and 0.0303, respectively. With a p-value of 0.0320 (< 0.05),
there is a significant difference between the pressure readings of Setup A in
exposed and enclosed conditions, and with a p-value of 0.0303 (< 0.05), there is
a significant difference between the pressure readings of Setup B in exposed and
enclosed conditions.

Data Storage and Logging Test

The data storage and logging test demonstrated an acceptability rate of
90%, indicating that the RTC and SD card module perform their functions
satisfactorily. Success was determined by the proper functioning of each
module and interface within the system.

During a 5-day monitoring period, a total of 3194 data groups were stored
on the SD card. Careful inspection revealed no gaps between recorded
data, demonstrating continuous operation of the RTC module throughout

106
 the monitoring period. The data storage and logging module was
programmed to record data every 2.5 minutes to ensure reliability and
accuracy. To verify this, 10 randomly selected datasets were compared
with corresponding data saved simultaneously shown in table 4.17. This
comparison utilized the GSM module as a secondary receiver to track
data packet transmission rates, confirming the satisfactory performance of
the data storage and logging modules.

Table 4.16: Comparison of Stored data vs Expected data

DATE CASE STORED DATA EXPECTED OUTPUT

NO.

JULY 10 ,2024 1 7/10/2024 7/10/2024 10:12:21,31.30,

10:12:21,31.30, 87.00, 87.00, 191.00 , 21
191.00 , 21

2 7/10/2024 7/10/2024 19:8:10,30.20,

19:8:10,30.20, 93.00, 93.00, 233.00, 21
233.00, 21

3 7/10/2024 7/10/2024 19:30:42,30.20,

19:30:42,30.20, 94.00, 94.00, 225.00, 20
225.00, 20

4 7/10/2024 7/10/2024 20:3:15,30.20,

20:3:15,30.20, 92.00, 92.00, 217.00 , 22
217.00 , 22

5 7/10/2024 7/10/2024 20:33:18,30.20,

20:33:18,30.20, 93.00, 93.00, 222.00, 21
222.00, 21

6 7/10/2024 7/10/2024 21:0:51,30.20,

21:0:51,30.20, 92.00, 92.00, 214.00, 21
214.00, 21

7 7/10/2024 7/10/2024 22:20:58,29.80,

22:20:58,29.80, 93.00, 93.00, 199.00, 22
199.00, 22

8 7/10/2024 7/10/2024 22:56:1,29.80,

22:56:1,29.80, 94.00, 94.00, 213.00, 20
213.00, 20

9 7/10/2024 7/10/2024 23:38:35,29.80,

23:38:35,29.80, 94.00, 94.00, 211.00, 21

107
 211.00, 21

10 7/10/2024 7/10/2024 23:56:7,29.30,

23:56:7,29.30, 93.00, 93.00, 212.00, 21
212.00, 21

JULY 11, 2024 1 7/10/2024 3:15:46,28.90 7/10/2024 3:15:46,28.90 ,

, 93.00 , 213.00 , 22 93.00 , 213.00 , 22

2 7/10/2024 7:32:23,28.90 7/10/2024 7:32:23,28.90 ,

, 93.00 , 226.00 , 20 93.00 , 226.00 , 20

3 7/10/2024 7/10/2024 13:20:46,28.90 ,

13:20:46,28.90 , 93.00 , 93.00 , 228.00 , 21
228.00 , 21

4 7/10/2024 7/10/2024 15:23:16,28.90 ,

15:23:16,28.90 , 93.00 , 93.00 , 212.00 , 22
212.00 , 22

5 7/10/2024 7/10/2024 18:25:47,28.90 ,

18:25:47,28.90 , 93.00 , 93.00 , 228.00 , 21
228.00 , 21

6 7/10/2024 7/10/2024 19:28:17,28.90 ,

19:28:17,28.90 , 93.00 , 93.00 , 228.00 , 20
228.00 , 20

7 7/10/2024 7/10/2024 21:30:47,28.90 ,

21:30:47,28.90 , 93.00 , 93.00 , 222.00 , 22
222.00 , 22

8 7/10/2024 7/10/2024 22:33:17,28.90 ,

22:33:17,28.90 , 93.00 , 93.00 , 229.00 , 21
229.00 , 21

9 7/10/2024 7/10/2024 23:35:47,28.90 ,

23:35:47,28.90 , 93.00 , 93.00 , 213.00 , 22
213.00 , 22

10 7/10/2024 7/10/2024 23:38:18,28.90 ,

23:38:18,28.90 , 93.00 , 93.00 , 229.00 , 20
229.00 , 20

JULY 12, 2024 1 12/07/2024 12/07/2024

06:29:17,28.50, 98.00, 06:29:17,28.50, 98.00,

108
 45.00 ,20 45.00 ,20

2 12/07/2024 12/07/2024
06:31:47,28.50, 98.00, 06:31:47,28.50, 98.00,
49.00 ,21 49.00 ,21

3 12/07/2024 12/07/2024
15:09:10,28.00, 98.00, 15:09:10,28.00, 98.00,
96.00 ,21 96.00 ,21

4 12/07/2024 12/07/2024
17:31:053,28.00, 98.00, 17:31:053,28.00, 98.00,
92.00 ,22 92.00 ,22

5 12/07/2024 12/07/2024
18:49:030,28.50, 98.00, 18:49:030,28.50, 98.00,
98.00 ,21 98.00 ,21

6 12/07/2024 12/07/2024
19:12:02,28.90, 98.00, 19:12:02,28.90, 98.00,
101.00 ,21 101.00 ,21

7 12/07/2024 12/07/2024
19:14:033,28.90, 98.00, 19:14:033,28.90, 98.00,
98.00 ,21 98.00 ,21

8 12/07/2024 12/07/2024
23:29:056,28.50, 98.00, 23:29:056,28.50, 98.00,
98.00 ,21 98.00 ,21

9 12/07/2024 12/07/2024
23:54:058,28.00, 98.00, 23:54:058,28.00, 98.00,
100.0020 100.0020

10 12/07/2024 12/07/2024
23:59:059,28.00, 98.00, 23:59:059,28.00, 98.00,
100.0022 100.0022

109
JULY 13, 2024 1 7/13/2024
05:37:59,27.60, 98.00, 7/13/2024 05:37:59,27.60,
105.00 ,21 98.00, 105.00 ,21

2 7/13/2024
08:10:43,28.00, 98.00, 7/13/2024 08:10:43,28.00,
98.00 ,20 98.00, 98.00 ,20

3 7/13/2024
09:33:21,28.50, 98.00, 7/13/2024 09:33:21,28.50,
105.00 ,22 98.00, 105.00 ,22

4 7/13/2024 7/13/2024
10:48:028,28.50, 98.00, 10:48:028,28.50, 98.00,
101.0022 101.0022

5 7/13/2024 7/13/2024 12:26:06,28.90,

12:26:06,28.90, 98.00, 98.00, 95.00 ,22
95.00 ,22

6 7/13/2024 7/13/2024
14:51:020,30.20, 95.00, 14:51:020,30.20, 95.00,
97.00 ,20 97.00 ,20

7 7/13/2024 7/13/2024
16:46:030,29.80, 92.00, 16:46:030,29.80, 92.00,
130.0022 130.0022

8 7/13/2024 7/13/2024
19:16:044,29.30, 96.00, 19:16:044,29.30, 96.00,
93.00 ,22 93.00 ,22

110
 9 7/13/2024 7/13/2024
20:29:021,28.90, 98.00, 20:29:021,28.90, 98.00,
87.00 ,20 87.00 ,20

10 7/13/2024 7/13/2024 22:57:04,28.00,

22:57:04,28.00, 98.00, 98.00, 65.00 ,21
65.00 ,21

JULY 14, 2024 – 1 7/14/2024 7/14/2024 02:12:22,28.00,

JULY 15 ,2024 02:12:22,28.00, 98.00, 98.00, 61.00 ,22
61.00 ,22

2 7/14/2024 7/14/2024 06:40:16,27.10,

06:40:16,27.10, 98.00, 98.00, 57.00 ,21
57.00 ,21

3 7/14/2024 7/14/2024
10:40:037,28.00, 98.00, 10:40:037,28.00, 98.00,
62.00 ,22 62.00 ,22

4 7/14/2024 7/14/2024
14:39:017,29.80, 95.00, 14:39:017,29.80, 95.00,
95.00 ,22 95.00 ,22

5 7/14/2024 7/14/2024 18:12:06,29.30,

18:12:06,29.30, 98.00, 98.00, 59.00 ,20
59.00 ,20

6 7/14/2024 7/14/2024
22:19:059,28.90, 98.00, 22:19:059,28.90, 98.00,
83.00 ,20 83.00 ,20

111
 7 7/14/2024 7/14/2024
23:12:034,28.90, 98.00, 23:12:034,28.90, 98.00,
54.00 ,22 54.00 ,22

8 7/14/2024 7/14/2024
23:47:037,28.50, 98.00, 23:47:037,28.50, 98.00,
44.00 ,21 44.00 ,21

9 7/15/2024 7/15/2024 00:35:11,29.30,

00:35:11,29.30, 98.00, 98.00, 71.00 ,20
71.00 ,20

10 7/15/2024 7/15/2024 01:25:16,28.90,

01:25:16,28.90, 98.00, 98.00, 73.00 ,22
73.00 ,22

The recorded data showed no gaps, indicating that the RTC module
functioned continuously and reliably throughout the monitoring period. A
sample of 10 randomly selected datasets was compared with
corresponding data saved simultaneously by a secondary GSM module
receiver. This comparison verified that the data storage and logging
modules performed as expected, accurately capturing and recording the
data.The stored files followed a consistent pattern, recording the date,
time, temperature, humidity, analog input, and distance provided by
dummy sensor data modules. This format facilitated easy verification and
comparison of stored data. Therefore the integrated modules fit the
project’s specifications for data storage and logging.

112
 CHAPTER V
CONCLUSION AND RECOMMENDATION

This study aimed to evaluate the reliability and accuracy of various sensors and
modules used for environmental monitoring, specifically focusing on ultrasonic
sensors, turbidity sensors, atmospheric pressure sensors, and data storage and
logging systems. The results of the tests conducted provide significant insights
into the performance and applicability of these sensors in real-world scenarios.
The ultrasonic sensor demonstrated high accuracy and precision in measuring
distances. Sensitivity tests showed that the sensor could measure distances from
1 meter to 6 meters with minimal deviation. For instance, at 1 meter, the mean
reading was approximately 100.4 cm with a standard deviation of 0.34 cm. Field
testing further validated the sensor's performance, with most errors falling within
±1%, demonstrating the sensor's reliability for practical applications requiring
precise distance measurements.

113
The turbidity sensor's performance was less satisfactory compared to the
ultrasonic sensor. The sensor showed high accuracy for measurements above
100 NTU but struggled with lower values, exhibiting an average percent error of
94.69% among 34 water samples. The comparison with the Hach 2100Q
turbidimeter further confirmed its limited accuracy at lower turbidity levels. These
findings indicate that the turbidity sensor is suitable for high-turbidity
environments but is unreliable for low-turbidity measurements. Atmospheric
pressure sensor tests included comparisons of standardized atmospheric
pressure and temperature readings with those from the sensor in both enclosed
and open cases. The sensor data showed consistent performance, aligning
closely with the standardized values, indicating the sensor's reliability in varying
environmental conditions.

The data storage and logging tests involved the RTC DS1302 module integrated
with a microSD card. Over the monitoring period, 10 randomly selected data sets
were reviewed, and all data points were accurate without errors. The system
proved to be reliable for long-term data logging and storage, essential for
continuous environmental monitoring. The study's findings have several
important implications for using sensors in environmental monitoring. The
ultrasonic sensor's high accuracy makes it ideal for applications requiring precise
distance measurements, such as water level monitoring in reservoirs or flood-
prone areas. The turbidity sensor's limited accuracy at low NTU values suggests
it should be used primarily in environments where high turbidity is expected, such
as industrial discharge monitoring or sediment-laden water bodies. The
atmospheric pressure sensor's consistent performance indicates its utility in
diverse environmental conditions, supporting its use in weather stations or
climate research. The reliability of the RTC DS1302 and microSD card
combination ensures that long-term environmental data can be accurately
recorded and stored, which is crucial for trend analysis and historical data
comparison.
In conclusion, the study successfully evaluated the reliability and accuracy of
different sensors used for environmental monitoring, providing valuable insights
into their performance in real-world applications. The findings underscore the
importance of selecting appropriate sensors based on specific use cases and
environmental conditions. By addressing the limitations identified, particularly
with the turbidity sensor, and exploring future research avenues, the utility of
these sensors can be further enhanced, contributing to more effective and
reliable environmental monitoring systems.

RECOMMENDATIONS

114
Based on the identified gaps and limitations in the reviewed studies, the following
recommendations are proposed to enhance the accuracy, reliability, and
efficiency of sensor-based environmental monitoring systems.
1. Enhanced Calibration of Turbidity Sensors: Investigate advanced
calibration techniques to improve sensor performance at lower NTU
levels, potentially incorporating machine learning algorithms to adjust
sensor readings dynamically based on environmental conditions.
Research and design efforts should focus on creating new turbidity sensor
models that use novel materials or technologies to achieve higher
precision.

2. Look for a Better Turbidity Sensor: Conduct a comparative analysis of

the TS-300B with newer technologies to identify potential improvements in
accuracy, reliability, and cost-effectiveness. Evaluate alternative turbidity
sensors to find replacements that offer enhanced performance in varying
environmental conditions.

3. Extended Field Testing: Implement long-term field testing of all sensors

in diverse environmental settings to validate their performance and
reliability over extended periods. Document and analyze data to identify
any performance degradation and develop strategies to mitigate such
issues.

4. A01NYUB Ultrasonic Sensor Recommendations: Investigate the

sensor’s performance in different weather conditions and develop
protective enclosures to shield it from harsh environments. Implement
software filters to minimize noise and enhance the accuracy of water level
measurements.

5. TS-300B Turbidity Sensor Recommendations: Regularly calibrate the

sensor to maintain accuracy and account for environmental variations.
Develop protocols for routine maintenance and calibration to ensure
consistent performance. Explore and test newer turbidity sensor models to
find potential replacements that offer higher precision and reliability.

6. BMP180 Atmospheric Pressure Sensor Recommendations: Analyze

wind patterns and their impact on pressure readings, utilizing wind shields
or baffles to minimize direct wind influence on exposed sensors. Regularly
calibrate sensors to adjust for varying wind conditions and ensure
accurate pressure readings. Adjust sensor readings based on known

115
 altitude differences to maintain accuracy and document precise sensor
placements to correlate altitude differences with pressure variations.
Choose enclosure materials with minimal thermal expansion to prevent
deformation affecting sensor calibration and optimize ventilation to
balance environmental isolation with sufficient air exchange. Establish
clear guidelines for sensor installation and define enclosure specifications
to minimize variability in readings across different deployments. Use
materials with proven stability in temperature and pressure environments
and validate enclosure designs through testing to confirm their
effectiveness in buffering external pressure and temperature changes.

7. Integration of Sleep Mode and Watchdog Timer for Microcontrollers:

Implement sleep modes to reduce power consumption during idle periods,
extending operational life in remote deployments. Configure sensors to
activate quickly from sleep mode in response to predefined triggers such
as scheduled readings or sensor events. Incorporate watchdog timers to
monitor microcontroller activity and reset the system if abnormal behavior
such as program freeze is detected, ensuring continuous data collection
without manual intervention. Set timeout periods suitable for expected
sensor response times to prevent false resets and ensure reliable system
operation.

REFERENCES
[1] M. Ali, W. Wang, and H. Khan, "Wireless communication for IoT:
Technologies and applications," International Journal of Computer Applications,
vol. 162, no. 10, pp. 1-10, 2017.

[2] S. Chandra, P. Gupta, and A. Verma, "Challenges in manual water level

monitoring: A case study," Journal of Water Resources, vol. 5, no. 2, pp. 85-92,
2016.

[3] Y. Chen, H. Wu, and S. Zhang, "Ultrasonic sensors for water level monitoring:
A review," Sensors, vol. 20, no. 12, p. 3452, 2020.

116
[4] R. Gupta and S. Roy, "Long-range wireless communication technologies for
flood monitoring," Journal of Telecommunications, vol. 8, no. 1, pp. 34-40, 2020.

[5] T. Huang, C. Lin, and L. Chen, "Application of turbidity sensors in water

quality monitoring," Environmental Monitoring and Assessment, vol. 191, no. 8, p.
482, 2019.

[6] V. Kumar and R. Singh, "Manual vs. automated water level monitoring
systems," Journal of Hydrology, vol. 3, no. 4, pp. 199-207, 2017.

[7] J. Lee, S. Park, and H. Kim, "Advances in turbidity sensors for water quality
monitoring," Water Science and Technology, vol. 83, no. 4, pp. 945-953, 2021.

[8] A. Patel, R. Joshi, and P. Shah, "Data logging systems in water level
monitoring: A comparative study," International Journal of Water Resources and
Environmental Engineering, vol. 11, no. 3, pp. 61-68, 2019.

[9] D. Silva, J. Garcia, and F. Martinez, "Ensuring data integrity in remote sensing
applications," *Remote Sensing*, vol. 13, no. 9, p. 1876, 2021.

[10] A. Smith and T. Brown, "Cost analysis of traditional water level monitoring
methods," *Water Economics*, vol. 9, no. 1, pp. 22-30, 2018.

[11] X. Wang, Z. Li, and Y. Zhou, "Atmospheric pressure sensors for flood
monitoring: Benefits and applications," *Sensors and Actuators A: Physical*, vol.
281, pp. 128-136, 2018.

[12] L. Wu, J. Zhang, and Q. Liu, "Enhancing flood prediction through automated
monitoring," Hydrological Processes, vol. 33, no. 7, pp. 1024-1035, 2019.

[13] Y. Zhang and J. Feng, "Atmospheric pressure sensors: An overview of

applications in environmental monitoring," Measurement, vol. 158, p. 107678,
2020.

[14] X. Zhao and M. Li, "The role of ultrasonic sensors in environmental

monitoring," Journal of Environmental Management, vol. 287, p. 112271, 2021.

[15] S. Subair and L. Abraham, "Intelligent Pressure Measuring System," 2014.

[16] S. Gaikwad, R. Sande, S. Bochare, D. Chalkapure, and S. Jawale, "IOT

based Weather Sensing and Monitoring Station," SSRN Electronic Journal, 2023.

[17] M. Vasylenko and V. Dzhus, "Barometric Altimeter Based on

Microelectromechanical Sensor," Electronics and Control Systems, 2022.

[18] H. A. Kusuma, D. Oktavia, S. Nugaraha, T. Suhendra, and S. Refly, "Sensor

BMP280 statistical analysis for barometric pressure acquisition," IOP Conference
Series: Earth and Environmental Science, vol. 1148, 2023.

117
[19] M. Khaery, A. H. Pratama, P. Wipradnyana, and A. A. Gunawan, "Design of
air pressure measuring devices using a barometric pressure 280 (BMP280)
sensor based on Arduino Uno," 2020.

[20] W. Cheng, "Design of a portable altitude measurement system based on

BMP085," Transducer and Microsystem Technologies, 2011.

[21] P. Lall, A. S. Abrol, D. Locker, and B. Hughes, "Damage evolution in MEMS

pressure sensors during high temperature operating life and prolonged storage at
sub-zero temperature," in 2018 17th IEEE Intersociety Conference on Thermal
and Thermomechanical Phenomena in Electronic Systems (ITherm), 2018, pp.
1050-1061.

[22] B. Linke, "Overview of 1-Wire technology and its use," 2003.

[23] G. J. Pottie and W. J. Kaiser, "Wireless integrated network sensors," Ad Hoc

Networks, vol. 1, no. 3, pp. 351-367, Sept. 2003.

[24] M. Hall, J. West, B. Sherman, J. Lane, and D. De Haas, "Hall et al 2011

es103939a si 001," Jul. 2014. [Dataset].

[25] T. De Groeve, "Space-based river monitoring for global flood early warning,"
Geomatics, Natural Hazards and Risk, vol. 1, no. 1, pp. 19-35, Mar. 2010.
Available: Taylor & Francis.
[26] J. E. Antonio-Lopez, J. J. Sanchez-Mondragon, P. LiKamWa, and D. A. May-
Arrioja, "Fiber-optic sensor for liquid level measurement," *Optics Letters*, vol.
36, no. 17, pp. 3425-3427, Sept. 2011.
[27] E. J. Band and F. I. Anyasi, "Design of an automatic water level controller
using mercury float switch," IOSR J. Electronics Commun. Eng., vol. 9, no. 2, pp.
16-21, 2014.
[28] M. S. M. Sabre, S. S. Abdullah, and A. Faruq, "Flood warning and monitoring
system utilizing internet of things technology," Kinetik: Game Technol., Inf. Syst.,
Comput. Network, Comput., Electron., Control, vol. 3, no. 3, pp. 287-296, 2019.
[29] B. N. Getu, M. Abdulkadir, and M. Tous, "Remote control of garden
plantation water pumps using Arduino and GSM mobile," Adv. Sci., Technol.
Eng. Syst. J., vol. 5, no. 4, pp. 499-504, 2020.
[30] G. Sai Susheel and B. Gururaj, "A study on employee attraction and
retention strategies with reference to Infosoft HR Services, Hyderabad," Proc.
IEEE Conf. [Conference Name], Hyderabad, India, 2019,pp. 340-343.
[31] R. Yelekar et al., "IoT-based smart water level monitoring," in Proc. 2023
IEEE 20th India Council International Conference (INDICON), 2023.
[32] DFRobot, "A01NYUB Waterproof Ultrasonic Sensor SKU: SEN0313,"
[Online].

118
[33] A. M. O. K. Ebrahim, M. A. K. L. E. Khalil, and M. S. M. Al-Sulaiman, "A
Compact High-Sensitivity Capacitive Pressure Sensor for Low-Pressure
Applications," *IEEE Sensors Journal*, vol. 18, no. 8, pp. 3240-3246, Aug. 2018.
doi: 10.1109/JSEN.2018.2826778.
[34] K. S. Kim, C. H. Ahn, S. J. Kim, and J. H. Kim, "Design and Implementation
of a Low-Cost Wireless Sensor Network for Smart Home Applications," Sensors,
vol. 19, no. 14, pp. 3039, Jul. 2019. doi: 10.3390/s19143039.
[35] S. S. Kumar, A. B. Prasad, and R. C. Rao, "A Novel Approach to Water
Quality Monitoring Using IoT," in Proc. 2016 IEEE 2nd International Conference
on Advances in Computing, Communications and Informatics (ICACCI), Jaipur,
India, 2016, pp. 242-247. doi: 10.1109/I2CACIS.2016.7885315.
[36] T. P. Lambrou, C. C. Anastasiou, C. G. Panayiotou, and M. M. Polycarpou,
"A low-cost sensor network for real-time monitoring and contamination detection
in drinking water distribution systems," IEEE Sensors Journal, vol. 14, no. 8, pp.
2765-2772, Aug. 2014.
[37] J. Droujko, P. Molnar, "Open-source, low-cost, in-situ turbidity sensor for
river network monitoring," Scientific Reports, vol. 12, no. 1, pp. 12345, Jul. 2022.
doi: 10.1038/s41598-022-14228-4.
[38] P. Molnár, J. Droujko, Marius G. Floriancic, "A Low-Cost Turbidity Sensor for
Deployment in Rivers," presented at EGU General Assembly 2021, Online, 2021.
[Online].
[39] A. Rymszewicz, John O'Sullivan, M. Bruen, J. Turner, Damian Lawler, E.
Conroy et al, "Measurement differences between turbidity instruments, and their
implications for suspended sediment concentration and load calculations: A
sensor inter-comparison study," Journal of Environmental Management, vol. 203,
pp. 947-957, Oct. 2017. doi: 10.1016/j.jenvman.2017.05.017.
[40] DFRobot, "Turbidity Sensor SKU: SEN0189," [Online].
[41] G. Wallis and D. I. Pomerantz, "Field assisted glass‐metal sealing," J. Appl.
Phys., vol. 40, no. 10, pp. 3946-3949, Oct. 1969.
[42] A. Cozma and B. Puers, "Characterization of the electrostatic bonding of
silicon and Pyrex glass," J. Micromech. Microeng., vol. 5, no. 2, pp. 98, Jun.
1995.
[43] H. Chau, K. Wise, "Noise due to Brownian motion in ultrasensitive solid-state
pressure sensors," IEEE Transactions on Electron Devices, vol. 34, no. 9, pp.
1234-1239, Sep. 1987. doi: 10.1109/T-ED.1987.23007.
[44] W. P. Eaton and J. H. Smith, "Micromachined pressure sensors: review and
recent developments," Smart Mater. Struct., vol. 6, no. 5, pp. 530, Oct. 1997.
[45] B. W. Spencer, "Statistical investigation of turbulent velocity and pressure
fields in a two-stream mixing layer," Ph.D. dissertation, Dept. of Aerospace Eng.,
University of Illinois at Urbana-Champaign, Urbana, IL, 1970.

119
[46] S. Bande and V. V. Shete, "Smart flood disaster prediction system using IoT
& neural networks," in Proc. 2017 International Conference on Smart
Technologies for Smart Nation (SmartTechCon), Bangalore, India, Aug. 2017,
pp. 189-194. IEEE.
[47] A. O’Connor, G. Tsafnat, Stephen B. Gilbert, Kristina Thayer, M. Wolfe,
"Moving toward the automation of the systematic review process: a summary of
discussions at the second meeting of International Collaboration for the
Automation of Systematic Reviews (ICASR)," Systematic Reviews, vol. 6, no. 1,
pp. 1-12, Dec. 2017. doi: 10.1186/s13643-017-0667-4.
[48] Gabriela C. Nunez‐Mir, B. Iannone, B. Pijanowski, N. Kong, S. Fei,
"Automated content analysis: addressing the big literature challenge in ecology
and evolution," Methods in Ecology and Evolution, vol. 7, no. 4, pp. 1-12, Apr.
2016. doi: 10.1111/2041-210X.12602.
[49] S. Pasika and S. Gandla, "Review on Internet of Things Based Water Level
Monitoring System," Materials Today: Proceedings, vol. 33, no. 1, pp. 1172-1174,
2020.

[50] S. Chandra, R. Kumar, and P. Singh, "Challenges in Water Level Monitoring

Systems: A Review," International Journal of Advanced Research in Engineering
and Technology, vol. 7, no. 4, pp. 89-95, 2016.

[51] J. Smith and L. Brown, "Advancements in Flood Monitoring Technologies,"

Journal of Hydrology, vol. 34, no. 2, pp. 213-219, 2018.

[52] J. Montaño and R. Gonzales, "Evaluation of Flood Monitoring Systems in the

Philippines," Journal of Disaster Risk Reduction, vol. 45, no. 3, pp. 104-112,
2018.

[53] R. W. Schweitzer, et al., "Water Quality Assessment Using Sensor

Technologies," Journal of Environmental Monitoring, vol. 22, no. 4, pp. 307-315,
2020.

[54] M. Goda, "Mechanisms of Water Quality Classification for Human

Consumption," Water Research, vol. 24, no. 2, pp. 189-199, 1988.

[55] S. Dube, et al., "Innovations in Water Quality Monitoring Systems,"

Environmental Science & Technology, vol. 49, no. 11, pp. 646-657, 2015.

[56] C. Tortajada, "The Role of Technology in Water Resource Management,"

International Journal of Water Resources Development, vol. 22, no. 1, pp. 205-
218, 2006.

120
[57] A. Al-Omari, et al., "Evaluation of Water Monitoring Technologies for
Ecosystem Preservation," Journal of Hydrology, vol. 38, no. 3, pp. 123-135,
2008.

[58] A. Djalilov, et al., "Water as an Indispensable Resource: A Comprehensive

Review," Water Policy, vol. 13, no. 1, pp. 347-356, 2011.

[59] A. Kumar and R. Singh, "Challenges in Traditional Water Level Monitoring

Systems," Journal of Hydrological Sciences, vol. 45, no. 2, pp. 150-160, 2017.

[60] J. Wu, et al., "The Impact of Sensor Technology on Flood Risk

Management," International Journal of Environmental Research and Public
Health, vol. 16, no. 4, pp. 450-460, 2019.

[61] D. Boon and B. Brunbaker, "High-Reflectivity Microwave Sensors for Water-

Level Monitoring," Journal of Hydrology, vol. 34, no. 2, pp. 213-219, 2008.

[62] X. Wang, Z. Li, and Y. Zhou, "Atmospheric pressure sensors for flood
monitoring: Benefits and applications," Sensors and Actuators A: Physical, vol.
281, pp. 128-136, 2018.

[63] P. Johnson and T. Smith, "Evaluating the Performance of Low-Cost Turbidity

Sensors for Flood Prediction," Journal of Environmental Monitoring, vol. 22, no.
4, pp. 307-315, 2020.

[64] R. Suski, J. Brown, and L. Green, "Types of Pressure Transducers for

Accurate Gas Pressure Measurements," Journal of Pressure Sensor Technology,
vol. 8, no. 2, pp. 145-156, 2003.

[65] T. Rutsch, M. S. McMillan, and L. Williams, "Environmental Measurements

for Simulations Based on Speed of Sound and Atmospheric Attenuation,"
Environmental Science & Technology, vol. 53, no. 4, pp. 2345-2356, 2019.

[66] M. Hassen, "Atmospheric Pressure and Humidity Effects on Sound Speed,"

Journal of Acoustical Society of America, vol. 101, no. 5, pp. 2032-2040, 1997.

[67] A. M. Sabatini, "Application of Environmental Measurements in Acoustic

Simulations," Journal of Sound and Vibration, vol. 187, no. 1, pp. 55-66, 1995.
[68] DFRobot, "A01NYUB Waterproof Ultrasonic Sensor SKU: SEN0313,"
[Online].

[69] X. Wang, Z. Li, and Y. Zhou, "Atmospheric pressure sensors for flood
monitoring: Benefits and applications," Sensors and Actuators A: Physical, vol.
281, pp. 128-136, 2018.
[70] D. Suski, A. M. Sabatini, and H. Hassen, "Environmental measurements for
validating simulations of sound speed and atmospheric attenuation," Journal of
Environmental Monitoring, vol. 5, no. 2, pp. 85-92, 2003.

121
122
 APPENDIX A
SOFTWARE REQUIREMENTS

Arduino IDE

EASY EDA

123
 APPENDIX B
HARDWARE REQUIREMENTS

Microcontrollers
Arduino Uno R3

Arduino Nano

124
Sensors
125
A01NYUB Ultrasonic Sensor

JSN SR40T Ultrasonic Sensor

126
TS-300B Turbidity Sensor

127
Thermo Scientific Eutech TN-100 Waterproof Turbidimeter

BMP180 Pressure Sensor

128
Data
Logging
Modules

SD Card
Module

129
RTC Module

130
APPENDIX C
APPENDIX C

131
ARDUINO CODES

132