Electronic Implementation of Sun Algorithm
Electronic Implementation of Sun Algorithm
Electronic Implementation of Sun Algorithm
A bstract
Implementation of Solar Position Algorithm (SPA) on an electronic platform is presented.
The system consists of a microcontroller and a real time clock. Solar angles according to
the SPA were calculated on a computer for 5 years (for each minute). The calculated solar
angles indexed according to the time and downloaded to secure digital (SD) through the
microcontroller based SD card read/write system. Solar angles for any given time can be
retrieved by providing the time index to the look up table recorded in the SD card. 5 MB
of space is required to store such a data set for one year. The performance of this system
was evaluated by interfacing to a manual sun tracking system of a parabolic trough solar
concentrator. It was found to be that the maximum variation of the focal line is 1.5 cm and
it is within the full acceptance angle of 3.8.
1. Introduction
Number of methods are being used to track the path of the sun for maximizing the
capturing of energy in most of the solar energy harnessing systems. Depending on the
technique used to find the position of the sun, solar trackers can be classified as active or
passive. The active trackers sense the solar radiation to locate the sun while the passive
trackers calculate solar angles relative to the earth using mathematical algorithms. Of the
two techniques, the passive solar angle calculation is the more reliable and effective
method since there is no effect for the calculation from environmental factors such as
cloud cover. There are different solar positioning algorithms that have been developed,
and the Solar Position Algorithm (SPA) for solar radiation applications developed at
National Renewable Energy Laboratory (NREL) is the most accurate SPA to the date. The
objective of this study is to implement the SPA on a microcontroller based electronic
system as a sun tracking system for a parabolic trough solar thermal energy concentrator
being developed at the Department of Physics, University of Colombo.
298
2. Methodology
The main controller of the tracking system is a PIC18F452 microcontroller. A Maxim
DS1305 Real Time Clock (RTC) and a SD memory card read/write system are connected
to the microcontroller in order to get the time and to store data respectively. Figure 1
shows the basic block diagram of the system.The RTC communicates with the
microcontroller via SPI interface. The RTC was preprogrammed with the local time and
was set to update the time at each minute. The SD card system communicates with the
main controller through RS232 interface. When this system is in the writing mode, the
data on its RS232 port is written into the SD card at the given specific memory
location.When the system is in reading mode the stored data can be read by providing the
address of the data array.
(SPI)
Timing System
The position of the sun relative to the surface of the earth can be expressed by two angles
called solar azimuth ( ) and elevation ( ). Equations (1) and (2) are used to calculate the
topocentric elevation and the azimuth angles using this algorithm.
(1)
(2)
where, is the topocentric local hour angle, is the topocentric sun declination, is the
REVHYHUVJHRJUDSKLFDOODWLWXGHDQG is the atmospheric refraction correction
299
3. Results
Solar angles were calculated using the above equations for a time period of 5 years with
one minute of time intervals by implementing the SPA on a computer. These angle values
were then used to derive the required tracking angle for the PTC and arranged them with
time as a lookup table. This lookup table is then downloaded to an SD memory card using
the SD memory controller through the RS232 interface of the computer. The memory
inside the SD card is indexed according to the time and any data from the memory can be
retrieved by the time value. The time produced by the RTC is read by the main controller
and fetch the corresponding solar angles from the SD card.
This system was then interfaced to a manualsun tracking system of a parabolic trough
concentrator (PTC). This PTC is aligned in the North-South direction and free to rotate
around its horizontal axis. The actuator system of this module consists of a DC servo
motor and a gear system and the motor is controlled by a PI controller. According to the
dimensions of this module, it has afull acceptance angle of 3.8 . Therefore the maximum
acceptable variation of thefocal line of the system can vary from the center of the
receiver tube. This deviation of the focal line was measured to evaluate the performance of
the tracking system as shown in Figure 2. According to the Figure 2, the maximum
deviation of the line is 1.5 cm which is in the acceptable range.
Variation of the solar beam Vs Time
2.5
1.5
Variation of the distance (cm)
0.5
-0.5
-1
-1.5
-2
-2.5
0 50 100 150 200 250 300 350
Time (s)
F igure 2: Variation of the solar beam relative to the center of the receiver
300
4. Discussion and Conclusions
In order to calculate solar azimuth and zenith angles using an electronic system, equations
(1) and (2) should be implemented on it and there are about 40 other equations that should
be solved before feeding data to those equations.Implementation of such a lengthy
calculation inside a microcontroller is not practical due to the limitation of program
memory and the processing power. Therefore pre-calculation of solar angles using a
computer and having them in a lookup table stored in a SD memory can be considered as
anefficient alternative method. Angle data for one hour can be stored in one data array of
512 bytes of the memory card with one minute time intervals. Therefore, the memory
requirement for one day is 13 kB and only 5 MB is needed for the storage of data for one
year. Hence, solar angle data for a long period can be stored in a memory with small
capacity.
A cknowledgement: Financial assistant by the International Science Programs of the Uppsala University,
Sweden is acknowledged.
A bstract
A novel, low cost and reliable method of using a microcontroller has been tested and
evaluated for measuring temperature. Most of the microcontrollers come with a Watch
Dog Timer (WDT) which is clocked by an internal Resistor Capacitor (RC) oscillator. The
oscillating frequency of the RC oscillator depends on the temperature. The time-out
periods (in counts) of WDT of Microchip PIC16F877A microcontroller were measured at
different temperatures by means of a LM35 sensor from National Semiconductor Inc. It is
found to give a good linear relationship (correlation coefficient R2=0.997) between time-
301