Control Loops & Tuning Methods for Industrial

Automation T3 Paper

By
Caleb Capps

Course: EGR 450

Date Submitted: April 3, 2024

Instructor: Prof. Terry Stevens

Contents
Executive Summary ........................................................................................................................ 1
1 Introduction ......................................................................................................................... 1
1.1 Bang-Bang Control ......................................................................................................... 2
1.2 Feed-Forward Control ..................................................................................................... 2
1.3 PIFF Control ................................................................................................................... 2
1.4 Fuzzy Logic Control ....................................................................................................... 3
1.5 PID Control ..................................................................................................................... 3
1.5.1 Tuning PID Controllers........................................................................................... 3
2 Control Loops in Industrial Automation ............................................................................. 4
2.1 Servo Motors ................................................................................................................... 5
2.2 Variable Frequency Drives ............................................................................................. 5
2.3 Closed-Loop Stepper Motors .......................................................................................... 6
3 Conclusion .......................................................................................................................... 6
4 Appendix A1 – Key Takeaways ......................................................................................... 6
5 Appendix B2 – T3 Components ......................................................................................... 7
6 Appendix C3 – References ................................................................................................. 7
7 Appendix D5 – List of Figures, Tables, Graphs, and Pictures ........................................... 8

Executive Summary

The current report gives a brief description of the most commonly used control systems in the
market. Five different controls systems, namely bang-bang control, feed-forward control,
proportional-integral feed-forward control, fuzzy logic, and PID (proportional, integral,
derivative) control are described in brief. Further exploration into how the control engineer
utilizes control systems in an industrial context are discussed, along with specific attention to
motion control applications. The use of control systems for servo motors, variable frequency
drives, and closed-loop stepper motors was investigated. The current report found that many
industrial motion systems come equipped with an integrated control system, with a significant
number having auto-tuning applications to reduce the time needed to tune the system before use.

1 Introduction

Control systems are ubiquitous to modern industrial automation, and come in two primary
categories: open-loop control, and closed-loop control. Control systems are used both for
monitoring the outputs of industrial processes as well as for tuning the response of components
inside of industrial machinery. While open-loop control does not utilize sensors to monitor the
status of the output parameter(s), closed-loop control utilizes any of a variety of sensors to

Page 1
monitor the status of the output parameter and employs a feedback algorithm to adjust the input
in order to reduce the difference between the desired and actual outputs (Fig. 1, [1]). When
discussing control systems, the controller refers to the algorithms used to respond to the sensor
feedback, the plant referes to disturbances to physical systems, and the sensor refers to the
methods used to obtain information from the environment about the system output. While many
controller types exist for responding to the plant disturbances, the most common of which is PID
(Proportional, Integral, Derivative) control. PID control is often used as it is well studied, easily
implemented, and can be applied to a broad range of controller use cases. Alternatives to PID
include feedforward compensation, bang-bang (on/off) control, PIFF (proportional Integral Feed-
Forward), and fuzzy logic.

Figure 1. General closed-loop feedback control system [1]

1.1 Bang-Bang Control

While most control systems that adjust the reference command signal continuously in order to
reach the desired system output, bang-bang control monitors the state of the feedback value and
turns the output of or on depending on if the value is above or below a predetermined setpoint
limit [2]. While bang-bang control is characterized by a less gradual approach to a setpoint in
general and a range of allowable system outputs, the simplicity of the controller design makes it
applicable when the system output does not need such tight process control. The most common
bang-bang application is for heating systems, including residential air conditioning, boilers, or
electric tea kettles. Pulse-width modulation, which is commonly used to simulate an analog
signal using digital outputs, is also a form of bang-bang control.
1.2 Feed-Forward Control
Feed-forward control involves using a mathematical model of the plant disturbancs and adjusting
the input parameter(s) in order to achieve the desired system output. This method is very useful
in situations where theoretical models very closely predict system outputs, such as in chemical
engineering [3].
1.3 PIFF Control
PIFF control, also called feed-forward augmentation, combines the use of estimations from a
feed-forward control with a PID controller in order to further reduce the controllers need to
respond to an induced error [4]. Because PID controllers respond to an induced error, the feed
forward controller attempts to reduce this lagging error signal before it occurs, which is then
compensated from the PID controller, resulting in less system output noise over time for a given
amount of disturbances from the plant. Because of the complexity of the feed-forward system in
addition to a PID controller, this controller system is often used only when extremely tight

Page 2
process control is necessary, and a suitable feed-forward prediction cannot be suitably found.
PIFF control is used most prevalently in motion control systems due to the necessity of high
precision in mechanically noisy environments.
1.4 Fuzzy Logic Control
Fuzzy Logic Control involves analyzing the analog input to the controller in terms of continuous
logical values between true (1) and false (0). In contrast to the more basic controllers described
above, “fuzzy are non-linear controllers [that] belong to a class of artificial-intelligence (AI)-
based control systems” [5]. Fuzzy logic has been widely applied in the automatic and industrial
control field, including image processing, motor control, and aircraft control [6].
1.5 PID Control
PID control involves obtaining the proportional, integral, and derivative values of the error signal
over time in order to multiply them by predetermined coefficients (KP, KI, KD) before adding
them to the reference command and finally producing an output. While PID control refers to all
three signal controls, in practice, many systems may only use PI, PD, or simply proportional
control depending on the response of the system observed. Figure 2 gives examples of the
performance of a P, PI, PD, and PID controller.
1.5.1 Tuning PID Controllers
The primary complication in the use of PID controllers is in figuring out the coefficients KP, KI,
and KD during the tuning process, as each coefficient changes the response of the system
dramatically for even slight changes in some cases. For the operator tuning a PID system, often
KP is tuned first as increasing this value has the effect of making the system more responsive to
signal error [7]. As the value of KP increases, oscillation will then start to occur in the system
output and KP should be divided in half, this is the starting KP value. Next, slowly increase the KI
value in small steps, watching the behavior of the system both during startup and when a
disturbance is induced. The goal of increasing the KI value is to reduce the controller response
time. Finally, if there is overshoot in the system, slowly increase the KD value, changing the
setpoint between runs and allowing the system to stabilize during each trial. Once these initial
values have been found, changing the KP, KI, KD values cannot be done according to a set
process, and experience is needed to develop an intuition in order to further optimize the
controller response. Figure 3 gives an example of methods for tuning PI control systems for
further refinement.

Page 3
 Figure 2. Feedback response compared to set value (SV) for a P, PI, and PID controller [8]

Figure 3. PI tuning map [9]

2 Control Loops in Industrial Automation

The most common use case for control systems in industrial automation is for motion control
devices such as Servos, VFDs (Variable Frequency Drives), and closed-loop stepper motors.
Variable frequency drives are used to control the frequency and thus the speed of an AC motor.
A closed-loop stepper motor is a stepper motor with an encoder attached to the rotor in order to
obtain a feedback response. Finally, a servo or servo motor is a DC motor with an integrated
controller circuit and potentiometer or similar feedback mechanism. Below is a brief description
of the methods used to tune each type of motor controller as well as common application uses. It
is important to note that before beginning the process of tuning any control system, the desired
output characteristics must be defined in order to effectively determine when the machine has
been tuned properly. In motion control systems. The desired output behavior is often described

Page 4
as a time to reach steady state and the corresponding steady state speed, torque, or position
desired.
2.1 Servo Motors
When tuning the transient response of a servo motor, the values that must be considered are:
• Reference value – also known as the set point, this is the target value for the motor.
• Rise time – measures how quickly the motor reached the reference value.
• Overshoot – occurs when the system briefly exceeds the reference value before settling.
• Peak value – the maximum value of the motor output during operation.
• Settling time – measures how quickly the motor reacts to a disturbance and returns to a
steady state.
Industrial servo motors utilize a servo controller specifically designed to match the selected
servo motor. The servo controller is fed a continuous current that can be modulated to produce a
continuously variable servo speed or torque. position, speed, and toque modes are selected from
the controller, and the performance of the system can be monitored to ensure performance is as
desired. Servo motors are used when high speed and high torque is required due to the use of DC
motors within the servo motor.
2.2 Variable Frequency Drives
In contrast to servo motors, variable frequency drives modulate the motor frequency in order to
produce a wide range of motor speeds and torques. Because there are a number of parameters
that affect the efficiency of motors connected to a VFD, many VFDs are equipped with an
autotuning application that will perform the same process as the operator, but with greater speed
and higher efficiency of the device after tuning. Examples of some of the parameters that may be
tuned for during the autotuning application include [10]:

• Line-to-line resistance
• Motor rated current
• Motor no-load current
• Motor rated slip
• Energy savings coefficient
• Leakage inductance
• Saturation compensation
• Motor iron loss
• Motor rated power

Page 5
 Figure 4. VFD autotuning program [10]

Because VFDs are monitoring speed as its output parameter, torque and position control are not
generally applicable for this type of motor control system.
2.3 Closed-Loop Stepper Motors
While standard stepper motors are open-loop devices that can lose positional accuracy over time,
closed-loop stepper motors utilize an encoder as well as a closed-loop control system to obtain
complete control of torque and position of the rotor. The most common control system for
closed-loop steppers is a simple PID system [11], which must be tuned to each application to
account disturbances in the installation. While closed-loop steppers are not nearly as common as
open-loop steppers, in conditions where maintaining knowledge of true position is critical to the
process and where a high positional accuracy is needed, closed-loop steppers offer superior
performance compared to servo motors. In recent years, the growth of the additive
manufacturing sector has begun to utilize closed-loop steppers for their no-loss positional
accuracy, leading to increasingly cheaper and more capable devices.

3 Conclusion

Control systems are a powerful tool used to control analog devices in industrial automation.
Numerous control system methodologies exist, with the most prevalent being bang-bang control,
feed forward control, PIFF control, fuzzy logic control, and PID control. Industrial automation
utilizes PID loops extensively, with many diverse types of motors offering integrated control
systems with internal PID controllers. Servo motors utilize a DC motor whose shaft position is
monitored and controlled using an internal PID controller, variable frequency drives often come
pre-programmed with a PID autotuning application for ease of operation, and closed-loop
stepper motors require bespoke tuning to each electromechanical system. While control systems
are an ever-growing area of study, the increasing necessity to interface with analog systems has
led to many parts of the tuning process to be abstracted away for the user, leading to reduced
development times for machines.

4 Appendix A1 – Key Takeaways

Analog control systems use a variety of control systems in order to obtain a tightly controlled
output response. PID control is used in the vast majority of cases because of its flexibility. Bang-
bang (on-off) control is used when tight output control is not as necessary. More advanced
control systems are available but come with increased time needed to develop a more accurate

Page 6
model of the system in order to be used effectively. Nearly all industrial automation components
make use of some form of integrated tuning in order to minimize the need for operators to tune
during installation.

5 Appendix B2 – T3 Components

Tech

• Servo Motors
• VFDs (Variable Frequency Drives)
• Closed-loop stepper motors
Terms

• Open-loop control
• Closed-loop control
• PID (Proportional, Integral, Derivative) control
Trends

• PID (Proportional, Integral, Derivative) control is the most common control system
• Bang-bang control is often used in heating systems
• Many VFDs are now being equipped with a PID autotuning application

6 Appendix C3 – References

1. Kluever, Craig A. 2015a. Dynamic Systems: Modeling, Simulation, and Control.

Hoboken, NJ: John Wiley and Sons, Inc.
2. “On/off Control (Bang-Bang Control) | Closed-Loop Control Systems | Textbook.” n.d.
Accessed April 3, 2024b. https://control.com/textbook/closed-loop-control/onoff-
control/.
3. Woolf, Peter et al. 2024c. Chemical Process Dynamics and Controls. LibreTexts.
https://eng.libretexts.org/Bookshelves/Industrial_and_Systems_Engineering/Che
mical_Process_Dynamics_and_Controls_(Woolf).
4. Natchwey, Peter. 2015d. “Feed Forwards Augment PID Control.” Control Engineering.
March 31, 2015. https://www.controleng.com/articles/feed-forwards-augment-
pid-control/.
5. Galan, Peter. n.d. “Control System Improvements: Feed-Forward, Adaptive, Fuzzy
Control.” Control Engineering. Accessed April 2, 2024e.
https://www.controleng.com/control-systems/pid-apc/.
6. Iswanto, Iswanto, and Irfan Ahmad. 2021f. “Second Order Integral Fuzzy Logic Control
Based Rocket Tracking Control.” Journal of Robotics and Control (JRC) 2 (6).
https://doi.org/10.18196/jrc.26142.
7. Explained, DC-PID. 2018g. “How to Tune a PID Controller • PID Explained.” PID

Page 7
 Explained (blog). November 24, 2018. https://pidexplained.com/how-to-tune-a-
pid-controller/.
8. Beniwal, N. S. 2012h. “Comparison of Conventional and Fuzzy P/PI/PD/PID Controller
for Higher Order Non-Linear Plant with High Dead Time.” International Journal
of Scientific and Research Publications 2 (8).
https://www.semanticscholar.org/paper/Comparison-of-Conventional-and-Fuzzy-
P-PI-PD-PID-Beniwal/ead1c4a69208d915e12688baf7bdb5f9de6ac1e5.
9. “PID Tuning in Distributed Control Systems | Yokogawa Electric Corporation.” 2010i.
2010. https://www.yokogawa.com/library/resources/white-papers/pid-tuning-in-
distributed-control-systems/.
10. Avery, Paul. 2021j. “Benefits of Auto-Tuning VFDs.” Control Engineering. September 5,
2021. https://www.controleng.com/articles/benefits-of-auto-tuning-vfds/.
11. Collins, Danielle. 2019k. “How Does Closed-Loop Stepper Control Work (and Why Not
Just Use a Servo)?” Linear Motion Tips. February 15, 2019.
https://www.linearmotiontips.com/how-does-closed-loop-stepper-control-work/.

7 Appendix D5 – List of Figures, Tables, Graphs, and Pictures

Figure 1. General closed-loop feedback control system [1] ........................................................... 2

Figure 2. Feedback response compared to set value (SV) for a P, PI, and PID controller [8] ....... 4
Figure 3. PI tuning map [9] ............................................................................................................. 4
Figure 4. VFD autotuning program ................................................................................................ 6

Page 8