PID Controller Tuning Using Different Optimization Algorithms



Lim, Hui Yan (2018) PID Controller Tuning Using Different Optimization Algorithms. Final Year Project (Bachelor), Tunku Abdul Rahman University College.

[img] Text
LIM HUI YAN - Full Thesis.pdf
Restricted to Registered users only

Download (2MB)


The purpose of the project is to compare different optimization methods to obtain optimum PID values for an AC servo motor. Different algorithms including Genetic Algorithm (GA), Particle Swarm Optimization (PSO) and Ant Colony Optimization (ACO) were being tested to identify the most convenient and accurate way to obtain the PID values. Using MATLAB, the transfer function of servo motor was coded and the optimized PID values were obtained by running the codes with the optimization algorithm. After the PID values were found, the step response of each set of PID was observed and analyzed to ensure the PID values obtained were safe to be applied into servo system. The rise time, settling time and overshoot of the step response was being analyzed. If the rise time and settling time exceeded the threshold, the algorithm would be run again after some modification. Based on the experiments conducted and the performance graph generated, the PSO has the highest accuracy as the PID values obtained is highly consistent followed by ACO and lastly GA. PSO has the highest consistency and accuracy as it searches thoroughly in the search space and communicate with all other particles which increase the efficiency. ACO has high accuracy but medium consistency which make it less efficient than PSO. As the PID values obtained are verified, it is written into the servo motor through MR Configurator 2. The performance of the servo motor after applying each set of PID values are recorded and being analyzed to determine which algorithm has higher accuracy. Based on the comparison of performance graph between auto-tuned PID and the other algorithms, the PID values obtained by PSO allow the servo motor has the most stable performance which is closest to the auto-tuned servo performance. In conclusion. PSO has the highest accuracy and consistency in finding PID values followed by ACO and then GA. The PID value obtained by PSO is the closest to the auto-tuned PID value in the servo motor and generate the least ripples in the performance graph. One of the limitations faced in this project is the results obtained by GA were inconsistent. As GA is a built-in tool in MATLAB, the programming of GA could not be not controlled by users. Thus, the problem of inconsistency was difficult to troubleshoot. In future, perhaps an upgraded version GA optimization toolbox can be developed to replace the current version in MATLAB for the result inconsistent problem.

Item Type: Final Year Project
Subjects: Technology > Mechanical engineering and machinery
Technology > Electrical engineering. Electronics engineering
Faculties: Faculty of Engineering and Technology > Bachelor of Engineering (Honours) Mechatronic
Depositing User: Library Staff
Date Deposited: 10 Oct 2018 08:03
Last Modified: 18 Apr 2022 07:09