Particle Swarm Optimization for Tuning PID Parameters to Control General Systems

The PID controller is a control system that is widely used in control engineering. PID stands for Proportional-Integral-Derivative, representing the three components of the controller. The controller can be used for various systems, from temperature control in heating devices to speed regulation of motors in robotics. The art lies in selecting the three parameters of the PID controller so that the system shows the desired control behaviour. This work investigates how Particle Swarm Optimization (PSO), an algorithm from the field of artificial intelligence, is capable of efficiently finding suitable control parameters for a PT3 system and a 2nd order-overshooting system. An analysis of the parameters found using PSO was conducted, which were assessed according to the Integrated Time Absolute Error (ITAE) criterion. These parameters are compared with those from previous studies to validate the reliability of PSO. For both systems examined, 300 simulations were carried out, the results of the simulations serve as a database for evaluation. The approach includes a detailed classification of the studied systems, an analysis of their transfer functions, and a discussion of the computational aspects of the ITAE criterion. Additionally, the operation of the PSO algorithm is explained. It discusses how the PSO algorithm was implemented and how the results were stored and evaluated. The evaluations show that PSO can find very precise parameters in a short time. The found parameters led to better control of the studied systems compared with traditional methods.