Tuning proportional integral derivative controller parameters by intuitive optimization methods

In parallel with technological innovations and developments, the demand for automatic control systems has rapidly increased in recent years. Control systems enhance the speed, accuracy, and reliability of production processes while reducing costs. Although various proportional control methods are available in the literature, the Proportional Integral Derivative (PID) control is one of the most widely used methods. PID control is a method based on system error that does not require the mathematical model of the controlled system to be known. In this method, the proportional action contributes to the control signal based on the error magnitude, integral component based on the error area, and derivative action based on the slope of the error. The integral component considers the past error, the proportional component reflects the current state, and the derivative component anticipates future errors, enabling the generation of appropriate control signals. In the PID control, the influence of these three actions on the system output is determined by the gain factors. The proper selection of gain factors ensures stable system operation, whereas incorrect selection may lead to instability. In this study, three different test systems were controlled using a PID controller. To improve the performance of the PID controller, the gain factors were optimized using Ziegler-Nichols step response, Particle Swarm Optimization (PSO), Radial Movement Optimization (RHO), Modified Radial Movement Optimization (DRHO), Differential Evolution (DE), and Genetic Algorithm (GA). During the gain-tuning process using heuristic optimization methods, four performance criteria were utilized: total absolute error, total squared error, time-weighted total absolute error, and time-weighted total squared error. Initially, the performance of the PID controller supported by these five heuristic optimization algorithms was compared based on specified error criteria. Subsequently, the performances of the optimization methods were analyzed. Numerous studies have focused on optimizing PID controller parameters using the PSO, DE, GA, and RHO methods. However, no studies have been conducted on optimizing PID controller gain factors using the DRHO algorithm. This innovation represents the most significant contribution of this master's thesis to existing literature.