Optimization of PID and sliding mode control parameters with improved artificial bee colony algorithm

Control is one of the most important concepts in the industry. Control systems are given importance in order to produce quality-oriented and cost-effective services on the industrial platform and to maintain the sustainability of this service. Due to the difficulty of mathematical modeling of external disturbances and uncertainties affecting production systems, heuristic algorithms have been used frequently to set optimum control parameters in recent years. Among these, the Artificial Bee Colony (ABC) Algorithm is an optimization algorithm that has proven itself in different areas in the literature. In this study, PID, the indispensable controller of industrial production systems, and the sliding-mode controllers ABC, which have been studied extensively in the literature, have been optimized due to the fast dynamic system response to disturbances and uncertainties. In general, while optimization of controller parameters in many studies is done with a fitness function, different suitability functions are defined considering both the effect of each controller on output and the contribution it makes to the control signal in this thesis research, and better results are achieved in optimization. Simulations performed on an electromechanical system model selected from the literature were evaluated both graphically and statistically. The results show that choosing different fitness functions specific to each parameter is more effective in finding the best values in ABC algorithm compared to the single fitness function selection.