Optimal PID Controller Design for AVR System Based on Multi-objective Optimization and Multi-attribute Decision Making

In this paper, a novel two-stage approach is proposed to design an optimal Proportional-Integral-Derivative (PID) controller for the Automatic Voltage Regulator (AVR) system by combining multi-objective optimization with multi-attribute decision making. First, the optimal PID controller design for the AVR system is formulated as a multiobjective optimization problem. Then a new multi-objective state transition algorithm based on decomposition is used to find the Pareto optimal solutions. In order to prevent integral saturation and large oscillation of the controller, the PID controller output is constrained by integral separation method. Finally, the best compromise solution is obtained by using a multi-attribute decision making method named AHP-Entropy-TOPSIS. The simulation results verify the effectiveness and superiority of the proposed approach.