Differential Evolution based intelligent control for speed regulation of a PMDC motor

Abstract

Differential Evolution (DE) is an evolutionary algorithm (EA) known for its simplicity, robustness and performance. Compared to other EAs, DE has shown better performance according to recent research. In this paper a DE algorithm is designed for controller optimization of a PMDC motor speed regulation system. Presenting a comprehensive description of the plant, architecture is designed for automatic speed regulation. Then the DE algorithm is applied on the system to optimize the controller parameters. Performance of the optimal controller is studied with simulations and performance and robustness of the DE algorithm has been analyzed. I. INTRODUCTION HE Differential Evolution (DE) is a global optimization EA which Storn and Price introduced in 1995 [1]. Like many other intelligent techniques DE also is based on population generation and search. Its operators for optimization are mutation, crossover and selection, similar to the Genetic Algorithm (GA). However the major difference between DE and GA is that DE largely depends on mutation for the generation of new solutions while GA depends on crossover [5].Having a very few governing parameters, DE is not only simple in implementation but also robust in performance [1]. Recent studies have acclaimed the superiority of DE over several other optimization techniques in terms of robustness and speed of convergence [1,[6][7][8].Intelligent optimization techniques have recently been applied for controller optimization in many electrical systems, such as induction motors, brushless DC motors and power systems [9][10][11]. Most of these optimization problems are solved with GA. In this paper controller optimization for a PMDC motor is considered and optimization is performed with DE which is much easier to implement and excellent in performance. In some recent research work [12] and [13], performance of DE is compared with other heuristic methods such as particle swarm optimization (PSO) and evolutionary algorithms and is found more robust, faster and Asim H. Syed is with the Electrical Engineering Department, King Fahd University of Petroleum and Minerals, Dhahran 31261, Kingdom of Saudi Arabia (email: syedasimhussain@gmail.com). M. A. Abido is with the Electrical Engineering Department, King Fahd

New robust controller design for voltage source inverters in microgrids

“…This can be accomplished by any intelligent control design method. We are going to use the Differential Evolution (DE) [55]- [57] method which has recently been shown to be a very fast, efficient and robust optimization technique.…”