Pid Control System Design Method Based On Evolutionary Algorithms

In automatic control field, PID control theory is one of the earliest control strategy been developed. Because of simple-description, high-dependability, etc. it has been used widely. Although traditional PID-design method can be used easily, but the controller been designed by them can' t meet engineering requirement when faced with non-liner, time-delay or time-varying system. So, if asked for better control performance, a more advanced algorithm is needed to design PID controller.Evolutionary Algorithm(EA) has been developed rapidly these days. Because it has more advantages than traditional optimization algorithms, it has been taken into use widely in automatic control which is tied up with optimization theory. Following this idea, in this paper, two evolution algorithms-Genetic Algorithm(GA) and Particle Swarm Optimization(PSO) have been used to design PID controller.The region of parameters must be restricted when using EA. If the region is initialized in a wrong, the result may fell to meet the demand, or even make the system unstable. To avoid this disadvantage, stability-region algorithm has been introduced into this paper and made to calculated PID-control system's stability-region, which is used to restrict PID parameters' range. As a result of this step, program has been simplified and more effective. Satisfied performance has been gotten in simulation test.A new algorithm, Particle Swarm Genetic Algorithm(PS-GA), which is based on the two before-mentioned algorithms, has been brought out in this paper. In simulation test, it also get good performance.Finally, in order to prove the performance of the PID controller designed by EA in real engineering circumstance, the off-line designed controllers have been used to control real-time temperature in a constant-temperature-chest. Compared with the one designed by traditional method, it has been concluded that the performance of controller designed by EA is far more better than the performance of the one designed by traditional method.