一种基于群智能算法的模糊控制器优化设计方法  

A Method for Fuzzy Controller Optimization Design Based on Swarm Intelligence

作　　者：张静华[1] 韩璞[1] 

机构地区：[1]华北电力大学河北省发电过程仿真与优化控制重点实验室,河北保定071003

出　　处：《计算机仿真》2016年第9期380-384,共5页Computer Simulation

摘　　要：模糊控制虽然有很多优点,但从实际工程应用角度看,模糊控制还有许多未解决的问题。其中模糊控制器的设计因素非常多,当受控对象比较复杂时,选择调整模糊控制器的各设计因素就比较困难。为避免模糊控制器设计中参数的复杂调试,并使其获得最佳控制性能,群智能算法常被用来对模糊控制器参数进行优化设计。以火电厂主气温控制为实验对象,研究基于群智能算法优化模糊控制器设计的几种方案,比较其稳定性、抗干扰性、对象变化的适应性,得出用群智能算法优化模糊控制器存在的问题,并给出了一个改进目标函数的多种群差分进化算法的优化设计方案。仿真结果表明,上述方法改善了模糊控制器的鲁棒性。Fuzzy control has a lot of merits, but there are many problems unresolved for engineering applications. One of the problems is that fuzzy controller has many factors to design. It＇ s difficult to tune the factors and parameters especially for complex objects. So swarm intelligence algorithm is often used for the optimization design of the fuzzy controller to avoid the complex adjustment of parameters and acquire the best performance. This paper took the main steam temperature control system as the research object, discussed several schemes of fuzzy controller optimization de- sign, compared the stability, anti-interference, adaptability, and then got the defect of the optimization design of fuzzy controller by using swarm intelligence algorithm. At last, the paper designed a scheme for the fuzzy controller optimization design by modified objective function and multi population deferential evolutionary algorithm. Simulation results show that the scheme can improve the robustness of the fuzzy controller.

关 键 词：群智能算法 模糊控制 优化设计 主气温控制系统 鲁棒性 

分 类 号：TP273.4[自动化与计算机技术—检测技术与自动化装置]