应用多模型粒子群优化算法的过热汽温模糊控制  

Fuzzy Control of Saturated Steam Temperature by Multi-Model Particle Swarm Optimization

作　　者：郝万君[1] 刘国良[1] 颜世佐[1] 强文义[1] 

机构地区：[1]哈尔滨工业大学航天学院

出　　处：《动力工程》2006年第5期650-655,共6页Power Engineering

基　　金：黑龙江省科技攻关项目(GC05A306)

摘　　要：针对单一模型的粒子群参数优化算法存在鲁棒性差的问题,提出一种应用多模型粒子群参数优化的方法,将其应用于模糊控制器参数的优化,有效地避免了模糊控制器设计中复杂的参数调试,使其获得良好的控制品质。通过对锅炉过热汽温系统的仿真实验,在负荷大范围变化的条件下,控制系统仍保持了良好的控制性能,并且具有较强的鲁棒性。仿真结果表明了所设计的控制器的有效性和所提出的优化算法的优越性。A multi-model particle swarm optimization （PSO） algorithm is being proposed for solving the problem of poor robustness of single-model PSO. By using it for optimizing the parameters of fuzzy controllers, complex adjustment requirements of parameters can be avoided in the design of the fuzzy controllers and thus improve their control properties. Control simulation results of a boiler＇ s superheater system show that the control system can still maintain good control properties, even during large load variations, exhibiting simultaneously strong robustness. This demonstrates the effectiveness of the designed controller. Figs 6, tables 2 and refs 11.

关 键 词：自动控制技术 模糊控制 多模型 参数优化 粒子群优化算法 

分 类 号：TK223[动力工程及工程热物理—动力机械及工程]