感应电机参数辨识三种智能算法比较.pdfVIP

14 11 Vol14 No11 2010 11 ELECTR IC MACH INES AND CONTROL Nov. 2010 陈振锋, 钟彦儒, 李洁 ( , 710048) : 针对感应电机参数辨识, 采用 3种智能优化算法, 即遗传算法蚁群算法微粒群算法感 应电机的实际输出电 和电气模型的观测电 之间的差值被作为目标函数不断对电气模型中的参 数进行更新, 从而辨识全部的感应电机参数变速运行实验是在电机不带负载的情况下进行的 通过实验, 对感应电机参数辨识 3种智能优化算法进行比较, 并得出结论遗传算法可以得到最准 确的电机参数, 微粒群算法次之, 蚁群算法最差蚁群算法所需时间最短, 遗传算法次之, 微粒群算 法最长 : 感应电机; 矢量控制; 参数辨识; 智能优化算法 : TU 313 : A : 1007 449X( 2010) 11 0007 06 Comparison of three intelligent optim ization algorithms for parameter identification of induction motors CHEN Zhenfeng, ZHONG Yanru, LI Jie ( School ofAutom ation and Information Engineering, X i an University of Technology, X i an 710048, Ch ina) Abstract: Genetic algorithm, ant colony optmi ization and particle swarm optmi ization were introduced and applied to the parameter identification of an inductionmotor for vector control. The errorsbetween the ac tual stator current output of an inductionmotor and the stator current output of themodelwere used as the criterion to correct themodel parameters, so as to identify all the parameters of an induction motor. Ex permi entswere conducted on speedvarying operation w ith no load. The three kinds of optmi ization algo rithm swere compared w ith each other and conclusionswere summarized. Genetic algorithm can acquire themost accurate parameters of inductionmotor. Particle swarm optmi ization takes second place, and ant colony optmi ization is theworst. In computing tmi e, the ant colony optmi ization is the fastest. Genetic al gorithm takes second place, and particle swarm optmi ization runs the most s