基于PSO算法的磁浮系统PID控制器优化与评价  

Optimization and Design of Maglev System PID Controller Based on Particle Swarm Optimization Algorithm

作　　者：宋荣荣[1,2] 陈滋利[3] 

机构地区：[1]西南交通大学电气工程学院,四川成都610031 [2]西南民族大学计算机科学与技术学院,四川成都610041 [3]西南交通大学数学学院,四川成都610031

出　　处：《西南交通大学学报》2015年第1期36-43,共8页Journal of Southwest Jiaotong University

基　　金：四川省科技支撑计划资助项目(2012GZ0103);教育部新世纪优秀人才支持计划资助项目(NCET-11-0712)

摘　　要：为了改善磁浮系统的非线性和不稳定性特点,利用微分几何方法将两个不同结构的非线性子系统转化为两个相同结构的线性子系统,设计了基于标准粒子群算法的比例积分微分控制器.从固定惯性权重、线性递减惯性权重和线性微分递减惯性权重中,选出适合电磁铁1和电磁铁2的固定惯性权重,得到电磁铁1控制器的固定惯性权重参数C为0.5,电磁铁2控制器的固定惯性权重参数C为0.49,并且通过建立模糊综合评价模型得出优化后的电磁铁1和电磁铁2的控制器抗干扰的能力是"好",且"好"的隶属度皆为0.561 9.实验结果表明,优化后的磁浮系统具有较好的鲁棒性.In order to improve the nonlinear and unstable characteristics of maglev systems, two nonlinear subsystems with different structures were transformed into linear ones with the same structure by differential geometry method, and then a PID controller that is based on the standard particle swarm optimization （PSO） was built. For the PSO algorithm, the fixed inertia weight （ FIW）, linear descend inertia weight （LIW）, and linear differential descend inertia weight （LDW） were comparatively studied through simulation; among them the FIW was found more suitable and hence selected for electromagnet 1 and electromagnet 2. After optimization of controller parameters, the value of the FIW parameter C for electromagnets 1 and 2 was set to 0.5 and 0.49, respectively. In addition, a fuzzy comprehensive evaluation model was built to evaluate the anti-interference performance of the electromagnets 1 and 2, revealing that the two electromagnets both have a good anti-interference ability, and the membership degree of ＂ good＂ is O. 561 9. Experimental results show that the optimized maglev system has a fairly good robustness.

关 键 词：磁浮系统 微分几何 比例积分微分控制器 粒子群优化算法 模糊综合评价 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]