改进粒子群优化的无刷直流电机模糊控制

摘要

A new adaptive simulated annealing particle swarm optimization ( ASAPSO ) algorithm is pro-posed to solve the problem that PSO may trap to local optimum. In this algorithm, each learning factor is automatically adjusted based on each particle 's fitness and the acceptable rule of simulated annealing is ap-plied in the process of the particle position updating. The optimization results of test function show that this algorithm has advantage of jumping out of local optimum and gets better overall accuracy. A PI type fuzzy controller based on ASAPSO design method is proposed for the characteristics of the BLDCM for time-var-ying,nonlinear,strong coupling. The PI type fuzzy controller based on ASAPSO is achieved by initialization of the fuzzy rules by PI fuzzy simulation and optimization of the weights of the fuzzy rules and quantitative scaling factor by ASAPSO algorithm. The simulation results show that the controller makes the brushless DC motor speed control system more rapid,stable and robust.%为了解决粒子群算法易陷入局部最优解等缺陷,提出一种根据各个粒子适应值自动调节学习因子的策略,并将模拟退火Metropolis准则引入粒子位置更新过程中,形成一种新的自适应模拟退火粒子群算法.经过测试函数检验,证明该算法跳出局部最优解的能力强,可达到更高的全局优化精度.针对无刷直流电机时变、非线性、强耦合的特性,提出一种通过PI模糊模拟进行模糊规则初始化,再以自适应模拟退火粒子群算法优化其模糊规则权值与量化比例因子的模糊控制器设计方法.仿真结果表明,该控制器使无刷直流电机调速系统具有良好的快速性,稳定性和鲁棒性.