基于改进粒子群算法的模糊神经网络PID控制器设计

Fuzzy Neural Network PID Controller Design Based on Improved Particle Swarm Optimization

针对模糊神经网络PID控制器中参数初始值的设置对控制器性能影响大的问题,提出一种改进的PSO算法优化模糊神经网络PID控制器参数的设计方法。该方法采用实数编码的方式对控制器参数进行优化,并以ITAT指标作为改进的PS0优化算法的适应度函数。实验仿真表明:经过改进的PSO算法优化的模糊神经网络PID控制器具有良好的动静态性能,响应速度更快,超调量更小,控制精度更高。

Because the setting of initial parameter values in fuzzy neural network PID controller has an important influence on the per- formance of the controller, this paper presents a design method of improved PSO algorithm to optimize the parameters of fuzzy neural network PID controller. The method uses real number encoding method to optimize the controller parameters and sets the ITAT index as the fitness function of the PSO algorithm. Simulation results show that the optimal fuzzy neural network PID controller using improved PSO algorithm has good dynamic and static performance, faster response, smaller overshoot, higher control precision.