基于遗传算法的间接自校正模糊神经网络控制

Indirect self-adaptive control based on fuzzy neural network using genetic algorithm

针对一般神经网络训练算法训练速度慢和易陷入局部极小点的不足 ,文章提出了一种基于模糊神经网络的间接自校正控制系统 ,控制器是以高斯函数为隶属度函数的径向基函数 (RBF)神经网络结构 ,利用改进的遗传算法 (GA)对结构和参数进行同步优化。神经网络模型 (NNP)利用弹性BP算法进行离线辨识。仿真与传统的模糊PID控制器控制进行比较 。

An indirect self-adaptive fuzzy-neural network controller (FNNC) has been presented with its parameters and the structure tuned simultaneously by GA. The structure of the controller is based on the Radical Basis Function (RBF) neural network with Gaussian membership functions. Dynamic crossover and mutation probabilistic rates have been used for faster convergence. Flexible BP has been applied for the neural network identification of the plant off-line. The performance of the proposed FNNC is compared with a conventional fuzzy-PID controller. Simulation results show that the FNNC presents encouraging advantages.