基于单神经元的参数自学习模糊控制器的研究

Research of Parameter Self-learning Fuzzy Controller Based on Single Neuron

为进一步改善永磁交流伺服系统的动静态性能,本文设计了一种基于单神经元的参数自学习模糊控制器,它在控制规则数与二维控制器相当的基础上,可实现三维模糊控制的效果。引入的单神经元采用改进的BP算法来实现比例因子的在线自学习。控制器具有结构及算法简单、易于解析实现的特点。为验证其有效性,本文通过仿真试验,将其与采用常规的PI调节器的控制系统进行比较,结果表明,这种模糊控制器具有较好的控制效果。

A parameter self-learning fuzzy controller based on single neuron is proposed. A three-dimensional fuzzy controller is implemented by simply using a two-dimensional fuzzy control rule-base without any increase of rules. The control parameters are self-tuned by introducing a single neuron together with a back-propagation learning algorithm. This method has simpler structure and control algorithms and can be realized online easily. Simulation results show that the performance is better than that of PI controller, and the system is much more robust.