基于RBF神经网络的隶属度函数学习算法

模糊规则的提取和模糊隶属度函数的学习是模糊系统设计中重要而困难的问题.针对当前开发模糊控制系统的一个难点--发现最优的隶属函数和模糊规则,研究了利用RBF神经网络的学习能力,从历史数据中发现隶属度函数,在一定程度上减轻了系统开发工作量,克服了由于缺乏经验而可能造成的偏差.文中探讨了一种用于提取模糊规则的RBF神经网络结构,提出了基于此网路结构的模糊隶属度函数学习算法,最后给出了用于验证该算法有效性的仿真实例.

Learning Algorithm of Membership Functions Based on RBF Neural Network

Extracting of fuzzy rules and learning of fuzzy membership functions are important and difficult problems in the design of fuzzy systems. Aiming at the problem finding the most optimal fuzzy rules and input/output membership functions in developing the fuzzy control system, RBF neural network is studied to decrease the correspondent work of system development and overcome some errors probably caused by lack of experience. The learning algorithm of membership function based on the RBF Neural Network is discussed and an example is given to demonstrate the validity of this algorithm.