【作者】 李战明； 康爱红；

【Author】 Li zhanming Kang aihong (School of Electrical and Information Engineering, Gansu University of Technology Gansu Lanzhou 730050)

【机构】 甘肃工业大学电气工程与信息工程学院；

【摘要】 针对多维模糊推理中的推理规则庞大和参数难辨识的问题,提出一种基于T-S模糊模型的RBF神经网络的自适应学习算法。该算法不仅能动态调节T-S型模糊RBF网络的隐节点数,还能使网络的数据中心值自适应变化,有较好的自学习能力和泛化能力。仿真结果验证了该算法是有效和可行的,表明此T-S型模糊RBF网络不仅可以快速逼近任意多变量非线性函数,而且具有良好的自适应能力。更多还原

【Abstract】 A adaptive learning algorithm of T-S fuzzy model based RBF neural network is proposed for the problems of enormous inference rules and difficult parameters identification in multi-dimension fuzzy inferences. In this method the number of hidden layer nodes of T-S fuzzy RBF net is not only modified dynamically, but also position of data centers of RBF net is changed adaptively during learning progress, moreover the algorithm has better self-learning ability and generalization ability. The simulation results show that the algorithm is effective and available , that the T-S fuzzy RBF neural network can fast approximate any multi-variable nonlinear function with any required degree of accuracy, furthermore has good adaptive ability .更多还原