摘要： 提出了一种Add-Mult型模糊神经网络模型(AMFNN)，给出了该模型的结构。根据 梯度下降算法，给出了AMFNN模糊神经网络的误差反传学习算法。与6种极具代表性的模糊推 理方法进行比较的结果表明，AMFNN模糊神经网络模型具有推理精度高、适用范围广、泛化 能力强以及实现容易等特点，因而具有广阔的应用前景。

关键词: 模糊推理, Add-Mult模糊神经网络, 模糊规则获取

Abstract: This paper presents a model of Add-Mult fuzzy neural network (AMFNN ) and the models architecture as well. Error back propagation algorithm for AMFN N is presented according to the gradient descent method. The result compared wit h six representative fuzzy inference methods shows that AMFNN has high reasoning precision, wide application scope, strong generalization capability and easy re alization characteristics. Consequently, AMFNN has vast application prospect.

中图分类号: 

• TP183