折线模糊神经网络的共轭梯度算法  

Conjugate Gradient Algorithm of the Polygonal Fuzzy Neural Networks

作　　者：何英[1] 王贵君[1] 

机构地区：[1]天津师范大学数学科学学院,天津300387

出　　处：《电子学报》2012年第10期2079-2084,共6页Acta Electronica Sinica

基　　金：国家自然科学基金(No.60974144)

摘　　要：为了近似实现模糊数的非线性运算及提高神经网络的逼近精度,引入折线模糊数和折线模糊神经网络,并依据折线模糊数的扩展运算对经典共轭梯度算法进行改进,使该算法在迭代过程中通过一维非精确Armijo-Goldstein线性搜索方法获得优化学习常数,进而在折线模糊神经网络环境下设计了折线模糊共轭梯度算法.最后,通过模拟实例说明了该算法具有计算复杂度低、收敛速度快等特性.In order to realize an nonlinear operations between fuzzy numbers and raise the accuracy of the approximation of fuzzy neural networks,the polygonal fuzzy numbers and a polygonal fuzzy neural network are introduced,and according to their ex tension operations to improve the classic conjugate gradient algorithm,in the iterative process,the optimal constants of the algorithm are obtained through the one-dimensional inexactitude A-G linear search method, and then a polygonal fuzzy conjugate gradient al- gorithm is designed under environment of polygonal fuzzy neural networks. Finally, utilizing a simulation examples, some good char- acteristics of this algorithm are illustrated,for example, low computational complexity, fast convergence and so on.

关 键 词：折线模糊数 折线模糊神经网络 共轭梯度算法 A-G算法 

分 类 号：TP183[自动化与计算机技术—控制理论与控制工程] O159[自动化与计算机技术—控制科学与工程]