模糊神经网络在噪声消除中的应用  

Application of fuzzy neural network to the noise cancellation

作　　者：姚宏伟[1] 梅晓榕[1] 庄显义[1] 

机构地区：[1]哈尔滨工业大学,黑龙江哈尔滨150001

出　　处：《电机与控制学报》1999年第1期50-52,61,共4页Electric Machines and Control

摘　　要：提出了一种基于改进模糊聚类算法的训练模糊神经网络的算法。该方法采用遗传算法改进传统的模糊聚类算法，并给出了一个衡量聚类有效性的函数以确定聚类算法中的聚类总数，从而确定模糊神经网络结构。仿真结果表明模糊神经网络可成功的应用于噪声消除。A learning algorithm based on an improved fuzzy clustering algorithm is proposed.The classical fuzzy clustering algorithm is improved with genetic algorithms. A function formeasuring clustering validity is defined with which the number of classes can be determined.The structure of fuzzy neural network is determined in the view of this method. The simulated results show that fuzzy neural network can be successfully applied to the field of the noise cancellation.

关 键 词：噪声消除 模糊神经网络 遗传算法 

分 类 号：TB53[理学—物理] TP18[理学—声学]