摘要: 为了克服单独应用BP算法时存在的缺陷,利用遗传算法(GA)对其进行了改进,建立了基于遗传算法与BP神经网络相结合的诊断模型,此外在二进制编码方法的基础上,讨论了十进制的编码方法与实现以及网络模型参数取值与学习次数间的相互影响等关键问题。
关键词:
神经网络,
BP算法,
遗传算法,
故障诊断
Abstract: A new method is proved to overcome the problems when BP only is used, in which the BP neural network is improved by genetic algorithm. On base of a binary encoding, a decimal-coded genetic algorithm, accomplishment and relationship between parameters and number of study are discussed .
中图分类号:
史永胜,宋云雪. 基于遗传算法与BP神经网络的故障诊断模型[J]. , 2004, 30(14): 125-127.
SHI Yongsheng,SONG Yunxue. Fault Diagnosis Model Based on Genetic Algorithm and BP Neural Network[J]. , 2004, 30(14): 125-127.