一种改进的神经网络机械故障诊断专家系统  

Improved expert system for mechanical fault diagnosis based on L-M neural network

作　　者：彭滔[1] 汪鲁才[1] 吴桂清[2] 张颖[1] 

机构地区：[1]湖南师范大学工学院,长沙410081 [2]湖南大学电气与信息工程学院,长沙410082

出　　处：《计算机工程与应用》2007年第1期232-234,共3页Computer Engineering and Applications

基　　金：湖南省自然科学基金资助项目(04JJ40042)。

摘　　要：针对传统BP神经网络训练中收敛速度较慢的缺点,提出一种基于L-M算法的神经网络应用于机械设备故障诊断的专家系统。论述了神经网络的专家系统结构,并以7216圆锥轴承试验研究为例,建立了基于该算法的故障诊断模型。仿真结果表明:该模型显著缩短了训练时间,具有较高的准确性。运用该神经网络专家系统进行机械故障诊断是有效的。An improved neural network based on L-M algorithm has been applied to fault diagnosis expert system against to the slow convergence rate of conventional BP neural network.The expert system structure based on neural network has been introduced,and a fault diagnosis model has been designed combining with 7216 tapered bearings experiment.Simulation results indicate this model can remarkably reduce the training time,with its relatively high accuracy,surpass the conventional BP neural network model.It is feasible for mechanical fault diagnosis.

关 键 词：神经网络 L—M算法 专家系统 故障诊断 

分 类 号：TP118[自动化与计算机技术—控制理论与控制工程]