基于蚁群神经网络的滚动轴承故障诊断  

Fault Diagnosis for Rolling Bearings Based on Ant Colony Algorithm-Neural Networks

作　　者：程加堂[1] 艾莉[1] 熊伟[1] 

机构地区：[1]红河学院工学院,云南蒙自661100

出　　处：《轴承》2012年第2期34-36,共3页Bearing

基　　金：红河学院科研项目(10XJY117)

摘　　要：为了提高滚动轴承故障诊断的准确性,将蚁群算法与神经网络相结合,根据轴承故障产生的机理,建立其BP神经网络的诊断模型,以网络的误差为目标函数,通过蚁群算法进行BP网络的权值优化,并用优化好的BP网络进行故障诊断。仿真结果表明,该方法具有较高的故障诊断准确度,具有较强的实用性。In order to improve the accuracy of rolling bearing fault diagnosis, a method of ant colony algorithm combined with neural network model is applied. According to the mechanism of bearing failure, the BP neural network diagnostic model is established. Taking the error as objective function, the weight of BP neural network is optimized by using multipie generation computation of ant colony, and then the fault diagnosis is accomplished via the optimized BP neural network. The simulation results show that the method has high accuracy of fault diagnosis, with a strong practical.

关 键 词：滚动轴承 蚁群算法 神经网络 故障诊断 

分 类 号：TH133.33[机械工程—机械制造及自动化] TP274[自动化与计算机技术—检测技术与自动化装置]