基于粗糙集-BP神经网络的机车滚动轴承故障诊断

论文提出了一种基于粗糙集理论与BP神经网络相结合的机车滚动轴承故障诊断方法.首先对原始故障诊断样本的连续属性进行离散化处理,然后利用粗糙集理论,对条件属性进行约简,删除冗余信息,最后将约简的最小属性集作为BP神经网络的输入,并设计BP神经网络对滚动轴承进行诊断.仿真结果表明粗糙集-BP模型不仅简化神经网络结构,而且提高了收敛速度和故障诊断正确率.

Fault Diagnosis of Locomotive Rolling Bearing Based on Rough Set Theory and BP Neural Network

A fault diagnosis method of the locomotive rolling bearing based on rough set theory and BP neural network is proposed. Firstly, the continuous attribute of original fault diagnosis sample is processed in the discrete way. Then, the conditional attribute is simplified by deleting redundant infernation based on the rough set theory. Lastly, considering the simplified minimum attribute set as the input of BP neural network, BP network is designed to diagnose the locomotive bear ing. The simulation result shows that rough BP neural network not only can simplify the structure of neural network, but al so can improve the convergence speed and accuracy of fault diagnosis.