基于GA-BP神经网络的滚动轴承故障诊断研究

Research on Fault Diagnosis of Rolling Bearing based on GA-BP Neural Network

为进一步降低滚动轴承设备故障率,保障机械设备安全运行,对实测滚动轴承振动信号进行集合经验模态分解(EEMD),并将分解得到的固有模态函数(IMF)分量能量熵作为BP神经网络的输入特征量。对BP神经网络初始权值和阈值用遗传智能优化算法(GA)进行优化,获得优化后的GA-BP网络。将提出的EEMD-GA-BP网络和EMD-GA-BP网络、EEMD-BP网络应用于滚动轴承内圈故障、外圈故障、滚动体故障以及正常状态下的诊断中。结果表明,EEMD-GA-BP网络对4种故障模式的识别准确率均在99%以上,优于传统的BP网络,可为滚动轴承故障诊断提供一定的参考价值。

In order to further reduce the failure rate of rolling bearing equipment and improve the safe operation of mechanical equipment, ensemble empirical mode decomposition(EEMD) was carried out on the measured vibration signals of rolling bearings. The energy entropy of the decomposed intrinsic mode function(IMF) component was used as the input characteristic of BP neural network. The initial weights and thresholds of BP neural network were optimized by genetic intelligent optimization algorithm(GA), and the optimized GA-BP network was obtained. The proposed EEMD-GA-BP network, EMD-GA-BP network and EEMD-BP network were applied to the diagnosis of rolling bearing inner race fault, outer race fault, rolling element fault and normal state. The research results show that the recognition accuracy of EEMD-GA-BP network for the four fault modes is more than 99%, which is superior to the traditional BP network. The research results can provide some reference value for rolling bearing fault diagnosis.