【摘要】 提出了基于Elman神经网络的变速器故障诊断方法,以克服传统方法不能用于时变系统的缺陷。由于加入了短时记忆环节和局部逆回互联功能,Elman神经网络具有优秀的时间序列预测能力,变速器故障就从特征信号平方预测误差的期望中检测出来。同时,引入了遗传算法来辅助神经网络的训练,以期获得最佳的检测性能。最后,通过变速器故障台架试验来验证该方法的有效性。更多还原

【Abstract】 In this paper,we propose an Elman neural network-based gearbox fault diagnosis scheme to address the difficulties that conventional methods can not be used in real time system. The Elman neural network has the advantageous time series prediction capability because of its memory nodes,as well as local recurrent connections. Gearbox faults are detected from the variants in the expectation of feature signal prediction error. A Genetic Algorithm(GA) aided training strategy for the Elman neural network is further introduced to achieve better detection performance. Experiments with a practical automobile transmission gearbox with an artificial fault are carried out to verify the effectiveness of our method.更多还原