基于Elman神经网络的齿轮故障诊断研究

在齿轮故障诊断过程中,针对传统的BP神经网络具有学习、记忆不稳定等缺点,提出了将Elman神经网络应用于齿轮故障诊断中,建立了Elman神经网络的应用结构模型,介绍了该网络的训练算法,阐明了齿轮故障诊断的实现过程;结果表明该神经网络学习记忆稳定,具有很好的学习功能,诊断方法具有高可靠性,达到了预期效果。

Research on Gear Fault Diagnosis Based on Elman Neural NET

It studies the disadvantages found in the application of gear fault diagnosis based on traditional BP neutral net which are low net convergence rate and low stability of learning and memory.It applies the Elman neural net to gear fault diagnosis and establishes Elman neural net structure,describes the training algorithm and the process of fault diagnosis.The result validates the net’s better learning function,high stability of learning and memory.This method has good reliability and the expected results are attained.