联想记忆神经网络在电机故障状态监测中的应用研究

Associative Memory Neural Networks and Its Application to Electric Motor Faults Detection

电机的状态监测和故障诊断存在多种不确定性特征,而神经网络在改善故障特征识别、提高信噪比、增强诊断结果的鲁棒性等方面有独特的优越性能。本文基于定子不平衡电流的在线检测,探讨了神经网络在电机故障诊断中的应用,模拟实验取得了与实际情况一致的结果,为电机故障的状态监测与早期诊断提出了一条新的有效的途径。

There exist many uncertain features in the state detection and fault diagnosis of electric motors/Neural networks has special advantages in the aspects of improving fault signs identification, enhancing the ratio of signal to noise, and strenthening the robustness of diagnostic results etc.. The application of neural networks to the fault diagnosis of electric motors based on the on line detection of the unbalanced stator currents is studied in this paper. Expected results which are in accord with real situation are obtained by simulating experiments. It provides a hew and effective method for the fault state detection and early diagnosis of electric motors.