基于局域波神经网络的柴油机故障诊断方法

Diagnosis of Diesel Engine Faults Based on Neural Network Approach of Local Waves

给出了基于局域波法的时频域局部能量的概念 ,提出了一种基于局域波法和 RBF神经网络的故障诊断方法。利用该系统对柴油机缸套活塞磨损程度进行估计 ,诊断结果与实际完全吻合 ,表明了该局域波神经网络故障诊断系统的有效性。

Vibration signals of a machine proved to be non-stationary ones and they usually carry the dynamic information of a machine that is very useful for fault diagnosis. The local wave method is a very good tool for signal identification. In this paper,a new and efficient fault diagnosis system based on the local wave method was developed,and a RBF network was introduced. Accordingly,a new local wave fault diagnosis approach was also developed. The proposed fault diagnosis approach was tested on diesel engine piston-liner wear estimation and the results showed that the system was effective.