Research On Fault Diagnosis Of Diesel Engine

Feature extraction is one of the most important and difficult topic in the fault diagnosis of mechanical equipments. It is directly related to the accuracy of fault diagnosis. Because of complicated structure and multiple excitations in the diesel engine, there are not a series of effective and applicable fault diagnosis methods for it at present. However, frequency characters of fault signal adopting vibration analysis are main information of fault equipment.The vibration diagnosis method is adopted by this paper. Based on analysizing structure and typical diagnosis of diesel engine, and analysizing the vibration signals by time domain, frequency domain and wavelet packet deeply, the paper puts forward a fault diagnosis flow of the diesel engine for the vibration signals of the body surface and cylinder head bolts on the TATRA815 diesel engine. The measured vibration signals are carried on correlation analysis firstly, according to frequency characteristics of body, the fault cylinder could be determined; secondly, the cylinder is analyzed by time domain analysis, it could derive that the kurtosis is a sensitive time-domain parameter of fault diesel. Thirdly, through frequency analysis, fault types of the diesel engine are determined by frequency spectrums increasing with rotate speed and typical fault analysis of the diesel. Finally, the characteristic frequency band of the fault is determined by wavelet packet analysis. The sensitive fault parameters determined by above analysis provide accurate characteristic parameters for pattern recognitions such as neural networks.