Investigation on diesel engine fault diagnosis by using Hilbert spectrum entropy

According to signal feature extraction,Hilbert spectrum entropy(HSE) by combining local wave time-frequency analysis and information entropy is used for diesel engine vibration signal feature extraction,then for the fault diagnosis.Firstly,vibration signal is decomposed by using local wave method.Then,Hilbert spectrum can be calculated according to several intrinsic mode functions.Finally,Hilbert spectrum entropy is obtained according to time-frequency distribution.It is used as feature characteristics for diesel engine pattern recognition.The diesel engine piston-liner wear is used as an example to testify the effectiveness of this method.Compared with time information entropy and frequency information entropy,it can be concluded that the HSE is effective to evaluate the condition of diesel engine.It puts forward an effective tool for diesel engine preventative maintenance.