reciprocating compressor annular valve, LMD, characteristic parameter factors, fault diagnosis
Taking the reciprocating compressor annular valve as the research object, the vibration signal of the reciprocating compressor ring valve was tested by the accelerometer vibration sensor. Through local mean decomposition (LMD), several PF components corresponding to the signal were obtained, The three characteristic parameter factors of these PF components are extracted, including the skewness coefficient (gi), kurtosis coefficient (qi) and total energy ratio (Ei/E). Then the valve is damaged to varying degrees, including sawing the valve plate, removing some springs from the valve and drilling the valve plate, the same analysis on the operating vibration signal of the damaged valve plate was done to obtain the corresponding parameters factors, and compared with corresponding parameter factors of the normal valve. The results show that in the valve sawing and part springs removing two states, the valve vibration signal obtained by the corresponding characteristic parameter factor will reflect the abnormal value of the fault, but the valve disc perforated state is not obvious. The above shows that although the LMD method has some limitations, it can accurately and effectively evaluate the reciprocating compressor valve vibration signal, and classify the working status and the type of fault of the valve, so it is a practical method to study the diagnosis of valve failure.