Fault Diagnosis Method Based on Modified Multiscale Entropy and Global Distance Evaluation for Valve Fault of Reciprocating Compressor

Abstract

According to the nonlinearity, nonstationarity and multi-component coupling characteristics of reciprocating compressor vibration signals, a fault diagnosis method of reciprocating compressor valve based on the modified multiscale entropy (MMSE) and global distance evaluation (GDE) is proposed. Firstly, to eliminate the noise interference, the VMD method with superior anti-interference performance was utilized to a strong non-stationarity vibration signal for all fault states. To overcome the obstacle of the conventional multiscale entropy (MSE) may yield undefined entropy or an imprecise estimation of entropy by the large scale factor, the MMSE method provided for moving-average procedure by replacing mean-average coarse-grained procedure was developed to the vibration signals, and then the GDE method of overall parameter selection was introduced to evaluate the extracted MMSE and select the optimal sensitivity scale feature. Finally, binary tree of Support Vector Machine (BTSVM), suitable for small sample classification, was selected as the classifier to identify the reciprocating compressor valve fault type. By analyzing the experimental data, it can be shown that the method can effectively identify the fault type of reciprocating compressor valve.