基于VMD-MSE与SSA-SVM的往复式压缩机气阀故障诊断

Fault diagnosis of reciprocating compressor air valve based on VMD-MSE and SSA-SVM

往复压缩机气阀故障振动信号具有较强的非线性和非平稳性。为了从往复压缩机气阀振动信号中提取故障特征用于故障诊断,提出一种基于变分模态分解(variational mode decomposition,VMD)与多尺度熵(multi-scale entrope,MSE)的故障特征提取方法,并与采用麻雀寻优算法(soarrow search algorithm,SSA)优化的支持向量机(suppot vector mackine,SVM)相结合,用于往复压缩机气阀故障诊断;通过对往复压缩机气阀信号进行VMD分解,选取合适的内禀模态分量(intrinsic mode function,IMF)进行信号重构,基于MSE熵值分析构成特征向量集,最后将其输入SSA-SVM训练并识别故障类型。试验结果表明,基于VMD-MSE与SSA-SVM的故障诊断模型能有效并准确的识别往复压缩机气阀故障。

Fault vibration signals of reciprocating compressor air valve have strong nonlinearity and non-stationarity.Here,to extract fault features from reciprocating compressor air valve vibration signals for fault diagnosis,a fault feature extraction method based on variational modal decomposition(VMD)and multi-scale sample entropy(MSE)was proposed,and it was combined with support vector machine(SVM)optimized using sparrow search algorithm(SSA)for reciprocating compressor air valve fault diagnosis.After reciprocating compressor air valve signal was decomposed with VMD,appropriate intrinsic mode functions(IMFs)were chosen for signal reconstruction,and feature vector set was formed based on MSE analysis.Finally,feature vector set was input into SSA-SVM for training and identifying fault types.The experimental results showed that the proposed fault diagnosis model based on VMD-MSE and SSA-SVM can effectively and correctly identify reciprocating compressor air valve faults.