基于多分量奇异熵的往复式压缩机故障分类

由于往复式机械振动信号的强烈非线性,对其进行特征提取较为困难.针对上述现象提出了一种计算信号多分量奇异熵的特征提取方法.通过局域波法提取出振动信号的各基本模式分量. 利用非线性动力学相空间重构理论适当选择嵌入维数与延迟时间,计算出往复机振动信号各基本模式分量的奇异熵值,提取出故障信息,并经自适应神经模糊推理系统(ANFIS)对故障特征进行分类.结果表明全部分类正确,达到了故障诊断的目的.

Classification of reciprocating compressor faults based on multi-component singular entropy

Condition monitoring of reciprocating machines through the analysis of their vibrations is recognized to be a difficult issue,essentially because of the strong nonlinearity of the vibration signals.A new extracting method of multi-component singular entropy is put forward to solve this problem.Local wave method is combined with singular entropy to extract the features from the intrinsic mode function(IMF) of the vibration signals of reciprocating machines.Dimensions and time lags are given by reconstruction theory of nonlinear dynamics.And the features will be used as the input of adaptive neuro-fuzzy inference system(ANFIS) to classify and recognize the fault mode.The fault results are classified correctly.The conclusion shows that this method is feasible.