基于支持向量机的往复压缩机示功图识别研究

Research on Reciprocating Compressor Indicator Diagram Fault Recognition Based on Support Vector Machine

提出了一种基于支持向量机的往复压缩机示功图识别方法。根据不同故障在示功图上反映的不同特征,进行故障特征提取。针对实际故障发生情况,构造了基于决策树的多分类支持向量机故障识别模型。使用不同核函数对计算机模拟与往复压缩机试验台实测的故障示功图进行识别,结果表明,该方法能有效应用于往复压缩机示功图故障识别。

A new method of indicator diagrams recognition of reciprocating compressor based on support vector machine was proposed.According to the different features in the indicator diagrams of different faults,the fault features was extracted.A fault recognition model was constructed based on multi-classification support vector machine and dicision tree with the actual situation.This method was used to recognise the fault indicator diagrams of computer simulation and reciprocating compressor experiments by different kernel functions.The results reflect that this method can be effectively applied in reciprocating compressor indicator diagrams fault recognition.