基于主元回归模型的挖掘机液压系统故障诊断

为提高挖掘机液压系统的可靠性,提出了基于主元回归(Principal Component Regression,PCR)模型和模糊C-均值(Fuzzy C-Means,FCM)聚类的挖掘机液压系统故障诊断方法.故障诊断方法将故障诊断分成故障特征提取和故障分类两个部分.在故障特征提取中,首先确定PCR模型的输入/输出结构,通过主元分析(Principal ComponentAnalysis,PCA)的累积贡献率得到故障样本的主元数目,建立相应的PCR模型并提取回归系数作为故障特征;在故障分类中,将FCM聚类作为故障分类器,对回归系数进行分类,判断系统的故障状态.仿真试验表明,提出的故障诊断方法能有效地应用于挖掘机液压系统.

Fault diagnosis for excavator's hydraulic system based on principal component regression model

In order to improve the reliability of the excavator's hydraulic system,a fault diagnosis approach based on principal component regression(PCR) model was proposed.This approach has two steps: fault feature extraction and fault classification.First,several input-output PCR models were established by the minimum number of principal components in terms of the total variance,and PCR parameters were regarded as the fault features.Second,FCM clustering was performed as the fault feature classifier to identify the fault patterns.Simulation results show that the proposed fault diagnosis approach can be effectively applied to excavator's hydraulic system.