基于可拓物元模型的故障诊断研究与应用

Research and Application of Fault Diagnosis Based on Extension Matter-Element Model

中文关键词:  主元分析  可拓学  物元模型  关联度  故障诊断

英文关键词:PCA  extension theory  matter-element model  relevance degree  fault diagnosis

中文摘要:

      制冷系统由于内部物质形态的多样性以及系统参数间的高度耦合，增加了故障检测及诊断的难度，为解决此问题，提出了一种基于主元分析（PCA）和可拓物元模型的诊断算法，利用主元分析法提取故障特征参数；建立以可拓物元模型算法为基础的故障诊断模型；该模型借助主元分析方法获取属性互不相关的训练集，通过建立其物元模型，利用关联函数定量计算待测对象对于每一种故障模式的关联程度，进而判断可能的故障模式；同时利用实验数据加以验证，结果表明：该模型有效地提高了故障诊断率，同时优于单纯的可拓物元模型，且该模型对小样本的处理能力优于BP神经网络模型，其诊断正确率更高，训练耗时较少。

英文摘要:

      The diversity of internal physical forms of refrigeration system and the deep coupling between the system parameters make the system more intricate and the detection and diagnosis more complicated.A diagnosis algorithm based on principal component analysis(PCA) and matter-element model of extension theory is proposed.PCA was employed to extract features from the vast data pool,and the extension matter-element model was established for the fault diagnosis.The qualitative description of matter-element model was built with the training dataset of independent characteristics that obtained by PCA method,with relevance function,the fault pattern was judged by the relevance degree of unknown object with each fault pattern.The hybrid model was presented and validated by the historical data from specially designed experiments.Results show that the model effectively improves the fault diagnosis rate,which is better than the extension matter-element model without PCA.It also has better performance on dealing with small sample problem than BP with higher diagnostic rate and much less training time.