论文部分内容阅读
基于因果拓扑图的工业过程故障诊断方法,将过程知识与数据驱动故障诊断方法结合,有效解决了故障定位和故障传播路径辨识问题。在因果拓扑图的基础上,基于偏相关系数提出一种相关性指标(correlation index,CI)定量衡量因果拓扑中变量间的相关性,实现变量间因果性和相关性的良好结合。为得到准确的故障检测结果,采用概率主元分析(PPCA)对CI指标进行监测。在检测出故障后,应用重构贡献图(reconstruction-based contribution,RBC)和因果拓扑图,并引入加权平均值的概念辨识出最可能的故障传播路径。将提出的方法用于带钢热连轧过程,结果表明,基于因果拓扑图的故障诊断方法能够准确地定位故障源,辨识故障传播路径。
Based on the causal topological map of industrial process fault diagnosis, the combination of process knowledge and data-driven fault diagnosis method can effectively solve the problem of fault location and fault propagation path identification. Based on the causal topology, a correlation index (CI) was proposed based on the partial correlation coefficient to quantitatively measure the correlation between variables in the causal topology and achieve a good combination of causality and relativity among the variables. To obtain accurate fault detection results, the PPCA is used to monitor the CI indicators. After detection of a fault, the most likely fault propagation path is identified using the concept of reconstruction-based contribution (RBC) and causal topography and the introduction of a weighted average. The proposed method is applied to the hot strip rolling process. The results show that the fault diagnosis method based on causal topology can accurately locate the fault source and identify the fault propagation path.