Condition Monitoring and Root Cause Diagnosis for Industrial Key Equipment based on an Adaptive Causal Index

Driven by the increasing degree of automation in the process industry, the production process is increasingly dependent on the normal operation of critical equipment. Therefore, it is important to achieve root cause diagnosis and propagation path identification of faults for critical equipment in the process. However, in practical applications, the complex structure of critical equipment leads to the fact that the causal relationship between components cannot be directly obtained through the mechanism. This makes it difficult to evaluate abnormal propagation pathways. Therefore, in view of the above problems, this paper uses the Granger causality test to analyze the relationship between equipment components. It is followed by an adaptive index for condition monitoring which is designed by the value of F-test. On this basis, the anomaly propagation network of the equipment is established and the abnormal states is traced through a dynamic Bayesian network. The effectiveness of the proposed method is demonstrated by a dataset collected from the actual industrial process. Results show that the adaptive index is effective and it can accurately monitor the condition of equipment.