Transfer Entropy-Based Automated Fault Traversal and Root Cause Identification in Complex Nonlinear Industrial Processes

Abstract

Root cause identification (RCI) of faults in industrial processes enables plant operators to pinpoint the source(s) of the fault and take appropriate corrective actions to prevent failures. Conventional techniques for RCI are not particularly suited for causal maps having cycles and time lags that are characteristic of industrial operations. We propose a fault traversal and root cause identification (FTRCI) algorithm for automatic identification of fault traversal pathways and root cause variables from causal maps. Multivariate time delayed transfer entropy between fault variable states is calculated to model the causal dependencies that are represented as a causal map. Then, using FTRCI, the fault traversal paths within the causal map are traced and the corresponding root cause variables are identified. The proposed methodology was applied to complex faults in the Tennessee Eastman process and an industrial coal pulverizer, and the root cause variables identified using this methodology were in accordance with the process knowledge.

Data-Driven Process Monitoring and Fault Diagnosis: A Comprehensive Survey

“…The streaming data literature offers potentially valuable works on clustering [767], pre-processing [768], outlier treatment [769,770], and event prediction [771], although we were unable to identify mutual references in the analyzed literature. Causality analysis is an example of an application based on a time series that has gained prominence in the PSE field, as evidenced by numerous studies [580,[772][773][774][775][776][777][778][779][780][781][782]. This technique uses statistical tests, such as the Granger causality test, to determine whether a given time series is useful in predicting another.…”