Industrial Process Health Assessment Using Causal Anomaly Propagation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current industrial process monitoring methods fail to accurately consider hierarchical relations and causality directions, leading to inadequate detection of overall health issues in complex industrial processes.
Innovation Solution
A method that determines the state of health of industrial processes by obtaining entity state variables, using machine learning or simulation models to predict health scores, and calculating propagation paths for anomalies based on material and energy flows, Granger Causality, and Transfer Entropy, to aggregate individual entity health importances into an overall process health assessment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If different statistical analysis methods are used to investigate data signals as independent from each other, then the analysis process is simple, but the detection of overall health issues is inadequate
Solution Approach 1:
The patent segments the industrial process into hierarchical levels (process level, unit level, equipment level) and analyzes data signals at each level separately before aggregating results. This allows independent statistical analysis of each segment while still capturing overall health issues through the hierarchical aggregation of findings from each level.
Solution Approach 2:
The patent introduces propagation paths as intermediary structures that connect individual equipment health states to unit health states and finally to overall process health. These propagation paths act as mediators that transmit and aggregate health information across hierarchical levels, enabling comprehensive detection without requiring direct analysis of all signals simultaneously.
2Measurement precision
If hierarchical relations and causality directions are considered in detecting overall health issues, then the detection accuracy is improved, but the analysis complexity increases
Solution Approach 1:
The patent dynamically determines propagation paths based on the specific industrial process being analyzed. Rather than using a fixed complex structure, the propagation paths are adapted to the actual hierarchical relationships and causality directions present in each process, allowing the system to capture complex dependencies without imposing unnecessary structural complexity.
Solution Approach 2:
The patent performs preliminary analysis at each hierarchical level (equipment level, unit level) before aggregating to the overall process level. This step-by-step preliminary action at each level simplifies the overall analysis by breaking down the complex task of analyzing all signals simultaneously into manageable sequential steps that respect hierarchical relationships.
Data Source
AI summary
A method for determining the state of health of an industrial process executed by at least one industrial plant comprising an arrangement of entities, and the state of each such entity, includes obtaining values of the entity state variables; providing the values to a model to obtain a prediction of the state of health; determining propagation paths for anomalies between said entities; determining importances of the states of health of the individual entities for the overall state of health of the process; and aggregating the individual states of health of the entities to obtain the overall state of health of the process.