Plant Abnormality Diagnosis Using Weighted Causal Event Trees
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing diagnostic systems for plants, such as power and chemical plants, face challenges in accurately predicting and preventing abnormalities by only considering past events, lacking comprehensive analysis of potential future events and data from similar plants.
Innovation Solution
A diagnostic device and method that incorporates a tree structure-based system to store and analyze past and potential abnormal events, with adjustable weighting of occurrence probabilities to estimate causes of abnormalities, including data from other plants for enhanced accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If only past abnormal events are considered in diagnostic analysis, then the diagnostic system is simple to implement, but the accuracy of cause estimation is insufficient
Solution Approach 1:
The system performs preliminary analysis by pre-storing second information about potential future abnormal events and their causes in a knowledge base before actual abnormalities occur. This allows the diagnostic device to proactively prepare diagnostic frameworks for anticipated issues, improving cause estimation accuracy without adding complex real-time analysis capabilities during operation.
Solution Approach 2:
The diagnostic device merges first information (past actual abnormal events) with second information (potential future abnormal events) in the knowledge base, combining historical data with predictive information. This integration allows the system to leverage both actual occurrence patterns and potential risk scenarios, enhancing diagnostic accuracy while maintaining a unified database structure.
2Reliability
If potential future abnormal events are included in the knowledge base, then the accuracy of predicting abnormality causes improves, but the amount of information to be managed increases
Solution Approach 1:
The system applies different quality characteristics to different types of information in the knowledge base. First information (actual past events) is marked with higher reliability weights since they represent verified occurrences, while second information (potential future events) is marked with lower reliability weights. This differential quality assignment allows the system to manage diverse data types effectively while improving prediction reliability through weighted analysis.
3Measurement precision
If weighting is applied to occurrence probabilities, then the precision of cause estimation improves, but the complexity of probability management increases
Solution Approach 1:
The system dynamically adjusts the weighting parameters of occurrence probabilities based on the reliability of information sources. When first information (actual past events) is available, higher weights are assigned to their occurrence probabilities. When only second information (potential future events) is available, lower weights are assigned. This parameter adjustment strategy enables precise cause estimation while automating the weighting management process.
Data Source
AI summary
A diagnostic device includes a storage unit that stores first information including a first abnormal event which occurred in the past in a plant, a first attribution event that is a cause of the first abnormal event, and a first occurrence probability of the first attribution event, in which a causal relationship between the first abnormal and attribution events is indicated by a tree structure, and second information including a second abnormal event which is supposed to occur in the plant but has not yet occurred, a second attribution event that is a cause of the second abnormal event, and a second occurrence probability of the second attribution event, in which a causal relationship between the second abnormal and attribution events is indicated by a tree structure; and an estimation unit that estimates the cause of the sign of the abnormality, based on the first and second information.