Control Loop Abnormality Detection Using Process-Command Fluctuations
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In complex plants with multiple control loops, detecting abnormalities in individual control loops is challenging due to interdependencies, making it difficult to identify signs of abnormality in process values, set points, or manipulated variables.
Innovation Solution
An abnormality detection system that acquires process values and command values for monitoring target devices, using the relationship between their fluctuation ranges to detect abnormalities, and a learning device that learns detection criteria from operational data to improve accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If individual control loops are monitored separately using process values, set points, or manipulated variables, then monitoring simplicity is maintained, but abnormality detection accuracy deteriorates due to interdependencies between control loops
Solution Approach 1:
The patent combines monitoring of multiple control loops into a unified system that analyzes interdependencies between loops. By merging individual loop monitoring with cross-loop correlation analysis, the system maintains operational simplicity while improving abnormality detection accuracy through holistic assessment of control loop relationships.
Solution Approach 2:
The system implements feedback mechanisms that continuously monitor process values, set points, and manipulated variables across multiple control loops, using the observed interdependencies to improve detection accuracy. The feedback loop analyzes correlations between different control loops to identify abnormalities that would be missed in individual monitoring.
2Measurement precision
If complex interdependencies between control loops are analyzed to improve abnormality detection, then detection accuracy improves, but system complexity increases
Solution Approach 1:
The patent segments the complex analysis task into manageable components: individual control loop monitoring, pairwise correlation analysis, and holistic interdependency assessment. This segmentation allows the system to analyze complex interdependencies through structured, modular processing steps, improving detection accuracy without overwhelming system complexity.
Solution Approach 2:
The system transitions from analyzing single control loops in isolation to examining multiple loops across additional dimensions of interdependency. By adding the dimension of cross-loop correlation analysis, the system improves detection accuracy while managing complexity through structured multi-dimensional assessment rather than brute-force complex modeling.
3Ease of manufacture
If traditional statistical processing of process data is used, then implementation simplicity is maintained, but detection capability for inter-loop abnormalities deteriorates
Solution Approach 1:
The patent creates a multi-functional monitoring system that performs both traditional statistical processing and interdependency analysis using the same data infrastructure. This universal approach maintains implementation simplicity by reusing existing data collection mechanisms while adding enhanced detection capabilities through correlation analysis of multiple control loops.
Solution Approach 2:
The system performs preliminary data collection and preprocessing for multiple control loops simultaneously, preparing the data in advance for both traditional statistical analysis and interdependency detection. This preliminary action simplifies implementation by establishing a unified data foundation that supports both simple and advanced detection methods.
Data Source
AI summary
An abnormality detection device includes: a process value acquirer that acquires, during operation of a plant including a plurality of devices, a process value of at least one monitoring target device among the plurality of devices; a command value acquirer that acquires a command value of a control operation amount for controlling the monitoring target device; and an abnormality detector that detects an abnormality of the monitoring target device on the basis of a relationship between a fluctuation range of a process value acquired by the process value acquirer and a fluctuation range of a command value acquired by the command value acquirer during a predetermined period.