Abnormality Determination Model Using Time-Series Signal Clipping
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In plant facility monitoring, setting upper and lower limits for signal data from multiple facilities is labor-intensive and costly, and existing multivariate analysis methods fail to utilize time-based information effectively, leading to reduced accuracy in abnormality determination.
Innovation Solution
A device and method that generate an abnormality determination model using time-series signal clipping and principal component analysis, creating multiple models based on correlation and operation patterns to accurately identify abnormalities across various facilities without manual parameter setting.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If upper and lower limits are individually set for each facility signal data, then abnormality determination can be performed for each facility, but manpower and cost increase significantly
Solution Approach 1:
The patent merges multiple facility signal data into a single multivariate dataset for simultaneous analysis. By combining signals from multiple facilities and applying multivariate analysis, the system determines abnormalities collectively rather than individually, reducing the need for separate limit settings for each facility while maintaining determination accuracy.
Solution Approach 2:
The patent creates a universal abnormality determination model that can be applied across multiple facilities simultaneously. The multivariate analysis framework serves multiple facilities with a single model, eliminating the need for facility-specific parameter tuning and reducing overall system complexity.
2Device complexity
If multivariate analysis is applied to convert process amounts into feature amounts, then analysis is simplified, but time-direction information is lost reducing determination accuracy
Solution Approach 1:
The patent performs preliminary extraction of time-directional features from the raw time-series signals before applying multivariate analysis. By pre-processing the data to capture temporal patterns (such as trends, rates of change, and periodicities) and then applying multivariate analysis on these extracted features, the system preserves time-direction information while benefiting from the dimensionality reduction and simplification provided by multivariate techniques.
Data Source
AI summary
An abnormality determination model generating device generates an abnormality determination model for determining an abnormality of a facility performing a predetermined operation, and includes: a time-series signal clipping unit configured to clip K times from one or more time-series signals indicating an operation state of the facility during normal operation of the facility; and an abnormality determination model generating unit configured to generate the abnormality determination model from the time-series signals during the normal operation clipped out by the time-series signal clipping unit, wherein the abnormality determination model generating unit is configured to clip L items per one time of clipping from the time-series signals during the normal operation clipped by the time-series signal clipping unit and configures an L-dimensional vector including L variables.