Anomaly Factor Estimation Using Sensor Propagation Order
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly diagnostic systems struggle to accurately estimate the factor of an anomaly in complex or large-scale facilities due to difficulties in grasping the propagation relationship of influences among facility components.
Innovation Solution
An anomaly factor estimating device that acquires time-series sensor data from multiple facility components, detects anomaly detection sensors, estimates the anomaly detection order and propagation order based on sensor data and a learned structure indicating dependence relationships between components, and calculates an anomaly factor score to identify the anomaly's source.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional anomaly diagnostic systems are used in complex or large-scale facilities, then the system can detect anomalies, but the accuracy of estimating the anomaly factor deteriorates due to inability to grasp propagation relationships
Solution Approach 1:
The patent introduces an intermediary mechanism (anomaly propagation model and detection order estimation algorithm) that mediates between the complex facility structure and the anomaly diagnosis process. This intermediary automatically infers propagation relationships from sensor data without requiring manual modeling, thereby maintaining high estimation accuracy even in complex facilities with many components and indirect influence paths.
2Measurement precision
If manual preparation of inter-detecting unit relationship information is required, then the propagation relationship can be defined accurately, but the ease of operation deteriorates due to difficulty in grasping propagation relationships in complex facilities
Solution Approach 1:
The system implements self-service by automatically inferring propagation relationships and detection orders from sensor data without requiring operator intervention. The anomaly propagation model automatically learns the influence relationships among facility components from the data, eliminating the need for operators to manually grasp and define complex propagation paths in large-scale facilities.
Solution Approach 2:
The patent transforms the propagation relationship definition from a manual parameter-setting task to an automatic data-driven inference process. By changing from fixed manual parameters to dynamic data-based parameters, the system adapts to complex facility structures without increasing operator workload.
3Reliability
If anomaly detection is performed in large-scale facilities with feedback control, then comprehensive monitoring is achieved, but the ability to estimate anomaly factors deteriorates due to difficulty in grasping propagation relationships
Solution Approach 1:
The patent segments the anomaly diagnosis process into distinct stages: anomaly detection, detection order estimation, propagation order estimation, and anomaly factor estimation. This segmentation allows the system to handle large-scale facilities with feedback control by processing information in manageable steps, where each stage builds upon the previous one to maintain high estimation accuracy despite the complexity and scale of the facility.
Data Source
AI summary
There are included a sensor data acquiring unit to acquire a plurality of pieces of time-series sensor data collected by a plurality of sensors provided in a plurality of facility components an anomaly detecting unit to detect a plurality of anomaly detection sensors in which an anomaly has occurred on the basis of a plurality of the pieces of sensor data, an anomaly detection order estimating unit to estimate an anomaly detection order in which occurrence of the anomaly is detected, an anomaly propagation path tracking unit to estimate an anomaly propagation order in which the anomaly has propagated on the basis of anomaly detection sensor information and an estimated structure indicating a dependence relationship between the facility components, and an anomaly factor estimating unit to estimate a factor of the anomaly on the basis of the anomaly detection order and the anomaly propagation order.