Dual-Model Anomaly Determination for Unknown Equipment Faults
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing anomaly determination methods overlook unknown anomalies in production equipment, mistakenly determining them as normal due to reliance on supervised learning with known failure patterns.
Innovation Solution
An anomaly determination device and method using a first determination model to identify predetermined anomalies and a second model to classify device states, outputting unknown anomalies when the anomaly is not predetermined, with a label setting unit adding new classes for unknown anomalies to update the models.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If machine learning using supervised data including failure patterns is used, then the ability to identify known anomalies is improved, but the ability to detect unknown anomalies deteriorates
Solution Approach 1:
The anomaly determination process is segmented into two distinct determination models: a first determination model for identifying known predetermined anomalies using supervised learning, and a second determination model for detecting unknown anomalies by classifying device states. This segmentation allows each model to specialize in its respective strength without compromising the other.
Solution Approach 2:
A label setting unit acts as an intermediary between the two determination models. It receives determination results from both models, compares them, and determines whether an anomaly is known or unknown by identifying discrepancies between the models' outputs. This intermediary component enables the system to distinguish between known and unknown anomalies effectively.
2Reliability
If supervised learning with known failure patterns is used, then the reliability for known anomaly detection is improved, but the risk of overlooking unknown anomalies increases
Solution Approach 1:
The system dynamically adapts by incorporating a label setting unit that learns from determination results. When unknown anomalies are detected (identified as cases where the first and second determination models disagree), the system can update its knowledge base, allowing it to evolve and recognize previously unknown anomaly patterns over time.
Solution Approach 2:
The label setting unit provides feedback by comparing determination results from both models and identifying unknown anomalies. This feedback mechanism allows the system to recognize patterns of unknown anomalies and potentially incorporate them into future detection processes, improving overall system reliability over time.
3Device complexity
If a single determination model is used, then the device complexity is reduced, but the measurement precision for distinguishing known and unknown anomalies deteriorates
Solution Approach 1:
The determination system is divided into two specialized models: one for known anomaly identification and another for state classification. This segmentation enables each model to focus on its specific function, achieving higher measurement precision than a single general-purpose model could provide.
Solution Approach 2:
The label setting unit serves as an intermediary that synthesizes information from both determination models. It compares their outputs and determines whether anomalies are known or unknown, achieving accurate anomaly classification without requiring either model to perform both functions independently.
Data Source
AI summary
An anomaly determination device and an anomaly determination method determine an anomaly of a device based on state data of the device, by using a first determination model configured to determine whether a predetermined anomaly has occurred in the device, and a second determination model configured to classify state of the device, and output the determined anomaly of the device as an unknown anomaly in a case where the anomaly of the device is not the predetermined anomaly.