Factory Abnormality Diagnosis with Dynamic Cluster Updating
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing factory automation diagnosis systems rely on manual interpretation of sensor data, leading to inefficiencies and inaccuracies in detecting abnormalities, as they do not effectively update clustering procedures when wrong labeling occurs.
Innovation Solution
A diagnosis system that classifies data into groups, extracts candidates for new groups, and learns a new model when appropriate, allowing for improved accuracy in diagnosing abnormalities by dynamically updating the clustering process based on user input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If manual diagnosis by skilled workers is performed, then diagnosis accuracy can be maintained through expert judgment, but diagnosis efficiency deteriorates due to time-consuming data checking
Solution Approach 1:
The patent replaces manual mechanical diagnosis by skilled workers with an automated computer-based diagnosis system that uses sensor data collection, feature quantity calculation, and clustering algorithms to automatically determine machine operating states and detect abnormalities, thereby maintaining diagnosis accuracy while significantly improving diagnosis efficiency
Solution Approach 2:
The diagnosis system enables machines to automatically determine their own operating states through real-time sensor data collection and analysis, allowing equipment to self-diagnose without requiring continuous manual inspection by skilled workers
2Productivity
If clustering results are used to determine operating states, then diagnosis speed is improved through automated classification, but diagnosis accuracy deteriorates when wrong clustering is performed and not updated
Solution Approach 1:
The patent implements a feedback mechanism where clustering results are continuously evaluated and the clustering model is updated based on new data and identified patterns, allowing the system to learn from previous diagnoses and improve accuracy over time while maintaining automated diagnosis speed
Solution Approach 2:
The diagnosis system dynamically adapts by updating clustering models and operating state determination criteria based on accumulated data and identified patterns, transforming the static clustering approach into a dynamic system that evolves and improves its accuracy continuously
3Device complexity
If fixed clustering models are used, then system complexity is reduced through straightforward classification, but adaptability deteriorates when new abnormality patterns emerge
Solution Approach 1:
The patent transforms the fixed clustering model into a dynamic system that automatically updates clustering parameters and operating state determination criteria based on new data patterns, enabling the system to adapt to emerging abnormality patterns while maintaining relatively simple system architecture through automated learning
Data Source
AI summary
A diagnosis system diagnoses presence or absence of an abnormality from data pieces collected in a factory. The diagnosis system includes (i) a diagnoser that diagnoses presence or absence of an abnormality by classifying, in accordance with a diagnosis model defining a plurality of groups, the collected data pieces into at least one of the plurality of groups, (ii) an extractor that extracts, from the collected data pieces, a candidate for a data piece to belong to a new group different from the plurality of groups, (iii) a reception device that provides candidate information relating to the candidate extracted by the extractor, (iv) and a learner that learns a new model including the new group. The diagnoser diagnoses presence or absence of an abnormality with the new model after the new model is learned.