Abnormality Detection Algorithm Updating From Operator Feedback
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing abnormality determination systems require significant operator input to create and update abnormal standards, leading to a heavy workload in defining and storing abnormality data for each type of abnormality.
Innovation Solution
An abnormality determination device with an input unit for signal input, a memory for storing a determination algorithm, an abnormality determination unit for determining abnormalities, an output unit for displaying results, an operation unit for receiving true-or-false feedback, and an update unit that adjusts the algorithm based on user feedback, reducing the need for extensive operator input.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If the operator manually inputs abnormality data one by one to create abnormal standards, then the determination algorithm can be updated, but the workload imposed on the operator becomes excessive
Solution Approach 1:
The system automatically learns abnormal standards by processing monitoring data without requiring manual data input from operators. The determination algorithm autonomously updates itself by identifying patterns in abnormal states from historical data, eliminating the need for operators to manually create and input abnormality standards while maintaining high determination accuracy
Solution Approach 2:
The system uses feedback from determination results to continuously improve the determination algorithm. By analyzing whether determination outcomes match actual abnormal states, the system automatically adjusts and refines the algorithm parameters, enabling self-improvement without operator intervention in the data input process
2Measurement precision
If comprehensive abnormality data is collected and stored for each type of abnormality, then the determination precision improves, but the time required for data input increases
Solution Approach 1:
The system performs preliminary learning by automatically analyzing historical monitoring data to pre-establish abnormal standards before actual abnormality determination is needed. This preliminary action captures and processes comprehensive abnormality patterns in advance, so when actual determination is required, the system can quickly reference pre-learned standards without requiring real-time data input
Solution Approach 2:
The patent replaces the mechanical process of manual data input and processing with an automated information processing system. The determination algorithm automatically collects, processes, and stores abnormality data from monitoring systems, substituting manual operator actions with automated computational processes that achieve the same data accumulation goal without time loss
Data Source
AI summary
An abnormality determination device includes an input unit to which a device-condition signal output from an operating device is input, a first memory configured to store a determination algorithm for determining a predetermined abnormality of the operating device, an abnormality determination unit configured to determine the predetermined abnormality of the operating device in accordance with the device-condition signal and the determination algorithm, an output unit configured to output a determination result regarding the predetermined abnormality made by the abnormality determination unit to a user, an operation unit configured to receive an input of true-or-false information of whether the determination result is correct from the user, and an update unit configured to update the determination algorithm in accordance with the true-or-false information.