Brake Release Abnormality Detection from Motor Time-Series Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing abnormality determination systems for motor drive mechanisms face challenges in accurately detecting abnormalities in the releasing operation of brakes, particularly when the variation in time-series data is small during abnormal conditions, and may require additional detection circuits, increasing costs and complexity.
Innovation Solution
An abnormality determination system that acquires time-series data before and after the motor command is output, using sample data created during abnormal events to determine brake operation abnormalities, and employs Mahalanobis distance calculations to differentiate between normal and abnormal states without additional detection circuits.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If sample data is created based on normal operation time-series data, then the abnormality determination system can detect abnormalities, but it fails to accurately detect abnormalities when the variation in time-series data is small during abnormal conditions
Solution Approach 1:
Instead of creating sample data from normal operation states and detecting deviations, the patent inverts the approach by creating sample data from abnormal operation states. The abnormality determination circuitry then determines whether current time-series data matches these abnormal sample data patterns, enabling accurate detection even when abnormal variations are small.
Solution Approach 2:
The patent changes the fundamental parameter used for abnormality determination from 'deviation from normal' to 'similarity to abnormal patterns'. By transforming the comparison basis from normal-state-centered to abnormal-state-centered, the system becomes sensitive to subtle abnormal variations that would be invisible in normal-state comparisons.
2Measurement precision
If additional detection circuits are added to improve brake operation detection, then detection capability is enhanced, but system cost and complexity increase
Solution Approach 1:
The patent makes the existing time-series data acquisition system multi-functional by enabling it to detect brake operation abnormalities through pattern matching against abnormal sample data. The same data acquisition circuitry that collects general operation data is now also used for specific brake abnormality detection, eliminating the need for separate detection circuits.
Solution Approach 2:
Instead of adding physical detection circuits, the patent creates virtual copies of abnormal brake operation patterns in the form of sample data. These digital copies serve as reference patterns that the system compares against current operation data, providing detection capability without additional hardware.
3Measurement precision
If Mahalanobis distance calculation is used to differentiate normal and abnormal states, then statistical analysis is performed, but false negatives occur when abnormal data variation is smaller than normal data variation
Solution Approach 1:
The patent inverts the Mahalanobis distance application by calculating the distance from current data to abnormal sample data patterns rather than from normal patterns. This reversal changes the detection logic to identify when data matches abnormal patterns, even if those patterns show small variations, thereby reducing false negatives.
Data Source
AI summary
An abnormality determination system includes first data acquisition circuitry configured to acquire time-series data relating to an operation of a device, sample data creation circuitry configured to create sample data based on abnormality time-series data which the first data acquisition circuitry acquires while an abnormality occurs in the operation of the device, and first abnormality determination circuitry configured to determine the abnormality in the operation of the device based on the time-series data and the sample data.