Dynamic Threshold Defect Detection for Rotating Arrangements
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting defects in rotating arrangements, such as bearings, face challenges in setting thresholds due to unstable parameter trends caused by vibrations, leading to either false positives or false negatives.
Innovation Solution
A device with a calculation unit that sets a threshold based on the mean value and standard deviation of measured parameters during a learning period, adjusted by a stability factor and alarm sensitivity factor to balance false positives and negatives, ensuring accurate defect detection.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the threshold is set close to the mean value to improve defect detection sensitivity, then the risk of false negatives is reduced, but the risk of false positives increases due to small deviations exceeding the threshold
Solution Approach 1:
The threshold is made dynamic by introducing a stability factor that adapts to the measured standard deviation. Instead of a fixed threshold close to the mean, the threshold dynamically adjusts based on the actual variability of the parameter, allowing it to be closer to the mean when stable and farther when unstable, thus resolving the contradiction between detection sensitivity and false positive rate
Solution Approach 2:
The invention changes the parameter of the threshold by introducing a multiplicative stability factor that modifies the base threshold calculation. This factor is derived from the standard deviation and mean value, transforming the threshold from a static value to one that adapts to changing conditions, thereby balancing sensitivity and reliability
2Reliability
If the threshold is set high to avoid false positives caused by vibrations, then the reliability increases, but the risk of false negatives increases as defects may not be detected in time
Solution Approach 1:
The threshold dynamically adapts to vibration levels through the stability factor, which is calculated from the standard deviation. When vibrations cause high variability, the threshold automatically increases to avoid false positives, and when conditions are stable, it decreases to maintain high detection sensitivity, thus resolving the contradiction between reliability and detection precision
Solution Approach 2:
The system uses feedback from the measured parameter values to continuously adjust the threshold. The standard deviation and mean value of recent measurements feed into the stability factor calculation, which in turn adjusts the threshold, creating a closed-loop system that adapts to changing vibration conditions and maintains optimal detection performance
3Measurement precision
If a learning period is used to initialize the threshold, then the threshold setting becomes more accurate, but the time required before reliable defect detection begins increases
Solution Approach 1:
The system performs preliminary measurements during a learning period to establish the mean and standard deviation, but the threshold becomes operational immediately with a default stability factor. This allows the system to start detecting defects right away while continuing to refine the threshold accuracy during the learning period, thus reducing the time loss without sacrificing precision
Data Source
AI summary
A device for detecting a defect of a rotating arrangement is provided. The rotating arrangement further provides one or more rotating elements. The device includes a measuring unit for measuring a parameter of the rotating arrangement for a predefined time period, a calculation unit for calculating a mean value and a standard deviation of the parameter of the predefined time period and for calculating a threshold based on a sum of the mean value and the standard deviation during the predefined time period, and a detection unit for detecting a defect by comparing a current value of the parameter with the calculated threshold. The calculation unit is configured to adjust the calculated threshold using a stability factor. The stability factor is a variable function of the standard deviation and the mean value. A method for detecting a defect of a rotating arrangement is also provided.