Self-Calibrating Fault Detectors for Asset Surveillance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing asset surveillance systems face challenges with static fault detectors that do not adjust for asset aging, leading to false alarms and missed alarms, and require manual recalibration, which is impractical for in-service assets.
Innovation Solution
A self-calibrating asset surveillance system that automatically adjusts fault detectors using statistical methods, such as sequential discounting expectation maximization, to differentiate between normal aging and failure data, reducing false and missed alarms without manual intervention.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If static fault detectors are used for monitoring aging assets, then the system structure remains simple, but false alarms increase and fault detection accuracy deteriorates
Solution Approach 1:
The patent implements dynamic fault detectors that automatically adapt their calibration parameters over time based on asset aging patterns. The system transitions from static thresholds to dynamic thresholds that evolve with the asset lifecycle, using sequential updating algorithms to maintain optimal detection sensitivity without manual intervention.
Solution Approach 2:
The fault detection system performs self-calibration by automatically learning asset aging patterns from historical data and adjusting its own parameters. The system uses unsupervised learning algorithms to identify normal aging behavior versus actual faults, eliminating the need for manual recalibration by operators.
2Reliability
If manual recalibration is performed periodically to maintain sensitivity, then false alarms are reduced, but operational time is lost and the process becomes impractical for in-service assets
Solution Approach 1:
The system implements continuous automatic recalibration that operates in the background without interrupting asset monitoring or requiring operational downtime. The calibration process runs continuously, updating fault detection parameters in real-time based on incoming asset data, ensuring uninterrupted surveillance.
Solution Approach 2:
The fault detection system autonomously performs its own recalibration using automated algorithms that learn from asset data. The system eliminates dependence on human operators by automatically adjusting sensitivity parameters, selecting appropriate calibration data, and updating detection thresholds without manual intervention.
3Object-generated harmful factors
If sensitivity is reduced to accommodate aging behavior, then false alarms decrease, but missed alarms increase and fault identification is delayed
Solution Approach 1:
The system employs dynamic sensitivity adjustment where detection thresholds adapt continuously to asset aging patterns. Rather than using fixed reduced sensitivity, the system maintains high sensitivity by dynamically distinguishing between normal aging variations and actual fault conditions through learned patterns and statistical methods.
Solution Approach 2:
The fault detection system applies different sensitivity levels to different aspects of asset monitoring. It uses localized adaptation where specific parameters are adjusted based on their individual aging patterns, allowing high sensitivity for critical fault detection while maintaining appropriate thresholds for normal variations.
4Ease of manufacture
If manual recalibration is required, then calibration can be performed, but the asset monitoring process must be suspended and operator intervention is needed
Solution Approach 1:
The system autonomously performs all calibration operations without requiring operator intervention. Automated algorithms handle data selection, parameter adjustment, and validation, making the calibration process independent of human operators and suitable for remote or continuous monitoring applications.
Solution Approach 2:
The monitoring process continues uninterrupted during calibration operations. The system performs real-time updates in the background while maintaining continuous asset surveillance, eliminating the need to suspend monitoring activities that would occur with manual recalibration procedures.
Data Source
AI summary
A computer-implemented asset surveillance system and method for self calibrating at least one fault detector providing asset surveillance by calibrating at least the one fault detector with statistics associated with expected asset behavior, acquiring observed data values from an asset, screening the observed data values based upon at least one defined criterion for obtaining screened data values, updating the statistics associated with expected asset behavior as a function of the screened data values for defining updated statistics, and recalibrating the at least one fault detector with the updated statistics.