Adaptive SPRT RUL Estimation Without Tripping Frequency Saturation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional RUL estimation techniques using SPRT tripping frequencies face challenges when tripping frequencies saturate, providing limited information on further degradation and leading to inaccurate forecasts, which is critical for operating assets without risking catastrophic failures, especially in industries where asset longevity is key.
Innovation Solution
The system employs an adaptive SPRT technique that adjusts sensitivity parameters to prevent saturation, using a logistic regression model to compute a risk index based on residuals and time-series signals, and incorporates a multivariate State Estimation Technique (MSET) for robust RUL estimation, allowing for continuous monitoring of asset degradation and proactive maintenance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional SPRT tripping frequency is used for RUL estimation, then early degradation detection is achieved, but saturation occurs providing limited information on further degradation
Solution Approach 1:
The patent applies dynamics by making the SPRT sensitivity parameters adaptive rather than fixed. The system dynamically adjusts the threshold and other parameters based on the current operational state and degradation level, allowing the monitoring system to remain sensitive across the entire degradation spectrum without saturating at any particular level.
Solution Approach 2:
The patent changes the parameters of the SPRT system from static to adaptive. By modifying sensitivity parameters based on observed degradation patterns and operational conditions, the system maintains optimal detection capability throughout the asset's lifecycle, preventing the information loss that occurs when fixed-parameter systems saturate.
2Measurement precision
If SPRT sensitivity is increased to detect subtle anomalies, then detection capability improves, but tripping frequency saturates reducing forecast accuracy
Solution Approach 1:
The system dynamically adjusts SPRT sensitivity parameters based on the current degradation state. When degradation is early-stage, higher sensitivity detects subtle anomalies. As degradation progresses, the parameters adapt to prevent saturation, maintaining reliable RUL forecasts throughout the asset's lifecycle rather than becoming overly sensitive and saturating.
Solution Approach 2:
The system uses feedback from ongoing monitoring to continuously adjust SPRT parameters. The tripping frequency and degradation patterns feed back into the parameter adjustment mechanism, allowing the system to self-tune its sensitivity to maintain both detection capability and forecast accuracy without manual intervention.
3Device complexity
If fixed SPRT parameters are used, then system complexity is reduced, but RUL estimation becomes inaccurate after saturation
Solution Approach 1:
The SPRT system performs self-service by automatically adjusting its own parameters based on observed data patterns. The adaptive parameter mechanism uses the monitoring data itself to tune the sensitivity and thresholds, eliminating the need for external calibration while maintaining accurate RUL estimation throughout the asset's operational lifecycle.
Solution Approach 2:
The patent transforms the parameter structure from fixed to adaptive, allowing parameters to change based on operational conditions and degradation patterns. This parameter evolution enables the system to maintain measurement precision across different stages of asset degradation without requiring complex external control mechanisms.
Data Source
AI summary
The system receives a set of present time-series signals gathered from sensors in the asset. Next, the system uses an inferential model to generate estimated values for the set of present time-series signals, and performs a pairwise differencing operation between actual values and the estimated values for the set of present time-series signals to produce residuals. The system then performs a sequential probability ratio test (SPRT) on the residuals to produce SPRT alarms with associated tripping frequency (TF). While the TF exceeds a TF threshold, the system iteratively adjusts sensitivity parameters for the SPRT to reduce the TF, and performs the SPRT again on the residuals. The system then uses a logistic regression model to compute a risk index for the asset based on the TF. If the risk index exceeds a threshold, the system generates a notification indicating that the asset needs to be replaced.