Adaptive RUL Estimation for Load-Bearing Structure Degradation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for predicting the remaining useful life (RUL) of load-bearing structures or equipment fail to provide reliable threshold values and probabilistic quantification, leading to potential premature shutdown or failure due to unreliable degradation measurement estimates.
Innovation Solution
A system employing a constraint convex regression model and particle-filtering techniques, including Kalman filtering, to estimate the total useful life (TUL) and predict the remaining useful life (RUL) based on degradation data from sensors, with the ability to recalibrate measurements and update estimates over time.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional degradation measurement with predetermined threshold is used, then system operation can be monitored, but reliable threshold value cannot be determined leading to premature shutdown or system failure
Solution Approach 1:
The patent transforms the fixed predetermined threshold approach into a dynamic threshold determined by constraint convex regression. The degradation threshold is no longer a static value but is adaptively calculated based on the fitted degradation curve and its statistical properties, allowing the system to distinguish between normal degradation fluctuations and true failure precursors.
Solution Approach 2:
The patent replaces the simple threshold comparison mechanism with a sophisticated statistical modeling approach. Instead of directly comparing degradation measurements against a fixed threshold, the system uses constraint convex regression to fit a degradation curve, calculates confidence intervals, and determines thresholds based on statistical significance, thereby substituting mechanical thresholding with statistical inference.
2Reliability
If conventional degradation measurement is used, then degradation can be tracked, but probabilistic quantification of system RUL cannot be provided
Solution Approach 1:
The patent implements feedback by continuously updating the degradation model with new measurements and recalculating the RUL probability distribution. The system uses the fitted degradation curve and its confidence intervals to provide feedback on the uncertainty of RUL predictions, allowing for dynamic adjustment of maintenance strategies based on the evolving probabilistic state of the system.
Solution Approach 2:
The patent transforms the static RUL prediction into a dynamic probabilistic distribution. Instead of providing a single fixed RUL value, the system continuously updates the probability distribution of RUL based on new degradation measurements and the evolving degradation curve, capturing the uncertainty and variability in the failure process.
3Reliability
If continuous monitoring is performed to avoid failure, then system safety is improved, but premature shutdown may occur due to unreliable threshold estimates
Solution Approach 1:
The patent performs preliminary action by fitting the degradation curve and calculating confidence intervals before making shutdown decisions. The system uses the constraint convex regression model to predict future degradation trajectories and assess the probability of exceeding failure thresholds, allowing for proactive maintenance planning that avoids both premature shutdown and unexpected failure.
Solution Approach 2:
The patent changes the decision-making parameter from a fixed threshold to a dynamic, statistically-derived threshold based on the degradation curve and its confidence intervals. This allows the system to adjust the effective threshold adaptively, maintaining safety while reducing premature shutdowns caused by unreliable fixed thresholds.
Data Source
AI summary
One embodiment can provide a system for estimating useful life of a load-bearing structure. During operation, the system can perform a degradation measurement on the structure to obtain degradation data for a predetermined time interval, apply a constraint convex regression model to the degradation data, estimate a total useful life (TUL) of the structure based on outputs of the constraint convex regression model, and predicting a remaining useful life (RUL) based on the TUL and a current time.