Equipment RUL Estimation with PCA Health Indicators and Adaptive Thresholds
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for estimating the remaining useful life (RUL) of equipment fail to account for dynamic degradation characteristics and environmental factors, often relying on static thresholds that are not unique to each component, leading to inaccurate predictions.
Innovation Solution
A method that uses sensor data to extract features, obtain a health indicator through principal component analysis, determine a critical time for degradation initiation, and predict a future degradation curve to establish a dynamic failure threshold based on degradation parameters, adapting to the unique characteristics of each equipment.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of manufacture
If a fixed static failure threshold is used based on prior knowledge, then the method is simple to implement, but the prediction accuracy deteriorates because it does not account for unique component degradation characteristics
Solution Approach 1:
The patent transforms the static failure threshold into a dynamic one by making it adaptable to each component's unique degradation characteristics. The dynamic threshold is determined through continuous monitoring of degradation parameters and environmental factors, allowing the threshold to evolve over time rather than remaining fixed. This resolves the contradiction by maintaining implementation feasibility while significantly improving prediction accuracy through adaptability.
Solution Approach 2:
The patent changes the parameter of the failure threshold from a fixed value to a dynamic variable that depends on degradation parameters and environmental conditions. By introducing these additional parameters and allowing the threshold to vary based on component-specific degradation patterns, the system achieves both practical implementability and high prediction accuracy.
2Measurement precision
If similarity-based approach is used with run-to-failure data, then the threshold determination is data-driven, but the method becomes infeasible in real-life scenarios where failure characteristics are not available
Solution Approach 1:
The patent enables the system to determine its own failure threshold without relying on external run-to-failure data from similar components. Each component serves itself by generating its unique threshold based on its own degradation parameters and environmental factors. This self-determination capability makes the method universally applicable in real-life scenarios while maintaining data-driven accuracy.
Solution Approach 2:
The patent applies local quality by determining failure thresholds specific to each individual component rather than using a general threshold for all components. Each component develops its own unique threshold based on its local degradation characteristics and environmental conditions, enabling the method to work effectively in real-life scenarios where component uniqueness cannot be ignored.
3Measurement precision
If prior knowledge of failure characteristics is required, then the threshold can be determined, but the method loses versatility as it cannot be applied when such knowledge is unavailable
Solution Approach 1:
The patent eliminates the dependency on prior knowledge by enabling each component to determine its own failure threshold through self-monitoring of degradation parameters. The component uses its own operational data and environmental factors to establish its threshold, making the method universally applicable regardless of whether prior failure knowledge exists.
Solution Approach 2:
The patent performs preliminary action by continuously monitoring and analyzing degradation parameters during the component's operation to determine the failure threshold before actual failure occurs. This proactive determination of thresholds based on real-time data enables the method to work without requiring prior failure characteristics.
4Measurement precision
If environmental factors and component uniqueness are considered, then prediction accuracy improves, but the system complexity increases
Solution Approach 1:
The patent manages system complexity by systematically incorporating environmental factors and component uniqueness through defined degradation parameters. Rather than adding arbitrary complexity, the system introduces specific parameters that capture these variations, allowing accurate prediction while maintaining a structured and manageable framework.
Data Source
AI summary
A system to predict a remaining useful life of an equipment includes a processor configured to receive signals from sensors of the equipment, extract features from sensor data of the sensors, and obtain a health indicator from the extracted features by principal component analysis. The processor is further configured to determine a critical time beyond which the degradation initiates in the equipment using pautas criteria, predict a future degradation curve, and determine a dynamic failure threshold based on degradation characteristics of the equipment. The dynamic failure threshold is determined in real time based on degradation parameters unique to the equipment.