Rotating Machinery Damage Assessment for Remaining Useful Life
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for determining damage and remaining useful life of rotating machinery, such as wind and water turbines, are inadequate due to reliance on assumed operating profiles, leading to inaccurate damage progression calculations and frequent failures within four to five years, despite a design life of over twenty years, and vibration analysis is prone to false alarms and lacks discrimination between indicative and non-indicative vibrations.
Innovation Solution
A method that calculates accumulated damage by using data from sensors monitoring operating conditions, including steady-state and transient states, and applying damage coefficients derived from both rated and field operating conditions, combined with a model-based approach to estimate component loads and predict remaining useful life, while employing meta-models for efficient damage prediction and maintenance scheduling.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If vibration threshold is set low to detect early damage, then measurement precision improves, but false alarms increase reducing reliability
Solution Approach 1:
The patent transforms the vibration monitoring approach by changing from absolute threshold monitoring to relative condition-based monitoring. It introduces damage coefficients that relate vibration magnitude to actual component damage, allowing the system to adapt thresholds dynamically based on operating conditions and component history rather than using fixed low thresholds that cause false alarms.
Solution Approach 2:
The patent introduces damage coefficients as an intermediary parameter between vibration measurement and damage assessment. These coefficients act as a mediator that translates raw vibration data into meaningful damage indicators, filtering out false alarms by considering the relationship between vibration characteristics and actual component degradation.
2Device complexity
If assumed operating profiles are used for damage calculation, then device complexity is reduced, but measurement precision of damage progression deteriorates
Solution Approach 1:
The patent implements feedback by continuously monitoring actual operating conditions and using this information to update damage calculations. Instead of relying on assumed profiles, the system feeds back real vibration data, operating parameters, and component response to refine damage coefficient estimates, creating a closed-loop system that adapts to actual component behavior.
Solution Approach 2:
The patent transitions from static assumed operating profiles to dynamic condition-based monitoring. The system continuously adapts its damage assessment based on real-time operating conditions, component history, and observed vibration patterns, making the damage calculation methodology flexible and responsive to actual component states rather than relying on fixed assumptions.
Data Source
Figure 1
Figure 2
Figure 3~4
AI summary
A method for predicting remaining useful life of a wind or water turbine or component determines in step (116) an accumulated damage for the turbine or component and compares this in step (118) to preset damage limit obtained in step (114). This provides a simple approach to estimating remaining useful life, giving the turbine operator an indication of the condition of turbines or farms under management.