Gas Path Analysis Tuning for Multi-Time-Scale Engine Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current gas turbine engine parameter estimation and health monitoring systems face limitations due to the inability to accurately account for multiple sources of uncertainty and varying time scales, leading to suboptimal tuning of onboard models, which affects the accuracy of engine performance metrics such as thrust and fuel efficiency.
Innovation Solution
An engine parameter estimation tuning system that includes a performance health monitor unit and a gas path analysis unit, utilizing a high-fidelity engine model to separate long-term and short-term deterioration parameters, allowing independent adjustment and tuning of estimated engine parameters, thereby improving accuracy across different time scales.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional parameter estimation methods are used without separating time scales, then the system structure remains simple, but the accuracy of engine parameter estimation deteriorates due to inability to account for different uncertainty time scales
Solution Approach 1:
The patent segments the parameter estimation process into two distinct time-scale components: a long-term deterioration model that captures slow drift in engine performance, and a short-term tuning model that handles rapid fluctuations. This segmentation allows each model to be optimized for its specific time scale, improving overall estimation accuracy while maintaining manageable complexity through dedicated processing pathways for each time scale.
2Measurement precision
If long-term and short-term parameters are adjusted independently, then the accuracy of parameter estimation improves, but the complexity of the tuning system increases
Solution Approach 1:
The tuning system is segmented into independent long-term and short-term adjustment modules. The long-term model estimates gradual deterioration trends using historical data, while the short-term model handles immediate operational variations. This segmentation enables independent tuning of each parameter set without interference, improving accuracy while the modular architecture keeps system complexity manageable.
Solution Approach 2:
The patent introduces an intermediary layer that reconciles the long-term and short-term parameter estimates. This intermediary component processes outputs from both models and generates a unified set of tuned parameters, allowing independent adjustment of long-term and short-term parameters while maintaining system coherence and preventing excessive complexity.
3Reliability
If traditional single-model approach is used, then the ease of operation is maintained, but the reliability of performance monitoring deteriorates due to confounding effects
Solution Approach 1:
The monitoring system is segmented into separate long-term and short-term analysis pathways. The long-term model focuses on gradual performance degradation, while the short-term model captures operational variations. This segmentation eliminates confounding effects by attributing changes to the appropriate time scale, improving reliability while the automated processing maintains ease of operation.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An engine parameter estimation tuning system (100) includes an engine parameter estimator unit (120) and a gas path analysis (GPA) unit (110). The engine parameter estimator unit (120) includes an onboard model (OBM) (122) configured to output estimated parameters based on operation of a gas turbine engine (102). The gas path analysis (GPA) unit (110) includes a performance health monitor unit (130) configured to adjust a long-term deterioration parameter (158) independently from adjustment of a short-term tuning parameter (152) to tune one or more targeted estimation parameters included in the estimated engine parameter (Pest). In this manner, the engine parameter estimation tuning system (100) can realize the different time scales associated with uncertainties in an engine (102) and accommodate them separately so that the estimated engine parameters (Pest) become much more accurate.