Gas Turbine Engine Model Calibration with Deterioration Separation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional dynamic model calibration techniques for gas turbine engines face challenges such as requiring large amounts of field data for model training, which delays engine entry into service, and self-tuning capabilities diminish as engines deteriorate, leading to reduced accuracy and reliability in health assessment.
Innovation Solution
A self-calibrating dynamic model that separates long-term deterioration from self-tuning functions, allowing for real-time calibration and maintenance of short-term tuning parameters, ensuring the self-tuning capability remains unaffected by engine deterioration and can be used upon engine entry into service.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If conventional self-tuning model calibration is used, then the model adapts to individual engine characteristics, but the self-tuning capability diminishes as the engine deteriorates, reducing accuracy
Solution Approach 1:
The patent segments the tuning parameters into two distinct components: short-term tuning parameters that capture transient operational variations, and long-term deterioration parameters that track progressive engine degradation. This segmentation allows the model to maintain accurate health assessment by separately managing adaptation to temporary conditions versus permanent degradation, preventing the conflation of deterioration with normal operational variability.
Solution Approach 2:
The patent implements dynamic adjustment of the model based on engine operating conditions. The system dynamically switches between using short-term tuning parameters for transient conditions and long-term deterioration parameters for sustained degradation, allowing the model to adapt its behavior to the current engine state and maintain reliability throughout the engine lifecycle.
2Measurement precision
If large amounts of field data are used for model training, then the model accuracy improves, but the time required for calibration increases, delaying engine entry into service
Solution Approach 1:
The patent performs preliminary calibration actions during factory testing and initial setup, establishing baseline model parameters before the engine enters service. This preliminary action reduces the need for extensive field data collection and iterative calibration, allowing the engine to be deployed with a pre-calibrated model that requires minimal field adjustment.
Solution Approach 2:
The system implements self-service calibration capabilities that automatically adjust the model using onboard sensors and embedded algorithms. The engine monitoring system continuously calibrates itself using operational data, eliminating the need for manual calibration processes and external expertise, thereby reducing calibration time while maintaining accuracy.
3Adaptability or versatility
If the model continuously adapts to engine deterioration, then it tracks current engine state, but it loses the ability to distinguish deterioration from normal operational variations
Solution Approach 1:
The patent segments the adaptation process into distinct pathways: one for capturing normal operational variations through short-term tuning parameters, and another for tracking genuine deterioration through long-term parameters. This segmentation preserves the ability to distinguish between temporary operational changes and permanent degradation by directing each type of change to the appropriate parameter set.
Solution Approach 2:
The system dynamically determines which parameter set to update based on the nature of the observed change. When transient operational variations are detected, only short-term parameters are adjusted; when sustained degradation patterns are identified, long-term parameters are updated. This dynamic approach prevents the model from losing deterioration detection capability while maintaining adaptability to normal variations.
Data Source
AI summary
An engine model calibration system includes a gas turbine engine, a sensor, a nominal engine estimation module for FD&I, and a self-tuning engine estimation module for engine parameter estimation. The nominal engine estimation module includes an open-loop nominal engine model configured to output estimated nominal engine parameters indicative of the one or more engine operating parameters, and selectively updates the open-loop nominal engine model based on one or more identified engine operating conditions. The engine model calibration system can perform calibration of a nominal engine model and a self-tuning engine model as they are used for FD&I and engine parameter estimation. The self-calibrating dynamic model also can separate the long-term tuning parameters and the short-term tuning parameters from one another so that the self-tuning capability for the self-tuning engine model is not compromised or diminished as the engine deteriorates.