Aircraft Engine Flight Models for Transfer Function Change Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Aircraft engine monitoring systems struggle to detect changes in the transfer function, especially in unexplored flight regions, due to noise and lack of data coverage, leading to false alarms and inadequate robustness to external conditions.
Innovation Solution
A device and method that acquire and learn from flight data to create individual flight models, allowing for the estimation of engine output variables and associated errors using reference values, enabling detection of transfer function changes and avoiding false alarms by comparing flights with different variable distributions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If an overall model is used to monitor engine state by comparing with measured variables, then the monitoring system is simple to implement, but it cannot detect changes in transfer function in little explored or unexplored regions of the flight envelope due to noise and insufficient data coverage
Solution Approach 1:
The patent segments the monitoring approach by creating individual flight models for different flights instead of using a single overall model. Each individual model is trained on data from its specific flight, allowing it to capture flight-specific characteristics and transfer function changes accurately, even in previously unexplored regions of the flight envelope.
Solution Approach 2:
The patent applies preliminary action by training individual flight models on historical flight data before actual monitoring occurs. This pre-training enables the models to learn from past flights and be ready to detect transfer function changes in real-time during new flights, including in regions that may not have been extensively explored during training.
2Measurement precision
If monitoring is performed on modeling parameters or residuals, then the system can detect anomalies, but false alarms occur due to changes in flight envelope, environment, or input variables
Solution Approach 1:
The patent applies local quality by creating flight-specific models that are tailored to each individual flight's characteristics. Each model adapts to the specific flight envelope, environmental conditions, and input variable distributions of its training flight, making it robust to variations and preventing false alarms that would occur with a generic overall model.
3Productivity
If data from remote flights is used to estimate parameters or residuals, then monitoring can be performed with limited current flight data, but detection capability is reduced when no data or little data exists over part of the flight envelope
Solution Approach 1:
The patent uses preliminary action by extensively training individual flight models on historical flight data before deployment. This pre-training ensures that when actual monitoring occurs with limited current flight data, the models already have learned representations that enable accurate detection even in regions where current flight data is sparse.
Data Source
AI summary
A device for monitoring a state of a propulsion engine includes an acquisition module that acquires data of flights of the propulsion engine, comprising, for each flight, values of input variables, environment variables, and output variables of the propulsion engine during the flight, a learning module that computes, by learning from the data of each flight, an individual flight model for the flight, a using module that computes, for each flight, estimates of the values of the output variables, by applying the individual flight model to reference values of the input variables and the environment variables, and an error associated with the estimates of the values of the output variables that is obtained by applying the individual flight model to the reference values of the input variables and the environment variables. The reference values belong to a set of reference data, which are identical for the individual flight models.