Aircraft Performance Modeling for Real-Time In-Flight Parameter Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Aircraft performance parameters are not available in real-time during flight, hindering pilot decision-making, and their calculation is time-intensive, relying on data that cannot be determined during flight.
Innovation Solution
An on-board computing device equipped with an aircraft performance model based on historical flight data and a recurrent neural network layer processes real-time time-series flight data to generate aircraft performance parameters, including recommended trim settings, enabling real-time computation and display.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional calculation methods are used to determine aircraft performance parameters, then accuracy can be maintained, but real-time availability is lost due to time-intensive calculations
Solution Approach 1:
The system performs preliminary actions by training the neural network model offline using historical flight data before actual flight operations. The model learns complex aerodynamic relationships in advance, enabling real-time predictions during flight without performing time-intensive calculations. This resolves the contradiction by shifting computational burden from runtime to training time.
Solution Approach 2:
The patent replaces traditional mechanical calculation methods (based on fundamental aerodynamic equations and iterative computations) with an intelligent system using neural networks. The neural network substitutes complex mathematical computations with pattern recognition based on learned relationships, achieving both speed and accuracy requirements.
2Ease of operation
If real-time performance parameters are computed using traditional methods, then pilot decision-making is improved, but computational resources are excessively consumed
Solution Approach 1:
The neural network model is trained offline using historical flight data, performing the heavy computational work before deployment. During actual flight operations, the pre-trained model requires minimal computational resources to generate real-time predictions, thus improving pilot decision-making while consuming minimal onboard computational energy.
Solution Approach 2:
The system creates a simplified computational copy of complex aerodynamic relationships through the neural network model. Instead of performing full physics-based calculations onboard, the system uses the trained network as a lightweight surrogate model that replicates performance parameter predictions with fraction of the computational cost.
3Measurement precision
If comprehensive flight data is collected for accurate parameter calculation, then measurement accuracy is improved, but data processing time increases
Solution Approach 1:
The neural network substitutes traditional data processing methods that require iterative calculations with a pattern recognition system. The network has learned to map input flight parameters directly to performance outcomes, eliminating the need for time-consuming computational algorithms while maintaining accuracy based on comprehensive historical data patterns.
Solution Approach 2:
The system performs preliminary analysis of comprehensive flight data during the offline training phase, extracting and encoding complex relationships in advance. During real-time operation, the pre-processed knowledge enables rapid predictions without re-processing the full data set, thus maintaining accuracy while minimizing processing time.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A device includes a memory, a network interface, and a processor. The memory is configured to store an aircraft performance model. The aircraft performance model is based on historical flight data of one or more aircraft. The aircraft performance model includes a recurrent neural network layer. The network interface is configured to receive real-time time-series flight data from a data bus of a first aircraft. The processor is configured to receive, via the network interface, the real-time time-series flight data. The processor is also configured to generate, based on the real-time time-series flight data and the aircraft performance model, one or more aircraft performance parameters. The processor is further configured to provide the aircraft performance parameters to a display device.