Aircraft Engine Fault Diagnosis Using Graph-Based Flight Data
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
The analysis of flight data for aircraft engines is complex and time-consuming, requiring improvements in diagnostic methods to efficiently identify faults during flight tests.
Innovation Solution
A method and system utilizing a graph-based representation and machine-learning algorithms to process flight data from multiple sensors, detecting faults by analyzing statistical deviations and determining the source of discrepancies in real-time, including identifying broken or incorrectly installed sensors and incorrect flight maneuvers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional manual analysis methods are used for flight data, then measurement precision can be maintained, but productivity is reduced and time consumption increases
Solution Approach 1:
The patent replaces manual mechanical analysis with automated electronic data processing systems. Flight data is automatically collected from sensors, processed through computational algorithms, and analyzed by computer systems to detect faults, replacing the traditional manual examination process while maintaining detection accuracy and significantly improving analysis efficiency
Solution Approach 2:
The system enables self-diagnostic capabilities where the flight data analysis system automatically processes and analyzes its own data without external intervention. The automated system collects, processes, and interprets flight data independently, identifying faults and generating reports without requiring manual analysis, thereby improving productivity while maintaining precision
2Measurement precision
If comprehensive flight data from multiple sensors is collected, then measurement precision improves, but device complexity increases
Solution Approach 1:
The patent segments the complex data processing task into distinct functional modules: data collection from multiple sensors, data validation and preprocessing, fault detection algorithms, and result generation. This modular segmentation allows comprehensive multi-sensor data to be processed systematically, improving detection precision while managing system complexity through organized functional breakdown
Solution Approach 2:
The patent implements a universal data processing framework that handles multiple sensor types and fault conditions through a single integrated system. The same processing architecture accommodates various sensor inputs and fault scenarios, reducing overall system complexity compared to having separate dedicated systems for each function while maintaining high detection precision across all parameters
3Productivity
If automated data processing is implemented, then productivity increases, but measurement precision may deteriorate due to processing errors
Solution Approach 1:
The patent incorporates feedback mechanisms where the automated processing system continuously monitors its own outputs and compares results against expected parameters. Validation algorithms verify data integrity at each processing stage, and the system can detect and correct processing errors, ensuring that automated high-speed processing maintains the same precision as manual analysis
Solution Approach 2:
The patent performs preliminary data validation and preprocessing before main fault detection analysis. Flight data is checked for completeness and consistency upfront, and invalid data is flagged or corrected before processing. This preliminary action prevents processing errors that could compromise precision while enabling rapid automated analysis of valid data
Data Source
Figure 1
Figure 2
Figure 3A~3D
AI summary
Systems and methods for diagnosing an engine (10) or an aircraft are described herein. Flight data of at least one of the engine (10) and the aircraft is obtained. A graph-based representation modeling a mathematical relationship between parameters of at least one of the engine (10) and the aircraft is obtained. The graph-based representation has a plurality of permutations (3001, 3002, 3003, 3004). Output data for the plurality of permutations (3001, 3002, 3003, 3004) is generated based on the flight data. The output data for the plurality of permutations (3001, 3002, 3003, 3004) is compared and a fault is detected based on a discrepancy in the output data. A signal indicative of the fault is outputted in response to detecting the fault.