Aircraft Sensor Segmentation for Targeted Predictive Maintenance
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current aircraft maintenance methods require extensive time and effort to determine if specific portions of an aircraft need maintenance, as they assess the entire aircraft rather than distinct operational components, leading to inefficiencies and increased costs.
Innovation Solution
A system that uses machine learning models trained on tailored sensor data from operationally distinct portions of an aircraft, allowing for targeted maintenance determinations by selecting groupings of sensors and parameters corresponding to unique forces and functions of each aircraft component, reducing the amount of data needed for training and improving efficiency.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If sensor data is collected from the entire aircraft to assess overall performance, then comprehensive monitoring is achieved, but the ability to identify specific portions requiring maintenance is lost
Solution Approach 1:
The patent divides the aircraft into multiple operationally distinct portions (wings, fuselage, tail, etc.) and assigns separate sensor groups to each portion. This segmentation allows the system to monitor each portion independently and identify specific areas requiring maintenance, resolving the contradiction between comprehensive monitoring and specific location identification.
Solution Approach 2:
The patent implements local quality by tailoring sensor groups and machine learning models to each specific aircraft portion's operational characteristics and expected forces. Each portion has customized monitoring parameters (e.g., G-forces for wings, vertical forces for fuselage), enabling precise local assessment while contributing to overall aircraft health monitoring.
2Measurement precision
If extensive sensor data from all aircraft portions is analyzed, then accurate maintenance determination is achieved, but computational time and processing power increase
Solution Approach 1:
The patent segments the data processing task by analyzing each aircraft portion independently using dedicated machine learning models. This allows parallel processing of multiple portions simultaneously and reduces the computational complexity compared to analyzing all sensor data from the entire aircraft as a single unified system.
Solution Approach 2:
The patent transforms raw sensor data into operationally relevant parameters specific to each aircraft portion (e.g., G-forces, vertical forces, lateral forces) before analysis. This parameter transformation reduces data dimensionality and focuses computational resources on the most relevant features for maintenance determination, improving both accuracy and efficiency.
3Adaptability or versatility
If machine learning models are trained on all sensor data from the aircraft, then comprehensive predictive capability is achieved, but training data requirements and computational resources increase
Solution Approach 1:
The patent trains separate machine learning models for each operationally distinct aircraft portion using sensor data specific to that portion. This segmentation reduces the volume of training data required for each model compared to training a single comprehensive model on all aircraft data, while maintaining predictive capability for each specific portion.
Solution Approach 2:
The patent applies local quality by customizing machine learning models to match the specific operational characteristics and sensor groups of each aircraft portion. Each model is trained on locally relevant data with parameters tailored to that portion's expected forces and operational context, improving training efficiency and model accuracy while reducing overall data requirements.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
In an example, a method for determining whether to perform aircraft maintenance is described. The method comprises selecting groupings of sensors and/or parameters associated with an aircraft type. The method comprises receiving feature data that corresponds to each grouping of sensors and/or parameters. The method comprises determining, from the feature data, values for predetermined operational metrics. The method comprises comparing the values to values for predetermined operational metrics that correspond to at least one other flight of the aircraft. The method comprises determining, based on comparing the values for the predetermined operational metrics, values of additional operational metrics. The method comprises training a machine learning model using at least the values for the additional operational metrics. The method comprises providing, based on an output of the machine learning model, an instruction to perform maintenance on one or more of the plurality of portions of the aircraft or another aircraft.