Aircraft Sensor Failure Localization Using Neural Network Comparison
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current aircraft systems lack efficient methods for detecting and localizing failures in sensors connected to flight control computers, limiting maintenance efficiency and increasing aircraft immobilization time due to the limited monitoring capabilities of flight control computers, which are primarily focused on detection rather than identification or localization of failures.
Innovation Solution
A method utilizing a module with electronic circuitry that conditions signals from and to sensors, employing a neural network and classifier to analyze and compare probabilities of failure, providing independent processing for failure detection and localization, thereby optimizing diagnosis and maintaining flight control processor capacity for safety and performance.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If flight control computer processor capacity is dedicated solely to aircraft control, then flight control safety and performance are improved, but failure detection and localization capability deteriorates
Solution Approach 1:
The system is divided into two independent parts: the flight control computer processor dedicated to control functions, and a separate failure detection module with its own processor for analyzing sensor signals and localizing failures. This segmentation allows each component to specialize in its function without interfering with the other, resolving the contradiction between control safety and failure detection capability.
Solution Approach 2:
A failure detection module acts as an intermediary between the sensors and the flight control computer. This intermediary analyzes sensor signals independently, providing failure localization information without requiring the main flight control processor to perform detection functions, thus maintaining both control safety and detection capability.
2Difficulty of detecting and measuring
If flight control computer performs both control and sensor monitoring, then failure detection capability is improved, but flight control performance deteriorates
Solution Approach 1:
The system separates control functions from monitoring functions into different processing units. The flight control computer processor handles control commands with high priority and speed, while the failure detection module independently processes sensor signals for failure detection, eliminating competition for processor resources and maintaining both detection capability and control performance.
Solution Approach 2:
The failure detection module serves as an intermediary that preprocesses sensor signals before they reach the flight control computer. By performing detection functions independently, it relieves the flight control processor from monitoring tasks, allowing the main processor to focus entirely on control operations with optimal performance.
3Device complexity
If maintenance operators manually search for failures, then system complexity is reduced, but maintenance time increases
Solution Approach 1:
The failure detection module provides self-service by automatically analyzing sensor signals and identifying failure locations without requiring manual intervention. The system independently generates diagnostic information that guides maintenance operators, reducing their workload and the time required to locate and repair failures, thus decreasing aircraft immobilization time.
Solution Approach 2:
The system performs preliminary failure detection and localization before maintenance operations begin. By pre-identifying the location and nature of failures through automated analysis, the system prepares diagnostic information in advance, allowing maintenance operators to directly address known issues rather than conducting extensive manual searches, thereby reducing maintenance time.
Data Source
AI summary
A method for detecting and localizing a failure of an input sensor of a flight control computer of an aircraft including comparing a first piece of information representative of a probability of occurrence of the failure, delivered by a neural network, with a second piece of information representative of a probability of occurrence of the failure, delivered by a classifier trained to carry out a detection and a localization of the failure. It is thus possible to optimize a failure diagnosis and a localization of a failure of an input sensor used in the determination of flight controls.