LRU Failure Diagnosis Using Sensor Lookup Probability Ranking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Aircraft maintenance crews face challenges in accurately diagnosing the cause of ambiguous symptoms of wear or failure in aircraft systems, often leading to unnecessary part replacements due to unclear fault identification.
Innovation Solution
A system comprising a controller connected to sensors that monitors operating conditions and uses a lookup table populated with domain expertise and reliability data to predict the likelihood of Line Replaceable Unit (LRU) failures, allowing for precise identification of faulty components.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If ground crews manually diagnose component failures based on ambiguous symptoms, then repair decisions can be made, but inaccurate diagnosis leads to unnecessary part replacements and increased costs
Solution Approach 1:
The patent introduces a diagnostic system as an intermediary between the ambiguous symptoms and the repair decision. This system uses sensors to collect data, a lookup table to map symptoms to potential failures, and a normalization process to rank likelihoods, thereby mediating the diagnosis process to improve accuracy and reduce unnecessary replacements.
Solution Approach 2:
The system implements feedback by continuously monitoring aircraft system data through sensors and using this information to update the diagnostic assessment. The normalization process provides feedback on the likelihood of each potential failure, allowing ground crews to make informed decisions based on quantified probabilities rather than ambiguous symptoms alone.
2Reliability
If ground crews replace multiple components to ensure the actual problem is fixed, then repair reliability improves, but downtime and repair costs increase
Solution Approach 1:
Instead of replacing all potentially faulty components (excessive action), the system uses the normalized likelihood rankings to identify and focus on the most probable failure causes (partial action). This approach maintains repair reliability by targeting the most likely issues while avoiding unnecessary replacements that would increase downtime and costs.
Solution Approach 2:
The lookup table is pre-populated with domain expertise and reliability data, and the system performs preliminary analysis of symptoms against this database before the actual repair decision is made. This preliminary action provides a ranked list of likely failures, enabling ground crews to prepare targeted repair plans that reduce both downtime and unnecessary part replacement.
3Difficulty of detecting and measuring
If comprehensive monitoring of all aircraft systems is implemented, then fault detection capability improves, but system complexity and diagnostic difficulty increase
Solution Approach 1:
The patent segments the diagnostic process into distinct components: sensor data collection, lookup table reference, likelihood calculation, and normalization ranking. This segmentation breaks down the complex task of diagnosing aircraft failures into manageable steps, improving fault detection capability while making the overall system more manageable despite the comprehensive monitoring required.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A system and method of identifying part failures is provided. The system may comprise monitoring a sensor (102, 104, 106, 108), detecting a failure mode, selecting a row from a lookup table corresponding to the failure mode, looking up a line replaceable unit (LRU) failure probability for the detected failure mode, and reporting the LRU failure probability for the detected failure mode.