Vehicle Fault Validation Using Physics-Based and Data-Driven Models
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current vehicle health monitoring systems generate nuisance fault indications due to environmental conditions, leading to unnecessary component removals and increased maintenance costs, as they do not consider the operating environment or component condition.
Innovation Solution
A combination of physics-based and data-driven models is used to predict the behavior of electronic components under various environmental conditions, assess the validity of fault indications, and reduce nuisance faults by analyzing historical maintenance data.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If fixed threshold levels are used for fault detection, then the system is simple to operate, but nuisance fault indications are generated under extreme environmental conditions
Solution Approach 1:
The patent implements dynamic threshold adjustment by continuously monitoring environmental parameters (temperature, humidity, vibration) and adapting the fault detection thresholds in real-time. Instead of using fixed thresholds, the system dynamically modifies acceptable parameter ranges based on current environmental conditions, preventing nuisance faults while maintaining operational simplicity.
Solution Approach 2:
The system changes the detection parameters by incorporating environmental context into the fault detection algorithm. When extreme environmental conditions are detected, the system adjusts the acceptable parameter ranges for voltage, current, and other electrical measurements, allowing the same component to have different threshold criteria under different environmental conditions.
2Reliability
If environmental conditions are considered in fault detection, then fault indication accuracy is improved, but system complexity increases
Solution Approach 1:
The patent makes the existing monitoring system multi-functional by enabling it to perform both traditional fault detection and environmental condition assessment using the same hardware infrastructure. The system universally handles multiple functions: monitoring electrical parameters, tracking environmental conditions, and adjusting detection criteria, all within a single integrated platform.
Solution Approach 2:
The system introduces an intermediary layer (software algorithm or processing unit) that sits between the raw sensor data and the fault detection logic. This intermediary processes environmental data and translates it into adjusted threshold criteria, shielding the complexity from the user while maintaining accurate fault detection.
3Reliability
If components are removed for testing when fault indications occur, then potential faults are identified, but vehicle inoperability increases
Solution Approach 1:
The system performs preliminary analysis of fault indications by evaluating environmental context and component history before triggering removal requests. By pre-assessing whether a fault indication is likely to be spurious based on environmental conditions and patterns, the system avoids unnecessary component removals and keeps vehicles operational longer.
Solution Approach 2:
The system implements feedback loops that learn from testing outcomes. When components are tested after removal, the results feed back into the system to refine future fault assessments. This feedback mechanism improves the accuracy of preliminary assessments, reducing unnecessary removals over time as the system becomes more sophisticated in distinguishing real faults from environmental artifacts.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
An off-board apparatus is provided for reducing nuisance fault indications from a vehicle. The apparatus is communicably coupled to a vehicle health monitoring (VHM) system, or one or more vehicle systems for at least collecting and communicating data thereto such as data that indicates fault generated by a line replaceable unit of the vehicle in response to a built-in test. The apparatus is also coupleable with a computerized maintenance management system configured to store data that describes a historical condition and maintenance of the vehicle. The apparatus is programmed to determine a probability of validity or invalidity of the fault indication, determine a condition indicator for the signal path based on the LRUs current and historical operational environment, and generate a notification, or output the fault indication for generation of a notification, only in an instance in which the fault indication, with a high probability is valid.