Vehicle Fault Modeling for Single Root Cause Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing diagnostic reasoners for vehicle health management systems face challenges such as inherent ambiguity, subjective probability determination, and ad hoc fault generation, leading to inefficient and extensive lists of potential faults, which complicate troubleshooting.
Innovation Solution
An enhanced fault model using design failure mode and effect analysis (DFMEA) information to generate faults at the subsystem level, reducing ambiguity and automating fault identification, and employing linear algebraic techniques to identify a single root cause, eliminating the need for Bayesian approaches.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If a comprehensive fault model is generated using design DFMEA information, then fault detection coverage is improved, but the number of potential faults increases making troubleshooting more complex
Solution Approach 1:
The patent segments the comprehensive fault model into subsystem-level faults, organizing the large number of potential faults from DFMEA analysis into manageable subsystem groups. This segmentation maintains complete fault detection coverage while reducing troubleshooting complexity by allowing technicians to isolate and diagnose faults at the subsystem level rather than evaluating all individual faults simultaneously.
Solution Approach 2:
The patent introduces an intermediary layer between the comprehensive fault model and the diagnostic reasoner. This intermediary automatically associates diagnostic trouble codes with subsystem-level faults and generates simplified fault representations, acting as a mediator that preserves the completeness of the original fault model while presenting a simplified view to the diagnostic system and users.
2Adaptability or versatility
If subjective probability determination is used in diagnostic reasoners, then flexibility in handling uncertain information is improved, but diagnostic accuracy deteriorates due to inherent ambiguity
Solution Approach 1:
The patent implements a self-service mechanism where the system automatically determines objective probabilities for fault occurrences and diagnostic code associations without requiring subjective human judgment. The diagnostic reasoner uses the enhanced fault model and available data to autonomously calculate probability values, eliminating the ambiguity of subjective determination while maintaining the ability to handle uncertain information through objective statistical methods.
Solution Approach 2:
The patent replaces the subjective, human-judgment-based probability determination mechanism with an automated computational mechanism. Instead of relying on experts to assign subjective probabilities, the system uses objective data from the enhanced fault model, historical failure data, and physics-based models to calculate probabilities automatically, substituting human cognitive processes with mechanical computation to eliminate subjectivity.
3Reliability
If extensive lists of potential faults are generated, then completeness of fault identification is improved, but the time required for troubleshooting increases
Solution Approach 1:
The patent performs preliminary actions by pre-generating the enhanced fault model with all potential faults from DFMEA information and pre-establishing the associations between diagnostic trouble codes and subsystem-level faults before actual troubleshooting occurs. This preliminary preparation allows the diagnostic reasoner to quickly query and evaluate only relevant faults based on observed symptoms, rather than manually reviewing extensive fault lists during troubleshooting, thus maintaining completeness while reducing time loss.
Solution Approach 2:
The patent extracts and isolates the most relevant faults from the extensive list of potential faults by using the diagnostic reasoner to evaluate and filter faults based on observed symptoms and their associations with diagnostic trouble codes. This extraction process removes irrelevant faults from the active troubleshooting consideration, maintaining complete fault identification capability while significantly reducing the time required to focus on and diagnose the actual problematic faults.
Data Source
AI summary
A method for vehicle fault management includes generating, for a vehicle system, a fault model including a plurality of faults using design failure mode and effect analysis information and parsing each of the faults of the plurality of faults by subsystem of the vehicle system. The method also includes determining, for a respective fault of the plurality of faults, whether the respective fault is associated with at least one existing diagnostic trouble code based on fault code requirement information. The method also includes, in response to a determination that the respective fault is associated with at least one existing diagnostic trouble code, associating, in a fault database, the at least one existing diagnostic trouble code with the respective fault.