Minimum Cardinality Candidate Diagnosis Generator for Model-Based Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Model-based diagnosis systems face challenges in efficiently reducing the search space and computational resources required to find minimum cardinality diagnoses, particularly in large systems with multiple simultaneous component failures, leading to excessive computational complexity and unnecessary conflicts.
Innovation Solution
The NGDE algorithm employs a minimum cardinality candidate diagnosis generator that exploits conflict set reduction rules and uses a hybrid truth maintenance system (HTMS) to transform conflict sets into simpler forms, combining depth-first search with branch and bound methods to identify minimal diagnoses efficiently.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If model-based diagnosis is applied to large systems with multiple simultaneous component failures, then diagnostic coverage is improved, but computational complexity increases excessively
Solution Approach 1:
The patent segments the diagnosis process into distinct phases: conflict generation, conflict set reduction, and candidate diagnosis generation. By dividing the complex diagnostic task into manageable segments with specific reduction rules applied at each stage, the computational complexity is controlled while maintaining comprehensive diagnostic coverage for large systems
Solution Approach 2:
The patent applies conflict set reduction rules as preliminary actions before generating candidate diagnoses. By pre-reducing the conflict sets using logical reduction rules, the search space for subsequent diagnosis generation is significantly diminished, allowing the system to handle large-scale systems without excessive computational complexity
2Measurement precision
If exhaustive search is used to find minimum cardinality diagnoses, then diagnostic accuracy is improved, but computational resources required increase excessively
Solution Approach 1:
The patent extracts and removes redundant conflicts from the conflict set using logical reduction rules before generating candidate diagnoses. By taking out unnecessary conflicts that do not contribute to finding minimum cardinality diagnoses, the computational resources required are reduced while maintaining diagnostic accuracy
Solution Approach 2:
The patent applies partial action by using targeted conflict set reduction rules rather than exhaustive processing of all possible conflicts. The reduction rules selectively process only the conflicts that are relevant to finding minimum cardinality diagnoses, reducing computational resource consumption while preserving diagnostic accuracy
3Adaptability or versatility
If conflict sets are not reduced, then all possible diagnoses are considered, but search space becomes unnecessarily large
Solution Approach 1:
The patent performs preliminary conflict set reduction using logical reduction rules before the candidate diagnosis generation phase. This preliminary action eliminates redundant conflicts that would otherwise expand the search space unnecessarily, while the reduction rules are designed to preserve all diagnoses that could potentially be minimum cardinality solutions
Solution Approach 2:
The patent segments the conflict set into reduced and non-reduced portions, applying reduction rules to eliminate only the redundant conflicts. This segmentation approach maintains diagnosis completeness by preserving essential conflicts while removing unnecessary ones that would bloat the search space
Data Source
AI summary
This NGDE algorithm incorporates a minimum cardinality candidate diagnosis generator which exploits conflict set reduction rules to dramatically reduce the search space and thus reduce the computational resources required to find minimum cardinality diagnoses.