Graph Database for Mechatronic Fault Diagnosis
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for diagnosing faults in mechatronic systems, which consist of numerous interconnected components and virtual parts, are inefficient and lack a systematic approach to identify the specific parts that need to be checked, leading to potential misidentification of error causes and inefficient repair plans.
Innovation Solution
A graph database structure is employed to represent all parts, functions, and malfunctions as nodes, with directed edges indicating relationships, allowing for 'backward reachability analysis' to identify faulty nodes and generate an efficient checking plan by inverting edges and using attributes like repair costs and frequencies to prioritize checks.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a systematic approach using graph database is implemented, then the precision of fault identification is improved, but the device complexity increases
Solution Approach 1:
The system segments the mechatronic system into multiple levels (first level with first nodes, second level with second nodes, third level with third nodes, fourth level with fourth nodes) and uses directed edges to represent relationships between them. This segmentation allows systematic analysis of fault propagation paths while maintaining manageable complexity through structured organization.
Solution Approach 2:
The graph database structure acts as an intermediary between the raw fault data and the diagnostic analysis. It mediates by organizing fault information into structured nodes and relationships, enabling systematic query processing and fault propagation analysis without directly exposing the complexity of the underlying mechatronic system.
2Reliability
If backward reachability analysis is performed to identify all potential fault causes, then the completeness of diagnosis is improved, but the time required for analysis increases
Solution Approach 1:
The system performs partial backward reachability analysis by querying the graph database with specific criteria (e.g., limiting the number of levels to search, filtering by fault type or component category). This allows obtaining sufficient diagnostic information without exhaustively analyzing all possible fault propagation paths, thereby reducing analysis time while maintaining adequate diagnosis completeness.
Solution Approach 2:
The graph database is pre-populated with relationships between components, functions, and faults before actual diagnosis is needed. This preliminary structuring of knowledge allows rapid query processing during actual fault analysis, reducing the time required for backward reachability analysis while maintaining comprehensive diagnostic coverage.
3Reliability
If all parts are checked to ensure thorough inspection, then the reliability of fault detection is improved, but the productivity of the repair process decreases
Solution Approach 1:
The system extracts and prioritizes the most likely fault candidates from the complete set of possible faults using the graph database query results. By identifying and focusing inspection efforts on the top prioritized components based on fault probability and impact analysis, the system achieves reliable fault detection without requiring exhaustive checking of all parts, thereby improving repair efficiency.
Solution Approach 2:
The system uses feedback from the graph database query results to dynamically adjust the inspection strategy. By analyzing the structure of fault propagation paths and the likelihood of various fault causes, it provides feedback that guides selective inspection of the most promising candidates, balancing detection reliability with inspection efficiency.
Data Source
AI summary
A method for determining a part to be checked of a mechatronic system includes providing a graph database having at least one first sub-level with first nodes, a second sub-level with second nodes, a third sub-level with third nodes and a fourth sub-level with fourth nodes, wherein directly adjacent sub-levels are connected by directed edges. The method also includes determining at least one of the fourth nodes which is output as faulty during a check of the mechatronic system, and inverting the directed edges. The method further includes determining at least one first node to be checked of the first nodes which is representative of at least one of the group consisting of at least one component and at least one part of the mechatronic system, starting from the determined fourth node, depending on a range.