Hierarchical Failure Diagnosis Using Binary Search Tree
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing systems for diagnosing failures in hierarchical systems often require extensive detection time, especially when dealing with transition failures, which can lead to increased downtime and inefficiencies.
Innovation Solution
A tree-building algorithm is employed to quickly diagnose the root cause of failures in hierarchical systems by utilizing a binary search tree structure to prioritize failure events based on dependencies, allowing for faster detection and differentiation between transition and permanent failures.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the system detects all error symptoms and analyzes them together to ensure correctness, then the accuracy of failure diagnosis is improved, but the detection time increases significantly
Solution Approach 1:
The patent segments the detection process into two distinct phases: a fast detection phase that quickly identifies whether a failure occurred and its general location, and a subsequent detailed analysis phase that only processes the specific segment identified as problematic. This segmentation allows the system to maintain high accuracy while dramatically reducing overall detection time by avoiding comprehensive analysis of all error symptoms in every case.
Solution Approach 2:
The patent applies partial action by performing complete detailed analysis only on the specific failure segment identified in the fast detection phase, rather than analyzing all error symptoms comprehensively in all cases. This selective approach maintains diagnostic accuracy for the actual failure while avoiding unnecessary analysis time for non-problematic segments.
2Measurement precision
If the system waits to judge whether an error is a transition failure or permanent failure, then the accuracy of failure classification is improved, but the detection time becomes several times longer
Solution Approach 1:
The patent performs preliminary classification of failures into transition and permanent categories during the fast detection phase using initial indicators and patterns. This preliminary action allows the system to immediately apply appropriate handling procedures without waiting for extended observation periods, thereby maintaining accurate classification while significantly reducing detection time for transition failures.
Solution Approach 2:
The patent implements a dynamic detection strategy that adapts the level of analysis based on the type of failure detected. For transition failures, the system uses dynamic, faster detection methods with preliminary classification, while for permanent failures, it applies more thorough analysis. This dynamic approach optimizes detection time while maintaining classification accuracy across different failure types.
Data Source
AI summary
An error detecting device for detecting failure of a hierarchical system comprises a detected signal receiving interface and a processor. The detected signal receiving interface receives a parent failure event occurring in the hierarchical system. If the processor diagnoses that the parent failure event is in a failure state, the detected signal receiving interface receives a first child failure event occurring in the hierarchical system. If the processor diagnoses that the parent failure event is not in the failure state, the detected signal receiving interface receives a second child failure event occurring in the hierarchical system. The parent failure event depends on the first child failure event. The second child failure event depends on the parent failure event. The processor sequentially diagnoses until the detected signal receiving interface receives one of the failure events at a bottom level in a binary search tree structure.