Hierarchical Fault Classification Framework for Large Systems
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing decision fusion techniques for fault diagnosis in large systems lack the capability to handle hierarchical subcomponent and subsystem interactions, and fail to integrate overlapping faults from diverse diagnostic models with heterogeneous information sources.
Innovation Solution
A hierarchical fault classification framework that acquires operational data, analyzes it using multiple diagnostic models, and derives an overall probability of fault by considering hierarchical relationships between subsystems and components, incorporating physics-based, experience-based, and regression-based models, and secondary evidential information.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If a single diagnostic model is used to isolate faults, then the system complexity is reduced, but the diagnostic accuracy and performance evaluation capability deteriorate
Solution Approach 1:
The patent divides the diagnostic system into multiple independent diagnostic models, each specializing in specific fault types or system components. This segmentation allows each model to focus on particular diagnostic tasks, improving overall diagnostic accuracy while maintaining manageable complexity through modular architecture.
Solution Approach 2:
The patent combines multiple diagnostic models into a unified decision fusion framework that integrates their outputs. This merging approach leverages the strengths of different models to achieve superior diagnostic performance compared to any single model alone.
2Device complexity
If a flat fault classification model is assumed, then the classification process is simplified, but the capability to capture subsystem hierarchy interactions is lost
Solution Approach 1:
The patent transitions from a flat, two-dimensional classification model to a hierarchical, multi-dimensional framework that incorporates subsystem relationships. This dimensional expansion enables the model to capture interactions between different hierarchy levels while maintaining structured organization.
Solution Approach 2:
The patent implements a nested hierarchical structure where component-level faults are embedded within subsystem-level classifications, which in turn are nested within system-level categories. This nesting preserves hierarchy interaction information while organizing complexity in a manageable nested framework.
3Reliability
If diverse diagnostic models with different techniques are used, then the coverage and reliability of individual models improve, but the difficulty of integrating their heterogeneous outputs increases
Solution Approach 1:
The patent introduces a decision fusion framework as an intermediary layer between diverse diagnostic models and the final diagnostic conclusion. This mediator standardizes and integrates heterogeneous model outputs, managing integration complexity while preserving the reliability benefits of diverse models.
Solution Approach 2:
The patent transforms outputs from different diagnostic models into a unified parameter space, converting diverse model results into comparable formats. This parameter transformation enables seamless integration of heterogeneous models while maintaining their individual reliability characteristics.
4Measurement precision
If decision fusion techniques are applied to combine multiple diagnostic models, then the diagnostic accuracy improves, but the computational complexity and data processing requirements increase
Solution Approach 1:
The patent segments the decision fusion process into distinct modular stages, including individual model execution, output standardization, fusion computation, and result interpretation. This segmentation reduces computational complexity by organizing processing tasks into manageable, potentially parallelizable segments.
Data Source
AI summary
A method for diagnosing and classifying faults in a system is provided. The method comprises acquiring operational data for at least one of a system, one or more subsystems of the system or one or more components of the one or more subsystems. Then, the method comprises analyzing the operational data using one or more diagnostic models. Each diagnostic model uses the operational data to determine a probability of fault associated with at least one of the one or more components or the one or more subsystems. Finally, the method comprises deriving an overall probability of fault for at least one of the system, the one or more subsystems, or the one or more components using the one or more probabilities of fault determined by the one or more diagnostic models and one or more hierarchical relationships between the subsystems and components of the system.