Diagnostic Model Divergence Analysis for Failure Cause Identification
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional diagnostic and prognostic systems for complex work machines often fail to identify specific causes of failures due to their inability to analyze operation conditions beyond simple classification into normal or abnormal categories, and require significant computational time, making it difficult to predict potential issues effectively.
Innovation Solution
A method and system that uses a computational model to derive output parameters from input parameters, detect divergences, and estimate future trends, while identifying desired statistical distributions to diagnose and predict failures by creating a diagnostic model that reflects interrelationships between input and output parameters, utilizing a zeta statistic to optimize input parameter distributions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If conventional diagnostic techniques classify operation conditions into normal or abnormal categories, then the system can identify failures, but it cannot identify specific causes of failures
Solution Approach 1:
The patent segments the diagnostic process into multiple independent steps: (1) obtaining actual output parameters from the system, (2) independently deriving expected output parameters using a computational model, (3) detecting divergences between actual and derived values, and (4) identifying specific input parameters causing the divergences. This segmentation allows the system to not only detect failures but also pinpoint their specific causes by analyzing which input parameters contribute most to the divergences.
2Reliability
If conventional techniques use differences between actual and model values as inputs to generate failure patterns, then failure classification is achieved, but computational time increases significantly
Solution Approach 1:
The patent extracts only the essential information needed for diagnosis by directly comparing actual output parameters with independently derived values from a computational model. Instead of using all difference values as inputs to generate comprehensive failure patterns, the system extracts specific divergences and traces them back to specific input parameters, significantly reducing computational requirements while maintaining diagnostic reliability.
Solution Approach 2:
The system performs preliminary computational work by pre-establishing the computational model that represents normal system behavior. This model is used to independently derive expected output parameters in real-time, allowing for rapid comparison with actual values without requiring complex post-processing or extensive computational analysis of failure patterns.
3Adaptability or versatility
If conventional systems use complex control systems with multiple input parameters, then operational flexibility is improved, but difficulty in identifying specific failure causes increases
Solution Approach 1:
The patent applies local quality analysis by examining the contribution of each individual input parameter to the overall system output. When a divergence is detected between actual and derived output parameters, the system analyzes which specific input parameters have the greatest influence on that divergence, allowing for localized diagnosis of specific component failures rather than requiring analysis of the entire complex system.
Data Source
AI summary
A method for diagnosing a system is provided. The method may include obtaining a plurality of input parameters from the system. The system may generate actual values of one or more output parameters based on the plurality of input parameters. The method may also include independently deriving values of the one or more output parameters, based on the plurality of input parameters, using a computational model, and detecting a divergence between the derived values and the actual values. Further, the method may include identifying respective desired statistic distributions of the plurality of input parameters, based on the actual values of the one or more output parameters, using the computational model.