Vibration Diagnosis Feedback Learning for More Reliable Machine Monitoring
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing predictive maintenance programs for machines, such as motors and pumps, rely on periodic data collection and analysis, which often requires specialized expertise and can result in inaccurate automated diagnoses due to the lack of trained technicians.
Innovation Solution
A machine diagnostic system that employs supervised learning to improve the accuracy and reliability of automated diagnosis results by using Bayesian probability models and neural networks to analyze vibration data and other diagnostic data from machines.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If automated diagnosis is implemented using programmed rules, then the need for specially trained technicians is reduced, but false positive and false negative results occur
Solution Approach 1:
The system implements feedback loops where automated diagnosis results are continuously evaluated and used to refine the diagnostic algorithms. Expert technicians review and validate automated findings, feeding this information back into the system to improve future diagnoses, thereby maintaining high reliability while preserving automation benefits
Solution Approach 2:
The patent introduces an intermediary layer between programmed rules and final diagnosis that includes multiple analysis algorithms and validation steps. This intermediary processing layer filters and refines automated results before presenting them, reducing false positives and negatives while keeping the system automated
2Reliability
If periodic data collection is used for predictive maintenance, then machine monitoring is achieved, but specialized expertise is required for data analysis
Solution Approach 1:
The system enables self-service diagnostics by automatically collecting, analyzing, and interpreting machine data without requiring specialized human intervention. Multiple algorithms work together to autonomously diagnose machine conditions, allowing the system to serve itself while maintaining predictive maintenance capabilities
Solution Approach 2:
The patent replaces the mechanical need for expert technician analysis with automated computational algorithms. Instead of relying on human expertise to interpret vibration and operational data, the system uses computer-based algorithms to perform the analysis function, substituting mechanical human skill with automated intelligence
3Adaptability or versatility
If multiple fault defect assumptions are monitored, then comprehensive machine diagnosis is provided, but the complexity of the diagnostic system increases
Solution Approach 1:
The diagnostic system is segmented into multiple independent algorithms, each specialized for detecting specific fault types. This modular segmentation allows comprehensive monitoring of multiple fault conditions while keeping individual algorithm complexity manageable, as each segment handles a specific aspect of diagnosis
Data Source
AI summary
Supervised learning is implemented to improve the accuracy of automated diagnoses performed by monitoring units installed at a machine. The monitoring units perform indicator acquisition and automated diagnoses based on a Bayesian model derived in accordance with the machine's known configuration. Raw data is collected, including machine vibration data and other diagnostic data. The data is analyzed to diagnose for specific fault defect assumptions so as to generate the automated diagnoses results and a rating for overall health of the machine. The results are uploaded to an external environment that can be accessed by an expert for review and correction. Based upon the expert's corrections, the Bayesian model is adjusted using supervised learning to improve the automated diagnoses performed by the monitoring units.