Engine Oil Wear Prediction via Multi-Source Data Fusion
Solution Overview
Problem
Current engine quality inspection methods, particularly ferrographic analysis, fail to accurately predict wear trends due to indirect characterization, leading to under-maintenance or over-maintenance issues, which are not suitable for modern lean production requirements.
Innovation Solution
A big data analysis system integrating an oil acquisition module, oil analysis module for spectral, ferrographic, and physicochemical data collection, a data fusion module using fuzzy logic and D-S evidence theory, and an oil prediction module based on genetic algorithms and BP networks to construct and train a prediction model for accurate engine wear prediction.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If ferrographic analysis technology is used to monitor wear conditions, then wear faults can be diagnosed, but the wear trend cannot be accurately predicted due to indirect characterization
Solution Approach 1:
The patent combines multiple data sources including spectral data, ferrographic data, and physicochemical data into a comprehensive data fusion module. This merging of multiple characterization methods enables both reliable wear fault diagnosis and accurate wear trend prediction by providing direct and indirect indicators simultaneously
Solution Approach 2:
The patent introduces an oil prediction model as an intermediary that processes fused oil analysis data to predict future wear trends. This intermediary translates indirect ferrographic characterization into accurate predictive information about engine wear progression
2Reliability
If preventive periodic maintenance is implemented, then equipment reliability is improved, but under-maintenance and over-maintenance occur simultaneously
Solution Approach 1:
The patent transitions from static periodic maintenance schedules to dynamic condition-based maintenance. The system continuously monitors oil condition parameters and adjusts maintenance timing based on actual wear trends, ensuring maintenance is performed neither too early nor too late, thereby improving both reliability and maintenance efficiency
Solution Approach 2:
The patent implements a feedback mechanism where wear prediction results from the oil prediction model continuously inform maintenance decisions. This closed-loop feedback enables precise timing of maintenance activities based on actual wear progression, eliminating both under-maintenance and over-maintenance scenarios
3Productivity
If engine operates in harsh environments, then productivity is maintained, but wear failure becomes extremely prone
Solution Approach 1:
The patent performs preliminary wear prediction and analysis before critical failures occur. By continuously monitoring oil condition and predicting wear trends, the system enables proactive maintenance actions that prevent wear failures while allowing the engine to operate continuously in harsh environments
Data Source
AI summary
The invention discloses a big data analysis system for engine quality detection and prediction, comprising an oil acquisition module for collecting oil in an engine; an oil analysis module for obtaining spectral data, ferrographic data, and physicochemical data of the oil; a data fusion module for fusing the spectral data, ferrographic data and physicochemical data based on a fuzzy logic and a D-S evidence theory to obtain oil fusion data; an oil prediction module for constructing an oil prediction model, training the oil prediction model based on the oil fusion data, and predicting the oil in the engine based on a trained oil prediction model to obtain oil prediction data; a quality detection module connected with the oil prediction module for obtaining a wear degree of the engine and completing a quality prediction of the engine based on the oil prediction data.