Particle Analysis for Engine Failure Prediction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Conventional methods for predicting engine failure through oil analysis are limited by their inability to detect debris at the part per billion level, lack of repeatability, and inability to analyze particles larger than 5 μm in diameter, making early detection of abnormal engine behavior impractical and costly.
Innovation Solution
The method involves analyzing oil samples from engines using a processor to categorize particles based on physical characteristics and chemical composition, comparing the data to historical datasets, and generating predictions of failure or failure mechanisms using predefined rules.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If optical atomic spectroscopy is used for oil analysis, then elemental analysis of total oil sample is achieved, but individual particles cannot be characterized and particles greater than 5 μm cannot be analyzed
Solution Approach 1:
The patent replaces optical atomic spectroscopy with a system combining a microscope and image sensor. This substitution enables direct visualization and measurement of individual particles, allowing characterization of particles across a broader size range including those greater than 5 μm in diameter, while maintaining the ability to perform compositional analysis.
2Reliability
If conventional oil analysis methods are used, then early detection of abnormal engine behavior is attempted, but detection sensitivity is insufficient for part per billion level debris
Solution Approach 1:
The patent segments the oil sample into individual particle images captured by the microscope and image sensor. This segmentation allows each particle to be analyzed independently for size, shape, and composition, enabling detection of individual debris particles at the part per billion level rather than relying on bulk sample analysis.
3Loss of information
If particle analysis is performed to identify wear sources, then comprehensive particle characterization is achieved, but analysis time and system complexity increase
Solution Approach 1:
The patent creates a multi-functional analysis system where a single microscope and image sensor setup performs multiple functions: particle visualization, size measurement, shape analysis, and compositional characterization. This universal system reduces overall complexity compared to using separate specialized equipment for each type of analysis while maintaining comprehensive wear mechanism information.
Data Source
AI summary
Methods and systems for failure prediction using analysis of oil or other lubricant. Raw data about feature(s) of each of a plurality of particles filtered from a fluid sample are used to categorize each particle into one of a plurality of categories, each category being defined by one or more of: chemical composition, size and morphology. Particle physical characteristics in each category are quantified to obtain a set of categorized data. The categorized data are compared with historical data. Results of the comparing are evaluated to generate a prediction of any failure or mechanism of failure.