AI Surface Defect Detection for Brake Disc Fatigue Testing
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for identifying and characterizing surface defects, particularly cracks on brake discs, are inefficient and costly due to reliance on human operators, leading to inaccurate measurements and incomplete data collection during fatigue tests, which are prolonged and resource-intensive.
Innovation Solution
A method utilizing artificial intelligence and machine learning algorithms to automate the identification and characterization of surface defects on brake discs, including crack detection and monitoring, by acquiring digital images and processing them to determine dimensional and positional parameters, enabling continuous monitoring and evaluation during dynamic tests.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If human operators visually inspect brake discs during fatigue tests, then crack detection can be performed, but the process is costly, time-consuming, and produces inaccurate measurements
Solution Approach 1:
The patent replaces the mechanical visual inspection process with an automated computer vision system using digital images and AI algorithms. The system captures images of brake discs during fatigue tests and automatically detects and measures cracks using machine learning models, eliminating the need for human operators to manually inspect and measure cracks during test stops.
Solution Approach 2:
The system enables self-service automation where the fatigue test equipment itself performs crack detection without external human intervention. The integrated imaging system and AI processing allow the test bench to automatically monitor crack evolution throughout the test, making the system self-sufficient in performing quality inspection functions.
2Loss of information
If periodic stops are made for visual inspection, then crack measurement is possible, but resource consumption increases and data collection becomes incomplete
Solution Approach 1:
The patent implements continuous crack monitoring throughout the fatigue test without requiring periodic stops. The imaging system captures images at multiple time points during the test, and the AI system continuously processes these images to track crack evolution in real-time, eliminating information loss associated with interrupted testing.
Solution Approach 2:
The system performs multiple functions simultaneously: it conducts the fatigue test, captures images of the brake disc surface, detects cracks using AI algorithms, measures crack dimensions, and monitors crack evolution over time. This multi-functionality consolidates what previously required separate manual inspection operations into a single automated system.
3Reliability
If manual crack measurement is performed by operators, then crack length can be measured, but measurement reliability varies and introduces error
Solution Approach 1:
The patent replaces manual measurement operations with an automated digital imaging and AI processing system. The system uses trained machine learning models to detect and measure crack lengths, positions, and orientations with consistent accuracy, eliminating the variability introduced by human operators' skills and attention.
Solution Approach 2:
The system incorporates feedback mechanisms where the AI model continuously processes new image data during the fatigue test, compares crack measurements against acceptance criteria, and provides real-time feedback on test status. This automated feedback loop ensures consistent, reliable measurements and immediate detection of critical crack conditions.
Data Source
AI summary
A method for identifying and characterizing surface defects on an object is described. Such a method comprises the steps of acquiring at least one digital image of the object or a part of the object on which the surface defects must be identified; then, providing the aforesaid at least one acquired digital image to an algorithm trained by means of artificial intelligence and/or machine learning techniques; then, identifying one or more surface defects present in the at least one acquired digital image, by means of said trained algorithm, and generating digital information related to each identified surface defect. The method then provides determining, for each identified surface defect, at least one respective dimensional parameter, representative of at least one dimension of the surface defect, and at least one respective positional parameter, representative of a position of the surface defect with respect to a reference point or line present in the image or to a two-dimensional spatial coordinate system associated with the aforesaid reference point or line. The aforesaid determining step is performed through a further processing of the aforesaid digital information, by electronic processing means. A method for identifying and characterizing cracks on a brake disc is also described.