Crack Tip Localization Using Digital Image Correlation and Neural Nets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for crack length measurement and propagation analysis in non-standardized samples and complex crack paths are limited, prone to errors due to temperature and material variations, and unsuitable for precise detection of irregular crack paths.
Innovation Solution
A computer-implemented method using a two- or three-dimensional digital image correlation system and a trained neural network, specifically a ParallelNets architecture combining convolutional and fully connected neural networks, for precise damage detection and crack tip localization, enabling accurate crack propagation analysis under varying load conditions.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If traditional crack length measurement methods (potential method, compliance method) are used, then measurement can be performed on standardized specimens, but the methods are unsuitable for non-standardized samples and complex crack paths
Solution Approach 1:
The patent replaces traditional mechanical/electrical measurement methods (potential drop, compliance) with an optical-based digital image correlation system. This substitution enables visualization and measurement of complex crack paths on non-standardized specimens by capturing displacement fields optically, rather than relying on electrical properties or mechanical compliance that require standardized geometries.
Solution Approach 2:
The patent transitions from one-dimensional crack length measurements to two-dimensional displacement field analysis. By capturing full-field displacement data across the specimen surface, the system can track crack propagation in complex geometries and irregular paths, not just straight-line crack lengths in standardized specimens.
2Reliability
If analytical formulas or simulation tools are used to determine crack tip stress, then calculations can be performed for standardized specimens, but friction losses, temperature influences, geometric inaccuracies, and material effects are not taken into account
Solution Approach 1:
The patent implements an iterative optimization process where the crack path and displacement fields are continuously refined. The system uses the measured displacement fields to update crack path estimates, recalculates stress intensity factors, and repeats the process until convergence. This feedback mechanism allows the system to account for complex boundary conditions, geometric variations, and material non-linearities that analytical formulas cannot capture.
Solution Approach 2:
The patent performs preliminary optical correction measurements to compensate for temperature influences and geometric inaccuracies before conducting the main crack propagation analysis. By pre-characterizing the specimen geometry and environmental conditions, the system can more accurately determine crack tip stress parameters while accounting for real-world variations.
3Productivity
If traditional measurement methods are used, then a single crack length curve can be generated, but the added value of current crack propagation tests is relatively small in relation to the test effort
Solution Approach 1:
The patent divides the specimen surface into multiple measurement zones and captures displacement fields across the entire surface simultaneously. This segmentation approach allows the system to track multiple crack tips, detect damage in different regions, and generate comprehensive crack propagation data from a single test, significantly increasing the information output relative to the test effort.
Solution Approach 2:
The digital image correlation system serves multiple functions: it measures displacement fields, detects crack initiation, tracks crack propagation paths, determines crack tip positions, and calculates stress intensity factors. This multi-functionality allows a single measurement system to provide comprehensive crack propagation data, increasing productivity without proportionally increasing device complexity.
4Measurement precision
If optical correction measurements are performed to compensate for temperature influences, then measurement accuracy can be maintained, but the testing process becomes more complex and time-consuming
Solution Approach 1:
The patent performs optical correction measurements during the initial setup phase before the main crack propagation test. By characterizing the specimen geometry and thermal expansion properties beforehand, the system can apply pre-calculated correction factors during the actual test, reducing the time required during critical measurement phases while maintaining accuracy.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enables precise detection and tracking of cracks and damage in complex loading conditions, overcoming limitations of traditional methods by providing high-resolution images and reliable crack tip localization.
Implementation Method 1
a first stationary imaging device formed by a two- or three-dimensional digital image correlation system
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A computer-implemented method for damage detection in a body under investigation, in particular comprising: calculating the current localization of at least one damage area and/or at least one area of a crack tip in the body using a trained neural network for damage detection, aligning a positionable imaging device based on the at least one damage area of the body and/or the at least one area of a crack tip in the body according to the procedure step, and generating at least one microscopic image of the at least one damage area using the second imaging device under applied load according to the procedure step and/or after unloading.