3D X-Ray Part Inspection With Iterative Projection Correction
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Non-destructive testing (NDT) of aerospace parts, particularly turbine blades, faces challenges with image artifacts and inter- and intra-examiner variability, leading to uncertain validation and high workloads due to the manual analysis of limited radiographic images, which are prone to beam hardening and Compton scattering, reducing reliability.
Innovation Solution
A method and system for non-destructive testing using volume modeling, involving X-ray radiography to acquire images from different angles, generate simulated projections based on a reference model, iteratively adjust transformation parameters to minimize discrepancies, and create an effective model for accurate 3D geometry characterization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Loss of time
If a limited number of radiographic images are used for NDT, then the acquisition time is reduced, but the reliability of validation deteriorates due to high weight of image artifacts
Solution Approach 1:
The patent creates a virtual copy of the part using CAD model and simulates radiographic images from this digital replica. By comparing simulated images (from CAD model) with actual acquired images, the system can validate the physical part without requiring extensive physical imaging, thus reducing acquisition time while maintaining validation reliability through the virtual-physical comparison approach
Solution Approach 2:
The patent introduces simulated radiographic images as an intermediary between the CAD model and the actual acquired images. These simulated images serve as a bridge that allows comparison and validation without directly relying on a large number of physical radiographs, thereby reducing the need for extensive image acquisition while maintaining validation accuracy
2Ease of operation
If manual analysis of radiographic images is performed by examiners, then the validation can be carried out, but the workload becomes arduous and inter- and intra-examiner variability reduces reliability
Solution Approach 1:
The patent replaces the manual mechanical analysis process performed by human examiners with an automated computational system. The system uses algorithms to compare simulated radiographic images with acquired images, automatically detecting discrepancies without human intervention. This substitution eliminates examiner variability and reduces workload while improving consistency and reliability of validation
Solution Approach 2:
The patent enables the validation process to be self-performing through automated comparison algorithms. The system independently compares simulated and acquired images, identifies discrepancies, and generates validation results without requiring human examiners. This self-service approach eliminates inter- and intra-examiner variability while maintaining ease of operation
3Loss of information
If image artifacts such as beam hardening and Compton scattering are present, then the gray levels are altered, but the validation becomes uncertain and difficult
Solution Approach 1:
The patent creates simulated radiographic images from the CAD model that inherently lack the artifacts present in physical imaging. By comparing these artifact-free simulated images with acquired images, the system can identify and isolate artifact effects, making validation more certain and less difficult
Solution Approach 2:
The patent converts the presence of image artifacts into a beneficial diagnostic tool. By comparing simulated images (without artifacts) with acquired images (with artifacts), the discrepancies reveal the presence and location of artifacts. This allows the system to distinguish between actual part features and artifact effects, thereby improving validation certainty
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 reliable 3D geometry characterization with a limited number of radiographic images by reducing image artifacts and variability, providing a high-level of reliability and efficiency in part validation.
Implementation Method 1
A normalized X-ray radiograph is interpreted as an image of the attenuation of the X-rays on traversing the part
Implementation Method 2
an attenuation itself related to the traversed thickness by a law which is often approximated to an exponential function, as is the case for the Beer-Lambert law
Implementation Method 3
the artifacts to be treated are essentially beam hardening and Compton scattering
Implementation Method 4
the artifacts to be treated are essentially beam hardening and Compton scattering
Data Source
AI summary
The invention relates to a non-destructive inspection method based on 3D modelling of a part, comprising: using an x-ray device to acquire images of the part at various projection angles; computing projections based on the images acquired at the various projection angles; in each of multiple iterations: generating simulated projections corresponding to the computed projections, based on a reference model of an external surface of the part and on a vector μ of transformation parameters of the reference model; modifying the vector μ with a view to reducing a discrepancy between the simulated projections and the computed projections; determining a corrected model of the external surface through transformation of the reference model by way of the vector μ resulting from the iterations; determining an effective model of the part by way of the corrected model.