Author: Ashish Ghanghash
Ghanghash, Ashish, 2023 Development of a Process for Converting Radiographic Images of metal into Finite Element Models: Assessing Accuracy and Validity, Flinders University, College of Science and Engineering
Terms of Use: This electronic version is (or will be) made publicly available by Flinders University in accordance with its open access policy for student theses. Copyright in this thesis remains with the author. You may use this material for uses permitted under the Copyright Act 1968. If you are the owner of any included third party copyright material and/or you believe that any material has been made available without permission of the copyright owner please contact copyright@flinders.edu.au with the details.
The objective of this research is to develop a process that converts radiographic images into finite element models (2D) by extracting geometry data from radiographic image. The ultimate purpose of this research is to elevate radiographic imaging to a new level by enabling the assessment of structural integrity through the creation of finite element models based on x-ray images, followed by the performance of tests on these models to verify their structural integrity. Additionally, a literature review was conducted, revealing that metals are the least utilized materials in the fabrication of a Finite Element Model (FEM). To further investigate this, four specimens with varying surface characteristics were meticulously crafted. The initial stage of this process entails conducting an X-ray analysis of the specimens. Subsequently, a MATLAB code is developed to extract the geometry data from the x-ray image, which is then imported into the finite element model using Ansys Workbench software. The results obtained from this research indicate the presence of minor errors in the extraction of specimen geometry. These errors stem from the non-uniform absorption of x-rays, primarily due to beam hardening. As a consequence, the edges of the specimen appear brighter in the x-ray image. Following the creation of the finite element model, tensile testing is
conducted on the specimen within Ansys and using Instron. The resulting strain values are then compared to validate the accuracy of the process. Given that this process is relatively new, there are opportunities for improvement. One potential avenue for enhancement involves utilizing x-ray intensity values instead of absorption values, as this has the potential to enhance the overall accuracy of the process. Despite these minor limitations, the developed process offers several advantages. Notably, it is a fast and efficient method that does not require significant computational power. Moreover, it demonstrates promising potential for success in the future. In light of these findings, further research could focus on refining the process and exploring additional applications for this innovative method. In conclusion, this research successfully develops a process for converting radiographic images into finite element models by extracting geometry data from x-ray contrast. The process is verified through tensile testing on specimens. Although minor errors are present in the extraction of specimen geometry, the process shows potential for improvement, such as utilizing x-ray intensity values. Overall, this process offers speed, efficiency, and
iv promising possibilities for future success. Further research can build upon these findings to enhance the process and explore its wider applications.
Keywords: Radiography, X-ray, FEM, Ansys, Noval.
Subject: Engineering thesis
Thesis type: Masters
Completed: 2023
School: College of Science and Engineering
Supervisor: Stuart Wildy