Inverse Identification of Residual Stress Distribution in Aluminium Alloy Components Based on Deep Learning
Residual stress within a structural component can significantly affect the mechanical performance and stability of a structure. Therefore, it is crucial to find a way to determine the residual stress distribution to maintain the normal working of structures. Conventional methods for residual stress...
| Κύριοι συγγραφείς: | , , , |
|---|---|
| Μορφή: | Άρθρο |
| Γλώσσα: | English |
| Έκδοση: |
MDPI AG
2022-01-01
|
| Σειρά: | Applied Sciences |
| Θέματα: | |
| Διαθέσιμο Online: | https://www.mdpi.com/2076-3417/12/3/1195 |
| _version_ | 1827661702386155520 |
|---|---|
| author | Tulin Xiong Lu Wang Xianzhi Gao Guangyan Liu |
| author_facet | Tulin Xiong Lu Wang Xianzhi Gao Guangyan Liu |
| author_sort | Tulin Xiong |
| collection | DOAJ |
| description | Residual stress within a structural component can significantly affect the mechanical performance and stability of a structure. Therefore, it is crucial to find a way to determine the residual stress distribution to maintain the normal working of structures. Conventional methods for residual stress determination primarily include experimental testing, finite element simulations and inverse identification. However, these methods suffer from disadvantages of high testing costs, long calculation time and low inverse efficiency. To avoid these shortcomings, this study developed a high-performance method based on a deep learning technique. In this method, an artificial neural network was used to replace the finite element calculation in the finite element model updating (FEMU) technique and the residual stress distribution of structural components was inversely obtained based on the measured residual stresses of a finite number of measuring points. Compared with the conventional FEMU technique, the calculation efficiency of the proposed method was considerably improved. Furthermore, the accuracy and efficiency of the method were verified by simulated four-point bending experiments considering an elastic-plastic material. |
| first_indexed | 2024-03-10T00:14:56Z |
| format | Article |
| id | doaj.art-4db176ec73fa4dd88bbabe4fc2fd1031 |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T00:14:56Z |
| publishDate | 2022-01-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-4db176ec73fa4dd88bbabe4fc2fd10312023-11-23T15:53:28ZengMDPI AGApplied Sciences2076-34172022-01-01123119510.3390/app12031195Inverse Identification of Residual Stress Distribution in Aluminium Alloy Components Based on Deep LearningTulin Xiong0Lu Wang1Xianzhi Gao2Guangyan Liu3School of Aerospace Engineering, Beijing Institute of Technology, Beijing 100081, ChinaSchool of Aerospace Engineering, Beijing Institute of Technology, Beijing 100081, ChinaSchool of Aerospace Engineering, Beijing Institute of Technology, Beijing 100081, ChinaSchool of Aerospace Engineering, Beijing Institute of Technology, Beijing 100081, ChinaResidual stress within a structural component can significantly affect the mechanical performance and stability of a structure. Therefore, it is crucial to find a way to determine the residual stress distribution to maintain the normal working of structures. Conventional methods for residual stress determination primarily include experimental testing, finite element simulations and inverse identification. However, these methods suffer from disadvantages of high testing costs, long calculation time and low inverse efficiency. To avoid these shortcomings, this study developed a high-performance method based on a deep learning technique. In this method, an artificial neural network was used to replace the finite element calculation in the finite element model updating (FEMU) technique and the residual stress distribution of structural components was inversely obtained based on the measured residual stresses of a finite number of measuring points. Compared with the conventional FEMU technique, the calculation efficiency of the proposed method was considerably improved. Furthermore, the accuracy and efficiency of the method were verified by simulated four-point bending experiments considering an elastic-plastic material.https://www.mdpi.com/2076-3417/12/3/1195residual stress distributionartificial neural networkfinite element model updatinginverse identification |
| spellingShingle | Tulin Xiong Lu Wang Xianzhi Gao Guangyan Liu Inverse Identification of Residual Stress Distribution in Aluminium Alloy Components Based on Deep Learning Applied Sciences residual stress distribution artificial neural network finite element model updating inverse identification |
| title | Inverse Identification of Residual Stress Distribution in Aluminium Alloy Components Based on Deep Learning |
| title_full | Inverse Identification of Residual Stress Distribution in Aluminium Alloy Components Based on Deep Learning |
| title_fullStr | Inverse Identification of Residual Stress Distribution in Aluminium Alloy Components Based on Deep Learning |
| title_full_unstemmed | Inverse Identification of Residual Stress Distribution in Aluminium Alloy Components Based on Deep Learning |
| title_short | Inverse Identification of Residual Stress Distribution in Aluminium Alloy Components Based on Deep Learning |
| title_sort | inverse identification of residual stress distribution in aluminium alloy components based on deep learning |
| topic | residual stress distribution artificial neural network finite element model updating inverse identification |
| url | https://www.mdpi.com/2076-3417/12/3/1195 |
| work_keys_str_mv | AT tulinxiong inverseidentificationofresidualstressdistributioninaluminiumalloycomponentsbasedondeeplearning AT luwang inverseidentificationofresidualstressdistributioninaluminiumalloycomponentsbasedondeeplearning AT xianzhigao inverseidentificationofresidualstressdistributioninaluminiumalloycomponentsbasedondeeplearning AT guangyanliu inverseidentificationofresidualstressdistributioninaluminiumalloycomponentsbasedondeeplearning |