Inverse Identification of Residual Stress Distribution in Aluminium Alloy Components Based on Deep Learning
Source abstract
Abstract
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.
Source: crossref · Quality: baseline
Research context
OpenAlex topics: Welding Techniques and Residual Stresses · Non-Destructive Testing Techniques · Advanced machining processes and optimization
SciVX discovery tags are terminology-based. OpenAlex topics are source-provided classifications.
Affiliations
School of Aerospace Engineering, Beijing Institute of Technology, Beijing 100081, China
Funding
National Natural Science Foundation of China
References (27)
Show source-provided references
1. Optimizing cold compression deformation to remove residual stresses in die forged disc of Al-Mg-Si alloy10.4028/www.scientific.net/KEM.778.53 ↗
2. Influences of thermal effects on residual stress fields of an aluminium-lithium alloy induced by shot peening10.1007/s00170-020-06557-3 ↗
3. Grain-orientation-dependent residual stress and the effect of annealing in cold-rolled stainless steel10.1016/S1359-6454(02)00021-6 ↗
4. A micro-macro confined compressive fatigue creep failure model in brittle solids10.1016/j.ijfatigue.2019.105278 ↗
5. Experimental validation of residual stress thermomechanical simulation in as-quenched superalloy discs by using diffraction and incremental hole-drilling methods10.1016/j.mtcomm.2021.102229 ↗
6. Residual stress measurement at the weld root of rib-to-deck welded joints in orthotropic steel bridge decks using the contour method10.1016/j.engstruct.2020.110946 ↗
7. Thermal relaxation of residual stress in shot-peened CNT/Al–Mg–Si alloy composites10.1016/j.jmrt.2019.01.023 ↗
8. A multireflection and multiwavelength residual stress determination method using energy dispersive diffraction10.1107/S1600576718004193 ↗
9. Residual stresses in additively manufactured AlSi10Mg: Raman spectroscopy and X-ray diffraction analysis10.1016/j.matdes.2021.109550 ↗
10. Residual stress determination in thin films by X-ray diffraction and the widespread analytical practice applying a biaxial stress model: An outdated oversimplification?
11. New framework for Bayesian statistical analysis and interpolation of residual stress measurements10.1016/j.mechrescom.2013.02.008 ↗
12. An approach to predict the distortion of thin-walled parts affected by residual stress during the milling process10.1007/s00170-017-0811-2 ↗
13. Residual stresses prediction in machining: Hybrid FEM enhanced by assessment of plastic flow10.1016/j.jmatprotec.2019.116332 ↗
14. An iterative technique for the reconstruction of residual stress fields in a butt-welded plate from experimental measurement, and comparison with welding process simulation10.1016/j.ijmecsci.2019.07.001 ↗
15. Evaluation of residual stresses and strains using the Eigenstrain Reconstruction Method10.1016/j.ijsolstr.2010.03.002 ↗
16. A review of characterisation and parameters identification of materials constitutive and damage models: From normalised direct approach to most advanced inverse problem resolution10.1016/j.ijimpeng.2017.01.028 ↗
17. Inverse identification of graphite damage properties under complex stress states10.1016/j.matdes.2019.108135 ↗
18. Parameter identification of an elasto-plastic behaviour using artificial neural networks–genetic algorithm method10.1016/j.matdes.2010.06.039 ↗
19. Application of artificial neural networks in micromechanics for polycrystalline metals10.1016/j.ijplas.2019.05.001 ↗
20. Identification of material parameters for aluminum foam at high strain rate10.1016/j.commatsci.2013.02.024 ↗
21. Human Tracking Using Convolutional Neural Networks10.1109/TNN.2010.2066286 ↗
22. Ronneberger, O., Fischer, P., and Brox, T. (2015). U-Net: Convolutional Networks for Biomedical Image Segmentation, International Conference on Medical Image Computing and Computer-Assisted Intervention, Springer.10.1007/978-3-319-24574-4_28 ↗
23. Simulation of hyperelastic materials in real-time using deep learning10.1016/j.media.2019.101569 ↗
24. An intelligent nonlinear meta element for elastoplastic continua: Deep learning using a new Time-distributed Residual U-Net architecture10.1016/j.cma.2020.113088 ↗
25. Estimates of the coverage of parameter space by Latin Hypercube and Orthogonal Array-based sampling10.1016/j.apm.2017.11.036 ↗
26. Chollet, F. (2021, December 07). Keras: The Python Deep Learning Library. Astrophysics Source Code Library. Available online: https://github.com/keras-team/keras.
27. A method for reconstruction of residual stress fields from measurements made in an incompatible region10.1016/j.ijsolstr.2014.02.008 ↗