The mechanical properties of wrought magnesium alloys depend on their microstructure, particularly the grain size and its distribution. Conventional modeling methods struggle to accurately predict the complex physical...
详细信息
The mechanical properties of wrought magnesium alloys depend on their microstructure, particularly the grain size and its distribution. Conventional modeling methods struggle to accurately predict the complex physical relationship between them. In the present study, a microstructure-property mapping model based on artificial neural network was proposed to accurately predict the micro-Vickers hardness, 0.2% proof stress and tensile strength of rolled as-casting AZ31 alloy. This method considered the average grain size and its dispersion as the main influencing factors. Firstly, the diversified data of microstructure and mechanical properties of cast-rolling AZ31 alloy was obtained through multi-process and different cross rolling experiments at different initial rolling temperatures. Consid-ering the promoting effect of nonlinear inertia weight and genetic operators on the convergence and search ability, the particle swarm optimization (PSO) algorithm was improved and subsequently introduced into the back propagation (BP) neural network to pre-training the initial weights and biases. Finally, a microstructure-mechanical propertymapping model based on the improved PSO-BP neural network was constructed and verified in detail. Results show that the combination of nonlinear inertia weight and se-lection and crossover process of genetic algorithm can significantly improve the population diversity and convergence performance of PSO algorithm. This can optimize the initial weights and biases required for the operation of BP neural network model. The optimi-zation ability of the model is remarkably improved in the process of establishing the microstructure-property mapping relationship.& COPY;2023 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY-NC-ND license (http://***/licenses/by-nc-nd/4.0/).
Predicting the physical property of a class of microstructures is crucial in material design, structural simulation, and design. property prediction may be conducted millions of times in these studies and is better de...
详细信息
Predicting the physical property of a class of microstructures is crucial in material design, structural simulation, and design. property prediction may be conducted millions of times in these studies and is better derived instantly for computational efficiency. This issue is addressed in this study via building a mapping from a 3D microstructure to its effective material property, or called structure-propertymapping, using a 3D convolutional neural network (CNN). Unlike the direct approach using labeled simulation data, the mapping is based on the physical knowledge of the structure-property relationship determined by its underlying PDE equations. The knowledge is embedded in the loss function of the CNN framework, which is designed and tested under several different formulations to improve its training convergence. Ultimately, the derived structure-propertymapping can instantly predict the associated material property for a given microstructure and has far better generalization ability than the data-labeled approach, as demonstrated via numerical examples.
暂无评论