<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>To address the shortcomings of the traditional BP neural network, this paper uses MEA to optimize the weights and thresholds in the traditional BP neural network, generates sufficient training samples to enhance the generalization ability of the model, and introduces a maturity judgment function to determine whether convergence is achieved. Based on the MEA-BP neural network, the effect of multiple aging on the microstructure of the aluminum alloy is investigated by numerical simulation method with the help of aging treatment to determine the heat treatment process of aluminum alloy. The results show that after the aging time exceeds 24 h, the hardness of the alloy tends to increase significantly, and the precipitation rate of the precipitated phase decreases. The peak hardness of the alloy at 75-120°C is the highest in the hardness curve at the fourth aging temperature of 90°C, which is maximum at the 7thh (119.4 HV). For the effect of the microstructure of the aluminum alloy, the T-phase was not found in the sweep diagrams of the specimens from the three aging states. This study can provide a theoretical basis and technical support for the formulation and optimization of the production process of aluminum alloy materials.</p>
</abstract>

To address the shortcomings of the traditional BP neural network, this paper uses MEA to optimize the weights and thresholds in the traditional BP neural network, generates sufficient training samples to enhance the generalization ability of the model, and introduces a maturity judgment function to determine whether convergence is achieved. Based on the MEA-BP neural network, the effect of multiple aging on the microstructure of the aluminum alloy is investigated by numerical simulation method with the help of aging treatment to determine the heat treatment process of aluminum alloy. The results show that after the aging time exceeds 24 h, the hardness of the alloy tends to increase significantly, and the precipitation rate of the precipitated phase decreases. The peak hardness of the alloy at 75-120°C is the highest in the hardness curve at the fourth aging temperature of 90°C, which is maximum at the 7thh (119.4 HV). For the effect of the microstructure of the aluminum alloy, the T-phase was not found in the sweep diagrams of the specimens from the three aging states. This study can provide a theoretical basis and technical support for the formulation and optimization of the production process of aluminum alloy materials.

Numerical simulation method of aluminum alloy heat treatment process based on BP neural network

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

To address the shortcomings of the traditional BP neural network, this paper uses MEA to optimize the weights and thresholds in the traditional BP neural...

Numerical simulation method of aluminum alloy heat treatment process based on BP neural network

Publicado en línea: 02 oct 2023

Páginas: -

Recibido: 21 dic 2022

Aceptado: 12 abr 2023

DOI: https://doi.org/10.2478/amns.2023.2.00431

Palabras clave
MEA-BP neural network, Aluminum alloy heat treatment, Aging treatment, Numerical simulation, Microstructure

© 2023 Xuewen Chen, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Numerical simulation method of aluminum alloy heat treatment process based on BP neural network

Publicado en línea: 02 oct 2023

Páginas: -

Recibido: 21 dic 2022

Aceptado: 12 abr 2023

DOI: https://doi.org/10.2478/amns.2023.2.00431

Palabras claveMEA-BP neural network, Aluminum alloy heat treatment, Aging treatment, Numerical simulation, Microstructure

© 2023 Xuewen Chen, published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Palabras clave
MEA-BP neural network, Aluminum alloy heat treatment, Aging treatment, Numerical simulation, Microstructure