锻压技术

基于MLP-PSO算法的锆合金支撑架成形质量优化

英文标题：Optimization on forming quality for zirconium alloy support frame based on MLP-PSO algorithm

分类号：TG386

出版年,卷(期):页码：2023,48(3):61-67

摘要：

为了提高锆合金支撑架的冲压成形质量，基于Dynaform软件和正交试验设计方法，将最大减薄率作为评价指标，研究了折弯半径、板料厚度、摩擦因数、凸凹模间隙、压边力和冲压速度等参数对支撑架成形质量的影响规律。通过数值模拟获得了样本数据，利用多层感知机神经网络训练出预测支撑架减薄率的模型，对各因素的相关性进行分析，并通过粒子群优化算法得到了最优参数方案。结果表明：多层感知机神经网络模型能够有效预测支撑架的减薄率。在影响支撑架冲压的各参数中，折弯半径和摩擦因数的影响较大，凸凹模间隙和冲压速度的影响较小。采用粒子群算法优化后的参数方案进行冲压成形，最大减薄率降低24.2%，可有效降低支撑架的破裂率，提高支撑架的冲压成形质量。

In order to improve the stamping quality of zirconium alloy support frame, based on Dynaform software and orthogonal experimental design method, taking the maximum thinning rate as the evaluation index, the influence laws of parameters such as bending radius, sheet thickness, friction factor, clearance between punch and die, blank holder force and stamping speed on the forming quality of support frame were investigated. Then, the sample data was obtained by numerical simulation, and the model for predicting the thinning rate of support frame was trained by using the multi-layer perceptron neural network. Furthermore, the correlation of each factor was analyzed, and the optimal parameter scheme was obtained by the particle swarm optimization algorithm. The results show that the multi-layer perceptron neural network model can effectively predict the thinning rate of support frame. Among the parameters affecting the stamping of support frame, the bending radius and friction factor have a greater influence, while the clearance between punch and die and the stamping speed have less influence. Using the parameter scheme optimized by particle swarm algorithm for stamping, the maximum thinning rate is reduced by 24.2%, which can effectively reduce the fracture rate of support frame and improve the stamping quality of support frame.

基金项目：

国家科技重大专项子课题（761215007）；湖南省教育厅资助科研项目（22C0951）

作者简介：楚岩（1998-），男，硕士研究生 E-mail：chuy0807@163.com 通信作者：毛建中（1963-），男，博士，教授 E-mail：maojianzhong66@163.com

参考文献：

[1]刘雪. 环形燃料组件定位格架外条带及栅元模具设计与工艺开发[D]. 长沙：湖南大学, 2018.

Liu X. Design and Process Development of Outer Strip and Cell Mold for Annular Fuel Assembly Spacer Grid[D]. Changsha: Hunan University, 2018.

[2]Lei C Y, Mao J Z, Zhang X M, et al. A comparison study of the yield surface exponent of the Barlat yield function on the forming limit curve prediction of zirconium alloys with M-K method[J]. International Journal of Material Forming, 2021, 14: 467-484.

[3]邓振鹏. 新锆合金薄板带材的可冲性及冲制工艺优化[D]. 长沙：湖南大学, 2019.

Deng Z P. Punching Property and Punching Process Optimization of New Zirconium Alloy Sheet Strip[D]. Changsha: Hunan University, 2019.

[4]李燕乐, 陈晓晓, 翟维东, 等. 基于响应曲面法的板料渐进成形最大减薄率预测与分析[J]. 吉林大学学报: 工学版, 2019,49(2):529-535.

Li Y L, Chen X X, Zhai W D, et al. Prediction and analysis of maximum thinning rate of sheet metal incremental forming based on response surface method[J]. Journal of Jilin University: Engineering and Technology Edition, 2019, 49(2): 529-535.

[5]何彦, 肖圳, 李育锋, 等. 使用CNN-SVR的汽车组合仪表组装质量预测方法[J]. 中国机械工程, 2022,33(7):825-833.

He Y, Xiao Z, Li Y F, et al. Assembly quality prediction method of automobile combination meter using CNN-SVR[J]. China Mechanical Engineering, 2022, 33(7): 825-833.

[6]鲍宏, 杨靖, 柯庆镝, 等. 基于支持向量回归的熔丝制造3D打印能效优化模型[J]. 中国机械工程, 2022,33(18): 2215-2226.

Bao H, Yang J, Ke Q D, et al. Energy efficiency optimization model of fuse filament manufacturing 3D printing based on support vector regression[J]. China Mechanical Engineering, 2022,33(18): 2215-2226.

[7]冯斌. 基于MLP的锆合金刚凸成形减薄率预测[D]. 长沙：湖南大学, 2020.

Feng B. Prediction of Thinning Rate of Zirconium Alloy Rigid-convex Forming Based on MLP[D]. Changsha: Hunan University, 2020.

[8]谢延敏, 孙新强, 田银, 等. 基于改进粒子群算法和小波神经网络的高强钢扭曲回弹工艺参数优化[J]. 机械工程学报, 2016,52(19):162-167.

Xie Y M, Sun X Q, Tian Y, et al. Optimization of process parameters for twisting springback of high-strength steel based on improved particle swarm optimization and wavelet neural network[J]. Journal of Mechanical Engineering, 2016, 52(19): 162-167.

[9]GB/T 228.1—2021,金属材料拉伸试验第1部分：室温试验方法 [S].

GB/T 228.1—2021, Metallic materials—Tensile test—Part 1: Test method at room temperature [S].

[10]童洲, 谈毅,段海峰,等. 基于正交试验和灰色关联的模块锻件热处理工艺优化[J]. 锻压技术,2021,46(8):186-192.

Tong Z，Tan Y，Duan H F，et al. Optimization of heat treatment process for module forgings based on orthogonal test and gray correlation[J]. Forging & Stamping Technology，2021,46(8):186-192.

[11]王格格, 郭涛, 李贵洋. 多层感知器深度卷积生成对抗网络[J]. 计算机科学, 2019,46(9):243-249.

Wang G G, Guo T, Li G Y. Multilayer perceptron deep convolutional generative adversarial networks[J]. Computer Science, 2019, 46(9): 243-249.

[12]王小川, 史峰, 郁磊, 等. Matlab神经网络43个案例分析[M]. 北京：北京航空航天大学出版社, 2013.

Wang X C, Shi F, Yu L, et al. Matlab Neural Network 43 Case Analysis[M]. Beijing: Beijing University of Aeronautics and Astronautics Press, 2013.

[13]孙恒, 耿金亮, 那凤祎, 等. 基于粒子群优化算法的双混合制冷剂液化工艺参数优化[J]. 天然气化工—C1化学与化工, 2022,47(2):116-121.

Sun H, Geng J L, Na F Y, et al. Optimization of process parameters for dual-mixed refrigerant liquefaction based on particle swarm optimization algorithm[J]. Natural Gas Chemical Industry, 2022,47(2): 116-121.

服务与反馈：

【本网站尚未开通全文下载服务】【加入收藏】