欢迎访问《兵工学报》官方网站,今天是 分享到:

兵工学报 ›› 2020, Vol. 41 ›› Issue (8): 1623-1632.doi: 10.3969/j.issn.1000-1093.2020.08.017

• 论文 • 上一篇    下一篇

基于翘曲度阈值约束的铝合金板材热轧工艺优化与实验验证

叶海潮1,2, 范武林1, 秦国华1, 林锋1, 郑许3,4, 杨扬1   

  1. (1.南昌航空大学 航空制造工程学院, 江西 南昌 330063; 2.南京航空航天大学 机电学院, 江苏 南京 210016;3.中南大学 材料科学与工程学院, 湖南 长沙 410083; 4.广西南南铝加工有限公司, 广西 南宁 530031)
  • 收稿日期:2019-11-15 修回日期:2019-11-15 上线日期:2020-09-23
  • 作者简介:叶海潮(1979—),男,讲师。E-mail:xxjy2000@163.com;
    秦国华(1970―),男,教授,博士生导师。E-mail: qghwzx@126.com
  • 基金资助:
    国家自然科学基金项目(51765047);江西省主要学科学术和技术带头人计划项目(20172BCB22013);江西省科技厅重点研发计划项目(20192BBEL50001);江西省优势科技创新团队建设计划项目(20181BCB24007);江西省教育厅科技项目(GJJ170572)

Optimization and Experimental Validation of Aluminum Alloy Plate Hot-rolling Based on Constraint of Warpage Threshold

YE Haichao1,2, FAN Wulin1, QIN Guohua1, LIN Feng1, ZHENG Xu3,4, YANG Yang1   

  1. (1.School of Aeronautical Manufacturing Engineering, Nanchang Hangkong University, Nanchang 330063, Jiangxi, China; 2.College of Mechanical & Electrical Engineering, Nanjing University of Aeronautics and Astronautics, Nanjing 210016, Jiangsu, China; 3.School of Materials Science and Engineering, Central South University, Changsha 410083, Hunan, China; 4.Alnan Aluminium Co., Ltd., Nanning 530031, Guangxi, China)
  • Received:2019-11-15 Revised:2019-11-15 Online:2020-09-23

摘要: 蛇形热轧是能够保证板材中心发生充分变形、破碎铸造状态下粗大晶粒为加工组织的铝合金厚板制备关键工艺,为此研究了板材在蛇形轧制工艺过程中产生的翘曲变形及其控制问题。通过研究铝合金板材轧制的机理模型,构建铝合金板材的蛇形轧制有限元仿真模型,得到有限元仿真值与参考值、测量数据吻合较好,应力、应变的最大相对误差仅为6.92%和7.33%,而翘曲因子也只有13%左右的误差;基于有限的训练样本组合,建立等效应变和翘曲因子的耦合预测模型,得到预测值与相应仿真值的误差不超过6%;在翘曲因子不超过规定阈值的约束条件下,建立使等效应变达到最大的轧制工艺参数优化模型,通过轧件等效应变构造个体的适应度函数,提出了轧制工艺参数优化模型的遗传算法求解方法。利用该方法不仅实现了铝合金板材轧制工艺参数的合理设计与选择,而且提高了板材轧制过程的仿真计算效率。

关键词: 铝合金板材, 蛇形轧制, 异速比, 翘曲度阈值, 耦合预测, 遗传算法

Abstract: The snake-like hot-rolling can divide the casted coarse grains into the machined microstructures by completely deforming the center of aluminum alloy plate. It is the key to prepare the thick aluminum alloy plates. Therefore, the warpage and its control in the process of snake-like rolling are studied. The mechanism model of hot-rolling of aluminum alloy plates is studied, and a finite element model is reasonably established for some training samples of neural network. The accuracy of the proposed model is verified by comparing the simulated and measured results. The maximum relative errors of stress and strain are 6.92% and 7.33%, respectively. Moreover, there exists about 13% relative error of warp factor. And a coupled prediction model of equivalent strain and warp factor is proposed according to the finite groups of training samples. The coupled prediction model is able to predict the simulated results within 6 % error margin. Under the condition that the warp factor is not more than the given threshold value, an optimal model with the objective of maximum equivalent strain is proposed for the rolling parameters. When an individual fitness is constructed to be a function of the average equivalent strain, the genetic algorithm can be used to solve the optimal model for rolling parameters. The proposed method can be used not only to improve the simulating calculation efficiency of rolling process, but also provide a basic theory of design and selection of rolling parameters. The propose method can not only improve the calculation efficiency of simulation of rolling process, but also provide a basic theory of design and selection of rolling parameters.

Key words: aluminumalloyplate, snake-likehotrolling, velocityratio, wrapthreshold, coupledprediction, geneticalgorithm

中图分类号: