锻压技术

车辆座椅侧板冲压工艺参数的蜂群算法优化

英文标题：Bee colony algorithm optimization on stamping process parameters for vehicle seat side panel

作者：程宝鑫1 曹玉苹2

单位：1.天津石油职业技术学院机械工程系 2.中国石油大学（华东）信息与控制工程学院

分类号：TG386. 3

出版年,卷(期):页码：2021,46(12):79-85

摘要：

为了减小车辆座椅侧板冲压成形回弹量，提出了基于交叉-变异蜂群算法的冲压工艺参数优化方法。分析了冲压工艺参数对车辆座椅侧板回弹量的影响规律，建立了以回弹量最小为目标的冲压优化模型。基于最优拉丁超立方抽样法在四维空间中抽取了120个点，使用Autoform软件对实验结果进行了仿真。使用单隐藏层神经网络拟合工艺参数与回弹量之间的非线性关系，经验证神经网络的拟合精度较高。为了加深人工蜂群算法的局部搜索深度，将交叉和变异思想引入到该算法中，从而给出了基于交叉-变异蜂群算法的参数优化方法。经实验验证，座椅侧板冲压件的减薄率和增厚率满足约束条件，回弹量仅为1.725 mm，说明提出的优化方法可有效减小座椅侧板的冲压回弹量。

In order to reduce springback amount of vehicle seat side panel after stamping, the optimization method of stamping process parameters based on crossover-mutation bee colony algorithm was proposed. Then, the influences of the stamping process parameters on the springback amount of vehicle seat side panel were analyzed, and a stamping optimization model was established by setting the goal of minimizing springback amount. Based on the optimal Latin hypercube sampling method, 120 points were extracted in 4-dimensional space, and the experimental results were simulated by software AutoForm. Furthermore, the non-linear relationship between process parameters and springback amount was fit by single hidden layer neutral network, which was clarified that the fitting accuracy of neutral network was very high. In order to deepen the local searching depth of artificial bee colony algorithm, the idea of crossover and mutation was introduced into the algorithm, and a parameter optimization method based on crossover-mutation bee colony algorithm was given. Experimental verification shows that the thinning rate and thickening rate of the stamping piece for seat side panel meet the constraint conditions, and the springback amount is only 1.725 mm, indicating that the proposed optimization method can effectively reduce the springback amount of seat side panel after stamping.

基金项目：

2018年度天津市高等职业技术教育研究会课题（XVⅢ4006）

程宝鑫（1983-），男，学士，副教授 E-mail：sindey19831025@163.com

参考文献：

[1]秦国华, 史双喜,刘跃峰,等.汽车左座椅摇臂冲压工艺分析及级进模设计[J].锻压技术,2019,44(9):113-117.

Qin G H, Shi S X, Liu Y F, et al. Analysis on stamping process of rocker arm for automobile left seat and progressive die design [J]. Forging & Stamping Technology, 2019,44(9):113-117.

[2]王章忠, 巴志新,李琦,等.热冲压工艺参数对22MnB5马氏体钢汽车B柱性能影响的有限元模拟[J].金属热处理,2020,45(5):221-228.

Wang Z Z, Ba Z X, Li Q, et al. Finite element simulation of influence of hot stamping process parameters on properties of 22MnB5 martensitic steel for automobile Bpillar [J]. Heat Treatment of Metals, 2020,45(5):221-228.

[3]林浩波, 刘军辉,吴立国. 基于遗传算法的防撞钢梁热冲压成形工艺优化[J].塑性工程学报,2019,26(5):65-69.

Lin H B, Liu J H, Wu L G. Optimization of hot stamping process for anticollision beam based on genetic algorithms [J]. Journal of Plasticity Engineering, 2019,26(5):65-69.

[4]刘强, 俞国燕,梅端. 基于Dynaform与RBFNSGAII算法的冲压成形工艺参数多目标优化[J].塑性工程学报,2020,27(3):16-25.

Liu Q, Yu G Y, Mei D. Multiobjective optimization of stamping forming process parameters based on Dynaform and RBFNSGAII algorithm [J]. Journal of Plasticity Engineering, 2020,27(3):16-25.

[5]邓振鹏, 周惦武,蒋朋松,等.基于正交试验的锆合金薄板带材冲压工艺参数优化[J].锻压技术,2019,44(9):12-17.

Deng Z P, Zhou D W, Jiang P S, et al. Optimization on stamping process parameters for zirconium alloy sheet strip based on orthogonal experiment [J]. Forging & Stamping Technology, 2019,44(9):12-17.

[6]莫明立. 某汽车后地板横梁冲压成形数值模拟及参数优化[J].锻压技术,2019,44(6):41-45.

Mo M L. Numerical simulation and parameters optimization on stamping for rear floor beam of an automobile [J]. Forging & Stamping Technology, 2019,44(6):41-45.

[7]谢晖, 沈云飞,王杭燕. 基于改进响应面模型的冲压回弹工艺稳健性优化[J].塑性工程学报,2018,25(4):26-32.

Xie H, Shen Y F, Wang H Y. Robustness optimization of stamping springback based on improved response surface model [J]. Journal of Plasticity Engineering, 2018,25(4):26-32.

[8]龚灯, 郭利,韩刚. S420MC高强度钢流变力学性能的实验研究[J].西安航空学院学报,2017,35(1):34-37.

Gong D, Guo L, Han G. Experimental study of rheological properties of high strength steel S420MC [J]. Journal of Xi′an Aeronautical University, 2017,35(1):34-37.

[9]季宁, 张卫星,于洋洋,等.基于最优拉丁超立方抽样方法和NSGAⅡ算法的注射成型多目标优化[J].工程塑料应用,2020,48(3):72-77.

Ji N, Zhang W X, Yu Y Y, et al. Multiobjective optimization of injection molding based on optimal latin hypercube sampling method and nsgaii algorithm [J]. Engineering Plastics Application, 2020,48(3):72-77.

[10]熊杰, 魏勇,严丹.基于BP神经网络的超声波温湿度补偿算法研究与应用[J].现代电子技术,2020,43(9):113-116.

Xiong J, Wei Y, Yan D. Research and application of ultrasonic temperaturehumidity compensation algorithm base on BP neural network [J]. Modern Electronics Technique, 2020,43(9):113-116.

[11]李志敏, 张伟.基于差分进化人工蜂群算法的云计算资源调度[J].计算机工程与设计,2018,39(11):3451-3455.

Li Z M, Zhang W. Clouding computing resource scheduling based on differential evolution artificial bee colony algorithm [J]. Computer Engineering and Design, 2018,39(11):3451-3455.

服务与反馈：

【文章下载】【加入收藏】