网站首页期刊简介编委会过刊目录投稿指南广告合作征订与发行联系我们English
基于卷积神经网络的板料挤压成形力预测
英文标题:Prediction on sheet metal extrusion forming force based on convolutional neural network
作者:赵震1 2 沈大为1 2 曹益旗1 2 向华1 2 庄新村1 2 
单位:1.上海交通大学 塑性成形技术与装备研究院   2. 上海交通大学 模具CAD国家工程研究中心 
关键词:板料挤压 成形力 精冲 卷积神经网络 凸凹模形状 
分类号:TG376.1
出版年,卷(期):页码:2021,46(9):76-84
摘要:

 板料挤压过程中的成形力计算是合理进行模具设计和压力机选择的重要依据。为了快速、准确地获取挤压力数值,提出了一种结合有限元仿真和卷积神经网络的板料挤压力预测模型。利用所建立的板料挤压有限元模型,结合知识模板技术,批量获取了不同工艺参数下的成形力数据集;在此基础上,针对凸凹模几何形状难以用参数统一表征的问题,以凸凹模的轮廓图像为直接输入量,基于凸、凹模基础形状类,采用混合卷积神经网络结构,构建了适用于不同工艺参数的挤压力预测模型。经过验证和评估,所建模型对于规则形状和组合形状板料挤压力均有较高的预测精度,可以满足工程计算的需求。

 The calculation of forming force in the extrusion process of sheet is an important basis for reasonable mold design and press selection. Therefore, in order to obtain the extrusion force value quickly and accurately, a prediction model of extrusion force for sheet metal was proposed by combining with the finite element simulation and the convolutional neural network. Using  the established finite element model for sheet metal and combining with the knowledge template technology, the data sets of forming forces under different process parameters were obtained in batches. On the basis of this, considering the problem that it was difficult to describe the geometric shape of punch and die with the uniform parameter, taking the contour image of punch and die as the direct input, based on the basic shapes of punch and die, the prediction model of extrusion force suitable for the different process parameters was established by the hybrid convolutional neural network structure. Through the verification and evaluation, the established model has high precision for extrusion force of the sheet metal with the regular shape or combined shape and meet the needs of engineering calculation.

基金项目:
国家自然科学基金资助项目(51875351)
作者简介:
赵震(1972-),男,博士,教授 E-mail:zzhao@sjtu.edu.cn
参考文献:

 [1]涂光祺. 精冲技术[M]. 北京:机械工业出版社,2005.


 


Tu G Q. Fineblanking Technology[M]. Beijing: China Machine Press,2005.


 


[2]周林, 刘艳雄, 金红,. 金属板料精冲挤压复合成形工艺相关国家标准研制的必要性及技术概要[J]. 锻压技术, 2019, 44(11): 186-190.


 


Zhou L, Liu Y X,Jin H,et al. The necessity and technical outline of developing related national standard for sheet metal fine blanking extrusion composite forming process[J]. Forging & Stamping Technology, 2019, 44(11): 186-190.


 


[3]Liu Y X,Shu Y W, Chen H, et al.Deformation characteristics analysis of the fineblanking-extrusion flanging process[J].Procedia Manufacturing,2020,50:129-133


 


[4]Schmidt R A, Birzer F, Hofel P, et al. Cold Forming and Fineblanking, a Handbook on Cold Processing, Steel Material Properties, Part Design[M]. Munchen: Carl Hanser Verlag, 2007.


 


[5]吕琳, 公冶凡娇, 邓雨辰,. 凸台板件挤压成形中的几个问题[J]. 热加工工艺, 2019, 48(11): 95-99.


 


Lyu L, Gongye F J, Deng Y C,et al. Some problems in extrusion forming of convex plate[J]. Hot Working Technology, 2019, 48(11): 95-99.


 


[6]庄新村, 赵震, 向华. 中厚板挤压成形力能分析及预测[J]. 塑性工程学报, 2009, 16(6): 33-38.


 


Zhuang X C, Zhao Z, Xiang H. Analysis and prediction of extrusion forming force for medium plate[J]. Journal of Plasticity Engineering, 2009, 16(6): 33-38.


 


[7]黄兴. 中厚板挤压成形力计算模型研究[D]. 上海:上海交通大学, 2009.


 


Huang X. Research on Calculation Model of Forming Force in Medium-thick Sheet Metal Extrusion Process[D]. Shanghai:Shanghai Jiao Tong University, 2009.


 


[8]冯怡爽,何霁,韩国丰,.金属板材塑性本构关系的深度学习预测方法及建模[J].塑性工程学报,2021,28(6):34-46.


 


Feng Y S, He J, Han G F, et al. Deep learning prediction method and modeling for plastic constitutive relation of sheet metal[J]. Journal of Plasticity Engineering,2021,28(6):34-46.


 


[9]柏阳,吴玉程,罗志勇, .基于Arrhenius方程和BP神经网络的2024Al/Al_(18)B_4O_(33)w复合材料热变形流变应力预测[J].锻压技术,2019,44(8):168-175181.


 


Bo Y, Wu Y C, Luo Z Y, et al. Flow stress prediction of 2024Al/Al_(18)B_4O_(33) w composites under thermal deformation based on Arrhenius equation and BP neural network[J]. Forging & Stamping Technology, 2019,44(8):168-175181.


 


[10]郭鹏,张新艳,余建波.基于深度强化学习与有限元仿真集成的拉深成形控制[J].机械工程学报,2020,56(20):47-58.


 


Guo P, Zhang X Y, Yu J B. Deep drawing forming control based on integration of deep reinforcement learning and finite element simulation [J]. Journal of Mechanical Engineering, 2020,56(20):47-58.


 


[11]Zhuang X C, Zhao Z, Li H Y, et al. Experimental methodology for obtaining the flow curve of sheet materials in a wide range of strains[J]. Steel Research International, 2013, 84(2): 146-154.


 


[12]曹益旗. 基于深度学习的板料挤压成形力预测研究[D].上海:上海交通大学, 2020.


 


Cao Y Q. Research on Prediction of Sheet Metal Extrusion Forming Force Based on Deep Learning[D]. Shanghai:Shanghai Jiao Tong University,2020.


 


[13]洪永放, 庄新村, 丁振文, . 不同挤压次数下板料挤压凸模磨损规律研究[J]. 塑性工程学报, 2018, 25(5): 130-135.


 


Hong Y F, Zhuang X C, Ding Z W, et al. Study on wear law of plate punch under different extrusion times [J]. Journal of Plasticity Engineering, 2018, 25(5): 130-135.


 

服务与反馈:
本网站尚未开通全文下载服务】【加入收藏
《锻压技术》编辑部版权所有

中国机械工业联合会主管 北京机电研究所有限公司 中国机械工程学会塑性工程分会主办
联系地址:北京市海淀区学清路18号 邮编:100083
电话:+86-010-82415085 传真:+86-010-62920652
E-mail: fst@263.net(稿件) dyjsjournal@163.com(广告)
京ICP备07007000号-9