锻压技术

宽腹高筋大锻件局部加载成形的狼群算法优化

英文标题：Optimization on wolf pack algorithm of local loading forming for wide belly high rib large forgings

作者：赵成喜陈金萍

单位：大连海洋大学

关键词：宽腹高筋大锻件局部加载成形晶粒尺寸筋部填充性最优拉丁超立方狼群算法 BP神经网络

分类号：TG316

出版年,卷(期):页码：2020,45(9):211-218

摘要：

为了提高宽腹高筋大锻件筋部填充性和晶粒尺寸的均匀度，提出了模拟狼群围捕猎物过程的工艺参数优化方法。分析了宽腹高筋大锻件的局部加载成形工艺，以坯料温度、压力机加载速度、单次压下长度作为优化参数，以筋部填充性和晶粒均匀性为目标，建立了优化模型。使用最优拉丁超立方法设计了实验，得到了不同实验条件下的有限元仿真结果。使用BP神经网络拟合质量参数与工艺参数之间的非线性关系，经测试，BP神经网络的拟合精度较高。模拟狼群围捕猎物过程，将最优参数组合视为食物浓度最大位置，则可以将参数寻优过程转化为狼群算法的搜索过程。经仿真和生产验证，优化后的筋部等效未填充高度减少了68.9%，晶粒尺寸均值减小了18.2%，晶粒尺寸标准差减小了27.5%，证明了该优化方法的有效性。

In order to improve the fillability of the wide belly high rib large forgings and the uniformity of grain size, the process parameters optimization method simulating wolf pack hunting process was proposed, and the local loading forming process of the wide belly high rib large forgings was analyzed. Then, the blank temperature, the loading speed of press and the once reduction lengh were chosen as the optimizing parameters, and the optimizing model was built by setting the ribs fillability and the grain uniformity as the goals. Next, the experiment was designed by optimal Latin hypercube method, and the finite element simulation result was acquired under the different experiment conditions. Furthermore, the nonlinear relationship of quality parameters and process parameters was fit by BP neutral network, and the fitting accuracy of BP neutral network was high by test. Finally, the process of wolves hunting was simulated, and considering the optimal parameter combination as the position of the maximum food concentration, the parameter optimization process was transformed into the search process of the wolf algorithm. The simulation and verification of production show that equivalent unfilled height of rib decreases by 68.9%, average grain size decreases by 18.2%, grain size standard deviation decreases by 27.5% after optimization, and the validity of the optimizing method is verified.

基金项目：

辽宁省教育厅2019年度科学研究项目（QL201914）

赵成喜（1978-），男，硕士，讲师 E-mail：32566778@qq.com

参考文献：

［1］刘一夫,任慧龙,张鑫,等. 几种截面形式铝合金加筋板极限强度数值研究
［J］.船舶工程, 2018,40(S1):243-247.

Liu Y F, Ren H L, Zhang X, et al. Numerical investigation of ultimate strength of stiffened plates with various cross-section forms
［J］. Ship Engineering, 2018,40(S1):243-247.

［2］魏科. 钛合金大型筋板构件等温局部加载成形不等厚坯料优化研究
［D］. 西安：西北工业大学,2018.

Wei K. Study on Unequal-thickness Billet Optimization for Isothermal Local Loading Forming of Titanium Alloys Large-scale Rib-Web Component
［D］. Xi′an: Northwestern Polytechnical University, 2018.

［3］张大伟.钛合金筋板类构件局部加载成形有限元仿真分析中的摩擦及其影响
［J］.航空制造技术,2017，(4):34-41.

Zhang D W. Friction and influence in FEM simulation of local loading process for titanium alloy rib-web component
［J］. Aeronautical Manufacturing Technology, 2017，(4):34-41

［4］肖娟,蔡展飞,韩金全,等. 两道次成形过程中拉伸速度及热处理对2A12-O铝合金晶粒尺寸的影响
［J］.精密成形工程,2019,11(3):51-58.

Xiao J, Cai Z F, Han J Q, et al. Effects of tensile speed and heat treatment on grain size of aluminium alloy in two-step deformation
［J］. Journal of Netshape Forming Engineering, 2019,11(3):51-58.

［5］罗俊鹏,杨成斌,丁凯,等.GH4169合金细长轴锻件局部锻造成形工艺研究
［J］.热加工工艺,2018,47(13):160-163.

Luo J P, Yang C B, Ding K, et al. Study on local forging process of GH4169 alloy slender shaft forgings
［J］. Hot Working Technology,2018,47(13):160-163.

［6］李萍,胡传鹏,杨卫正,等.带横向外凸筋多楔轮旋压成形规律及工艺研究
［J］.哈尔滨工业大学学报,2018,50(11):153-159.

Li P, Yang C P, Yang W Z, et al. Study on spinning formingprocess of multi-wedge belt pulley with transverse outer ribs
［J］. Journal of Harbin Institute of Technology, 2018,50(11):153-159.

［7］刘左发,冯文杰,陈莹莹.离合器外齿毂冷挤压成形工艺参数多目标优化
［J］.塑性工程学报,2020,27(4):13-20.

Liu Z F, Feng W J, Chen Y Y. Multi-objective optimization of cold extrusion forming process parameters for clutch outer gear hub
［J］. Journal of Plasticity Engineering, 2020,27(4):13-20.

［8］莫洪武,塔金星.基于神经网络的机械油泵轴锻造工艺优化
［J］.锻压技术,2018,43(9):21-25.

Mo H W, Ta J X. Forging process optimization on mechanical oil pump shafts based on neural network
［J］. Forging & Stamping Technology, 2018,43(9):21-25.

［9］万鹏. 基于田口算法的汽车翼子板冲压成形过程模拟及其成形工艺参数优化
［D］.芜湖：安徽工程大学,2017.

Wan P. Simulation of Forming Process and Optimization of Forming Process Parameters of Automobile Fender Based on Taguchi Algorithm
［D］. Wuhu: Anhui Polytechnic University, 2017.

［10］季宁,张卫星,于洋洋,等.基于径向基函数神经网络和多岛遗传算法的注射成型质量控制与预测
［J］.工程塑料应用,2020,48(4):62-68.

Ji N, Zhang W X, Yu Y Y, et al. Quality control and prediction of injection molding based on RBF neural network and MIGA
［J］. Engineering Plastics Application, 2020,48(4):62-68.

［11］韩建福,杜昌平,叶志贤,等. 基于双BP神经网络的扑翼飞行器气动参数辨识
［J］.计算机应用,2019,39(S2):299-302.

Han J F, Du C P, Ye Z X, et al. Identification of aerodynamic parameters of flapping-wing micro aerial vehicle based on double BP neural network
［J］. Journal of Computer Applications, 2019,39(S2):299-302.

［12］周璟.混沌狼群围捕算法的车间机器人导航路径规划
［J］.机械设计与制造,2020，(1):251-255.

Zhou J. Workshop used robot navigation path planning method based on chaotic wolf pack besieging algorithm
［J］. Machinery Design & Manufacture, 2020，(1):251-255.

服务与反馈：

【文章下载】【加入收藏】