锻压技术

基于小波神经网络和粒子群算法的铝合金板冲压回弹工艺参数优化

英文标题：Parameters optimization of stamping and springback for aluminum-alloy sheet based on wavelet neural network and particle swarm optimization algorithm

作者：孙新强谢延敏田银何育军

单位：西南交通大学

关键词：铝合金回弹小波神经网络粒子群算法

分类号：

出版年,卷(期):页码：2015,40(1):137-142

摘要：

针对铝合金复杂件冲压后出现的较大回弹缺陷，同时为减少冲压成形工艺参数的优化时间，使用有限元仿真软件DYNAFORM对冲压成形及回弹过程进行数值模拟，在确保数值模拟与试验结果基本一致的基础上，利用代理模型对回弹进行了优化研究。以NUMISHEET′96 S梁为研究对象，凸模圆角半径、凹模圆角半径、压边力、板料厚度作为影响因素，成形后最大回弹值作为成形目标，运用拉丁超立方抽样，通过数值仿真获得样本数据，建立影响因素与成形目标之间的小波神经网络代理模型，利用粒子群算法对该模型迭代寻优获得最优工艺参数。结果表明：小波神经网络能较好地描述板料工艺参数与回弹之间的映射关系，优化后成形件的回弹量大大减小。

For the large springback appeared after the stamping process of aluminum-alloy sheet, and for reducing the time of optimizing process parameters, the stamping process and springback were numerically simulated based on the finite element analysis software DYNAFORM. On the basis of the numerical simulation consisting with experimental results, the agent model was used to the optimization research of springback. The S-rail of NUMISHEET′96 was taken into account, with the fillet radius of punch, the fillet radius of die, the blank holder force and the sheet thickness as influencing factors, and the maximum springback value after stamping was regarded as forming target. By using latin hypercube to sample, and the simulation was carried to get the samples, and the wavelet neural network agent model between influencing factors and forming target was built. Then the optimal solution was obtained by iterations of particle swarm optimization algorithm. The results show that the wavelet neural network agent model can describe the input-output relationship between the sheet forming process parameters and the springback, and the springback can be remarkably reduced after the optimization.

基金项目：

国家自然科学基金资助项目（51005193）

孙新强（1991-），男，硕士研究生

参考文献：

［1］张宇，朱平，陈关龙，等. 基于有限元法的轿车发动机罩板轻量化设计[J］. 上海交通大学学报，2006，40(1)：163-166.Zhang Y, Zhu P, Chen G L，et al. The lightweight design of bonnet in auto-body based on finite element method[J］. Journal of Shanghai Jiaotong University, 2006，40(1):163-166.

［2］谢延敏，王新宝，王智，等. 基于灰色理论和GA-BP的拉延筋参数反求[J］. 机械工程学报，2013，4(49): 44-49.Xie Y M, Wang X B, Wang Z, et al. Reverse of drawbead parameters based on grey theory and GA-BP [J］.Journal of Mechanical Engineering,2013,4(49):44-49.

［3］苏娟华，张文琼，李丽春. 神经网络与正交试验法结合优化板料成形工艺[J］. 锻压技术，2009，34(1)：154-146.Su J H, Zhang W Q, Li L C. Application of network and orthogonal experiment method in optimization of sheet metal forming process technology[J］. Forging & Stamping Technology, 2009，34(1):154-156.

［4］郑小洋，叶仲泉. Legendre小波神经网络 [J］. 重庆大学学报，2003，26(8)：48-51.Zheng X Y, Ye Z Q. Legendre wavelets neural network[J］. Journal of Chongqing University, 2003，26(8):48-51.

［5］Hamed Khodadadi, Ehsan Razavi S, Hossein Ahmadi-Noubari. A comparison between neural networks and wavelet networks in nonlinear system identification [J］. Engineering and Technology, 2012, 4(9): 1021-1026.

［6］Ferrante Neri, Ernesto Mininno, Giovanni Iacca. Compact particle swarm Optimization [J］. Information Sciences, 2013, 239:96-121.

［7］Zhan Z H, Zhang J, Li Y, et al. Adaptive particle swarm optimization [J］. IEEE Transactions on Systems, 2009, 39(6):1362-1381.

［8］王波，王灿林，梁国强. 基于粒子群寻优的D-S算法[J］. 传感器与微系统，2007，26(1)：84-86.Wang B, Wang C L, Liang G Q. D-S algorithm based on particle swarm optimizer[J］. Transducer and Microsystem Technologies, 2007, 26(1):84-86.

［9］Min-Shyang Leu, Ming-Feng Yeh. Grey particle swarm optimization [J］. Applied Soft Computing, 2012, 12(9):2985-2996.

［10］傅建，刘挺，彭必友，等. 基于计算机辅助实验的铝合金发动机罩外板成形工艺参数优化[J］. 锻压技术，2014，39(2)：31-36.Fu J, Liu T, Peng B Y, et al. Forming parameters optimization for aluminum-alloy hood outer based on computer aided test[J］. Forging & Stamping Technology, 2014, 39(2):31-36.

［11］王巍，李东升，李小强. 铝合金板弯曲回弹试验与有限元仿真[J］. 锻压技术，2013，38(3)：160-164.Wang W, Li D S, Li X Q. Test and FEM simulation of springback for aluminum alloy sheet bending[J］. Forging & Stamping Technology, 2013, 38(3):160-164.

［12］Marco Schwarze, Ivaylo N Vladimirov, Stefanie Reese. Sheet metal forming and springback simulation by means of a new reduced integration solid-shell finite element technology[J］. Comput. Methods Appl. Mech. Engre., 2011, (200):454-476.

［13］Lee S W, Yoon J W, Yang D Y. Comparative investigation into the dynamic explicit and the static implicit method for springback of sheet metal stamping[J］. Springback of Sheet Metal Stamping, 1998, (12):347-373.

［14］郭超群，陈军，陈劼实，等. 超高强度钢薄板扭曲回弹特性的数值模拟与实验分析[J］. 上海交通大学学报，2010，44(4)：468-472.Guo C Q, Chen J, Chen J S，et al. Numerical simulation and experimental validation of distortional springback of advanced high-strength st

eel sheet metal forming[J］. Journal of Shanghai Jiaotong University, 2010，44(4):468-472.

服务与反馈：

【本网站尚未开通全文下载服务】【加入收藏】