| 基于RBF神经网络模型的板料成形变压边力优化 |
| |
| 引用本文： | 谢延敏,何育军,田银.基于RBF神经网络模型的板料成形变压边力优化[J].西南交通大学学报,2016,29(1):121-127. |
| |
| 作者姓名： | 谢延敏 何育军 田银 |
| |
| 基金项目： | 国家自然科学基金资助项目(51275431) |
| |
| 摘 要： | 为了解决变压边力优化过程中RBF(radial basis function )神经网络隐层节点训练难的问题,利用人工智能算法的优越性,建立了基于人工免疫算法的RBF神经网络,并将其用于非线性函数的逼近中.结合分块压边圈与改变压边力控制技术,通过Dynaform软件进行数值模拟获得成形数据,建立了变压边力与成形质量之间的RBF神经网络近似模型.利用人工免疫智能算法对该近似模型进行优化,获得最优压边力参数.将该方法应用于S形梁冲压成形中,与优化前的结果进行比较,采用优化后最优变压边力可以抑制起皱,最大起皱量减少了89.53%.
|
| 关 键 词： | 板料成形 变压边力 数值模拟 RBF神经网络 |
| 收稿时间： | 2015-06-16 |
Optimization of Variable Blank Holder Forces in Sheet Metal Forming Based on RBF Neural Network Model |
| |
| XIE Yanmin,HE Yujun,TIAN Yin.Optimization of Variable Blank Holder Forces in Sheet Metal Forming Based on RBF Neural Network Model[J].Journal of Southwest Jiaotong University,2016,29(1):121-127. |
| |
| Authors: | XIE Yanmin HE Yujun TIAN Yin |
| |
| Abstract: | In order to solve the difficulty of training the hidden layer nodes in radial basis function (RBF) neural network during the optimization of variable blank holder forces,a RBF neural network based on the artificial immune algorithm was established by taking advantages of artificial intelligence algorithms, and then used to approximate a nonlinear function. Using both the block blank holder technology and the variable blank holder force control technology, numerical simulations were conducted in Dynaform to obtain the forming data, and an approximate model of RBF neural network was established between the variable blank holder forces and the forming quality. The approximate model was optimized by artificial immune algorithm to obtain the optimal blank holder force parameters. In addition, the method was applied to the S-rail stamping. The results show that compared with that before optimization, the maximum wrinkle amount was reduced by 89.53%, and wrinkles could be effectively controlled by the optimized variable blank holder forces. |
| |
| Keywords: | |
|
| 点击此处可从《西南交通大学学报》浏览原始摘要信息 |
|
点击此处可从《西南交通大学学报》下载全文 |
|
