基于灰色关联的铝合金板拉深成形扭曲回弹工艺参数优化

发布时间：2018-03-15 01:25

本文选题：铝合金板　切入点：灰色关联　出处：《锻压技术》2015年03期 　论文类型：期刊论文

【摘要】：针对铝合金板成形工艺参数如何选取的难点,利用灰色关联准则对铝合金板成形质量进行分析,通过因子关联度的方差分析,获得铝合金板成形工艺中的主要影响因子。为了减少板料成形工艺参数优化时间,以铝合金板成形中主要影响因子为设计变量,以板料成形后扭曲回弹、增厚、减薄为成形目标,使用拉丁超立方抽取样本,通过Dynaform软件进行数值模拟获得训练样本,利用人工免疫算法训练RBF神经网络,建立主要影响因子与成形目标之间的RBF神经网络近似模型,最后采用人工免疫算法对该模型进行优化,获得最优工艺参数。以Numisheet'96 S梁为研究对象,利用本文所提出的方法进行拉深成形研究,通过对比分析优化前后的成形结果,证明了该方法能极大地提高铝合金板的成形质量。
[Abstract]:Aiming at the difficulty Aluminum Alloy plate forming process of how to select the parameters of the Aluminum Alloy plate forming quality was analyzed using grey correlation criterion, through variance factor correlation analysis, obtained Aluminum Alloy plate forming process of the main influence factor. In order to reduce the parameter optimization of sheet metal forming, the main influencing factor Aluminum Alloy plate forming as the design variables, with the sheet forming twist springback, thickening and thinning for forming the target, using the Latin hypercube sampling, numerical simulation is conducted to obtain the training samples by Dynaform software, using artificial immune algorithm to train RBF neural network, RBF neural network approximation model is established between the main factors and forming target, finally the artificial immune algorithm to optimize the model, obtain the optimal parameters. The Numisheet'96 S beam as the research object, was carried out using the method proposed in this paper By comparing and analyzing the forming results before and after the optimization, it is proved that this method can greatly improve the forming quality of the aluminum alloy plate.

【作者单位】：西南交通大学机械工程学院;
【基金】：国家自然科学基金资助项目(51275431)
【分类号】：TG389

【参考文献】