论文部分内容阅读
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。
According to the difficulty of how to select the forming parameters of the aluminum alloy plate, the forming quality of the aluminum alloy plate is analyzed by the gray relational criterion. The main influencing factors in the forming process of the aluminum alloy plate are obtained by the variance analysis of the factor correlation degree. In order to reduce the optimization of sheet metal forming process parameters, the main influence factors in the forming of aluminum alloy sheet are design variables. After forming the sheet, the material is distorted, rebounded, thickened and thinned. The samples are drawn with Latin hypercube and analyzed by Dynaform The training samples were obtained by numerical simulation. The RBF neural network was trained by artificial immune algorithm. The RBF neural network approximation model between the main influence factors and the forming target was established. Finally, the artificial immune algorithm was used to optimize the model to obtain the optimal process parameters. Taking the Numisheet’96 S beam as the research object, the drawing method of this paper was used to study the forming process. The comparison of the forming results before and after the optimization shows that this method can greatly improve the forming quality of the aluminum alloy plate.