三维注塑成型的数值模拟与预测优化研究

项目名称： 三维注塑成型的数值模拟与预测优化研究

项目编号： No.11426170

项目类型： 专项基金项目

立项/批准年度： 2015

项目学科： 数理科学和化学

项目作者： 栗雪娟

作者单位： 西安建筑科技大学

项目金额： 3万元

中文摘要： 注塑成型是聚合物加工成型的主要方式之一。在成型过程中，聚合物熔体发生了复杂的热流变行为，成型的工艺参数对制品性能有显著影响。因此，研究注塑成型过程及相应的物理现象具有重要的理论和实际意义。本项目首先建立描述聚合物熔体流动的三维粘弹性模型，并运用Level Set方法追踪熔体前沿界面，对三维注塑成型充填过程进行数值模拟研究；其次，对注塑成型中的三维熔接痕进行动态追踪和显示，并讨论工艺参数对熔接痕形成及其位置的影响；最后，基于注塑成型加工工艺参数，运用人工智能优化算法对熔接痕的形态及产生位置进行预测优化研究。

中文关键词： 注塑成型；熔接痕；多目标优化；粒子群优化；混合神经网络

英文摘要： Injection molding is one of the most widely employed methods in polymer processing. Since polymer melt usually have complex thermorheological properties in polymer processing, and processing parameters have a significant impact on product properties, so it is of great theoretical and realistic significance to study numerically the injection molding and corresponding physical phenomena in injection molding. In this project, firstly the viscoelastic flow model will be established to describe polymer melt flows and the Level Set method is employed to trace the moving interfaces in filling stage of the three dimensional injection molding. Secondly, the three dimensional weldlines in injection molding process are dynamically traced, and the effects of processing parameters on the weldlines and their positions are researched. Lastly, weldlines optimization is implemented by using artificial intelligence optimization algorithm based on processing parameters.

英文关键词： Injection molding；Weld line；Multi-objective Optimization；Particle swarm Optimization；Hybrid neural network