中文摘要：

现有多型腔注塑模流道优化设计通常对该复杂问题进行简化处理，仅将熔体流动平衡或低生产成本作为优化目标，缺乏准确而全面的设计方案优劣评价指标，难以获得真正最优的流道设计方案。针对此问题，提出采用变粒度策略来处理注塑模流道优化设计这一多目标多约束的复杂工程优化问题，在保证各型腔制品质量均衡性的同时兼顾注塑生产的成本和效率，建立不同粒度的注塑模流道多目标优化设计数学模型，在设计初期利用粗粒度模型快速获取通向整体最优解的方向，在此基础上逐渐细化模型的粒度，最终利用细粒度模型获取能使综合指标最优的设计方案。提出基于流动模拟和Pareto优于关系准则的注塑模流道多目标优化设计问题求解算法，并利用该算法求取不同粒度下流道优化设计问题的Pareto最优解。鼠标组合型腔流道的多目标优化设计实例验证了此方法的有效性。

英文摘要：

Present approaches for the runner optimization of multi-cavity injection molds usually simplify the complex problem and aim at achieving either balanced melt flow or low production cost. The optimal schemes in the point of view of injection molding practice can hardly be obtained due to the lack of comprehensive and exact criteria for evaluating different schemes. A variable-grain strategy is proposed to tackle the complex problem with multiple objectives and restrictions involved. Mathematical models of various grains are established, which include cost, efficiency and the uniformity of product quality as criteria for evaluating different runner schemes. The coarse-grain model is utilized in the early design to efficiently determine the direction to the globally optimal solution. Then more accurate models can be utilized during the optimization procedure and finally the optimal schemes evaluated from comprehensive point of view can be acquired on the basis of the fine-grain model. An algorithm is put forward for fast locating the Pareto optimal sets of multi-objective optimization problem based on flow simulation and superiority-of-feasible-points criterion, which is used to obtain the Pareto optimal sets of the runner optimization models in various grains. The feasibility of proposed strategy and algorithm is demonstrated through the runner optimization of a compound-cavity mold for producing mice.