中文摘要：

传统的冷却时间的估算都是采用简单的计算公式,并没有考虑模具钢的种类、塑料的种类以及冷却系统的设计等,其结果与实际情况偏差较大。采用HsCAE等软件模拟出的冷却时间比较准确,但需要进行详细的冷却系统的设计和很长的分析时间。通过综合考虑影响冷却时间的主要因素,利用BP神经网络建立冷却时间的快速预测模型,能快速预测出冷却时间,并通过具体实例验证了该模型的优越性。

英文摘要：

Traditionally,the estimation of injection mold cooling time was based on simple formula,which did not considered the type of steel mold,plastic type,the design of cooling system and so on,thus,there was larger deviation with the actual ones.Simulation using software such as HsCAE can provide accurate cooling time,but needed detailed cooling system design and very long time for analysis.Based on a comprehensive consideration of the main factors affecting cooling time,a BP neural network model was established for predicting cooling time quickly.Results of examples showed the superiority of the presented model.