文献类型：期刊文献

中文题名：医用TPU薄壁微管的挤出工艺参数优化

英文题名：Optimization of Extrusion Process Parameters for Medical TPU Thin-walled Microtubes

作者：胡伟康[1];刘焕牢[1];何优[1]

机构：[1]广东海洋大学机械工程学院,广东省小家电创新设计及制造工程技术研究中心,广东湛江524088

年份：2024

卷号：52

期号：11

起止页码：78

中文期刊名：塑料工业

外文期刊名：China Plastics Industry

收录：北大核心2023、CSTPCD、、CSCD2023_2024、北大核心、CSCD

基金：国家自然科学基金项目(52175458);广东省普通高校重点领域专项(2022ZDDX3006);广东省科技厅高新技术专题(2021A05186)。

语种：中文

中文关键词：挤出成型;薄壁微管;数值模拟;RBF神经网络;多目标优化

外文关键词：Extrusion Molding;Thin-walled Microtube;Numerical Simulation;RBF Neural Network;Multi-objective Optimization

中文摘要：针对热塑性聚氨酯弹性体(TPU)薄壁微管挤出成型工艺参数问题,以直径在2 mm以内的TPU薄壁微管为研究对象,通过数值模拟方法研究薄壁微管的挤出过程,并结合RBF神经网络和NSGA-Ⅱ多目标优化算法对特定尺寸的工艺参数进行寻优。首先,建立口模流道的初始三维模型,并根据薄壁微管挤出特性结合数值模型方法对流道进行优化。在此基础上研究各挤出工艺参数对挤出成型的影响,发现熔体温度对管材尺寸影响较小,挤出流率、牵引速度、注气压力更适合用于对管材尺寸的控制。然后使用最优拉丁超立方实验设计方法选择样本点,建立基于RBF神经网络的挤出工艺代理模型,通过对比发现RBF神经网络模型精度高,结合NSGA-Ⅱ算法获得特定尺寸的最优工艺参数,经实际生产验证得到的挤出制品符合尺寸要求。通过上述对TPU薄壁微管的工艺优化方法能够获得合格产品,对实际挤出成型加工过程有一定指导意义。

外文摘要：To solve the problems of the process parameters of thermoplastic polyurethane elastomer(TPU)thin-walled microtube extrusion,the extrusion processes of thin-walled TPU microtubes with a diameter of less than 2 mm were studied by the numerical simulation method,and the process parameters of specific sizes were optimized by combining RBF neural network and NSGA-Ⅱmulti-objective optimization algorithm.Firstly,the initial three-dimensional model of the orifice flow channel was established,and the flow channels were optimized according to the characteristics of thin-walled microtube extrusion combined with the numerical model methods.On this basis,the influences of extrusion process parameters on the extrusion molding were studied.The results show that melt temperature has little effect on pipe size.The extrusion flow rate,traction speed and gas injection pressure are more suitable for the pipe size control.Then,the optimal Latin hypercube experimental design method was used to select the sample points,and the extrusion process agent model based on RBF neural network was established.Through comparison,the results show that the RBF neural network model has high accuracy,and the optimal process parameters of specific sizes are obtained by combining NSGA-Ⅱalgorithm.The above optimization methods for TPU thin-walled microtubes could obtain qualified products,showing guiding significances for the actual extrusion processes.