摘要
通过优化工艺参数,提高注塑件的成型质量。以航标灯外壳为例,建立CAD模型并利用有限元分析软件Moldflow进行注塑成型数值仿真,运用正交试验设计方法与注塑工艺数值模拟相结合,通过对正交试验仿真结果进行分析,综合评估了注塑过程中的熔体温度、模具温度、保压压力和保压时间对注塑成型关键质量特性翘曲的影响规律。并以仿真试验数据训练BP神经网络(BPNN)模型表征工艺参数与翘曲关系,利用遗传算法(GA)以翘曲最小为约束条件优化注塑工艺参数。结果表明,各因素对航标灯外壳翘曲影响的顺序为熔体温度>保压时间>保压压力>模具温度,其中熔体温度与保压时间为显著因素。通过BPNN-GA方法优化后获得的工艺条件为:熔体温度为215.01℃,保压时间为5.31 s,保压压力为50.37 MPa、模具温度为78.75℃。在此工艺条件下,航标灯外壳最大翘曲Moldflow仿真值为1.736 mm,BPNN-GA预测值为1.748 mm,二者吻合度高,说明所建立的神经网络模型能较好地表征工艺参数与翘曲之间关系。
The molding quality of injection parts was improved by optimizing the process parameters.The beacon light cover was taken as an example,a CAD model was established and the finite element analysis software Moldflow was used for the numerical simulation of injection molding.Combining the orthogonal test design method with numerical simulation of injection molding process,the effect rules of melt temperature,mold temperature,packing pressure and packing time on warpage,which is the key quality characteristics of injection molding,were evaluated comprehensively by the analysis for orthogonal test simulation results.The BP neural network(BPNN)model was trained with simulation data to characterize the relationship between process parameters and warpage,and the genetic algorithm(GA)was used to optimize the injection molding process parameters under the constraint of minimum warpage.The results show that the influence order of each factors on warpage of beacon light cover is melt temperature>packing time>packing pressure>mold temperature,and the melt temperature and packing time are the significant influence factors.The process conditions optimized by the proposed BPNN-GA method were obtained:the melt temperature of 215.01℃,the packing time of 5.31 s,the holding pressure of 50.37 MPa and the mold temperature of 78.76℃.Under this process condition,the maximum warpage of beacon light cover is 1.736 mm by Moldflow simulation and 1.748 mm by BPNN-GA prediction.The two results are in good agreement,indicating that the established neural network model can represent the relation between process parameters and warpage well.
作者
刘强
陈洪荣
梅端
陆卓凡
何梓烽
钟崇铭
LIU Qiang;CHEN Hong-rong;MEI Duan;LU Zhuo-fan;HE Zi-feng;ZHONG Chong-ming(School of Mechanical and Power Engineering,Guangdong Ocean University,Zhanjiang 524088,China;Southern Marine Science and Engineering Guangdong Laboratory(Zhanjiang),Zhanjiang 524088,China;Guangdong Marine Equipment and Manufacturing Engineering Research Center,Zhanjiang 524088,China;Faculty of Mathematics and Computer Science,Guangdong Ocean University,Zhanjiang 524088,China)
出处
《塑性工程学报》
CAS
CSCD
北大核心
2020年第7期123-129,共7页
Journal of Plasticity Engineering
基金
广东省普通高校重点科研项目(2018KZDXM038)
湛江市科技项目(2018A01019,2017A03005,2019B01004,2019B01006)
广东海洋大学科研启动资助项目(R19016)
广东海洋大学大学生创新创业训练计划项目(S201910566065)
南方海洋科学与工程广东省实验室(湛江)资助项目(ZJW-2019-01)。
作者简介
通信作者:梅端,女,1989年生,硕士,讲师,主要从事人工智能算法研究,E-mail:meid@gdou.edu.cn;第一作者:刘强,男,1984年生,博士,讲师,主要从事制造系统工程研究,E-mail:liuqiang@gdou.edu.cn。
