首页|基于GA-BP-PSO算法的薄壁注塑件翘曲变形优化

基于GA-BP-PSO算法的薄壁注塑件翘曲变形优化

扫码查看
以聚对苯二甲酸乙二酯(PET)的塑料瓶胚零件为例,通过Moldflow软件设计浇注系统和冷却系统并进行有限元分析以优化零件的翘曲变形量.选定熔体温度、模具温度、保压压力、保压时间和注射时间为5个影响因素,设计了L16(45)的正交试验表.对正交实验数据进行了极差分析,得出了各因素对翘曲变形量的影响程度并获得较优工艺参数.通过GA-BP-PSO算法对工艺参数进一步优化,得到最佳工艺参数:熔体温度265℃、模具温度60℃、保压压力125 MPa、保压时间12.867 1 s、注射时间0.340 5 s.上述工艺参数对应的零件翘曲变形量为0.137 3 mm.最后通过Moldflow软件进行数值模拟,得到翘曲变形量为0.139 5 mm,较优化前的翘曲变形量0.179 6 mm,降低了22.33%.软件模拟值和经GA-BP-PSO算法得到的预测值仅相差1.60%,将优化后的工艺参数组合应用于实际生产中,所获得的产品符合生产要求,验证了GA-BP-PSO算法的准确性与可行性.
Optimization of warpage deformation of thin-walled injection molded parts based on GA-BP-PSO algorithm
Taking the plastic bottle preform made of polyethylene terephthalate(PET)as an example,the gating system and cooling system were designed by Moldflow software and finite element analysis was carried out to optimize the warping deforma-tion of the workpiece.Five factors including melt temperature,mold temperature,holding pressure,holding time and injection time that influenced the warping deformation were identified,the orthogonal experiment table of L16(45)was designed,and the data were analyzed by range,so as to obtain the degree of influence of various factors on the amount of warpage and to obtain better process parameters.The process parameters were further optimized through the GA-BP-PSO algorithm,and the best process parameters are obtained:the melt temperature is 265℃,the mold temperature is 60℃,the holding pressure is 125 MPa,the holding time is 12.867 1 s,the injection time is 0.340 5 s.The warping deformation of the workpiece corresponding to the above process parameters is 0.137 3 mm.Finally,the numerical simulation was carried out by Moldflow software,and the warpage deformation is 0.139 5 mm,which is 22.33%lower than the warpage deformation before optimization of 0.179 6 mm.The difference between the software simu-lation value and the predicted value obtained by the GA-BP-PSO algorithm is only 1.60%,and the optimized process parameter combination is applied to actual production.The obtained products meet the production requirements,verify the accuracy and feasi-bility of the GA-BP-PSO algorithm.

warping deformationMoldfloworthogonal experimentGA-BP neural networkparticle swarm algo-rithmoptimization

陈忠杭、胡燕海、沈加明、倪德香、王舟挺

展开 >

宁波大学机械工程与力学学院,浙江宁波 315211

宁波华美达机械制造有限公司,浙江宁波 315000

翘曲变形 Moldflow 正交试验 GA-BP神经网络 粒子群算法 优化

国家自然科学基金

51705263

2024

工程塑料应用
中国兵器工业集团第五三研究所 中国兵工学会非金属专业委员会 兵器工业非金属材料专业情报网

工程塑料应用

CSTPCD北大核心
影响因子:0.371
ISSN:1001-3539
年,卷(期):2024.52(3)
  • 16