基于RSM-GA与BP-PSO的钛合金筋板特征构件筋槽充填优化  

Optimization of rib-groove filling of titanium alloy rib-web eigenstructure based on RSM-GA and BP-PSO

作　　者：丁潼 魏科[1,2] 董显娟 黄龙[1] 徐俊楠 侯勇 DING Tong;WEI Ke;DONG Xian-juan;HUANG Long;XU Jun-nan;HOU Yong(School of Aeronautical Manufacturing Engineering,Nanchang Hangkong University,Nanchang 330063,China;Department of Materials Science and Engineering&RIAM,Seoul National University,Seoul 08826,Korea)

机构地区：[1]南昌航空大学航空制造工程学院,江西南昌330063 [2]首尔国立大学材料科学与工程学院,韩国首尔08826

出　　处：《塑性工程学报》2024年第7期52-62,共11页Journal of Plasticity Engineering

基　　金：国家自然科学基金资助项目(52005241);国家留学基金资助项目(202208360107);江西省自然科学基金资助项目(20232BAB204050)。

摘　　要：以TA15钛合金筋板特征构件为研究对象,首先,采用图像感知的实时监测方法,通过对图像进行捕捉获得任意时刻下的筋槽充填状态,并结合有限元模拟研究了筋槽中的材料流动及充填规律;其次,基于RSM-GA构建了双响应面回归模型,以此优化设计坯料尺寸参数,并利用GA优化算法开展智能决策;同时,建立不同坯料对筋槽充填的BP神经网络智能预测模型,配合PSO算法的智能决策进行稳健优化求解;随后,通过对比两个模型的优化解,得出RSM-GA模型精度较高且优化解的充填效果更好;最后,利用图像感知进行RSM-GA优化结果的物理实验模拟验证,证实了该研究在获得最佳筋槽充填的同时,可弱化坯料优化设计中不确定性因素波动的影响。Taking TA15 titanium alloy rib-web eigenstructure as the research object,firstly,the real-time monitoring method of image perception was used to capture the filling state of the rib-groove at any time,and the material flow and filling rule in the rib-groove were studied by finite element simulation.Secondly,the dual response surface regression model was constructed based on the RSM-GA to optimize and design billet size parameters,and GA optimization algorithm was used to carry out intelligent decision-making.At the same time,the BP neural network intelligent prediction model of different billets for rib-grooves filling was established,and the robust optimization solution was carried out with the intelligent decision of PSO algorithm.Then,by comparing the optimal solutions of the two models,it is concluded that the RSM-GA model has higher precision and better filling effect of the optimal solutions.Finally,the physical simulation experiment validation of the RSM-GA optimization results was carried out using image perception.It is confirmed that the influence of uncertain factor fluctuating in the optimization design of billet while obtaining the optimum rib-groove filling.

关 键 词：钛合金筋板特征构件 筋槽充填 稳健优化 实时监测 算法决策 

分 类 号：TG316.8[金属学及工艺—金属压力加工]