摘要
为了提高差压铸造的铸件性能,以A356铝合金转向节为研究对象,以凝固时间、二次枝晶臂间距和孔隙率为优化指标,通过数值模拟与优化算法的结合对冷却参数进行优化。首先通过分析确定不同管路冷却开启时间作为优化变量;然后基于正交试验安排进行数值模拟,得到支撑向量机回归算法(SVR)的训练样本,确定优化变量与三个优化指标之间的映射关系,通过鱼群算法寻找最优组合参数。最后对得到的最优组合参数进行数值模拟验证。结果表明模拟结果与预测结果基本吻合。在最优冷却参数组合下,凝固时间由208.2 s减小为175.6 s,二次枝晶臂间距由47.25μm变为44.18μm,孔隙率由12.79%减小为0.923%。
In order to improve the casting performance of differential pressure casting,the A356 aluminum alloy knuckle was used as the research object,the solidification time,the secondary dendrite arms spacing(SDAS)and porosity were used as optimization indicators,and the cooling parameters were optimized through the combination of numerical simulation and optimization algorithms.Firstly,the cooling opening time of different lines was determined as the optimization variable by analysis.Numerical simulation was carried out based on the orthogonal experiment arrangement,the training samples of the support vector machine regression algorithm(SVR)were obtained,and the mapping relationship between the optimization variables and the three optimization indicators was determined;the fish-swarm algorithm was used to find the optimal combination parameters.Finally,the obtained optimal combined parameters were verified by numerical simulations.The results show that the simulation results are basically consistent with the predicted results.Under the optimal cooling parameter combination,the solidification time is reduced from 208.2 s to 175.6 s,the distance between SDAS is changed from 47.25μm to 44.18μm,and the porosity is reduced from 12.79%to 0.923%.
作者
李智
陈阵
张健
LI Zhi;CHEN Zhen;ZHANG Jian(Hubei Key Laboratory of Power System Design and Test for Electrical Vehicle,Hubei University of Arts and Science,Xiangyang 441053,China;School of Automotive and Traffic Engineering,Hubei University of Arts and Science,Xiangyang 441053,China)
出处
《热加工工艺》
北大核心
2023年第9期83-87,共5页
Hot Working Technology
基金
湖北省教育厅科学技术研究项目(Q20212602)
“新能源汽车与智慧交通”湖北省优势特色学科群(XKTD012022)
湖北文理学院科研启动费项目(2059131)。
关键词
差压铸造
正交试验
支撑向量机回归算法(SVR)
人工鱼群算法
differential pressure casting
orthogonal experiments
support vector machine regression algorithm(SVR)
artificial fish-swarm algorithm