首页|超高强钢U形件热冲压的NSGA-Ⅱ多目标优化方法

超高强钢U形件热冲压的NSGA-Ⅱ多目标优化方法

扫码查看
为了减小超高强钢U形件热冲压成形的回弹角和生产周期,提出了基于二元耦合选择NSGA-Ⅱ算法的多目标优化方法.介绍了超高强钢温度和微观组织随热冲压成形过程的变化情况,以减小成形回弹角和冲压周期为目标建立了多目标优化模型,选择了坯料初始温度、冲压速度、保压压强作为优化的试验因素.在优化空间中随机抽取了50组采样点,根据试验得到了试验指标参数值.使用单个自适应神经元网络对试验指标和试验因素间的模型进行了回归,提出了二元耦合选择NSGA-Ⅱ算法进行优化模型求解.对优化后的参数组合进行生产验证,优化后回弹角均值比厂家产品减小了27.59%,单件的生产周期减小了3.52%,且回弹角和生产周期的标准差略有减小,说明优化后的生产质量、生产效率、生产稳定性均有所提高.
Ultra High Strength Steel U-Shaped Hot Stamping Multi-Objective Optimization Method Based on NSGA-Ⅱ
In order to reduce the springback angle and production cycle of hot stamping U-shaped parts of ultra-high strength steel,a multi-objective optimization method based on NSGA-Ⅱ algorithm was proposed.The variation of temperature and microstructure of ultra-high strength steel with hot stamping process was introduced.A multi-objective optimization model was established to reduce springback angle and stamping cycle.The initial temperature of blank,stamping speed and packing pressure were selected as the ex-perimental factors for optimization.50 groups of sampling points are randomly selected in the optimization space,and the test index parameters are obtained according to the test.A single adaptive neural network is used to regress the model between test indexes and test factors,and a binary coupling NSGA-Ⅱ algorithm is proposed to solve the optimization model.The results show that the average springback angle is27.59%less than the manufacturer's product,the production cycle of single piece is 3.52%,and the standard de-viation of springback angle and production cycle is slightly reduced,which indicates that the production quality,production efficien-cy and production stability are improved after optimization.

U-Shaped Hot Stamping WorkpieceUltra High Strength SteelBinary Coupling ChoosingSingle Adap-tive Neural NetworkMulti-Object Optimization

周梅、段辉

展开 >

山东建筑大学,山东 济南 250101

U形热冲压件 超高强钢 二元耦合选择 单个自适应神经元网络 多目标优化

国家自然科学基金

51105233

2024

机械设计与制造
辽宁省机械研究院

机械设计与制造

CSTPCD北大核心
影响因子:0.511
ISSN:1001-3997
年,卷(期):2024.400(6)