欢迎访问《轻工机械》稿件在线采编系统!设为首页 | 加入收藏    
信息公告:  
文章检索:
稿件处理系统
期刊信息
  • 中国标准连:ISSN1005-2895
  • 续出版物号: CN 33-1180/TH
  • 主管单位:轻工业杭州机电设计研究院有限公司
  • 主办单位:轻工业杭州机电设计研究院有限公司、中国轻工机械协会、中国轻工业机械总公司
  • 社  长:刘安江
  • 主  编:黄丽珍
  • 地  址:杭州市余杭区高教路970号西溪联合科技广场4-711
  • 电子邮件:qgjxzz@126.com
理事单位          MORE>>
李亚东, 张泽鹏, 陈旭栋, 姜星辰.基于补充极值点和组合代理模型的零件表面建模[J].轻工机械,2025,43(2):36-41
基于补充极值点和组合代理模型的零件表面建模
Surface Modeling of Parts Based on Supplementary Extremum Points and Combined Surrogate Models
  
DOI:10.3969/j.issn.1005 2895.2025.02.005
中文关键词:  零件表面建模  补充极值点  组合代理模型  克里金模型  径向基函数模型
英文关键词:parts surface modeling  supplementary extremum points  combined surrogate models  Kriging model  radial basis function model
基金项目:西安市科技计划项目(2023JH GXRC 0008)。
作者单位
李亚东, 张泽鹏, 陈旭栋, 姜星辰 西安工程大学 机电工程学院 陕西 西安710600 
摘要点击次数: 14
全文下载次数: 7
中文摘要:
      针对目前零件表面模型无法满足精密零件表面建模的精度要求,课题组提出了一种基于补充极值点和组合代理模型的零件表面建模方法。首先,根据皮尔逊相关系数将零件表面分成多个较平滑的区域,在每个区域里设置少量初始点即可找到零件表面上的所有极值点,将这些极值点作为补充采集点有助于提升零件表面模型的精度;然后,提出了一种基于克里金模型和径向基函数的零件表面建模方法,通过计算模型的预测残差平方和来不断调整加权系数直至零件表面模型符合精度要求;最后,用3种模型对补充极值点前、后的各组数据进行拟合得到零件表面模型,对比各模型的均方根误差可知补充极值点和组合代理模型均可有效提高零件表面建模的精度。
英文摘要:
      In response to that the current parts surface models could not meet the accuracy requirements of precision part surface modeling, a parts surface modeling method based on supplementary extremum points and combined surrogate models was proposed. Firstly, based on the Pearson correlation coefficient the surface of the part was divided into multiple smoother regions, and a small number of initial points were set in each region to find all the extreme points on the surface of the part. Using these extreme points as supplementary collection points could help to improve the accuracy of the parts surface model. Then, a surface modeling method for parts based on Kriging model and radial basis function was proposed. The weighting coefficients were continuously adjusted by calculating the sum of squared prediction residuals of the model until the surface model of the parts met the accuracy requirements. Finally, three models were used to fit the data of each group before and after supplementing the extremum points to obtain the parts surface model. Comparing the root mean square errors of each model, it can be concluded that both supplementing the extremum points and the combined surrogate model can effectively improve the accuracy of surface modeling of parts.
查看全文  查看/发表评论  下载PDF阅读器
关闭