摘要
针对空间惯性传感器检验质量释放残余速度,提出一种基于Kriging代理模型的预测方法,分别采用Kriging代理模型,RBF代理模型及SVM代理模型对试验数据进行拟合,并以相关系数R为模型精度指标对三种代理模型进行评价。通过对比得出Kriging代理模型对捕获定位释放机构检验质量释放残余速度的预测精度最高。
In view of the residual velocity of inspection mass release of the spatial inertial sensor,a prediction method based on Kriging surrogate model is proposed.The Kriging surrogate model,RBF surrogate model and SVM surrogate model are respectively applied for test data fitting,and the correlation coefficient R is adopted as the indicator of model accuracy for evaluation of these 3 kinds of surrogate models.Through comparison,it is found that the Kriging surrogate model has the highest prediction accuracy for residual velocity of inspection mass release of the grabbing positioning and release mechanism.
作者
王日龙
刘博林
Wang Rilong;Liu Bolin
出处
《一重技术》
2024年第6期66-70,共5页
CFHI Technology
基金
辽宁省教育厅高校科研基金重点攻关项目(项目编号:LJKZZ20220037)。
