摘要
近年来机器视觉、机器人等技术在在线装配质量控制中快速发展,对实时、准确且能考虑工艺不确定性的装配偏差控制模型的需求愈发强烈,并且如何基于所构建的模型,结合在线测量零部件检测结果,实现装配工艺参数的在线优化控制是亟待解决的工程问题。针对此问题这里提出了基于随机Kriging代理模型的在线装配工艺的优化控制方法—基于实测、仿真等历史数据构建随机Kriging模型。然后,在该模型基础上,结合装配零部件的在线测量数据对装配工艺参数进行实时的前馈优化控制。最后通过某车型后车门装配案例,对不同模型的预测精度进行对比,分析随机Kriging模型的预测精度,并对车门装配工艺进行在线优化控制,验证了这里提出方法的有效性,为车身不确定装配条件下的在线工艺控制提供理论依据。
In recent years,machine vision,robotics and other technologies have developed rapidly in the on-line assembly quality control,there is a growing demand for an assembly deviation control model that is real-time,accurate and can take process uncertainty into account,and how to realize the on-line optimization control of assembly process parameters is an urgent engineering problem based on the model built,combined with the on-line measurement component test results.To solve this problem,this paper proposes an optimal control method of online assembly process based on random Kriging proxy model,and constructs a random Kriging model based on historical data such as field measurement and simulation.Then,based on the model,combined with the online measurement data of assembly parts,the real-time feedforward optimization control of assembly process parameters is carried out.Finally,through the case of rear door assembly of a model,the prediction accuracy of different models is compared,and the prediction accuracy of the random Kriging model is analyzed,and on-line optimization control of door assembly process,verification of the effectiveness of the proposed method,it provides a theoretical basis for the on-line process control under the condition of uncertain assembly of the body.
作者
郭允明
张恃铭
GUO Yun-ming;ZHANG Shi-ming(School of Mechanical Engineering,University of Shanghai for Science and Technology,Shanghai 200093,China)
出处
《机械设计与制造》
北大核心
2020年第9期176-181,186,共7页
Machinery Design & Manufacture
基金
国家自然科学基金项目(51875362)。