弱刚性大型航天构件数字孪生精密装配技术

中国科学院沈阳自动化研究所创新提出数字孪生工艺闭环，通过孪生模型实现不可测变量的虚拟感知与预测，支持对物理装配过程在线优化。突破几何误差与装配力耦合作用下装配精度预测、大型构件高精度装配干涉检测、螺栓组残余预紧力预测等技术，形成米级工件、微米级精度、百毫秒级实时性的数字孪生装配技术体系。研发行业首套机器人化数字孪生装配系统，在航天装备企业应用，变革人工“两试一装”工艺，实现机器人一次装配成功，大幅提升关键工艺环节效率，破解企业安全减人与效率提升的矛盾，节约企业新增投资数亿元。
 

SIA has proposed the digital twin based closed-loop for assembly process that leverages twin model for

virtual sensing and prediction of unmeasurable variables, enabling online optimization of physical assembly

process. This technology has achieved breakthroughs in areas such as assembly precision prediction under

the coupled effects of geometric errors and assembly forces, high-precision interference detection for large component during assembly, and residual preload prediction for bolt groups, establishing a digital twin assembly technology system characterized by meter-scale work pieces, micron-level precision, and real-time performance at the scale of hundreds of milliseconds. The team developed the industry’s first digital twin-equipped robotic assembly system, which has been widely applied in aerospace equipment enterprises for batch production. One-time assembly success using robots has been achieved, replacing “two trials followed by final assembly”with skilled workers. It significantly improves the efficiency of the key process, resolves the conflict between workforce reduction and efficiency improvement, and saves hundreds of millions RMB in new investments for enterprises.

精密装配，数字孪生，弱刚性构件，在线优化

Precision Assembly; Digital Twin; Low Stiffness Components; Online Optimization
 

中国科学院沈阳自动化研究所

Shenyang Institute of Automation, Chinese Academy of Sciences