Research on the construction process modeling and dynamic control algorithm of wear-resistant ground of super-large workshop based on digital twin

This study focuses on the construction process of wear-resistant ground of super-large workshop, and proposes a modeling and dynamic control algorithm based on digital twin (DT). Aiming at the shortcomings of traditional construction methods in crack control, flatness assurance and resource scheduling, a three-layer DT modeling framework integrating BIM, GIS and CAE technology is constructed to realize multi-scale dynamic simulation. By integrating multi-source data, the incremental Kalman filter algorithm is used to drive the real-time dynamic update of the model to ensure the synchronization between the virtual model and the physical entity. The crack risk prediction and avoidance algorithm based on XGBoost classifier and the resource scheduling model of multi-objective integer programming are designed to form a real-time control closed loop of "perception-analysis-decision-execution". The field verification in a new energy automobile battery factory project shows that the DT system can significantly reduce the number of cracks, improve the flatness control accuracy and improve the equipment utilization rate. Although the construction period and cost are slightly higher than the traditional methods, it has obvious advantages in construction quality and crack control, which provides a new intelligent solution for the construction of super-large wear-resistant ground.