Dynamic scheduling strategy of mine auxiliary transportation robot based on digital twin
摘要
Mine auxiliary transportation material scheduling is an important part of coal mine safety and efficient production. The transportation environment in coal mines is complex, accidents occur frequently, and the demand for materials is widely distributed. The transportation of materials mainly relies on manual shift scheduling, which is difficult to dispatch vehicles and has a low degree of information transparency. It has become an important direction for the intelligent construction of coal mines. With the research and application of auxiliary transportation robots, material distribution is gradually achieving continuous transportation and is moving towards intelligent dispatching. Aiming at the problem of auxiliary transportation material scheduling in mines, it is mapped to the horizontal transportation of automated terminals, and a centralized dynamic scheduling method for auxiliary transportation robots based on digital twins is proposed. To minimize the maximum transportation time of auxiliary transportation robots while ensuring conflict-free path planning, a two-stage scheduling model is proposed, which integrates balanced task allocation and addresses constraints such as robot endurance, load capacity, and potential malfunctions. A two-layer algorithm is designed by combining the cultural genetic algorithm with the Dijkstra algorithm incorporating time window constraints to solve the model. Once the generated dispatching plan is simulated and validated through a digital twin system, the dispatching system centrally controls the transportation robots and fixed equipment to collaboratively execute the transportation tasks. Based on the actual auxiliary transportation operations of a mine in Guizhou Province, a digital twin system for auxiliary transportation was designed and implemented. Following the verification of virtual-physical consistency, simulation analyses and experimental tests were conducted to evaluate the performance of the proposed model and alg