Mine large model driven trackless rubber-tyred vehicle global transport capacity balanced scheduling method

Abstract

This invention relates to the field of trackless rubber-tired vehicle scheduling technology in coal mines, and discloses a global capacity balancing scheduling method for trackless rubber-tired vehicles driven by a large-scale mine model. The method includes: collecting multi-source data from multiple mining areas across the entire mine and preprocessing it to generate real-time state vectors; using a large-scale mine model to infer the global capacity balance and supply-demand gap; constructing a matching matrix based on supply-demand priorities when capacity is imbalanced, determining the optimal number of vehicles to be dispatched, and generating cross-mining area dispatch instructions; constructing a comprehensive path cost model during execution, and dynamically optimizing driving paths based on congestion prediction; and finally, iteratively updating the model decision weights based on execution feedback data using a reinforcement learning algorithm. This invention achieves dynamic cross-mining area allocation of vehicle resources and intelligent path planning, effectively alleviating the contradiction between local capacity shortages and global resource idleness, and improving the overall efficiency and safety of the mine transportation system.

Images