Research And Application Of Vehicle Path Planning Method In Complex Coal Mine Environment

Coal mining has evolved through mechanization and automation stages,presently advancing towards intelligent development,with the establishment of smart mines becoming a widely accepted industry objective.Within the entire coal production system,auxiliary transportation systems in coal mines hold a pivotal role,as its technical prowess and operational efficiency directly impact the realization of personnel reduction and productivity enhancement goals.Owing to the intricate structure of subterranean tunnels,traditional transportation methods struggle to fully harness the capabilities of existing auxiliary coal mine transportation systems,resulting in excessive occupation of limited track lines and mine cart operational blockages,thus considerably diminishing transport efficiency.Therefore,this thesis,premised on evolutionary algorithms and vehicle routing planning theoretical foundations,explores path planning methodologies for the distinctive underground environment,with the primary research encompassing the following aspects:First,this thesis studies the underground multi-objective transportation path planning problem and design a method combining NSGA-II(Nondominated Sorting Genetic Algorithms II)multi-objective optimization algorithm and idealized minimum value point-based multi-objective decision algorithm.A new chromosome coding method and cross-variance operator are designed on the basis of NSGA-II to calculate a set of Pareto optimal solution set and find the "optimal solution" by multi-objective decision algorithm,which realizes the multi-objective optimization for the route planning of underground transportation.The multi-objective decision algorithm is used to find the "optimal solution" and achieve a multi-objective optimization approach to route planning for downhole transportation.By conducting experiments on a standard test set,it is demonstrated that the method proposed in this thesis can effectively solve the underground transportation path planning problem and optimize the travel distance of mine cars and the number of rail cars used,thus improving transportation efficiency and reducing transportation costs.Second,this thesis studies the downhole multi-vehicle yard transportation path planning problem and design an improved genetic algorithm that incorporates a clustering strategy and an elite retention model.First,a simple clustering strategy is used to assign each task point to an initial yard during the initialization of the genetic algorithm.Then,a set of chromosomes is randomly generated and transformed into a set of transportation paths using a two-stage path construction method.Next,individuals are selected from the population using a tournament selection strategy with an elite retention model,the chromosomes are crossed using an improved path crossing operator,and the initially assigned task points are dynamically adjusted to the appropriate vehicle yard by combining the cross-vehicle yard variation operator to improve the task points contained in each vehicle yard during the evolutionary process.Finally,experiments show that the improved algorithm proposed in this thesis can effectively solve the reasonable assignment of task points to yards and the downhole multi-vehicle yard transportation path planning problem.Finally,the underground transportation path planning system for coal mines is designed and implemented and applied to actual mine production.Through the steps of feasibility analysis,business requirement analysis and functional analysis,the specific functional,performance,safety and usability requirements of the system are determined.And according to the object-oriented programming concept and modular development strategy,the underground transportation path planning system for coal mines is developed with the support of the theoretical research and results related to this topic.