Prediction Analysis And Scheduling Of Truck Fuel Consumption In Open-pit Mine Based On Machine Learning

In the process of open-pit mining production,how to achieve the goal of green development,cost reduction and efficiency increase is an important problem for the development of mining enterprises.Fuel consumption index is an important parameter to evaluate the green and economic of open-pit mine.Accurate prediction of fuel consumption index can realize the accurate control of truck fuel consumption in open-pit mine,and it is also the key factor to realize the cost scheduling of truck control in open-pit mine.Through the prediction of truck fuel consumption in open-pit mine,the truck scheduling model of open-pit mine is optimized,so as to improve the utilization rate of open-pit mine resources,reduce the production cost of enterprises,improve the economic benefits of enterprises,promote the economic,green and intelligent development of mines,and realize the efficient and intelligent production of mines.The specific work mainly includes the following aspects:(1)This paper expounds the machine learning theory and algorithm involved in the prediction of truck fuel consumption in open-pit mine,combs and analyzes the truck scheduling optimization theory,which provides the basis for solving the scheduling model problem based on the prediction of truck fuel consumption.(2)The factors affecting the fuel consumption in truck transportation process of open-pit mine are analyzed,and the characteristic samples of the truck fuel consumption prediction model are determined,and the collection and processing of sample data are described.(3)Aiming at the problem of fuel consumption prediction of open-pit trucks,machine learning support vector machine algorithm(Support Vector Machine,SVM)combined with swarm intelligence optimization algorithm(particle swarm optimization algorithm,PSO;genetic optimization algorithm,GA)is used to establish the prediction of open-pit truck fuel consumption Model,and use relevant characteristic data to carry out simulation experiments.The experimental results show that: PSOGA-SVM has better predictive performance.(4)In this paper,the fuel consumption of trucks in each transportation process is predicted,and the result of fuel consumption prediction is taken as the initial parameter into the truck scheduling model of open-pit mine,and the scheduling model based on the fuel consumption prediction of trucks in open-pit mine is constructed.And through the intelligent management and control system of an open-pit mine in Henan Province to collect the actual production data,after processing,the improved hybrid ant colony algorithm is used to solve the model,and the actual production scheduling plan is obtained.The research of this paper provides help for mine to realize cost control scheduling and improve the utilization rate of mine resources,and has certain significance for the economic and green development of open-pit mine.