Research On Multi-Agent System Based Low Carbon Operation Optimization On Automated Container Terminals

Since the 21st century,"smart port" and "green port" have become the new development concept of container terminals.In this context,how to realize the automated and efficient operation of container terminal and how to realize the energy saving and emission reduction of operation are the key problems urgently to be solved in the current and future terminal operation.This problem belongs to the research field of low-carbon operation scheduling of automated container terminals.At present,the main research idea is to optimize the operation plan from the perspective of AGV assignment and route planning,and then add energy consumption factors.The main problem is that the influence of AGV's track and speed changes on operation time and energy consumption is not considered,resulting in some errors.Therefore,this dissertation attempts to establish an optimization model of automated container terminal operation scheduling considering the details of AGV operation process(AGV operation behavior)and energy consumption.The model is solved by the bi-level programming model and multi-agent technology.The main research work is summarized as follows:(1)The multi-objective AGV scheduling problem considering energy consumption.Aiming at the automated guided vehicle scheduling problem(AGVSP),a multi-objective bi-level programming model is established.The upper model is a bi-objective integer programming model aiming at the overall energy consumption and the shortest completion time.The lower model is a bi-objective route planning model aiming at the single AGV energy consumption and the shortest driving route of AGV.A bi-level hybrid heuristic algorithm with genetic algorithm(GA)and ant colony algorithm(ACA)is designed to solve the bi-level programming model.At the same time,in order to eliminate the error caused by the estimation of the task time of each AGV,multi-agent technology is used to model the operation process of each AGV,which greatly improves the calculation accuracy.The validity of the model is verified by experimental analysis.The results show that the model can reduce the overall energy consumption by 11.65%when it has little effect on the overall efficiency.(2)The multi-objective AGV scheduling problem considering operation behaviorsOn the basis of(1),the operation details of AGV are further considered.Considering the fact that the travel time and energy consumption of AGV are not consistent due to the difference of the actual operation route and speed,the traditional AGV route planning problem is transformed into the combination optimization problem of AGV operation behavior,and a multi-objective AGVSP bi-level programming model considering the operation behavior is established.The upper model is a bi-objective model aiming at the overall energy consumption and the shortest completion time.The lower model is a combined optimization model aiming at the single AGV energy consumption and the shortest operation time of AGV.It can not only realize the detailed analysis and optimization of the AGV operation behavior and energy consumption,but also realize the optimization of the overall operation efficiency and energy consumption.A bi-level GA is designed to solve the bi-level programming model,and the validity of the model is verified by numerical experiments.The results show that the model can effectively reduce the energy consumption under the same operation mode,considering the waiting time of the shore bridge and the overall efficiency.(3)The multi-objective integrated operation scheduling considering the overall efficiency and energy consumption on the container terminalsOn the basis of(1)and(2),aiming at the requirements of improving the overall operation efficiency and reducing the overall operation energy consumption of the automated container terminal,considering the influence of the operation scheduling optimization of the quay crane operation and the operation plan of the yard crane on the horizontal transportation operation,a two-stage and bi-objective automated container terminal bi-level scheduling optimization model is established.The first stage is to optimize the operation scheduling of the quay crane reducing the waiting time of the auxiliary trolley of the double trolley quay crane and improving the operation efficiency of the quay crane;the second stage is based on the study of the multi-objective AGV scheduling problem considering the operation behavior,and introduces the operation plan of the yard crane as the influencing factor into the establishment of the multi-objective bi-level programming model.A greedy algorithm is designed to solve the first stage model,and a bi-level GA is designed to solve the second stage model.The validity of the model is verified by numerical experiments.The results show that the model can effectively reduce the energy consumption in the process of horizontal transportation taking into account the waiting time of quay crane and yard crane under the condition that the overall efficiency is unchanged.The balance point of efficiency and energy consumption in the process of horizontal transportation is found by adjusting the Pareto coefficient of the upper and lower layers.The validity and feasibility of the model are verified.The theoretical significance of this study is combining the bi-level programming model of AGV scheduling with the agent model of analyzing the details of AGV operation process,which provides a new modeling and optimization method for further accurate solution of low-carbon operation scheduling problem of automated container terminal.The application significance is providing a new analysis and solution idea for solving the low-carbon operation scheduling problem of automated container terminal,which conforms to the idea of lean management,and adapts to the development direction of accurate real-time online control of automated terminal operation.