| Marine debris refers to the waste existing in the marine and coastal environment,of which plastic debris accounts for the vast majority of marine debris.With the emphasis on environmental protection and health,the harm of marine debris to the environment and human beings is increasingly valued by people.For large pieces of debris floating on the sea surface,the most effective way is to send vessels to collect them.Therefore,how to collect marine debris efficiently and environmentally is the focus of this paper.The research framework of this paper is: using remote sensing technology to locate the initial position of debris and predict its drift trajectory through GNOME software,using vessels equipped with photovoltaic,battery and diesel generators as hybrid energy systems to collect debris.In order to better reflect the drift of marine debris in the mainland of the day,this paper sets a floating time window for marine debris,that is,sets multiple time windows on the drift trajectory of marine debris,and the location of marine debris points in each time window is different,so this paper defines the problem in the logistics space-time network.Due to the complexity of simultaneously optimizing the route and energy management strategy,in order to simplify the problem,this paper proposes a two-stage optimization method for vessel route and energy management strategy.In the first stage,a mixed integer linear programming model is established to minimize the total travel time of the vessel.The problem is to consider the vessel routing problem with continuous speed and floating time window,and a hybrid algorithm combining elite strategy and adaptive large-scale neighborhood search algorithm(ALNS algorithm)is proposed to solve the problem.In the second stage,a mixed integer linear programming model is established to minimize the total cost,including the power generation cost of diesel and photovoltaic,the charging and discharging cost of battery and the carbon tax cost.The model considers the power load balance,the power limit of each part of the hybrid energy system and the charging and discharging state of the battery,and solves the problem by CPLEX.After the model and solution method are determined,this paper analyzes the solution results through a set of examples.In the first stage of vessel route,the sensitivity of floating time window is analyzed.By expanding the scale of the example and comparing with other algorithms,the performance of the proposed algorithm is analyzed,and its superiority and effectiveness are verified.In the second stage of energy management strategy,this paper analyzes the power flow between the hybrid energy system and the vessel load,and compares the cost and carbon emissions under three light intensities.This paper also compares hybrid energy vessels with pure diesel vessels.In order to carry out the joint optimization of vessel route and energy management strategy,this paper simplifies the problem by assuming that the recovery time of marine debris is not less than one hour,combines the above two stages,and establishes a hybrid energy vessel considering the floating time window and continuous speed.The bi-objective model of joint optimization of marine debris collection route and energy management strategy minimizes the total travel time and total cost of vessels.In this paper,the multiobjective evolutionary algorithm based on decomposition(MOEA/D algorithm)is combined with the ALNS algorithm to solve the model.The proposed algorithm uses the MOEA/D algorithm as the framework and the ALNS algorithm as the neighborhood new solution generation method.The decomposition method and sub-problem update mechanism of the MOEA/D algorithm are improved.In this paper,the effectiveness of the algorithm is verified by a set of examples,and the sensitivity analysis of photovoltaic output power and floating time window is carried out.The analysis results show that the photovoltaic output power has a great influence on the total cost,but has no obvious influence on the total travel time,and the floating time window has a certain influence on the total travel time and total cost.In addition,this paper also analyzes the velocity distribution and energy flow distribution of each solution in the Pareto solution set,and concludes that the total cost is proportional to the proportion of high speed and inversely proportional to the proportion of photovoltaic output.Finally,by expanding the scale of the example and comparing the proposed algorithm with the NSGA II algorithm,the performance of the proposed algorithm is analyzed,and the effectiveness of solving large-scale problems and the superiority compared with the NSGA II algorithm are verified. |
