| Greenhouse gas emission has become one of the major problems,which threatens sustainability and economy of human society.Due to the increase of urban passenger and freight demand,transportation has been considered as one of the main sources of urban greenhouse gas emission.In order to reduce the urban greenhouse gas emission,the researchers from all around the world are focusing on new energy technology in transportation,among which battery electric bus(BEB)has been deemed as the most potential one by global community as well as Chinese government due to its “zero direct tailpipe”,“low noise”,“zero direct greenhouse gas emissions” and “better energy efficiency” features.However,some concerns(e.g.high battery cost and limited service range)hinder its further development.Recently,several institutes have focused on the study of the fast charging technologies.Among them,wireless charging technology(WPT)is considered as the most effective one to alleviate “range anxiety” problem referring to the unsuitability of the electric range for daily activities.Compared to traditional charging technology,WPT can be installed underground to wirelessly recharge BEBs during operation,thus significantly reducing the area occupied by fast charging facilities.Combined with operational modeling method and intelligent optimization algorithm,this paper intends to study the optimization of urban transit networks(e.g.conventional bus network,feeder bus network and flexible feeder bus network)and the application of WPT in urban transit systems.Firstly,this paper studies the optimization problem of the conventional bus system powered by WPT.Considering the energy constraints and battery service life,a mixed integer non-linear optimization model is proposed to simultaneously optimize two major decision variables(e.g.locations of WPT devices and battery capacity)in the conventional bus network powered by WPT.A tangible genetic algorithm(GA)combined with simulated binary crossover(SBX)and polynomial mutation(PM)is developed to efficiently search for the optimal solution.The result analysis suggests that WPT outperforms terminal charging technology in terms of the least total cost and greenhouse gas emission.The sensitivity analysis shows that the battery capacity is more sensitive to the change in the bus energy consumption rate,and that larger battery would be used if battery unit price is lower,charging rate is higher and WPT cost is higher.Secondly,this paper studies the optimization problem of the feeder bus system powered by WPT.Considering the bus routing constraints and energy constraints,a mixed integer nonlinear optimization model is proposed to simultaneously optimize four major decision variables(e.g.feeder bus routes,service frequencies,locations of WPT devices and battery capacity)in the feeder bus network powered by WPT.A nested genetic algorithm(NGA)is developed,in which the bus routes are optimized in the internal genetic algorithm(GA),and service frequency,locations of WPT and battery capacity are jointly optimized in the external GA.The computational efficiency of NGA is demonstrated through numerical comparisons to the solutions founded by LINGO and original GA.The result analysis suggests that WPT outperforms terminal charging technology in terms of the least total cost.The sensitivity analysis shows that the increase of bus routes in the study feeder system results in greater fleet size,reduced service frequency and larger battery capacity,and that the number of stops with WPT devices increases as the increase of battery unit cost,charging rate and energy consumption rate.Thirdly,this paper studies the optimization problem of the flexible feeder bus system powered by WPT.Considering the passengers’ time constraints and energy constraints,a mixed integer non-linear optimization model is proposed to simultaneously optimize two major decision variables(e.g.flexible bus routes and bus type)in the flexible feeder bus network powered by WPT.A tangible hybrid algorithm(HVNS-SA)combining simulated annealing algorithm(SA)and variable neighborhood search(VNS)is developed to efficiently search for the optimal solutions.The computational efficiency of the proposed HVNS-SA is demonstrated by the numerical comparisons with GAMS(DICOPT solver)and original VNS.The result analysis suggests that the WPT shows better performance than terminal charging technology with respect to the least total cost.The sensitivity analysis shows that the selection of bus type is more sensitive to energy consumption rate,and that if the slack time and recharging time increase,the flexible bus system will require the bus equipped with larger battery capacity. |
PDF Full Text Request