Construction Of Representative Operating Cycles And Optimization Of On-Board And Stationary Energy Storage Systems For A Catenary-Free Tramline

Catenary-free trams are considered a great prospect for future development,due to their salient characteristics of large passenger capacity,high energy efficiency,no greenhouse gas emissions,no visual pollution and low price.Popularizing the catenary-free trams can promote an environment-friendly development of cities,since it not only enriches urban public traffic but also alleviates energy shortage and environmental pollution.In China,the catenary-free tramline has been operated in several cities.The on-board energy storage system(OESS),charging station and power supply network are main components of energy conversion of the catenary free tramline.Issues that need to be addressed for an economic operation of the catenary-free tramline are to understand operating cycles of the trams and optimize of OESS and energy supply simultaneously.To the best of our knowledge,the optimization of single tram has been mentioned in many studies,but the research involving the OESS and stationary energy supply of the tramline has not been discussed in the field of the catenary-free tramline.In this paper,according to actual operating data of an existing catenary-free tramline,i.e.Haizhu line in Guangzhou,China,representative operating cycles of the tramline are constructed,an optimization of a stationary energy storage system(SESS)is formulated,a stochastic optimization of the SESS for actual tramline and a convex optimization of the OESS and the stationary energy supply for design tramline are introduced.The research results of this paper have theoretical guidelines and practical value for the optimal design of the catenary-free tramline.Four contributions are made as follows:(1)A construction method of the representative operating cycles of the catenary-free tramline is proposed.The actual operating data of the trams are originated from the Haizhu line.The representative operating segment at each station is then constructed by first selecting microtrip candidates by velocity-related characteristic parameters,and then extracting the representative one from the preselected microtrip candidates by minimizing a Euclidean distance function described by energy-related characteristic parameters.Since the economic operating of the OESS and the stationary energy supply is related to station dwell time and reversed waiting time at the starting and final stations,we need to extract these two characteristic parameters for the representative segments,respectively.After this,the representative operating cycles can be constructed by inserting average dwell time and arranging the extracted representative segments based on the station order.Comparing the representative operating cycles and the original operating data in terms of relative error of the characteristic parameters,probability distributions of velocity and power,and relative distributions of velocity and power,results show that the representative operating cycles can simultaneously characterize the characteristics of velocity and energy consumption of the catenary-free tram.(2)A convex optimziation of a catenary-free tramline equipped with stationary energy storage systems is introduced.Since a shared electric grid is suffering from power superimposition when several trams charge at the same time,we propose to install stationary energy storage systems(SESSs)for power supply network to downsize charging equipment and reduce operational cost of the electric grid.To evaluate the tradeoff between component cost and operational cost,an optimization problem,which integrates storage type selection,sizing,energy management and different installation configurations of the SESSs,is introduced and formulated in terms of an annual cost.Disciplined convex modelling is applied to obtain a computationally tractable solution for a case study on an existing line in China.Results verify the effectiveness of the introduced optimization and the economy of installing the SESS.(3)A stochastic optimization of a stationary energy storage system for a catenary-free tramline is proposed.As the tram operation may not be fully aligned with a predetermined timetable,an economical coordination of the electric grid and the SESS under uncertain charging demands needs to be investigated.To this end,a chance-constrained program is formulated for the SESS optimization,where demanded charging of the tramline is satisfied in a probabilistic sense.The introduced chance-constrained program is translated into a robust and deterministic mixed-integer second-order cone program(MISOCP)by first saturating charging power to a stochastic upper limit and then prolonging charging periods until entire energy is delivered for all charging scenarios that are being investigated.A case study for the Haizhu line in Guangzhou,China,shows that a considerable cost-benefit can be obtained by installing an SESS when charging power is fully delivered for all scenarios,while a significant cost-saving can be achieved when charging power is delivered 99%of the time.(4)An optimization involving the OESS and the stationary energy supply for the catenary-free tramline is introduced.Joint optimization of different types of the OESS and the corresponding energy supply mode can effectively realize the economic operation of the catenary-free tramline.Four types of the OESS are considered,including supercapacitor-only storage,lithium-ion capacitor-only storage,lithium-ion battery-only storage,and hybrid storage of supercapacitor and lithium-ion battery.To find the optimal sizing and energy management for the OESS,as well as the optimal site selection for the charging stations,a convex optimization that includes an investment cost of the OESS and a construction cost of the charging station is formulated.A case study for the Haizhu line is presented,where a sensitivity analysis is performed to quantitatively compare a cost-benet of utilizing different types of the OESS under different dwell time,different reversed waiting time at starting and final stations and different construction prices of the charging stations.Besides,the effectiveness of the optimized configuration,i.e.OESS sizes and charging station sites,is verified by taking the representative operating cycles as input.