Research On Joint Operation Technology For Vehicle-grid System For Promoting Cleaner Application Of Electric Vehicles

Electric vehicles(EVs)can replace fuel vehicles to effectively reduce emission from the traffic sector,and thus,it can offer opportunities for improving ambient air quality.However,such opportunities could become unattainable if the electricity for charging EVs is mainly supplied by coal-fired thermal units(hereinafter referred to as thermal units),which is the cases in China.In this situation,from a long-term perspective,it is necessary to develop wind and solar power to make EVs get cleaner electricity widely.On the other hand,from a short-term perspective,it is necessary to focus on technologies that can immediately promote the cleaner application of EVs.Based on modeling charging demand and battery swapping demand(BSD)of EVs,this dissertation mainly focuses on the latter one and studies the joint operation technology for the vehicle-grid system in promoting the cleaner application of EVs,relying on distributed photovoltaic(PV),wind farms and generation scheduling strategy that can strengthen emission control for thermal units.Results from this study will benefit the development of EVs in China.Modeling charging demand and BSD for EVs is the basis of research including planning and scheduling issues for EVs.This dissertation proposes a modeling method based on simulating the behavior process sequences of EVs.It continuously and cyclically simulates the behavior sequence of EVs and gradually aggregating the regular characteristics of EVs,and in this way,it solves the problem that the existing statistical data is difficult to provide the state of charge(SOC)distribution information of EVs.Compared with the modeling methods that hypothesize and simplify SOC distribution with a specific probability distribution function,the proposed method can obtain more consistent modeling results and provide better data support for the subsequent researches.Distributed PV is widely used in urban regions,and thus,it has the advantage of being interconnected with adjacent battery swapping stations(BSSs).This dissertation studies a capacity sizing problem for a nanogrid via a flexible robust optimization approach,and then it analyzes the potentials for networking nanogrids to serve a BSS.And then,it presents a joint operation system which allows sharing generation and storage resource by exchanging batteries.The proposed system consists of multiple networked nanogrids and a BSS,which mutually swap batteries via specialized trucks.This dissertation presents the main components and discuss the operation of the proposed architecture and considers a mixed-integer linear programming(MILP)formulation for the optimal capacity sizing model of units in the proposed system which is solved with stochastic optimization.The results show that the proposed method offers economic benefits,ensures reliability,and prevents the underutilization of capital-intensive distributed PV generation and battery storage.Networked nanogrids can exploit diversities of supply and demand patterns,by sharing energy and storage,for achieving mutual benefits.This dissertation is able to explore cleaner electricity for BSSs in urban regions and have the potential to promote the cleaner application of community-level EVs.Construction of sizeable centralized charging stations(CCS)in cities is limited by lack of land,power grid capacity,and other factors.Also,it is challenging to guarantee whether the electricity is clean enough.This dissertation studies the centralized changing and unified delivering(CC&UD)system based on multiple wind farms.It establishes a joint scheduling model which considers planned wind power generation,battery swapping demand,battery discharging and charging,battery delivering,etc.A two-stage random optimization method is employed to expand the model to consider the randomness in wind power.The solution method is designed based on the genetic algorithm.Simulation results demonstrate the effectiveness of the proposed model,indicating that the CC&UD system based on multiple wind farms can benefit the local consumption of wind power and reduce the pressure of the power grid to absorb wind power.It has high potentials in promoting the cleaner application of city-level EVs.In quite a long time,most EVs in China will mainly rely on the power grid dominated by thermal units.Since emission from thermal units is easy to be controlled,this dissertation studies the methods of adjusting generation scheduling to reduce the impact of fuel emission on air quality and human health.It can improve the clean level of electricity production in the power grid,and thus,it can use relatively clean electricity to meet the energy demand for large-scale EVs.This dissertation proposes a methodology to analyze the differential marginal impacts of emissions at various times and locations on ambient air pollutant concentration(AAPC)and human health damages.A Gaussian puff model addresses the source-receptor relationship between air pollutant emission and the resulting increment in AAPC(ΔAAPC)in densely populated regions,and concentrationresponse(C-R)functions quantify health impacts due to changes in ΔAAPC.In this way,the proposed model can rapidly capture approximated responses of spatially resolved health impacts to changes in power generation emission.This dissertation integrates such estimates into a linearized model for generation scheduling with CCS operations.Numerical results illustrate that the emission-associated AAPC and health-impacts can be cost-effectively reduced by adjusting the integrated generation scheduling,which will enhance the environmental benefits of adopting EVs in densely populated regions.The proposed method has potentials in promoting the cleaner application of large-scale EVs.