Market-oriented Energy Storage System Operation Strategy And Optimal Sizing Technology For Its Distributed Form

In recent years,countries around the world have put forward ambitious decarbonization plans for accelerating energy transition,improving energy security,and reducing dependence on fossil energy.However,the connection of large-scale renewable energy(such as photovoltaics,wind power,etc.)brings great uncertainty to the operation of power system.From the perspective of power uncertainty,the output of renewable generation fluctuates greatly due to environmental changes.And the fluctuations of renewable generation output might lead to insufficient system backup,which might further cause system frequency fluctuations and node voltage fluctuations.From the perspective of capacity uncertainty,with the increasing of the renewable generation penetration rate,the power grid may fail to fully absorb the output of renewable generation,resulting in large-scale abandonment of renewable power.With the ability to move energy in time and space,energy storage system can effectively deal with the power fluctuation of renewable generation.Besides,it provides the possibility for renewable generation to be efficiently consumed.However,the energy storage system is expensive,which affects its application in real systems.To solve the economic problems of energy storage,the research of this work is followed by two aspects: planning and operation.In order to solve the economic problem of energy storage,it is necessary to comprehensively consider the multiple uncertainties of renewable energy and the physical constraints of energy storage itself to carry out energy storage planning and operation research facing the existing market mechanism.Among them,the multiple uncertainties of renewable energy include three aspects: uncertainty of scale,uncertainty of output,and uncertainty of extreme scenarios.The physical characteristics of energy storage include the characteristics of charge/discharge life and the characteristics of limited energy.The uncertainty of the scale growth of renewable energy makes it not economical to allocate energy storage centrally at one time.Therefore,how to coordinate the multi-stage optimization configuration of energy storage in the growth trend of renewable energy is one of the research questions of this article.In addition,a very small number of extreme scenarios caused by the uncertainty of energy output may lead to redundancy in the configuration of energy storage.Therefore,it is also one of the research questions of this paper to develop the optimal energy storage plan that meets the investor’s risk appetite.It is worth noting that distributed energy storage and available energy sources operate in the distribution network.It participates in the time-of-use electricity price market and is also a passive responder of prices.In order to obtain greater profit margins,decentralized energy storage intensively participates in the power market competition and profit through the form of aggregators.Taking into account the characteristics of energy storage’s own charging/discharging life,by formulating a reasonable charging and discharging strategy,the operating life of energy storage can be increased and operating costs can be reduced.Therefore,how to develop the random scheduling decision model of the energy storage day-ahead market considering the charging and discharging life is one of the research questions in this paper.Energy storage can participate in the reserve market to enrich the income source of energy storage.Therefore,how to build a model for energy storage to participate in the energy market and the reserve market,and to tap more potential profit space of energy storage,is one of the research questions of this paper.For the questions mentioned above,the details of this work are as follows:1)Research on multi-stage optimal allocation of distributed energy storage facing the scale uncertainty of renewable generationThe dynamic growth of distributed photovoltaic capacity connected to the grid will bring a series of challenges to the safe and stable operation of the grid.Therefore,it is necessary to carry out research on the multi-stage optimal sizing of energy storage considering the growth trend of photovoltaics.Firstly,the proposed multi-stage investment of energy storage is aimed at maximizing the multi-stage investment income.The investment income is made up of investment costs,arbitrage income,environmental benefits,and costs reduction via delaying grid construction.Then,Monte Carlo simulation method can use the historical data to generate photovoltaic and load calculation scenarios.The k-means method is used to obtain typical operating scenarios,and the solution rate can be improved by reducing the number of scenarios.Next,the nested particle swarm algorithm is used to calculate the operation strategies for the short-period operation and sizing strategies for long-period planning.Finally,in the IEEE test system,the multi-stage optimization configuration strategies and profits of energy storage under different photovoltaic growth modes are discussed.The results prove the economic advantages of multi-stage investment,and further clarify the dominant position of construction cost,arbitrage income and residual value in energy storage investment profits.2)Research on distributed energy storage planning considering the risk caused by renewable generation outputFluctuations of renewable energy output and load may lead to the existence of extreme scenarios,causing systemic risks to the system.And the tail risk loss caused by extreme scenarios is often difficult to reflect.Therefore,in this paper,a financial analysis tool,Conditional value at risk,is introduced to assess the tail risk of extreme scenarios.Considering the uncertainty of renewable energy output and load during energy storage planning period,a two-stage planning model for hybrid energy storage based on conditional value-at-risk is proposed.The upper layer aims to minimize energy storage investment costs and optimizes storage’s capacity.The lower layer aims to minimize intra-day operating costs and optimizes intraday operations.At the technical level,hybrid energy storage is introduced to cooperate with the renewable generation.Based on the status switch constraints of energy storage,a universal energy storage model for power-based type and energy-based type is proposed.At the mathematical level,equivalent transformations are carried out for the lower-level model parts that are difficult to be solved by commercial software optimization,including: 1)second-order cone model transformation;2)absolute value model linearization transformation;3)chance constraint model linearization transformation.Finally,the results of energy storage optimization configuration under different hybrid energy storage systems,different confidence levels,and different wind power permeability are compared and analyzed,which proves the feasibility of the model.At the same time,the results show that the proposed model can provide a decision-making basis for the selection of hybrid energy storage and investment decisions considering risk preference.3)Research on the day-ahead market scheduling model of energy storage considering the constraints of charge and discharge lifeRelying on the advanced measurement system and energy management system,distributed energy storage can realize its large-scale management through the business model of aggregators.Large-scale energy storage participates in the dayahead market scheduling in the form of aggregators,which can strive for higher profits.First,a unit commitment model for day-ahead energy market considering the life of energy storage charging and discharging cycles is proposed.This model can ensure that when the energy storage system participates in the energy market,energy storage can avoid frequent switching of charging and discharging behavior while pursuing short-term economic benefits.However,the proposed model that considers the number of charge and discharge switching times of the energy storage system will further aggravate the difficulty of solving the problem.Secondly,to speed up the solution of the problem,an enhanced model was developed for energy storage and generators.The enhanced version of the model further compresses the solution space and approaches the optimal convex hull.Finally,the IEEE standard test system is used to verify the computational performance of the proposed model.The results show that the proposed enhancement model has obvious effects in shortening the calculation time.4)Research on the stochastic scheduling model of energy storage system under the energy and reserve marketOnly participating in the day-ahead energy market is inconsistent with the diversified benefits of energy storage.Therefore,to further expand the income of energy storage,this paper discusses the stochastic scheduling model of energy storage participating in the day-ahead energy and reserve market at the same time.First,power reserve model of energy storage system proposed in this paper covers all the possible condition that energy storage can provide reserve,which is more complete than the existing one.Next,considering the energy limited feature of energy storage,this paper proposes the concept and model of energy reserve for the first time,to guarantee the energy storage can supply power in energy and reserve markets continuously for several hours.Then,to solve the large-scale unit commitment problem,a heuristic progressive hedging decomposition algorithm is proposed to decompose large-scale stochastic problem into several sub-problem,which can be solved by iteration.Finally,several test cases are used to prove the advantages of the proposed method.The results show that the proposed energy and reserve model for energy storage can provide more reserve services,and can ensure the continuous supply of energy.