| The large-scale utilization of fossil energy promotes the rapid development of China’s economy,but intensifies the contradiction between energy and environment at the same time.The efficient utilization of renewable energy has become the main direction of energy structure optimization and an important support for sustainable development.In order to further rationalize the energy structure and explore the decisive role of the market in resource allocation,China’s new round of power system reform will give full play to the market-oriented function,establish a fair,reasonable and competitive power market,and tap the vitality of multi-source competition on the generation side.Distributed renewable energy has the characteristics of small single capacity,regional dispersion and output fluctuation,which increases the difficulty of unified dispatching of power grid and brings risks to the operation of distribution network.Therefore,it is an important foundation for the future development of distributed renewable resources to realize the aggregate management of Distributed Renewable Resources and improve the level of renewable power consumption and market competition.Based on this,this paper combined with the virtual power plant technology,carried out the aggregation management of distributed resources,and modeled and optimized the internal operation of the virtual power plant with multi distributed resources aggregation.At the same time,combined with the reform path of China’s power market,this paper analyzed the transaction optimization modeling of the virtual power plant participating in the medium and long-term power market,day ahead market and day real-time market,Based on the actual situation of China’s areas without electricity spot market,this paper analyzes the optimization of virtual power plants participating in ancillary service market transactions,thus forming the optimization research of virtual power plants participating in multi-level electricity market transactions.The main contents of this paper are as follows(1)In this section,the basic concept,characteristics,typical project mode and function of virtual power plant are summarized.Firstly,the definition and characteristics of virtual power plant are introduced in detail,and the typical structure of virtual power plant is analyzed.Secondly,typical virtual power plant projects are selected from both domestic and foreign sides to carry out research.Typical virtual power plant projects in Germany,EU and other foreign countries are selected to analyze their operation mode and purpose.At the same time,combined with the virtual power plant projects in Shanghai,northern Hebei,Jiangsu and Tianjin,the implementation contents and outstanding benefits of typical virtual power plant projects in China are summarized.Finally,based on the basic characteristics and typical implementation projects of virtual power plants,the categories of virtual power plants are analyzed,and the main feasible paths for virtual power plants to participate in power market transactions are proposed,which paves the way for the research on the optimization model construction of virtual power plant power transaction in the following chapters.(2)A low-carbon optimal operation model of virtual power plant considering electrical mutual conversion is proposed.Firstly,combined with the policy guidance of "carbon emission reduction",considering the change of energy flow in the virtual power plant after the introduction of P2G technology,a virtual power plant structure connected with "P2G" equipment is constructed.Secondly,combined with the analysis and preliminary calculation of the flow direction of electric energy and natural gas in the virtual power plant,carbon trading is introduced to achieve the full supply of carbon raw materials.Based on the "zero carbon emission”goal and economic goal,a multi-objective optimization operation model of virtual power plant considering the "electricity gas" mutual transformation is constructed.Finally,the example analysis shows that after the introduction of P2G equipment and carbon emissions trading,the virtual power plant operation can achieve "zero carbon emissions" template,and the economic benefit is positive,that is,the virtual power plant with "P2G" has profit potential.In addition,the impact of carbon trading price on the operation of virtual power plant is analyzed,Furthermore,the price transmission mechanism of the impact of carbon trading price on the operation of virtual power plant is analyzed.(3)The trading strategy of virtual power plant in medium and long term market is proposed.Firstly,the two modes of China’s electricity market are analyzed,and the transaction varieties and modes in the medium and long-term market are summarized;Secondly,this paper combs the relevant laws of the decomposition of medium and long-term electricity contracts in the current market,and puts forward the revenue sources of virtual power plants in the medium and long-term market under different market modes,that is,the revenue models of virtual power plants under fixed price contracts and price difference contracts;Then,combined with the renewable energy quota system and green certificate trading mechanism,an optimization model of medium and long-term contract trading of virtual power plant considering renewable energy derivatives is constructed;Finally,on the basis of green certificate transaction,contract transaction and output cost of each unit,different renewable energy output scenarios are designed,and the benefits of virtual power plant participating in medium and long-term market contract transaction under centralized power market transaction rules and decentralized power market transaction rules are considered.(4)The transaction optimization model of virtual power plant in day ahead market is proposed.Firstly,the sources of uncertainty for virtual power plants to participate in day ahead market trading are analyzed by modeling;Secondly,an optimization method for the uncertainty of the output of virtual power plants in day ahead market is proposed.The prediction method is optimized and improved from the aspect of power generation prediction,and the wind and solar output prediction model of eemd-cs-elm is constructed.Combined with CVaR theory,the uncertainty deviation between the actual output and the planned output is further overcome,Based on the optimization of forecasting method and CVaR,the trading optimization model of virtual power plant day ahead market is constructed;Finally,a typical area is selected for example analysis to verify the effectiveness of the improved prediction method and the feasibility of the model.(5)A three-stage transaction optimization model of virtual power plant based on master-slave game is proposed.Firstly,the coupling relationship among day ahead market,day ahead market and real-time market is analyzed,and it is proposed that virtual power plants can achieve optimal equilibrium through game in day trading.Secondly,combined with the three stages of day ahead,time ahead and real-time,taking the maximum revenue of virtual power plant as the goal,considering the revenue composition of different stages,the relevant optimization model is constructed by stages.Finally,referring to the spot market transaction data of Denmark market in northern Europe in April 2020,the paper designs a virtual power plant market transaction example to verify the effectiveness of the three-stage transaction optimization model.(6)A transaction optimization model of virtual power plant based on information gap decision theory is proposed.Firstly,the transaction path and internal physical model of virtual power plants with P2G are proposed;Secondly,considering the load uncertainty in the market transaction,the transaction optimization model of virtual power plant is proposed under the two preconditions of not considering the load uncertainty and considering the uncertainty;Finally,the Latin hypercube sampling scene generation method and distance measurement scene reduction method are combined to deal with the source side uncertainty.The multi-objective particle swarm optimization algorithm,Pareto optimal solution screening model and fuzzy theory are combined to solve the proposed model,and a numerical example is designed to verify it. |
PDF Full Text Request