Research On Optimal Combination And Operation Strategy Of Multi-type Resources Of Virtual Power Plant

Distributed energy resources(DER)will be an important part of the future power system with the growing deployment of distributed renewable energy generation,electric vehicles,demand response and other technologies.In smart grid,the management of DER involves such aspects as active distribution system,microgrid,dynamic electricity price,virtual power plant and so on.Among them,VPP is a flexible way to integrate geographically dispersed,heterogeneous,and multi-type DER.VPP participates in the power market and grid management,also facilitates the agregated benefits of DERs.The main contributions of the thesis work are as follows:Firstly,this thesis reviews the basic structure and key issues of the above VPP management methods.Focusing on VPP,it points out that virtual power plants have good applicability in information and communication infrastructure,power market and DER integration.The research and application of VPP is currently in its early stage,but it has broad application prospects.It provides a solution for the transformation from centralized power production to the distributed energy utilization.Secondly,considering the heterogeneity of DER,a universal power generation model based on VPP interactive resource is established.Subsequently,DER is classified as distributed generations and flexible loads.The five types of resources,photovoltaics,wind turbines,air-conditioning,electric vehicles,and energy storage system,are used as typical representatives,and their power characteristics are analyzed in detail.In addition,compared with traditional loads,flexible loads have schedulable characteristics.The regulation ability of charging loads tends to be underestimated if analyzed with traditional reliability definitions.Therefore,based on the energy demand of flexible loads,this thesis proposes new definition of “loss of load” and develops reliability index for reducible load and transferrable load according to the response characteristics of flexible loads.Thirdly,the response capacity of VPP is determined by the composition and responsiveness of the members of the internal alliance.Considering that the user’s response behaviors are influenced by the user’s willingness to respond,based on the level of control authority of the VPP to the interactive resources,the response power of interactive resource is divided into risky assets and risk-free assets.Then the resource’s response scheduling problem is mapped to the asset weight allocation problem in the portfolio theory.Therein,the user’s response power uncertainty is measured by the portfolio risk;the optimal combination model is established with the expected maximum benefit as the objective;and the resource combination of hybrid VPP is provided.Fourthly,a high proportion of distributed renewable energy generation is prone to cause power flow congestion.Taking advantage of DER’s dispersion characteristics,VPP can propose alternatives to provide congestion management services to the grid based on user protocols.In order to deal with the uncertainty caused by power forecast error in operation,a two-stage optimization model is established by using the day-ahead and intra-day coordinated optimal scheduling.In the first stage,the scheduling model is established to describe the uncertainty of renewable energy forecast by scenario set.In the second stage,the resource load forecast is continuously updated to minimize the uncertainty caused by the prediction error.Based on the aggregation and flexible control of multiple flexible resources,this thesis constructs a VPP optimized operation model for the optimization of multiple controllable resources and economic operation problems,and provides support for the practical application of VPP technology.