Research On Two-stage Optimal Scheduling Of Virtual Power Plant Considering The Uncertainty Of Photovoltaics

The virtual power plant integrates and manages the new energy,controllable loads,and energy storage system installed in the distribution network through distributed energy management systems and advanced communication means.It participates in the operation of the power grid as a special power plant,so as to better solve the problem of power imbalance between source and network loads,to fully tap the advantages of distributed energy to bring value and benefits to the grid and users.With the implementation of my country’s "30·60" carbon peak and carbon neutral strategic goals,and speeding up the construction of a new pow er system with new energy as the main body,virtual power plants will gather more and more new energies such as photovoltaics and wind power.The inherent uncertainty of new energy output is bound to bring deviations to the traditional single-stage economic dispatch.Therefore,it is very important to propose a flexible and comprehensive virtual power plant optimization scheduling method for carbon neutrality to deal with the impact of new energy uncertain output.Therefore,this study takes the virtual power plant optimal dispatch method under the background of carbon neutrality as the subject,and proposes a two-stage virtual power plant optimal dispatch method that takes into account the photovoltaic uncertainty.The main work of this paper is as follows:First,in view of the problem that the inherent uncertainty of photovoltaic output will bring deviations to the traditional single-stage economic dispatch,this paper analyzes the uncertainty from the two perspectives of day-ahead and intra-day,and designs a two-stage optimization scheduling framework that takes into account the uncertainty of photovoltaic.Based on the basic structure of the virtual power plant with photovoltaic power generation,a mathematical model of each unit of the virtual power plant including photovoltaic power generation,electric load,aggregate temperature control loads and micro-turbine is established,and the operating characteristics of each unit are analyzed.On this basis,a two-stage optimization scheduling method program was designed.In the day-ahead scheduling stage,the robust optimization method and the method of minimizing conditioned value at risk were used to solve the problem of lack of reliability of traditional day-ahead economic scheduling caused by photovoltaic uncertainty.In the intra-day scheduling phase,the aggregate temperature control loads and the reserve capacity of the micro gas turbine are used to respectively suppress the high-frequency and low-frequency components of fluctuating power,so as to solve the problem of large gap between the day-ahead scheduling plan and the intra-day operation strategy of the tie-line power caused by photovoltaic uncertainty.Secondly,in view of the traditional day-ahead economic dispatch model of virtual power plants,which only considers economics and ignores issues such as the balance of interests of internal units during operation,the impact on th e distribution network,and environmental pollution,a multi-objective master-slave game day-ahead scheduling model oriented to carbon neutrality and taking into account photovoltaic uncertainty is established for virtual power plant.Based on the master-slave game,a one-master-multi-slave day-ahead optimization scheduling model that takes into account the virtual power plant revenue,carbon emissions and the degree of fluctuation of tie-line power and the uncertainty of photovoltaic power generation is built by robust optimization method and minimizes conditioned value at risk method.At the same time,based on the established one-master-multi-slave model,the existence and uniqueness are proved,indicating that the model has a unique Stackelberg-Nash equilibrium solution.Then the model is solved by a distributed equilibrium solving method of genetic algorithm joint quadratic programming,and obtains a reliable day-ahead optimal scheduling scheme that takes into account the photovoltaic uncertainty.The simulation results show that,compared with the traditional day-ahead economic dispatch model,the multi-objective master-slave game dispatch model for carbon neutrality and photovoltaic uncertainty proposed in this paper can guarantee the comprehensive inco me of the virtual power plant operator,the income of the generation side and users tend to be balanced.And indicators such as carbon emissions,tie line power mean square error,and load peak-to-valley ratio have been improved.Finally,in view of the complex calculation and low efficiency of the traditional intra-day optimal dispatch power control method of virtual power plant in dealing with nonlinear problems,a virtual power plant intra-day optimal dispatch control method based on improved model predictive control is proposed.The Lyapunov stability is used to improve the traditional model predictive control,and the p ower control algorithms for the aggregate temperature control loads and the micro-turbine are respectively derived.The aggregate temperature control load and the reserve capacity of the micro-turbine are used to respectively smooth the high frequency and low frequency components of the fluctuating power to achieve intra-day response.And an intra-day optimal scheduling plan that takes into account the photovoltaic uncertainty can be obtained.The simulation results show that compared with the traditional model predictive control method,the improved model predictive control method proposed in this paper has better performance for aggregate temperature control loads and power control of micro-turbine,and the calculation time is shorter.Through the above work,the revenue of the virtual power plant power trading operator,the generation side and the user side under the conditions of photovo ltaic uncertainty is improved,the reliability of dispatch results is improved,the carbon emissions is reduced.And it is conducive to the stability of the external operating characteristics of the virtual power plant.The research results of this paper c an help my country achieve "30·60" carbon peak and carbon neutral strategic goals,and provide new ideas for the construction of a new type of power system with new energy as the main body.