Research On Optimal Operation Of Virtual Power Plant Under Carbon Peaking And Carbon Neutrality Background

Currently,considering the serious impact of carbon emissions on global climate change,how to promote clean and low-carbon transition has become a key concern in various fields.As one of the fields that generate a large amount of carbon emissions,the transformation of the power system is imminent.With the proposal of carbon peaking and carbon neutrality and the advancement of the construction process of new power system,there will inevitably be a large number of renewable energy connected to the power system,which will pose a certain threat to the safe and stable operation of the system.In view of this,exploring and researching methods and paths for power system safety,environmental protection,and economic transformation is one of the important supports for the carbon peaking and carbon neutrality and the construction of new power system.As a new business form that can break regional constraints and widely gather distributed resources,virtual power plants can provide a unified access and control platform for distributed power generation units,thereby reducing the impact on the power grid caused by a large number of decentralized access of distributed power generation units;It can also fully tap the potential of load resources,effectively improve system flexibility,and provide sufficient flexible resource protection for the large-scale and safe consumption of renewable energy.In response to the above issues,this paper conducts corresponding research on the optimization of virtual power plants under the carbon peaking and carbon neutrality background.The main contents and conclusions are as follows:In response to the above issues,this paper conducts corresponding research on the optimization operation of Virtual Power Plant(VPP)under under Carbon Peaking and Carbon Neutrality background.Firstly,based on the analysis of typical elements and forms of virtual power plants,the supporting mechanism of virtual power plants for Carbon Peaking and Carbon Neutrality goal is discussed,the connotation and basic technology of situation awareness for virtual power plants are analyzed,and the operation mode of virtual power plants and the regulation mechanism based on cloud-edge-device collaboration are constructed.Secondly,four key characteristic indicators are selected as the basis for the division of the output operation status of distributed wind turbines.On this basis,a distributed renewable energy output prediction model combined with the Convolutional Neural Network(CNN)and Long Short Term Memory(LSTM)neural network is constructed,And through case analysis,it has been proven that the combined model proposed in this paper has superior performance in predicting distributed renewable energy compared to using CNN or LSTM models alone.Once again,a load classification method based on the principle of responsive time scales is proposed,and a multi time scale optimization operation model for virtual power plants is constructed.Through empirical analysis,conclusions are drawn:firstly,adopting differentiated load incentive and regulation models can effectively improve the comprehensive efficiency of virtual power plants;secondly,changes in the capacity ratio of the two types of loads will have a significant impact on the renewable energy consumption capacity and revenue of virtual power plants.Finally,the potential problems and key development tasks that virtual power plants may face in the future development process are analyzed,and the development guarantee mechanism of virtual power plants under the dual carbon target is explored.