Research On Optimal Operation Method Of Renewable-Dominated Power System Driven By Data Value

In recent years,the increasingly serious environmental pollution problem and the urgent need for energy in economic development have led countries around the world to gradually attach importance to the application of renewable energy.However,for the power system,the output of renewable energy sources,such as wind power and photovoltaic,has strong uncertainty and volatility.Accepting the integration of renewable energy sources into the grid while maintaining system security and stability is a major challenge faced by the renewable-dominated power system.With the development of smart grid technology and the advancement of communication technology,data has become an important asset for the operation of the power system.Data can effectively eliminate the uncertainty of renewable energy output,assist in the planning and operation of renewable energy power systems,and effectively promote the development of the power grid.However,the vast power system generates massive amounts of data at every moment,and the value of the data is difficult to determine.Therefore,the power grid is unable to filter redundant data and locate high-value information,resulting in low efficiency in information utilization.This article studies the utility value and economic value of renewable energy data quality in power system scheduling scenarios,assists in efficient identification of required data in the power grid,promotes the consumption of renewable energy,improves resource utilization efficiency,and improves the economic efficiency of power grid operation,and achieves the following three innovative results:(1)This paper proposes a universal data value calculation model based on the basic characteristics of power data.The characteristics and economic value calculation methods of data as a special asset are analyzed,which differ from traditional physical assets.The quality dimensions of data and the application of data quality in different fields are researched.Based on the utility based data value calculation method,combined with the characteristics of power data and the application scenarios of power system operation,a value calculation model suitable for power system data is proposed.(2)This paper researches the utility value of renewable energy prediction data based on the dimension of data quality.The information value of renewable energy generation power data in power system operation problems is analyzed.The data quality of renewable energy power is calculated by selecting specific quality dimensions,such as information entropy and non noise ratio.A power prediction model is established to analyze the utility of renewable energy data,that is,how additional photovoltaic power generation data can help improve the accuracy of daily prediction,and explore the utility value of renewable energy data based on information quality.(3)Based on the day ahead scheduling model of unit commitment,the economic value calculation model of new energy data is proposed.The day ahead dispatching of power system is selected as the application scenario of new energy information.A twostage stochastic unit commitment model is built to quantify the economic value of data,and a data value driven optimal dispatching model of power system is established.The impact of different quality new energy data on the accuracy of daily prediction is explored,and then the economic value of these data is analyzed based on scheduling scenarios.Based on case studies of IEEE 30 and 118 node systems,the impact of photovoltaic data quality on prediction accuracy is explored,and the functional relationship between data quality and system economic cost is researched.The proposed data value calculation method can quickly and effectively help the power grid determine its economic value in scheduling scenarios based on the quality attributes of the data itself.