Study On Water And Wind Complementary Coordination Method Of Provincial Power Grid With High Proportion Of Renewable Energy Sources

With the gradual deepening of country’s energy structure transformation,the proportion of clean energy such as wind power and photovoltaic in the provincial power grids has increased year by year.Reasonably deal with the uncertainty of clean energy power generation such as wind power and photovoltaic,and increase the proportion of clean energy consumption and stable access to the grid.The scale of installed capacity has become the key to whether my country can achieve clean and efficient power generation in the field of electric energy,and then achieve the goal of "carbon neutrality" in the 14 th Five-Year Plan.Hydroelectric power generation has no pollution emissions and flexible power generation,which is easy to be used as a complementary power source for strong volatility and clean energy such as wind power and photovoltaic.This paper focuses on the provincial grid with a high proportion of hydropower and wind power,and conducts research on complementary and coordinated dispatch between cascade hydropower and regional wind power clusters.The main research contents are as follows:(1)Proposed a long-term dimensionality reduction optimization scheduling method for large-scale hydropower systems.This method uses the principal component analysis of the series of dispatch samples of the hydropower station group to identify the characteristic value of the reservoir water level change in the dispatch process and its corresponding characteristic function.The KL method is used to describe the reservoir water level sequence as a linear function of the aforementioned water level change characteristic item,and determines the power generation dispatch process corresponding to the specific incoming water through the combination of the random coefficients of each item.Kullback-Leibler divergence is introduced to optimize the probability distribution of the random coefficients of the characteristic items according to the characteristics of the problem,and a two-stage iterative optimization strategy is established to achieve efficient and stable solution of the coefficients of the characteristic items.(2)A method for generating typical scenarios of regional wind power clusters based on the C-Vine Copula theory is proposed.This method performs cluster analysis on the historical forecasted wind speed sequence of the region,and divides the historical output data of wind farms in the four main wind power concentrated areas according to the typical meteorological classification of each region.The correlation analysis of the output process between wind farms is carried out on each historical output sequence data set classified according to the wind speed process,and the Copula joint probability distribution model is established for the output of the wind farm stations in the area at each time period using the C-Vine Copula function generation method.A set of typical output scenarios of each wind power station in the region is generated through the C-Vine Copula joint distribution function of wind power plants in various regions,and the typical scenarios and corresponding probabilities are obtained through further dimensionality reduction clustering.(3)Based on the actual hydropower and wind power grid data in Yunnan,combined with the above-mentioned hydropower dimensionality reduction scheduling method and wind power cluster scenario generation method,the joint optimization scheduling model of hydro-wind power in Yunnan Province is solved.From a long-term scale,while ensuring the absorption of wind power,the system aims to maximize the power generation of the system,and optimize the dispatch of Lancang River cascade hydropower based on the actual output of wind power clusters.From a short-term scale,the effectiveness of the C-Teng Copula theory method in generating typical output scenarios of regional wind power clusters for short-term hydro-wind peak shaving of power grids is verified.Finally,the various research parts involved in this article are summarized,and the parts that have not yet been fully studied are prospected.