Research On Dynamic Scenario Generation Of Wind Farm Output And Unit Combination

With the development of social economy,human demand for energy is increasing.Large-scale wind power has been integrated in grid operation.Compared with conventional fossil fuels such as coal and oil,the uncertainty of wind power is very obvious,and the wind power can't be accurately predicted under the current technical conditions.At the same time,due to the rapid fluctuation of wind speed,the wind power fluctuates frequently,which can't be scheduled as traditional energy.How to describe the uncertainty of wind power and take it into account in power system scheduling and operation decision are the key question.In this paper,the theory of scenario analysis is applied to describe the uncertainty of wind power and the stochastic optimization method is applied to the scheduling and operation decision of power system with wind power.In this paper,the dynamic scenario generation method of single wind farm output is studied.Firstly,the randomness of wind power is modeled.By analyzing the actual distribution model,normal distribution model and versatile distribution model,these three models are used to describe the actual probability distribution of wind power,and then analyze the wind power fluctuation.The versatile distribution of wind power is used to describe the fluctuation characteristics of wind power.Finally,according to the inverse transformation sampling method and the use of multiple standard normal distribution to describe the correlation of random variables across time series,a method of randomly generating a large number of simulated wind power uncertainties based on versatile distribution is proposed.This paper demonstrates the effectiveness of generating a wind power dynamic scenario method based on versatile distribution.In this paper,the dynamic scenario generation method of multi-wind farm is studied.Compared with the single wind farm,correlation between the wind farms should be taken into account when modeling the output uncertainty of multi-wind farm.In this paper,several common correlation measures are introduced to analyze the correlation of multi stochastic variables.Then the Copula function theory is introduced to connect the multivariate random variable joint distribution function with its respective edge distribution function,with the correlation between random variables considered.The method to model multi-wind farm combined output based on the Copula function is proposed,and then combined with the scenario method,this paper proposes two types of multi-wind farm combined output scenario generation method,scenario generation method based on correlation analysis and dynamic scenario generation method based on empirical Copula function.This paper proves the validity of the dynamic scenario generation method of multi-wind farm output based on empirical Copula function.In this paper,how to consider the uncertainty of wind power in the dispatch and operation of power system is studied.The unit combination problem of power system with large-scale wind power is formulated,and the stochastic unit combination model is transformed to a mixed integer linear programming(MILP)problem.Case study is designed in IEEE 118 node bus system.The accuracy of solving the versatile distribution model in the single wind farm unit combination model and the necessity of using the dynamic data of the multi-wind farm based on the empirical Copula in unit combination model are verified by comparative examples.Finally,this paper summarizes the dynamic scenario generation method of wind farm output and its application in the stochastic unit combination.The prospects of future research work are summarized.