Researches On The Characteristics Modeling And Application Of Regional Wind Farm Clusters Considering Correlation And Randomness

With the increasing energy crisis and environmental pollution,the application of clean energy to alleviate the predicament of energy and environment has gradually become the central issue.In recent years,China has vigorously developed new energy for power generation,and wind power has become the main force in the development of new energy.With the large-scale access of wind farms into the power grid,it will bring great challenges to the dispatch and operation of power systems.Wind power generation itself has strong stochasticity,volatility and intermittence characteristics.At the same time,in the actual power system,there are certain correlation between outputs of wind farms and wind farms due to the similarity of regional climate,which will affect the safe and stable operation of the power system.Therefore,how to accurately describe the stochasticity and correlation of regional wind farm outputs is of great significance for power system planning and dispatching operation.At present,wind power is developing in the direction with large scale and high concentration in china.With the number and capacity of wind farms outputs increasing,more and more accurate wind power models are needed.Therefore,the models of multiple wind farm outputs are proposed in this paper.The main research contents are as follows:(1)This paper proposed the multivariate wind power models considering correlation based on Copula function.It mainly includes the multivariate wind power modeling method based on Copula model,then illustrates the method of selecting the optimal Copula function from three different method.The example shows that outputs of wind farms in Hainan province are suitable for describing their characteristics by using the Clayton Copula function.(2)By further analyzing the temporal characteristics of wind power,this paper proposed a spatiotemporal correlated multivariate wind power model that takes diurnal and intraday variation feature of wind power into account.In order to embody the correlated variation of daily wind power at different sites,daily typical scenarios are obtained from historical wind power data using clustering algorithm.A univariable MCMC model is then established to describe the transition process of these scenarios.Next,multivariable MCMC models considering intraday variation are developed to describe the correlation existed in the intraday time sequences of multiple wind farms.The daily typical scenario Markov chain and intraday wind power time sequences for each typical scenario state are simulated successively and then complete multivariate wind power time sequences are generated.(3)Based on the historical measured power data of Gaopai wind farm and Sigeng wind farm in Hainan Province,the proposed models are verified three aspects: statistical characteristics,temporal characteristics and spatial correlation.The results show that the proposed time series model can accurately describe the daily and intraday variation characteristics of historical wind power,and has higher fitting accuracy in probability distribution,average wind power,auto-correlation function and cross-correlation function.The probability distribution statistics of Copula model is slightly lower,but its modeling process is relatively simple.Therefore,in the actual power grid planning and operation dispatching research,the models can be selected according to the acceptability of engineering errors and computational efficiency.