Research On Probability Modeling Method Considering Multiple Distributed Output Dependencies

As more and more distributed new energy accessed to the grid,due to their respective output has a certain randomness and regularity,the traditional deterministic trend model which grid research used has already can't meet the demand of research,the probability flow algorithm was developed on the basis of the original flow calculation to described the characteristic of their output.When various kinds of probabilistic trend,such as PV,Wind and ES,was injected into the same node of the gird,due to the output of these various kinds of distributed power generation and load is influenced by the same scope of geographic space's climatic factors such as temperature and environment,so its tidal current model must have a certain correlation between.The independent of each variable is inevitable during probabilistic power flow calculation.Both at home and abroad,the research about the calculation of the probability considering correlationa between variety of probability grid trend variables injecting into the same node of the gird is lack of accurate and effective processing method,therefore,the research to solve this problem is of great significance.In this paper,to solve the problem mentioned above,the main work is as follows.Considering the establishment of a comprehensive model which can deal with a variety of distributed energy probabilistic's tidal current correlations,they are treated as the comprehensive node injection to carry out the probabilistic trend calculation.There the distributed power treated as a negative load.First of all,this study selected the nonparametric kernel density estimation method to obtain the probability distribution of all kinds of distributed energy output,getting ride of the prior knowledge of the exact form of probability distribution assumption,using the data sample itself,the data distribution,to obtain the output capacity of PV and Wind's nonparametric estimation model,comparing with the parameter estimation method,the nonparametric estimation method has a better effect in the description of the edge scenery of the variety of distributed energy.Secondly,on the photovoltaic and wind power output,on the basis of accurate description of distributed energy output,five commonly used two-dimensiona copulas are expanded to three dimension,and hybrid models contain a variety of three-dimensional copulas are structured,based on the above structured results,the joint probability distribution models are respectively established.Using the EM algorithm to solve the problem of multiparameter estimation in the established models,the simulation results show that the three dimensional mixed copulas model fitting the actual distribution better.Finally,the comprehensive probability model is analyzed in frequency,constructing the three-dimensional coupla-MC method to obtain the joint probability distribution modl of the comprehensive load's probability distribution model.Providing feasible steps for the pretreatment before using the existing probability flow algorithm calculating method.