Dynamic Economic Dispatch Of Power System Considering High Proportion Of Wind Power And Its Spatial-temporal Correlation

The research content of this paper is the dynamic economic dispatching model of power system considering the high proportion of renewable energy and its time-space correlation.On the research object,this paper explains the application physics background and engineering background of specific renewable energy,and determines the necessity and important economic value of studying the spatial correlation of wind power.This paper also analyzes the economic dispatching field of power system,and studies the broad application prospect and practical engineering value of this problem.The advantages and development potential of wind power for China’s energy security,technology upgrading and environmental protection are clarified,which have important strategic value.In terms of correlation feature modeling,taking wind power as the specific illustration object,the physical basis for the existence of the correlation feature was analyzed.By mining the existing active power data of wind power,some typical statistical indicators used to judge the correlation feature were proposed.Get correlation classification basis,using the statistical indicators,this paper proposes a data filtering method considering correlation characteristics,will be mutual correlation characteristic of data validation,improve the existing method when handling data of redundant under-pressure sieve analyzer,will not be able to reflect the correlation data,accurate correlation of the data set.Based on this data set,a versatile distribution function modified was proposed to perform modeling and characterization,to achieve a better fitting of high-dimensional random variables.On the economic dispatch problem,this paper proposes a new method considering correlation characteristic of economic operation;the method is further extension of the original scheduling method.On the basis of the past,joined in on the investigation to the correlation characteristics,namely when introducing spatial correlation in the objective function,the constraint conditions to restrict,also get the original scheduling model considering the correlation characteristics of economy.Furthermore,based on this model,this paper proves that this kind of problem has exactly the same mathematical form,that is,after considering the time-space correlation,only the part describing the probability distribution needs to be replaced by the marginal distribution function of the high-dimensional distribution function.By this method,the original theoretical system is perfectly extended to the optimization problem with high-dimensional random variables,and the unity of its mathematical form is confirmed.Thus,the optimization problem of multiple random variables is more completely described.The above contents are verified in the IEEE-30 node and 118 node system,and have obvious advantages in scheduling cost,scheduling plan output and other aspects.In order to solve this kind of problem,an efficient algorithm is proposed.The algorithm is based on the original model the objective function of the secondary features;on the basis of classic sequence quadratic programming is improved on the criterion of interval,the original problem can be approximated by a series of higher accuracy of quadratic programming problem,to approximate the original problem effectively.Raising the precision of problem solving,it also promises the improved algorithm with higher flexibility.In the IEEE-30 nodes and 118 node system,the proposed method is verified,which proved the effectiveness of the proposed method in this paper.