Research On The Uncertainty Day-ahead Dispatching Method Of Power System Considering Multivariate Random Variables

With the continuous development of new energy including wind power and photovoltaic,its large-scale access has brought profound changes to power grid,significantly increasing the system’s uncertainty.The concept of "demand response" has become a new way to solve the problem of large-scale new energy integration.However,it also makes the power system more complex and brings new challenges to the dispatching and operation control of the power grid.Therefore,it is urgent to study the scheduling methods of power systems that can consider multivariate random variables.This paper focus on UHVDC power peaking balance,the scheduling method of power system considering multivariate random variable correlation,and the scheduling method considering the multiple types of random variables.1)A new evaluation method is proposed,and two new evaluation indexes are introduced,which are peak shaving margin and abandoned electricity quantity.The piecewise evaluation model is set up:when the peak surplus is positive,the goal is to maximum of peak shaving margin,when the peak surplus is negative,the goal is to minimize the amount of abandoned electricity.The balance of peak adjustment is checked month by month,and the optimal peak regulation mode of UHVDC is worked out according to the results of the check.This chapter will provide theoretical support for the study of optimal scheduling method of power system on the follow-up before.2)A robust scheduling method is proposed,which considers the uncertainty of load,wind power and photovoltaic output and the probable correlation.Firstly,an improved robust optimal scheduling model considering multivariate random variable is constructed.Then,stochastic samples of correlation are transformed into independent random samples by Cholesky decomposition method,and the worst case is determined based on the sample feature.Finally,the model is solved.Compared with the traditional method,it can improve the economy of the system operation while ensuring the robustness of the scheduling decision,and has a high efficiency.3)Based on scheduling model of power system considering multivariate random variable correlation,the demand response into the scheduling model,construction of power system scheduling model considering the multiple types of random variables.The Benders decomposition is used to decompose the model into a main problem and two sub problems,solve them respectively.The result of IEEE 30-bus system indicated the scheme combined power side and demand side based on demand side interaction may conductive to the safety and economic operation in power system.