An Improved Modeling Method for Uncertainty of Renewable Energy Power Generation Considering Random Variables Correlation

Abstract

With the increasing proportion of renewable energy access, accurate modeling of renewable energy uncertainty plays an important role in power system operation, planning and decision-making. In this paper, an improved method of scene generation and scene reduction is proposed to realize the uncertainty modeling of wind power generation and photovoltaic power generation. To deal with the correlation between renewable energy sources, wind-solar joint distribution function is firstly established to generate the power output scenes based on the Frank-Copula function. Considering the solving difficulty of the inverse transformation of cumulative probability distribution function, random sampling method and inverse transformation algorithm are then improved to simplify the modeling process while ensuring modeling accuracy. In addition, cubic spline interpolation method is also used to fit the cumulative output curve of renewable energy generation, and the backward scenario reduction technique is applied to reduce the huge sample number of renewable energy random variables. By comparing with the parameter fitting scenario generation method that does not consider wind-solar correlation, the proposed modeling method is lastly verified to better reflect the uncertainty, correlation and complementarity of wind power generation and photovoltaic power generation.