Robust optimization scheduling method considering wind power multivariate correlation ellipsoid set

Abstract

The invention discloses a robust optimization scheduling method considering a wind power multivariate correlation ellipsoid set. The method comprises the steps of obtaining edge probability distribution of a wind power true value and a wind power prediction value according to historical data of a wind power plant; initializing a cyclic variable; calculating a covariance matrix; calculating parameters of multivariate normal condition distribution; sampling the condition distribution to obtain an actual value sample; constructing a multi-ellipsoid set; calculating a multi-ellipsoid integrity index and a high-efficiency index with the dimension TR of historical data on the dth day; traversing all dates, and calculating comprehensive indexes; traversing all TR to obtain an optimal correlationtime scale; constructing a multi-dimensional ellipsoid uncertainty set according to the latest day-ahead prediction value; and based on the multi-dimensional ellipsoid uncertainty set, constructing amicrogrid two-stage robust scheduling model, and finding an economic optimal scheduling scheme in the worst uncertainty scene. The method is used for solving the uncertainty of wind power in a microgrid, improving the economy of a robust scheduling scheme and reducing unbalanced power.

Description

technical field [0001] The present invention relates to an optimal scheduling method, more specifically, relates to a robust optimal scheduling method considering wind power multiple correlation ellipsoid sets. Background technique [0002] With the depletion of fossil fuels and increasing environmental pollution, renewable energy such as wind power has played an increasingly important role, accounting for an increasing proportion of power system energy supply. However, the inherent randomness and volatility of renewable energy sources such as wind power pose challenges to power system scheduling. When optimizing scheduling, it is necessary to consider the possible impact of this randomness in order to achieve the economy and stability of scheduling. [0003] Robust optimal scheduling considers that random variables fluctuate within the range of an uncertain set, and the obtained solutions can resist the interference of this uncertainty. Robust optimization can deal with t...

AI Technical Summary

Problems solved by technology

However, the time correlation of forecast error is only significant within a limited time scale, and the time correlation is weak for periods far apart, so its time correlation should be ignored, but there is no clear criterion for this time scale

Images