A new energy uncertainty ensemble modeling method based on spatiotemporal correlation

Abstract

The invention discloses a new energy uncertain set modeling method based on spatial-temporal correlation, and belongs to the technical field of electrical engineering. The method comprises the following steps of firstly, collecting historical data of all new energy field station output, wherein historical data of each day is a historical scene; then solving a minimum closed-package high-dimensional ellipsoid capable of surrounding all historical scenes; selecting a plurality of vertexes of the ellipsoid as the top points of the uncertain set, namely, obtaining the expression of the generalizedpolyhedron defined by the vertexes to describe the uncertain set, and finally introducing a zooming coefficient modification uncertain set. According to the method, a new uncertain set is formed by convex hulls surrounded by limit scenes of a limited number, and can be easily expanded to the existing two-stage robust optimization method. The number of the limit scenes and the number of the randomvariables are linearly increased, and the solving speed of the robust optimization method is guaranteed. The method has the advantages that the economy of the set can be adjusted by modifying coefficient according to the actual demand.

Description

technical field [0001] The invention belongs to the technical field of electrical engineering, and more specifically relates to a new energy uncertain set modeling method based on time-space correlation. Background technique [0002] With the aggravation of energy crisis and environmental pollution, the penetration rate of photovoltaic wind power and other new energy in the grid is increasing, and new energy shows strong randomness, intermittent and volatility, and its prediction accuracy is low. This undoubtedly brings huge challenges to the traditional power grid operation mode. From the perspective of dispatching, the traditional dispatching often describes the economic dispatching problem and the day-ahead unit combination problem as a deterministic mathematical optimization problem. After the energy is introduced, the deterministic problem gradually becomes a stochastic optimization problem. In view of this problem, the existing literature mostly adopts methods such as...

AI Technical Summary

Problems solved by technology

However, in the existing robust optimization methods, box sets are often used to describe uncertain sets, that is, the spatial correlation between new energy stations and the time correlation of a single station are not considered, which undoubtedly increases the uncertainty Determining the volume of the set leads to the overly conservative scheduling strategy obtained by the traditional robust optimization solution, which leads to economic decline, and even the situation that no scheduling strategy satisfies the constraints

Images