Method for generating and reducing wind power multi-period time sequence scene

Abstract

The invention discloses a method for generating and reducing a wind power multi-period time sequence scene. The method comprises the following steps of: 1, establishing a single-period asymptotic optimal scene generation strategy based on a Wasserstein probability distance index and a multi-period scene reduction strategy based on an improved tabu search algorithm; 2, defining a distance index torepresent a spatial feature, and defining a correlation loss and probability similarity index to represent a random feature retention degree before and after reduction of a high-dimensional scene; and3, solving a wind power multi-period time sequence scene by using an improved tabu algorithm. According to the method, the problem of large feature information loss during scene reduction is solved from the two directions of the space and random features of the scene, and the random features of the wind power can be effectively expressed while the wind power simplified scene is accurately and efficiently generated.

Description

Technical field[0001]The invention relates to the field of power system distribution network planning, in particular to a method for generating and reducing wind power multi-period time series scenarios.Background technique[0002]At present, various random variables inevitably exist in the power system. The uncertainty of random variables will greatly affect the economic, stability, and reliability evaluation results of the power system, which also makes stochastic programming have a wide range of application prospects in decision-making, optimization and scheduling problems. As a breakthrough point, scenario analysis provides a good solution for the uncertainty expression of various random variables.[0003]As the proportion of renewable energy generation in the power system gradually increases, generating representative sequence scenarios of renewable energy is an indispensable link in power system planning and decision-making. For example, wind power has strong randomness and volati...

AI Technical Summary

Problems solved by technology

Most of the existing wind power sequence scene generation methods assume the probability equivalent value of the original scene. Although the calculation efficiency has been improved, when the scene is reduced, the data set is generally only divided according to a certain distance criterion, ignoring the high-dimensional Problems such as the similarity between scenes and the correlation between random variables in the scene sequence lead to a large loss of scene feature information

Images