Lateral error translation modification method of wind electricity power forecast system

Abstract

The invention provides a lateral error translation modification method of a wind electricity power forecast system. The method includes the steps that wind electricity power historical forecast and actual measurement data sequences are collected and are classified according to four seasons, and data of each season every year are divided into three kinds of power trend sections, that is a continuous power increasing section, a continuous power lowering section and an unchanged power section; duration time sequences of each power trend section are obtained according to power trend duration models, probability statistic analysis is carried out to obtain a macroscopic centralized distribution value, a wind electricity power forecast lateral error model is input to obtain a lateral error of each season every year, and a weighted average value of the lateral errors of each identical season of N years serves as a lateral error modification value of each season; translation modification is carried out on forecast power sequences supplied to power grid dispatching by the wind electricity power forecast system by means of the weighed average value, and the translation modification effect is evaluated on the whole. The method has common adaptability for power short-term or exceedingly-short-term forecast systems of wind electricity, photoelectricity and the like and can further improve precision of various power forecast systems.

Description

technical field [0001] The invention belongs to the technical field of power system prediction and control, and in particular relates to a lateral error translation correction method of a wind power prediction system. Background technique [0002] Predicting the output power of wind farms for a period of time in the future (short-term or ultra-short-term) with high accuracy and reliability is an important tool for power system dispatching agencies to formulate wind power dispatching plans and realize "controllable, adjustable" and large-scale parallelization of wind power generation. The premise of network goals and the problems that must be addressed. At present, most of the researches on wind power forecasting focus on forecasting methods and their improvement. The existing technology has carried out in-depth discussion and research on different wind power forecasting methods, and the national standard requires that grid-connected wind farms should be equipped with wind p...

AI Technical Summary

Problems solved by technology

[0005] Aiming at the lateral error problem existing in the existing wind power prediction system caused by the prediction data sequence leading or lagging behind the measured data sequence on the horizontal time axis, the invention discloses a horizontal error translation correction method of the wind power prediction system, It is universally applicable to various power prediction systems such as wind power and photovoltaic power generation

Images