Research on the mapping model for provincial wind power prediction

Author

Chan, Zhibao ; Zhou, Hai ; Ding, Jie

Author_Institution

Comput. Math., Electr. Power Res. Inst. of China, Nanjing, China

fYear

2012

fDate

21-24 May 2012

Firstpage

1

Lastpage

3

Abstract

This paper presents three methods for the mapping model for provincial wind power prediction. After correlation analysis of the historical data, several wind farms´ output power are found to be principal related to the global provincial wind power. For the first method, curve fitting and weighted average values are used to establish the mapping model. The second method is based on multiple linear regression. The third method depends on radial base function networks to search the unpredictable mapping. Finally, these methods are tested by forecasting horizon of 24h ahead and compared with their performance, which shows the validity of them.

Keywords

correlation methods; curve fitting; power system simulation; radial basis function networks; regression analysis; wind power plants; curve fitting; global provincial wind power; historical data correlation analysis; mapping model; multiple linear regression; provincial wind power prediction; radial base function networks; time 24 h; weighted average values; wind farms; Equations; Forecasting; Mathematical model; Predictive models; Radial basis function networks; Wind farms; Wind power generation; Wind power prediction; curve fitting; mapping model; multiple linear regression; radial base function networks;

fLanguage

English

Publisher

ieee

Conference_Titel

Innovative Smart Grid Technologies - Asia (ISGT Asia), 2012 IEEE

Conference_Location

Tianjin

Print_ISBN

978-1-4673-1221-9

Type

conf

DOI

10.1109/ISGT-Asia.2012.6303283

Filename

6303283