By Topic

Short-term wind power prediction based on wavelet transform-support vector machine and statistic characteristics analysis

Sign In

Cookies must be enabled to login.After enabling cookies , please use refresh or reload or ctrl+f5 on the browser for the login options.

Formats Non-Member Member
$33 $13
Learn how you can qualify for the best price for this item!
Become an IEEE Member or Subscribe to
IEEE Xplore for exclusive pricing!
close button

puzzle piece

IEEE membership options for an individual and IEEE Xplore subscriptions for an organization offer the most affordable access to essential journal articles, conference papers, standards, eBooks, and eLearning courses.

Learn more about:

IEEE membership

IEEE Xplore subscriptions

4 Author(s)
Jie Shi ; North China Electric, Power University, 2 Beinong Road, Changping, Beijing, 102206, China ; Yongqian Liu ; Yongping Yang ; Wei-jen Lee

The prediction algorithm is an important key factor in wind power prediction. However, there are pros and cons on different forecasting algorithms. Based on the principles of wavelet transform (WT), support vector machine (SVM) as well as characteristics of wind turbine generation systems, two prediction methods are presented and compared in this paper. In method 1, the time series of model input are decomposed into different frequency composes and models are set up separately based on SVM. The results are combined together to obtain the final wind power output. In method 2, the wavelet kernel function is applied in place of RBF kernel function in SVM training. To supply more valuable suggestions, the means of evaluating prediction algorithm precision is proposed. The operation data from two wind farms both in North China and U.S.A are used to test the usability of the method. The mean relative error of WT-SVM model (method 1) is less than that of traditional SVM model.

Published in:

2011 IEEE Industrial and Commercial Power Systems Technical Conference

Date of Conference:

1-5 May 2011