Title :
Data mining for prediction of wind farm power ramp rates
Author :
Kusiak, Andrew ; Zheng, Haiyang
Author_Institution :
Dept. of Mech. & Ind. Eng., Univ. of Iowa, Iowa City, IA
Abstract :
In this paper, multivariate time series models are built to predict the power ramp rate of a wind farm. The power changes are predicted at ten-minute intervals. Multivariate time series models are built with data-mining algorithms. Five different data-mining algorithms are tested using data collected at a wind farm. The support vector machine regression algorithm performed best of the five algorithms studied in this research. It provided predictions of the power ramp rate for a time horizon of 10 to 60 minutes. The boosting tree algorithm selected predictors enhancing the prediction accuracy. The data used in this research originated at a wind farm of over 100 turbines. The test results of various models are presented in the paper. Suggestions for future research are provided.
Keywords :
data mining; power engineering computing; regression analysis; support vector machines; time series; wind power plants; data mining; multivariate time series models; prediction accuracy; support vector machine regression algorithm; ten-minute intervals; time 10 min to 60 min; wind farm power ramp rates; Data mining; Power generation; Power system modeling; Predictive models; Weather forecasting; Wind energy generation; Wind farms; Wind forecasting; Wind power generation; Wind speed; Power ramp rate prediction; data-mining algorithms; multivariate time series model; parameter selection; wind farm;
Conference_Titel :
Sustainable Energy Technologies, 2008. ICSET 2008. IEEE International Conference on
Conference_Location :
Singapore
Print_ISBN :
978-1-4244-1887-9
Electronic_ISBN :
978-1-4244-1888-6
DOI :
10.1109/ICSET.2008.4747170
