Hourly Solar Radiation Forecast Based on k-NN Nonparametric Regression Model

Article Preview

Abstract:

It is hard to accurately predict solar radiation due to its very prominent nonlinear feature, so four major models based on the k-NN model, which have time, sunshine hours, temperature, wind speed and humidity as input data, are presented in the paper for hourly solar radiation forecasting. The models forecasting instantaneous solar radiation values in the Xining region have been created through experiments carried out in Qinghai University. The accuracy of the optimum model is up to 88.7% on average and 97.01% at most on the premise and the allowed absolute percentage error is lower than 20%.

You might also be interested in these eBooks

Info:

Periodical:

Pages:

217-222

Citation:

Online since:

June 2014

Export:

Price:

Permissions CCC:

Permissions PLS:

Сopyright:

© 2014 Trans Tech Publications Ltd. All Rights Reserved

Share:

Citation:

* - Corresponding Author

[1] Rongqiang Cui. Chinese Journal of Nature [J]. 2010, 32(3): 149-155. In Chinese.

Google Scholar

[2] Christopher Stewart, Kai Shen. Some Joules Are More Precious Than Others: Managing Renewable Energy in the Datacenter. In Proceedings of the Workshop on Power Aware Computing and Systems, (2009).

Google Scholar

[3] Investment Advisor. In 2013-2017 Photovoltaic Industry Investment Analysis and Forecast Report of Qinghai Province . 2012. 9. In Chinese.

Google Scholar

[4] Liangliang Cheng, Hang gGuo, Xiaoyan Guan and so on . BP model in deformation monitoring data analysis and prediction [J]. Science of Surveying and Mapping, 2005, 33(4): 124-226. In Chinese.

Google Scholar

[5] An Cheng, Ning Cheng Xianglong Zhou. Data mining technology and application [M] Beijing: Science Press, 2006: 161-163. In Chinese.

Google Scholar

[6] Danzhi Zhou . Kernel Estimation of Regression Function and Its Application in the economic analysis [J]. Journal of Zhejiang Wanli University, 2002(1): 78-80.In Chinese.

Google Scholar

[7] Tom M. Mitchell. Machine Learning [M]. Beijing: Machinery Industry Press, 2003: 166-167. In Chinese.

Google Scholar