Date

2018-7

Author

Kasht, Hana

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

142
views

110
downloads

Cite This

Sky conditions classification is important for solar radiation predictions and solar energy systems design. Sky conditions are generally classified into clear, partly cloudy, and overcast sky using clearness index. However, there is no clear-cut clearness index values to define sky conditions, and there is a need for robust methods for reliable sky conditions classification. This work presents sky conditions classification for Kalkanli region using Support Vector Machines. Global horizontal solar irradiation data are used to obtain daily clearness index. The global horizontal solar irradiation data are obtained from TMY2 dataset. The daily clearness index values are then used to classify sky conditions by utilizing Support Vector Machines Classifier. Although many data analysis techniques have been employed for sky conditions classification, this is the first time Support Vector Machines is used for this purpose. After determination of clear sky days based on the classification results, a simple correlation model is proposed to estimate the hourly global solar irradiation from hourly temperature. The proposed model for estimation of global solar irradiation is an improved version of a commonly used model. Using statistical indicators results, the proposed model gives better results for hourly global solar irradiation estimations, in comparison to the commonly used model.

Subject Keywords

Solar Radiation, Clearness Index, Data Analysis, Support Vector Machines, Sky Conditions

URI

https://hdl.handle.net/11511/69807

Collections

Northern Cyprus Campus, Thesis