Identification of bad solar power generation data Hatali Güneş Enerjisi Üretim Ölçümlerinin Tespit Edilmesi

DUYMAZ, Erencan
Göl, Murat
In this paper, a novel method has been presented for the had data identification of the data recovery process from a solar power measurement data set. Due to failures in solar power measurement logging devices, a long period of solar power generation data might necessary to be recovered. In order to ensure that the recovered data is the correct one, the daily solar generation curve has to be investigated. Due to the high data amount, this control process has to be done automatically. It is observed that the daily solar generation curve is much similar to cosine wave. In this study, the recovered data is compared to reference cosine wave and it is aimed to identify the suspicious data group. In this context, this work is based on the normalized residuals test.