Κατεύθυνση Φυσική Περιβάλλοντος (ΕΦΑΡΜΟΣΜΕΝΗ ΦΥΣΙΚΗ)
Library of the School of Science

2012-03-14

2012

Γεωργίου Αναστάσης

Ιακωβίδης Κωνσταντίνος Αναπλ. Καθηγ.

Μοντελοποίηση ημερήσιων τιμών συνιστωσών ολικής ολοφασματικής, φωτοσυνθετικής και υπεριώδης με την τεχνική των νευρωνικών δικτύων

Greek

The object of this paper is to investigate and modeling of solar radiation for
the region of Cyprus. The radiation is studied solar, total photosynthetic and
UV-total radiation. To study the parameters of solar radiation, the data used
provided by the Meteorological Service of Cyprus. The database refers to the
years 2004 to 2006. Trained artificial neural networks (ANN) for the
calculation of the investigated parameters of the radiation. The calculation of
three parameters using neural networks, was done by using some other parameters
for training the models. These are temperature (T), relative humidity (r),
external atmospheric radiation (Eoh), precipitation water (w), clearness index
Kt and sunshine (n - N) .For models were the statistical tests MBE and RMSE,
which combined with the yield R2 gave the most accurate model. This technique
has proven effective in measuring the three parameters of solar radiation. The
results are compared and discussed. Also carried out empirical models to be
compared with a method of ANN models to find what is most accurate for each
radiation.

Global, Photosynthetical, UV-total, ANN models, Radiation

No

0

Yes

72

183

File access is restricted only to the intranet of UoA.

https://pergamos.lib.uoa.gr/uoa/dl/object/1319071