Neural network based study of PV panel performance in the presence of dust
by Moh'd Sami Ashhab; Omar Akash
International Journal of Embedded Systems (IJES), Vol. 11, No. 1, 2019

Abstract: Neural networks are utilised to study PV solar panels' performance dependence on accumulated dust and tilt angle. For every location in the world there is an optimum tilt angle of dust-free PV panels for best performance. However, in some areas of the world dust is an obstacle for PV panels operation. Examples of such areas are the Middle East and the Arabian Gulf countries. Accumulation rate of dust depends on the location as well as the tilt angle of the PV panel. In this paper, relevant data are collected for two identical sets of PV panels in the city of Ras Al Khaimah in UAE where significant amounts of dust are present. Each set includes four different PV panels with various tilt angles. One set is kept clean while the other set is left unclean. The data includes information about PV panel cleaning condition, tilt angle, date and output power. Available data are used to train a neural network model which predicts the PV panel power under the given conditions. Furthermore, the neural network model serves as a key for adjusting the tilt angle of the PV panels at an optimum value for best performance taking into account the dust factor.

Online publication date: Tue, 29-Jan-2019

The full text of this article is only available to individual subscribers or to users at subscribing institutions.

 
Existing subscribers:
Go to Inderscience Online Journals to access the Full Text of this article.

Pay per view:
If you are not a subscriber and you just want to read the full contents of this article, buy online access here.

Complimentary Subscribers, Editors or Members of the Editorial Board of the International Journal of Embedded Systems (IJES):
Login with your Inderscience username and password:

Username:        Password:          Forgotten your password?

Want to subscribe?
A subscription gives you complete access to all articles in the current issue, as well as to all articles in the previous three years (where applicable). See our Orders page to subscribe.

If you still need assistance, please email subs@inderscience.com