Reliable Photovoltaics Output Power Prediction in Qatar
Author | Lari, Ali Jassim |
Author | Egwebe, Augustine |
Author | Touati, Farid |
Author | Gonzales, Antonio S. |
Author | Khandakar, Amith Abdullah |
Available date | 2024-07-24T10:13:53Z |
Publication Date | 2021 |
Publication Name | Proceedings - International Conference on Developments in eSystems Engineering, DeSE |
Resource | Scopus |
ISSN | 21611343 |
Abstract | Renewable energy is gradually becoming the most promising type of power generation that could replace fossil fuels in the future. One of the most widely used form of renewable energy is solar/PV energy. To examine the impacts of different climatic circumstances and maintain solar power converters' optimal performance while meeting peak demand via diverse environmental conditions, accurate PV generating power prediction models are required. Air temperature, relative humidity, Photovoltaics (PV) surface temperature, irradiance, dust, wind speed, and output power are among the environmental parameters examined and addressed in this study. The model suggested in this study optimises and trains three prediction algorithms: the Artificial Neural Network (ANN), the Multi-Variate (MV), and the Support Vector Machine (SVM). To choose the best PV generating power forecast, the model uses three well-known prediction algorithms plus a voting method. Furthermore, given the environmental circumstances, the voting system predicts the output power with great accuracy. The MSE for Artificial Neural Network (ANN), Multi-variate (MV), and Support Vector Machine (SVM) is 98, 81, and 82, respectively. In comparison, the voting algorithm's Mean Squared Error (MSE) is only slightly higher than 53. With respect to the environmental circumstances in Qatar, the suggested PV power generation forecast algorithm produces trustworthy results. The suggested voting algorithm is anticipated to aid in the design process of photovoltaic (PV) facilities when energy output is very predictable. |
Language | en |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
Subject | Artificial Neural Network Eenvironment Machine Learning PV Qatar |
Type | Conference |
Pagination | 546-551 |
Volume Number | 2021-December |
Files in this item
Files | Size | Format | View |
---|---|---|---|
There are no files associated with this item. |
This item appears in the following Collection(s)
-
Electrical Engineering [2685 items ]