Prediction of solar direct irradiance in Iraq by using artificial neural network
Global solar irradiance is one of the main significant factors for designing and considering the volume of any solar station beside of it is usage in agricultural and building issue. Due of lack a precise information about the irradiance in Iraq metrological organization and seismology, this study i...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Salahaddin University-Erbil
2021-10-01
|
| Series: | Zanco Journal of Pure and Applied Sciences |
| Subjects: | |
| Online Access: | https://zancojournals.su.edu.krd/index.php/JPAS/article/view/4058 |
| _version_ | 1819227552689946624 |
|---|---|
| author | zana Saleem Gzing Adil Mohammed |
| author_facet | zana Saleem Gzing Adil Mohammed |
| author_sort | zana Saleem |
| collection | DOAJ |
| description | Global solar irradiance is one of the main significant factors for designing and considering the volume of any solar station beside of it is usage in agricultural and building issue. Due of lack a precise information about the irradiance in Iraq metrological organization and seismology, this study is aimed to adopt the historical global data, build numerical analysis via using artificial neural network and predicting hourly irradiance. The test is applied over three locations Erbil, Bagdad, and Basra for being references to their closest locations. A foreword neural network (FNN) is the learning algorithm that is used in this study with relying on seven input variables consisting of Temperature, Precipitation, Humidity, Wind speed, Wind direction Sunshine duration and Date. After normalizing and standardizing data, an iteration method is used for determining the optimum number of neuron(s) in a hidden layer. It yields a least Root Mean square error (RMSE) between 2.5 to 3. The computed correlation coefficients are between 0.94 -0.96 for the mentioned locations. |
| first_indexed | 2024-12-23T10:43:12Z |
| format | Article |
| id | doaj.art-420a10300bb247ca875e65ba2ca71a0b |
| institution | Directory Open Access Journal |
| issn | 2218-0230 2412-3986 |
| language | English |
| last_indexed | 2024-12-23T10:43:12Z |
| publishDate | 2021-10-01 |
| publisher | Salahaddin University-Erbil |
| record_format | Article |
| series | Zanco Journal of Pure and Applied Sciences |
| spelling | doaj.art-420a10300bb247ca875e65ba2ca71a0b2022-12-21T17:50:07ZengSalahaddin University-ErbilZanco Journal of Pure and Applied Sciences2218-02302412-39862021-10-01335435010.21271/ZJPAS.33.5.5Prediction of solar direct irradiance in Iraq by using artificial neural networkzana Saleem0Gzing Adil Mohammed1Lafargeholcim, Kurdistan region of Iraq, Sulaymaniyah.Department of Oil, Gas and Energy administration, Public Administration and Natural Resources, Charmo University.Global solar irradiance is one of the main significant factors for designing and considering the volume of any solar station beside of it is usage in agricultural and building issue. Due of lack a precise information about the irradiance in Iraq metrological organization and seismology, this study is aimed to adopt the historical global data, build numerical analysis via using artificial neural network and predicting hourly irradiance. The test is applied over three locations Erbil, Bagdad, and Basra for being references to their closest locations. A foreword neural network (FNN) is the learning algorithm that is used in this study with relying on seven input variables consisting of Temperature, Precipitation, Humidity, Wind speed, Wind direction Sunshine duration and Date. After normalizing and standardizing data, an iteration method is used for determining the optimum number of neuron(s) in a hidden layer. It yields a least Root Mean square error (RMSE) between 2.5 to 3. The computed correlation coefficients are between 0.94 -0.96 for the mentioned locations.https://zancojournals.su.edu.krd/index.php/JPAS/article/view/4058renewable energysolar systemartificial neural networkprediction. |
| spellingShingle | zana Saleem Gzing Adil Mohammed Prediction of solar direct irradiance in Iraq by using artificial neural network Zanco Journal of Pure and Applied Sciences renewable energy solar system artificial neural network prediction. |
| title | Prediction of solar direct irradiance in Iraq by using artificial neural network |
| title_full | Prediction of solar direct irradiance in Iraq by using artificial neural network |
| title_fullStr | Prediction of solar direct irradiance in Iraq by using artificial neural network |
| title_full_unstemmed | Prediction of solar direct irradiance in Iraq by using artificial neural network |
| title_short | Prediction of solar direct irradiance in Iraq by using artificial neural network |
| title_sort | prediction of solar direct irradiance in iraq by using artificial neural network |
| topic | renewable energy solar system artificial neural network prediction. |
| url | https://zancojournals.su.edu.krd/index.php/JPAS/article/view/4058 |
| work_keys_str_mv | AT zanasaleem predictionofsolardirectirradianceiniraqbyusingartificialneuralnetwork AT gzingadilmohammed predictionofsolardirectirradianceiniraqbyusingartificialneuralnetwork |