SOLAR RADIATION FORECAST by USING MACHINE LEARNING METHOD for GAZIANTEP PROVINCE

Abstract

Renewable energy sources have become a popular topic all over the world in terms of cost, efficiency and environmental pollution. Solar energy is the most significant of the renewable energy sources. Solar energy, which was used only as heat and light energy in the past, is widely used in electrical energy production with the advancement of today's technology. Traditionally used photovoltaic cells are semiconductor materials that are produced in various chemical structures and convert the energy they receive from sunlight directly into electrical energy. The research and development of photovoltaic cells is moving forward at an accelerating pace. With this development process and relying on the today's technology, it is aimed to increase the efficiency of photovoltaic cells and to produce more electrical energy as a result of various trials. By analysing the energy production of photovoltaic cells, efficiency-enhancing situations are examined according to solar radiation values. In this study, a model was constructed using the regression approach, which is a method of machine learning. This model has been developed using the MATLAB program of the meteorological data of 2021 from Gaziantep. In addition, a variety of error analysis tests were utilized in order to evaluate the effectiveness of the model that was built. As a consequence, the model created using the linear regression method yields successful results in estimating solar radiation in Gaziantep province. This is demonstrated by the coefficient of determination (R2) value of 0.98, the Mean Absolute Error (MAE) value of 0.023, the Root Mean Square Error (RMSE) value of 0.028, and the Mean Square Error (MSE) value of 0.0008.

Keywords

• Solar energy
• Energy production
• Photovoltaic cell
• Cell efficiency
• Machine learning

References

1. [1] Sayin, S., Koc, İ., (2011), Photovoltaic (PV) Systems used in active utilization of solar energy and usage methods in building, Selcuk University Journal of Engineering, Science and Technology, 26, 89-106.
2. [2] Sahan, M., Sahan, H., and Yegingil, I., (2014), Measuring annual total and ultraviolet (UV) solar energy data, Journal of Suleyman Demirel University Graduate School of Natural and Applied Sciences, 14, 10–16.
3. [3] (2022), Solar Energy Potential Atlas. [Online]. Available: https://gepa.enerji.gov.tr/Mycalculator/Default.Aspx
4. [4] Canka Kilic, F., (2015), Latest status in Turkey and production technologies, Engineer and Machine, 56, 28-40.
5. [5] Kaplan, A.G., (2022), A new approach based on moving least square method for calculating the Weibull coefficients, Environmental Progress & Sustainable Energy, 41(4), e13934.
6. [6] Atique, S., Noureen, S., Roy, V., and Macfie, J., (2020), Time series forecasting of total daily solar energy generation: A comparative analysis between ARIMA and machine learning techniques, 2020 IEEE Green Technologies Conference, 1-3 April 2020, 175-180.
7. [7] Aksoy, B., and Selbas, R., (2019), Estimation of wind turbine energy production value by using machine learning algorithms and development of implementation program, Energy Sources Part A: Recovery, Utilization and Environmental Effects, 43, 692-704.
8. [8] Abdelhafidi, N., Bachari, N.I., and Abdelhafidi, Z., (2021), Estimation of solar radiation using stepwise multiple linear regression with principal component analysis in Algeria, Meteorology and Atmospheric Physics, 133, 205-216.

9. [9] Nath, N.C., Sae-Tang, W., and Pirak, C., (2020), Machine learning-based solar power energy forecasting, Journal of the Society of Automotive Engineers Malaysia, 4, 307-322.
10. [10] [Besharat, F., Dehghan, A.A., and Faghih, A.R., (2013), Empirical models for estimating global solar radiation: A review and case study, Renewable and Sustainable Energy Reviews, 21, 798-821.
11. [11] Kaplan, Y.A., Sarac, M.S., and Unaldi, G.G., (2022), Developing a new model in solar radiation estimation with genetic algorithm method, Environmental Progress & Sustainable Energy.
12. [12] Sahan, M., (2021), Global solar radiation forecast for Gaziantep, Antakya and Kahramanmaras using artificial neural networks and Angström-Prescott equations, Suleyman Demirel University Faculty of Arts and Science Journal of Science, 16, 368-384.
13. [13] Roger, A., and Ventre, M.J., (2003), Photovoltaic Systems Engineering, 2nd ed., CRC Press.
14. [14] Killada, P., (2017), Data analytics using regression models for health insurance market place data, MSc. Thesis in Engineering: Computer Science, The University of Toledo.