Analysis and Forecasting of Solar Power Generation using Machine Learning

Todizara Andrianajaiana; Tsivalalaina DavidRazafimahefa; Haba Cristian-Gyozo; Dorin DumitruLucache

doi:10.59429/ifr.v1i1.128

Todizara Andrianajaiana Doctoral school EDT-ENRE, University of Antsiranana, Antsiranana, Madagascar
Tsivalalaina DavidRazafimahefa Doctoral school EDT-ENRE, University of Antsiranana, Antsiranana, Madagascar
Haba Cristian-Gyozo Faculty of Electrical Engineering, Technical University of Iasi, Iasi, Romania
Dorin DumitruLucache Faculty of Electrical Engineering, Technical University of Iasi, Iasi, Romania

DOI: https://doi.org/10.59429/ifr.v1i1.128

Keywords: Power generation, Artificial inteligence, Fore-casting, Solar power generation, Data processing

Abstract

Current photovoltaic systems are equipped with a monitoring system. The data is recorded with a predefined time base. The operator has a large amount of data. This article is a contribution to the analysis and understanding of this data. It proposes to estimate Generation from this data using machine learning. The study was conducted at the 1Mw generation site in Miroslava, Iasi, Romania. The visualization and interpretation of the generation data is presented in this article. Then, prediction techniques are presented to obtain an estimate of the plant’s generation. These techniques are: Simple Exponential Smoothing (SES), Autoregression (AR), Seasonal Auto-Regressive Integrated Moving Average with eXogenous factors (SARIMA), Holt Winter’s Exponential Smoothing (HWES), Long Short Term Memory neural (LSTM) and Convolutional Neural Network (CNN).

References

1. A. Khatibi, M. H. Jahangir, F. Razi Astaraei, and F. Mohabbati, Predicting the renewable energy consumption in 2026 by using a recursive moving average model, International Journal of Ambient Energy,pp. 1–8, 2022.

2. A. Tomar et al.,Machine learning, advances in computing, renewable energy and communication, 2022.

3. J. Quinonero-Candela, M. Sugiyama, A. Schwaighofer, and N. D. Lawrence, Dataset shift in machine learning.Mit Press, 2022.

4. L.-V. Vlǎducu and C.-F. Lǎceanu, The impact of renewable energy production on greenhouse gas emissions in romania, Gas,vol. 21, p. 54, 2022.

5. R. Prǎvǎlie, I. S ırodoev, J. Ruiz-Arias, and M. Dumitras¸cu, Using renewable (solar) energy as a sustainable management pathway of lands highly sensitive to degradation in romania. a countrywide analysis based on exploring the geographical and technical solar potentials, Renewable Energy,2022.

6. C. Donnees Mondiales, Roumanie: Données et statistiques du pays, 2022. [Online]. Available: https://www.donneesmondiales.com/europe/roumanie/index.php

7. T. Transelectrica, https://www.transelectrica.ro/ro/web/tel/home,2022.[Online].Available: https://www.transelectrica.ro/ro/web/tel/home

8. A. Ibrahim, R. Kashef, and L. Corrigan, Predicting market movement direction for bitcoin: A comparison of time series modeling methods, Computers & Electrical Engineering,vol. 89, p. 106905, 2021.

9. Z. Cui, J. Wu, Z. Ding, Q. Duan, W. Lian, Y. Yang, and T. Cao, A hybrid rolling grey framework for short time series modelling, Neural Computing and Applications,vol. 33, no. 17, pp. 11 339–11 353, 2021.

10. G. Millán, R. Osorio-Comparán, and G. Lefranc, Preliminaries on the accurate estimation of the hurst exponent using time series, in 2021 IEEE International Conference on Automation/XXIV Congress of the Chilean Association of Automatic Control (ICA-ACCA).IEEE, 2021, pp. 1–8.

11. C. Janiesch, P. Zschech, and K. Heinrich, Machine learning and deep learning, Electronic Markets,vol. 31, no. 3, pp. 685–695, 2021.

12. M. Ramesh, C. Mani, B. Reddy, and M. Venkataramanaiah, Forecasting of bse sensex using simple exponential smoothing (ses) method, ACA-DEMICIA: An International Multidisciplinary Research Journal,vol. 11, no. 3, pp. 656–665, 2021.

13. M. H. P. Swari, I. P. S. Handika, and I. K. S. Satwika, Comparison of simple moving average, single and modified single exponential smoothing, in 2021 IEEE 7th Information Technology International Seminar (ITIS).IEEE, 2021, pp. 1–5.

14. G. Zhu, L. Li, Y. Zheng, X. Zhang, and H. Zou, Forecasting influenza based on autoregressive moving average and holt-winters exponential smoothing models, Journal of Advanced Computational Intelligence and Intelligent Informatics,vol. 25, no. 1, pp. 138–144, 2021.

15. M. Pleños, Time series forecasting using holt-winters exponential smoothing: Application to abaca fiber data, Zeszyty Naukowe SGGW w Warszawie-Problemy Rolnictwa Swiatowego´ ,vol. 22, no. 2, pp. 17–29, 2022.

16. S. Karlsson, S. Mazur, and H. Nguyen, Vector autoregression models with skewness and heavy tails, arXiv preprint arXiv:2105.11182,2021.

17. F. R. Alharbi and D. Csala, “A seasonal autoregressive integrated moving average with exogenous factors (sarimax) forecasting model-based time series approach, Inventions,vol. 7, no. 4, p. 94, 2022.

18. R. Huang, C. Wei, B. Wang, J. Yang, X. Xu, S. Wu, and S. Huang, Well performance prediction based on long short-term memory (lstm) neural network, Journal of Petroleum Science and Engineering,vol. 208, p. 109686, 2022.

19. J.-Y. Lim, S. Kim, H.-K. Kim, and Y.-K. Kim, “Long short-term memory (lstm)-based wind speed prediction during a typhoon for bridge trafficcontrol, Journal of Wind Engineering and Industrial Aerodynamics,vol. 220, p. 104788, 2022.

20. G. Alotaibi, M. Awawdeh, F. F. Farook, M. Aljohani, R. M. Aldhafiri, and M. Aldhoayan, Artificial intelligence (ai) diagnostic tools: utilizing aconvolutional neural network (cnn) to assess periodontal bone level radiographically—a retrospective study, BMC Oral Health,vol. 22, no. 1, pp. 1–7, 2022.

21. M. Zak, A. Haliuc, S. Cheval, B. Antonescu, A. Tis¸covschi, M. Dobre, F. Tatui, A. Dumitrescu, A. Manea, G. Tudorache et al., Meteorological information at the end of 19 th century from romanian newspapers-an intro to the database,” in AGU Fall Meeting Abstracts,vol. 2019, 2019, pp. PP43D–1622.

22. A.-I. Albu, G. Czibula, A. Mihai, I. G. Czibula, S. Burcea, and A. Mezghani, Nextnow: A convolutional deep learning model for the prediction of weather radar data for nowcasting purposes, Remote Sensing,vol. 14, no. 16, p. 3890, 2022.

23. R. Bosneagu, C. E. Lupu, E. Torica, S. Lupu, N. Vatu, V. M. Tanase, C. Vasilache, D. Daneci-Patrau, and I. C. Scurtu, Long-term analysis of air temperatures variability and trends on the romanian black sea coast, Acta Geophysica,vol. 70, no. 5, pp. 2179–2197, 2022.

24. T. Chai and R. R. Draxler, Root mean square error (rmse) or mean absolute error (mae)?–arguments against avoiding rmse in the literature, Geoscientific model development,vol. 7, no. 3, pp. 1247–1250, 2014.

25. M. Elsaraiti and A. Merabet, “Solar power forecasting using deep learning techniques,” IEEE Access,vol. 10, pp. 31 692–31 698, 2022.

26. D. Chicco, M. J. Warrens, and G. Jurman, The coefficient of determination r-squared is more informative than smape, mae, mape, mse and rmse in regression analysis evaluation,” PeerJ Computer Science,vol. 7, p. e623, 2021.

27. D. S. K. Karunasingha, Root mean square error or mean absolute error? use their ratio as well, Information Sciences,vol. 585, pp. 609–629, 2022.

28. C.-C. Wang, H.-T. Chang, and C.-H. Chien, Hybrid lstm-arma demand-forecasting model based on error compensation for integrated circuit tray manufacturing,” Mathematics,vol. 10, no. 13, p. 2158, 2022.

29. A. Rodríguez Sánchez, R. Salmerón Gómez, and C. García, The coefficient of determination in the ridge regression, Communications in statistics-simulation and computation,vol. 51, no. 1, pp. 201–219, 2022.