Solar Power Prediction with an Hour-based Ensemble Machine Learning Method

Abstract

I n recent years, the share of solar power in total energy production has gained a rapid increase. Therefore, prediction of solar power production has become increasingly important for energy regulations. In this study we proposed an ensemble method which gives competitive prediction performance for solar power production. This method firstly decomposes the nonlinear power production data into components with a multi-scale decomposition technique such as Empirical Mode Decomposition EMD . These components are then enriched with the explanatory exogenous feature set. Finally, each component is separately modeled by nonlinear machine learning methods and their results are aggregated as final prediction. We use two different training approaches such as Hour-based and Day-based, for predicting the power production at each hour in a day. Experimental results show that our ensemble method with Hour-based approach outperform the examined machine learning methods

Keywords

• Solar power,Time series forecasting,Machine learning,Ensemble methods,Empirical mode decomposition

References

1. Khashei M, Bijari M. A Novel Hybridization of Artificial Neural Networksand ARIMA Models for Time Series Forecasting. Appl. Soft Comput., 11, 2664–2675, 2011, doi:10.1016/j. asoc.2010.10.015.
2. Buyuksahin UC, Ertekin S. A feature-based hybrid ARIMA- ANN model for univariate time series forecasting. Journal of the Faculty of Engineering and Architecture of Gazi University 35:1 (2020) 467-478 doi.org/10.17341/ gazimmfd.508394.
3. Huang J, Korolkiewicz M, Agrawal M, Boland J. Forecasting solar radiation on an hourly time scale using a coupled autoregressive and dynamical system (cards) model. Solar Energy; 87, 136-149, Jan. 2013, doi.org/10.1016/j. solener.2012.10.012.
4. Glasbey CA, Allcroft DJ. A spatiotemporal auto-regressive moving average model for solar radiation. Appl Stat, 57, 343-355, 2007.
5. Mellit, A. Artificial Intelligence Technique for Modelling and Forecasting of Solar Radiation Data: A Review. Int. J. Artif. Intell. Soft Comput., 1, 52–76, 2008.
6. Martin L, Zarzalejo LF, Polo J, Navarro A, Marchante, R, and Cony M. Prediction of Global Solar Irradiance Based on Time Series Analysis: Application to Solar Thermal Power Plants Energy Production. Solar Energy, 84, 1772-1781, Oct. 2010, doi.org/10.1016/j.solener.2010.07.002.
7. Mohamed A, Chowdhury C. Solar Power Forecasting Using Artificial Neural Networks. In 2015 North American Power Symposium (NAPS), 1–5, 2015. Hamid E, Himdi K. Artificial Neural Network for Forecasting One Day Ahead of Global Solar Irradiance. SSRN Scholarly Paper. Rochester, NY: Social Science Research Network, May 29, 2018.
8. Daniel O, Kubby J. Feature Selection and ANN Solar Power Prediction. Research Article. Journal of Renewable Energy, 2017. https://doi.org/10.1155/2017/2437387.

9. Xu X, Qi Y, Hua Z. Forecasting demand of commodities after natural Disasters. Expert Systems with Applications, Volume 37, Issue 6, June 2010, Pages 4313-4317, doi. org/10.1016/j.eswa.2009.11.069.
10. Yu W, Mu-Chen C. Forecasting the short-term metro passenger flow with empirical mode decomposition and neural networks, Transportation Research Part C, Volume 21, Issue 1, April 2012, Pages 148-162, https://doi. org/10.1016/j.trc.2011.06.009.
11. Huang NE, Shen Z, Long SR, Wu MC, Shih HH, Zheng Q, Yen NC, Tung CC, Liu HH. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proceedings of the Royal Society of London A: Mathematical, Physical and Engineering Sciences 454 (1998) 903-995. doi:10.1098/rspa.1998.0193.
12. Barnhart BL, Eichinger WE. Empirical mode decomposition applied to solar irradiance, global temperature,sun spot number and co2 concentration data. J Atmos Solar Terr Phys 2011; 73:1771.
13. Majumder I, Behera MK, Nayak N. Solar Power Forecasting Using a Hybrid EMD-ELM Method. International Conference on Circuit, Power and Computing Technologies (ICCPCT), 20-21 April 2017, doi:10.1109/ICCPCT.2017.8074179.
14. Monjoly S, Andre M, Calif R, Soubdhan T. Hourly forecasting of global solar radiation based on multiscale decomposition methods: A hybrid approach. Energy, Volume 119, 15 January 2017, Pages 288-298, doi.org/10.1016/j. energy.2016.11.061.
15. Calif R, Schimtt FG, Huang Y, Soubdhan T. Intermittency study of high frequancy global solar raiation sequences under a tropical climate. Solar Energy, 98, 349-365, 2013.
16. Huang NE, Wu ML, Qu W, Long SR, Shen SP. Applications of Hilbert Huang transform to non-stationary nancial time series analysis. Applied Stochastic Models in Business and Industry, 19(3), (2003), 245-268.
17. Angela Z, Faltermeier R, Keck I, Tomé A, Puntonet C, Lang E. Empirical Mode Decomposition - an Introduction. Proceedings of the International Joint Conference on Neural Networks, 1–8, 2010.
18. Kutlu, C, Li J, Su Y, Wang Y, Pei G, Riffat S. Annual Performance Simulation of a Solar Cogeneration Plant with Sensible Heat Storage to Provide Electricity Demand for a Small Community: A Transient Model. Hittite Journal of Science & Engineering, 6(1), (2010), 75-81.
19. Epias, 2018, Gerçek Zamanlı Üretim, retrieved October 26, 2018, from: https://seffaflik.epias.com.tr/transparency/ uretim/gerceklesen-uretim/gercek-zamanli-uretim.xhtml