Improvement of the quality of short-term forecast of the solar potential using hybrid model: application to Dakar region
Keywords:
Solar, Intermittent, Segmentation, Prediction, Hybrid, EM, Viterbi.Abstract
This study addresses the critical need to enhance short-term forecasting of solar potential for photovoltaic
power generation in Senegal, focusing on the Dakar region. A hybrid forecasting model is developed by integrating
a Gaussian Mixture Model (GMM), a Hidden Markov Model (HMM), and a Kalman Filter (KF) to characterize the
variability of solar radiation influenced by meteorological factors. The GMM identifies distinct fluctuation states of
solar radiation using the EM algorithm, while the HMM captures temporal dependencies between these states
through Forward Backward and Viterbi algorithms. Subsequently, a Kalman Filter refines state-based predictions
within a 20-minute horizon. Comparative results demonstrate that the hybrid model [NRMSE=0.0093,
NMAE=0.044 and NMBE=0.0006] significantly outperforms the standalone Kalman Filter [NRMSE =0.048,
MBE=2.718 and NMAE=0.004] 0.004], achieving lower normalized root mean square errors and bias metrics. This
innovative approach offers a robust probabilistic and deterministic framework to improve intermittent solar energy
forecasting, supporting optimized energy planning in the Sahelian zone.
References
P.Lauret, M.H.Diagne, M.David, A.Rodler, M.Muselli,C.Voyant, «A Bayesian model committee approach toforecastingglobal solar radiation», In: World renewable energy forum,WREF2012, including world renewable energy andColorado renewable energy society(CRES) annual conference, 4354–4359, (2012).
J.M.Vindel and J.Polo,«Markov processes and Zipf’s law in daily solar
irradiationat earth’s surface », Journal of Atmospheric and Solar Terrestrial Physics, 107, pp.42-47, (2014).
D.W.Vander Meer, « comment on verification of deterministic solar forecasts: Verification of probabilistic solar forecasts», Sol Energy 2020.http:// dx.doi. Org/10.1016/j.solener.2020.04.015.
P.Lauret and al, «Probabilistic Solar Forecasting Using Quantile Regression Models Forecasting», Energies 2017, 10, 1591; doi: 10.3390/en 10101591.
T. Hong and S. Fan, «Probabilistic electric load forecasting: a tutorial
review », International Journal of Fore casting, (2016); doi.org/10.1016/j.ij forecast.2015.11.011.
D.W.Vander Merr, J.Widen and J.Munkammar «Review on probabilistic forecasting of photovoltaique power production and electricity consumption
Renewable and Sustainable Energy Reviews, Volume81, Part1, 2018, PP-1484-1512, ISSN1364-0321, (2018); doi.org/10.1016/j.rser.2017.05.212.
G.P.Zhang and al, «Time Series Forecasting using a hybrid ARIMA and neural network model», Neurocomputing, Volume 50, 2003, Pages 159-175, ISSN09252312, https://doi.org/10.1016/S0925-2312 (01)00702-0.
C.Voyant, M.Muselli, C.Paoli and M.L.Nivet, «Numerical weather pre diction (NWP) and hybride ARMA/ANN model to predict global solar radia ion», Energy 2012; 39(1):341-341-355.
W.Ji and K.C .Chee, «Prediction of hourly solar radiation using a novel hybrid model of ARMA and TDNN», Solar Energy, 85(5), 808-817, (2011).
G.Reikard and al, «Predicting solar radiation at high resolutions:A comparison of time series forecasts», (March 2009), Solar Energy 83 (3): 342-349, DOI: 10.1016/j.solener.2008.08.007.
M.Diagne andal, «Postprocessingof solar irradiance forecasts from WRFmodelat Reunion Island », Solar Energy, Volume105, July2014,Pages99-108, https://doi.org/10.1016/j.solener.2014.03.016.
A.Mellit, M.Benghanem and M.Bendekhis, «Artificial neural network model for prediction Solar radiation data: Application for sizing stand-alone photovoltaic power system », July 2005, IEEE Xplore, Conference: Power EngineeringSocietyGeneralMeeting, (2005), DOI:10.1109/PES.2005.1489526.
S.A.P.Kani and al, «Very short-term wind speed prediction: A new artifici al neural network-Markov chain model », Energy Conversion and Management 52 (1):738-745, (2011).
F.O.Hocaoglu and al,«A novel hybrid (Mycielski Markov) model for hourly solar radiation Forecasting, Renewable Energy, Volume 108, (2017), pages 635643, https://doi.org/10.1016/j.renene.2016.08.058
Shuai Li and al, «Typical Solar radiation year construction using k-means clustering and discrete-time Markov chain», Applied Energy, Volume 205, 1 November 2017, Pages720-731.
Duda R. O., Hart P. E., «Pattern Classification and Scene Analysis », John Wiley and Sons, New York, USA, 1973.
L.R.Rabiner, «A tutorial on Hidden Markov Models and selected and selected applications in speech recognition», Proc. IEEE, vol. 77, pp.257-286, (1989).
B.O.Ngoko, H.Sugihara and T.Funaki, «Synthetic generation of high tem-poral resolution solar radiation data using Markov models », Sol. Energy 103, 160-170, (2014).
A.Mbaye, J. Ndong, M.L. NDiaye and al, «Kalman filter model as a tool for short-term forecasting of solar potential: case of the Dakar site», EDP Sci- ence, vol.57, p 2267 (2018).
A.Mbaye, M.L.Ndiaye, D.M.Ndione and al, «ARMA model for short term forecasting of solar potential: application to a horizontal surface on Dakar site», OAJ Mater Device 4 (1):1–8, (2019).
L. R. RABINER and B.H. JUANG, «An introduction to hidden Markov Models»,IEEE ASSP Magazine, p.4-16,(1986).
Forney G. D, « The Viterbi Algorithm, Proceedings of the IEEE », vol 61, no 3, pp 263-27, 1973.
Dempster A. P., Laird N. M., and Rubin D. B., « Maximum likelihood from incomplete data via the EM algorithm», Journal of the Royal Statistical Society B, 39:1-38, 1977.
N.Morgan and H. Bourlard, «Continuous Speech Recognition: An Introduction to the Hybrid HMM /Connectionist Approach», IEEE Signal Proces- sing Magazine, Vol. 12, n°3, pp. 25-42, Mai 1995.
W. PIECZYNSKI, « Chaînes de Markov triplets et segmentation des images », Chapitre4 ; PP127, (janvier 2009).
W.Pieczynski, Modèles de Markov en traitements d’images Markov models in image processing, Traitement du Signal », Vol. 20, No. 3, pp.255-278, (2003).
P.Brémaud and al, «Initiation aux Probabilités et aux chaînes de Markov », Springer, (2009).
F.C.Kaminsky, R.H.Kirchhoff, Syu C.Y. Manwell J.F. (1990b), A
comparison of alternative approaches for the synthetic generation of a wind speed time series». Wind Engineering, 9ème symposium, 1-8.
Kirchhoff R.H., Kaminsky F.C., Syu C.Y., (1988), «A Markov Chain ana lysis of wind speed at windsor, Massachusetts». Wind Engineering, 5, 9-16.
S.E.Moon, S.B.Ryoo, J.G.Know, (1994), «A Markov chain model for dai- ly precipitation in South Korea». International Journal of Meteorology, 4, 1009 -1016.
P.Sparis ,J.Antonogiannakis, D.Papadopoulos, (1995),«Markov Matrix coupled approach to wind speed and direction simulation», Wind Engineering, Vol 19(3), 121-133.
Ted Soubdhan,Joseph Ndong andal,«A robust forecasting framework based onthe Kalman filtering approach with a twofold parameter tuning procedure:Applicationto solar and photovoltaic prediction»,SolarEnergy, 2016;131:246 -259.
E.Lorenz, J.Remund, S.C.Muller, W.Traunmuller, G.Steinmaurer, D.Pozo andal, «Benchmarkingof different approaches to forecast solar irradiance»,24th European Photovoltaic Solar Energy Conference Hamburg, Germany, vol . 21, 25, (2009).
H. Jiang, Y. Dong and L. Xiao, «A multi stage intelligent approach basedon an ensemble of two-way interaction model for forecasting the global horizontal radiation of India», Energy Conversion andManagement, vol.137, p 142 (2017).
A. MBAYE, M.L. NDIAYE, J.Ndong and P.A.S.Ndiaye, «Impact of meteorological parameters on short-term forecasting: Application to the Dakar site », in Proceedings of the 2019 IEEE 2nd International Conference on Power and Energy Applications 2019 (ICPEA 2019), Singapore, April 2019.
Downloads
Published
How to Cite
Issue
Section
License
Copyright (c) 2025 AMY MBAYE, DIDIER MARIA NDIONE, Mouhamed Cherif Aidara, Fatou Ndiaye, Amadou Ndiaye, Mamadou Lamine Ndiaye, Joseph Ndong
This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License ( Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License) that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- To the extent transferable, copyright in and to the undersigned article is hereby assigned to Collaborating Academics and Open Access Journal Materials and Devices (ISSN: 2495-3911) for publication in the website of the journal and as part of a book (eventually a special volume) that could be produced in a printed and/or an electronic form.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).
- Figures, tables, and other information present in articles published in the OAJ Materials and Devices may be reused without permission, provided the citation of original article is made in figure's or table's caption.
