Day-ahead photovoltaic power production forecasting methodology based on machine learning and statistical post-processing

Theocharides, Spyros; Makrides, George; Livera, Andreas; Theristis, Marios; Kaimakis, Paris; Georghiou, George E.

doi:10.1016/j.apenergy.2020.115023

Day-ahead photovoltaic power production forecasting methodology based on machine learning and statistical post-processing

Journal Article · Tue Apr 21 00:00:00 EDT 2020 · Applied Energy

DOI:https://doi.org/10.1016/j.apenergy.2020.115023· OSTI ID:1617300

Theocharides, Spyros ^[1]; Makrides, George ^[1]; Livera, Andreas ^[1]; Theristis, Marios ^[2]; Kaimakis, Paris ^[3]; Georghiou, George E. ^[1]

Univ. of Cyprus, Nicosia (Cyprus)
Sandia National Lab. (SNL-NM), Albuquerque, NM (United States)
Univ. of Central Lancashire Cyprus, Pyla (Cyprus)

A main challenge towards ensuring large-scale and seamless integration of photovoltaic systems is to improve the accuracy of energy yield forecasts, especially in grid areas of high photovoltaic shares. The scope of this paper is to address this issue by presenting a unified methodology for hourly-averaged day-ahead photovoltaic power forecasts with improved accuracy, based on data-driven machine learning techniques and statistical post-processing. More specifically, the proposed forecasting methodology framework comprised of a data quality stage, data-driven power output machine learning model development (artificial neural networks), weather clustering assessment (K-means clustering), post-processing output optimisation (linear regressive correction method) and the final performance accuracy evaluation. The results showed that the application of linear regression coefficients to the forecasted outputs of the developed day-ahead photovoltaic power production neural network improved the performance accuracy by further correcting solar irradiance forecasting biases. The resulting optimised model provided a mean absolute percentage error of 4.7% when applied to historical system datasets. Finally, the model was validated both, at a hot as well as a cold semi-arid climatic location, and the obtained results demonstrated close agreement by yielding forecasting accuracies of mean absolute percentage error of 4.7% and 6.3%, respectively. Finally, the validation analysis provides evidence that the proposed model exhibits high performance in both forecasting accuracy and stability.

View Accepted Manuscript (DOE)

Research Organization:: Sandia National Laboratories (SNL-NM), Albuquerque, NM (United States)

Sponsoring Organization:: European Union; USDOE National Nuclear Security Administration (NNSA)

Grant/Contract Number:: AC04-94AL85000; NA0003525

OSTI ID:: 1617300

Report Number(s):: SAND--2020-4201J; 685417

Journal Information:: Applied Energy, Journal Name: Applied Energy Journal Issue: C Vol. 268; ISSN 0306-2619

Publisher:: ElsevierCopyright Statement

Country of Publication:: United States

Language:: English

References (40)

The Application of Cluster Analysis in Strategic Management Research: an Analysis and Critique KETCHEN Jr., David J.; Shook, Christopher L. Strategic Management Journal, Vol. 17, Issue 6 https://doi.org/10.1002/(SICI)1097-0266(199606)17:6<441::AID-SMJ819>3.0.CO;2-G	journal	June 1996
Solar and photovoltaic forecasting through post-processing of the Global Environmental Multiscale numerical weather prediction model: Solar and photovoltaic forecasting Pelland, Sophie; Galanis, George; Kallos, George Progress in Photovoltaics: Research and Applications, Vol. 21, Issue 3 https://doi.org/10.1002/pip.1180	journal	November 2011
Energy yield prediction errors and uncertainties of different photovoltaic models: Energy yield prediction errors and uncertainties Makrides, George; Zinsser, Bastian; Schubert, Markus Progress in Photovoltaics: Research and Applications, Vol. 21, Issue 4 https://doi.org/10.1002/pip.1218	journal	November 2011
Early Stopping - But When? Prechelt, Lutz Lecture Notes in Computer Science https://doi.org/10.1007/3-540-49430-8_3	book	January 1998
A forecast error correction method in numerical weather prediction by using recent multiple-time evolution data Xue, Hai-Le; Shen, Xue-Shun; Chou, Ji-Fan Advances in Atmospheric Sciences, Vol. 30, Issue 5 https://doi.org/10.1007/s00376-013-2274-1	journal	August 2013
Accurate photovoltaic power forecasting models using deep LSTM-RNN Abdel-Nasser, Mohamed; Mahmoud, Karar Neural Computing and Applications, Vol. 31, Issue 7 https://doi.org/10.1007/s00521-017-3225-z	journal	October 2017
Making data structures persistent Driscoll, James R.; Sarnak, Neil; Sleator, Daniel D. Journal of Computer and System Sciences, Vol. 38, Issue 1 https://doi.org/10.1016/0022-0000(89)90034-2	journal	February 1989
Numerical weather prediction Kimura, Ryuji Journal of Wind Engineering and Industrial Aerodynamics, Vol. 90, Issue 12-15 https://doi.org/10.1016/S0167-6105(02)00261-1	journal	December 2002
Prediction of short-term PV power output and uncertainty analysis Liu, Luyao; Zhao, Yi; Chang, Dongliang Applied Energy, Vol. 228 https://doi.org/10.1016/j.apenergy.2018.06.112	journal	October 2018
Another look at measures of forecast accuracy Hyndman, Rob J.; Koehler, Anne B. International Journal of Forecasting, Vol. 22, Issue 4 https://doi.org/10.1016/j.ijforecast.2006.03.001	journal	October 2006
The origins of computer weather prediction and climate modeling Lynch, Peter Journal of Computational Physics, Vol. 227, Issue 7 https://doi.org/10.1016/j.jcp.2007.02.034	journal	March 2008
Solar forecasting methods for renewable energy integration Inman, Rich H.; Pedro, Hugo T. C.; Coimbra, Carlos F. M. Progress in Energy and Combustion Science, Vol. 39, Issue 6 https://doi.org/10.1016/j.pecs.2013.06.002	journal	December 2013
Potential of photovoltaic systems in countries with high solar irradiation Makrides, George; Zinsser, Bastian; Norton, Matthew Renewable and Sustainable Energy Reviews, Vol. 14, Issue 2 https://doi.org/10.1016/j.rser.2009.07.021	journal	February 2010
Solar position algorithm for solar radiation applications Reda, Ibrahim; Andreas, Afshin Solar Energy, Vol. 76, Issue 5 https://doi.org/10.1016/j.solener.2003.12.003	journal	January 2004
Quality of performance assessment of PV plants based on irradiation maps Drews, A.; Beyer, H. G.; Rindelhardt, U. Solar Energy, Vol. 82, Issue 11 https://doi.org/10.1016/j.solener.2008.04.009	journal	November 2008
Online 24-h solar power forecasting based on weather type classification using artificial neural network Chen, Changsong; Duan, Shanxu; Cai, Tao Solar Energy, Vol. 85, Issue 11 https://doi.org/10.1016/j.solener.2011.08.027	journal	November 2011
Assessment of forecasting techniques for solar power production with no exogenous inputs Pedro, Hugo T. C.; Coimbra, Carlos F. M. Solar Energy, Vol. 86, Issue 7 https://doi.org/10.1016/j.solener.2012.04.004	journal	July 2012
Short-term forecasting of power production in a large-scale photovoltaic plant Mellit, A.; Massi Pavan, A.; Lughi, V. Solar Energy, Vol. 105 https://doi.org/10.1016/j.solener.2014.03.018	journal	July 2014
Review of photovoltaic power forecasting Antonanzas, J.; Osorio, N.; Escobar, R. Solar Energy, Vol. 136 https://doi.org/10.1016/j.solener.2016.06.069	journal	October 2016
Summarizing multiple aspects of model performance in a single diagram Taylor, Karl E. Journal of Geophysical Research: Atmospheres, Vol. 106, Issue D7 https://doi.org/10.1029/2000JD900719	journal	April 2001
Learning representations by back-propagating errors Rumelhart, David E.; Hinton, Geoffrey E.; Williams, Ronald J. Nature, Vol. 323, Issue 6088 https://doi.org/10.1038/323533a0	journal	October 1986
Short-term solar forecasting based on sky images to enable higher PV generation in remote electricity networks Schmidt, Thomas; Calais, Martina; Roy, Eric Renewable Energy and Environmental Sustainability, Vol. 2 https://doi.org/10.1051/rees/2017028	journal	January 2017
The Royal London Space Planning: An integration of space analysis and treatment planning Kirschen, Robert H.; O'Higgins, Elizabeth A.; Lee, Robert T. American Journal of Orthodontics and Dentofacial Orthopedics, Vol. 118, Issue 4 https://doi.org/10.1067/mod.2000.109031	journal	October 2000
Machine learning algorithms for photovoltaic system power output prediction Theocharides, Spyros; Makrides, George; Georghiou, George E. 2018 IEEE International Energy Conference (ENERGYCON) https://doi.org/10.1109/ENERGYCON.2018.8398737	conference	June 2018
Photovoltaic system power forecasting based on combined grey model and BP neural network Wang, Shouxiang; Zhang, Na; Zhao, Yishu 2011 International Conference on Electrical and Control Engineering (ICECE) https://doi.org/10.1109/ICECENG.2011.6057634	conference	September 2011
Gauss-Newton approximation to Bayesian learning Dan Foresee, F.; Hagan, M. T. Proceedings of International Conference on Neural Networks (ICNN'97) https://doi.org/10.1109/ICNN.1997.614194	conference	January 1997
Voltage stability and sensitivity analysis of grid-connected photovoltaic systems Xue, Yaosuo; Manjrekar, Madhav; Lin, Chenxi 2011 IEEE Power & Energy Society General Meeting, 2011 IEEE Power and Energy Society General Meeting https://doi.org/10.1109/PES.2011.6039649	conference	July 2011
The Solar Forecast Arbiter: An Open Source Evaluation Framework for Solar Forecasting Hansen, Clifford W.; Holmgren, William F.; Tuohy, Aidan 2019 IEEE 46th Photovoltaic Specialists Conference (PVSC) https://doi.org/10.1109/PVSC40753.2019.8980713	conference	June 2019
Forecasting Power Output of Photovoltaic Systems Based on Weather Classification and Support Vector Machines Shi, Jie; Lee, Wei-Jen; Liu, Yongqian IEEE Transactions on Industry Applications, Vol. 48, Issue 3 https://doi.org/10.1109/TIA.2012.2190816	journal	May 2012
A Weather-Based Hybrid Method for 1-Day Ahead Hourly Forecasting of PV Power Output Yang, Hong-Tzer; Huang, Chao-Ming; Huang, Yann-Chang IEEE Transactions on Sustainable Energy, Vol. 5, Issue 3 https://doi.org/10.1109/TSTE.2014.2313600	journal	July 2014
Solar Power Prediction Based on Satellite Images and Support Vector Machine Jang, Han Seung; Bae, Kuk Yeol; Park, Hong-Shik IEEE Transactions on Sustainable Energy, Vol. 7, Issue 3 https://doi.org/10.1109/TSTE.2016.2535466	journal	July 2016
Pearson's correlation coefficient Sedgwick, P. BMJ, Vol. 345, Issue jul04 1 https://doi.org/10.1136/bmj.e4483	journal	July 2012
Bayesian Interpolation MacKay, David J. C. Neural Computation, Vol. 4, Issue 3 https://doi.org/10.1162/neco.1992.4.3.415	journal	May 1992
Information-Based Skill Scores for Probabilistic Forecasts Ahrens, Bodo; Walser, André Monthly Weather Review, Vol. 136, Issue 1 https://doi.org/10.1175/2007MWR1931.1	journal	January 2008
Machine Learning: Algorithms and Applications Mohammed, Mohssen; Khan, Muhammad Badruddin; Bashier, Eihab Bashier Mohammed https://doi.org/10.1201/9781315371658	book	January 2016
PVWatts Version 5 Manual Dobos, A. https://doi.org/10.2172/1158421	report	September 2014
Day-Ahead Photovoltaic Forecasting: A Comparison of the Most Effective Techniques Nespoli, Alfredo; Ogliari, Emanuele; Leva, Sonia Energies, Vol. 12, Issue 9 https://doi.org/10.3390/en12091621	journal	April 2019
A Physical Hybrid Artificial Neural Network for Short Term Forecasting of PV Plant Power Output Dolara, Alberto; Grimaccia, Francesco; Leva, Sonia Energies, Vol. 8, Issue 2 https://doi.org/10.3390/en8021138	journal	February 2015
A New Sub-topics Clustering Method Based on Semi-supervised Learing Xu, Xiaodan Journal of Computers, Vol. 7, Issue 10 https://doi.org/10.4304/jcp.7.10.2471-2478	journal	October 2012
Updated world map of the Köppen-Geiger climate classification Peel, M. C.; Finlayson, B. L.; McMahon, T. A. Hydrology and Earth System Sciences, Vol. 11, Issue 5 https://doi.org/10.5194/hess-11-1633-2007	journal	January 2007

Similar Records

Comparative Analysis of Machine Learning Models for Day-Ahead Photovoltaic Power Production Forecasting

Journal Article · Wed Feb 17 19:00:00 EST 2021 · Energies · OSTI ID:1769936

Data-Driven Day-Ahead PV Estimation Using Autoencoder-LSTM and Persistence Model

Journal Article · Mon Sep 21 20:00:00 EDT 2020 · IEEE Transactions on Industry Applications · OSTI ID:1810249

Day-Ahead Probabilistic Forecasting of Net-Load and Demand Response Potentials with High Penetration of Behind-the-Meter Solar-plus-Storage

Technical Report · Thu Feb 27 23:00:00 EST 2025 · OSTI ID:2560048

Related Subjects

42 ENGINEERING
artificial neural networks
clustering
forecasting
machine learning
performance
photovoltaic

Day-ahead photovoltaic power production forecasting methodology based on machine learning and statistical post-processing

Citation Formats

References (40)

Similar Records

Related Subjects