Multivariate Models for Photovoltaic Power Forecasting with Non-climatic Exogenous Variables
Keywords:
multivariate models, photovoltaic power, forecasting, renewable energyAbstract
Forecasting electricity generation from renewable resources is crucial for the efficient planning and operation of power systems. The development of forecasting models based on local meteorological variables is common, however, sometimes this information is unavailable. This study explores the use of multivariate models that do not incorporate meteorological variables, but use historical power-generated data from eight PV plants located in the same region to predict the future value of a target plant. This allows for improved forecasting when meteorological variables are unavailable and the only information available is the generation of the PV plants. The performance of LSTM and BiLSTM networks is compared for different time horizons, considering various lags of the power series itself for estimating future values. The main contributions of this study include the introduction of power time series from other plants as model inputs, the use of spatial interpolation to fill in missing data and the application of causality tests between time series for the selection of predictor variables, and the uncertainty associated with the predictions is analyzed using quantile regression techniques.
Downloads
References
R. Haas, N. Duic, H. Auer, A. Ajanovic, J. Ramsebner, J. Knapek y S. Zwickl-Bernhard, «The photovoltaic revolution is on: How it will change the electricity system in a lasting way,» Energy, vol. 265, p. 126351, 2023.
M. Godoy, Z. García, H. Hernández, J. A. Cueto, J. I. Silva y G. Crespo, «Frequency response analysis under faults in weak power systems.,» International Journal of Electrical and Computer Engineering IJECE, vol. 12, nº 2, pp. 1077-1088, April 2022.
Y. Gallego, L. Casas, Z. García y Y. Rivas, «Impacto de la implementación de paneles fotovoltaicos en el sistema eléctrico Cayo Santa María,» Energética, vol. 38, nº 2, 2017.
Y. M. Atwa y E. F. El-Saadany, «Optimal Allocation of ESS in Distribution Systems with a High Penetration of Wind Energy,» IEEE Transaction on Power Systems, vol. 25, nº 4, pp. 1815-1822, 2010.
G. Delille, B. Franois y G. Malarange, «Dynamic Frequency Control Support by Energy Storage to Reduce the Impact of Wind and Solar Generation on Isolated Power System’s Inertia.,» IEEE Transaction on Sustainable Energy, vol. 3, nº 4, pp. 931-939, 2012.
A. Saleh, E. Ozkop, R. J. Meng, Z. García y O. A. Betancourt, «Selecting Locations and Sizes of Battery Storage Systems Based on the Frequency of the Center of Inertia and Principle Component Analysis,» IEEE Transactions on Industry Applications, vol. 56, nº 2, pp. 1040-1051, 2020.
O. A. Betancourt, Z. García, A. Saleh, E. F. Hill, X. Zhao y F. P. Sánchez, «Battery Energy Storage Systems for Primary Frequency Regulation in Island Power Systems,» de Proceeding 2020 IEEE/IAS 56th Industrial and Commercial Power Systems Technical Conference (I&CPS), Las Vegas, 2020.
X. Yang, M. Xu, S. Xu y X. Han, «Day-ahead forecasting of photovoltaic output power with similar cloud space fusion based on incomplete historical data mining,» Applied Energy, vol. 206, pp. 683-696, 2017.
M. Pierro, F. Bucci, M. De Felice, E. Maggioni, D. Moser, A. Perotto, F. Sapada y C. Cornaro, «Multi-Model Ensemble for day ahead prediction of photovoltaic power generation,» Solar Energy, vol. 134, pp. 132-146, 2016.
A. Mellit, A. Massi Pavan y V. Lughi, «Deep learning neural networks for short-term photovoltaic power forecasting,» Renewable Energy, vol. 172, pp. 276-288, 2021.
A. Muñoz, «Modelos de predicción a corto plazo de la generación eléctrica en instalaciones fotovoltaicas,» 2014.
A. J. Rocha y A. P. Alves, «Feature Extraction via Multiresolution Analysis for Short-Term Load Forecasting,» IEEE Transaction on Power Systems, vol. 20, nº 1, pp. 189-198, 2005.
N. Amjady y F. Keynia, «A New Neural Network Approach to Short Term Load Forecasting of Electrical Power Systems,» Energies, pp. 488-503, 2011.
K. K. M. Sajjad y F. Keynia, «Applying Wavelet to ANN based short-term load forecasting: A case study of Zanjan power pystem.,» International Journal of Electrical and Electronics Engineering Research, vol. 3, nº 2, pp. 209-2016, 2013.
J. R. Gómez, K. G. Di Santo y M. A. Saidel, «Linear and non-linear methods for prediction of peak load at University of São Paulo,» Measurement, vol. 78, pp. 187-201, 2016.
J. A. Peña, P. R. Viego, J. R. Gómez y E. A. Padrón, «Pronóstico del consumo pico para la gestión energética de la Universidad de Cienfuegos,» Revista Universidad y Sociedad, vol. 11, nº 4, pp. 220-228, 2019.
P. Ramsami y V. Oree, «A hybrid method for forecasting the energy output of photovoltaic systems,» Energy Conversion and Management, vol. 95, pp. 406-413, 2015.
M. A. Gómez, J. R. Gómez, J. V. Lorenzo, R. Fonte y Z. García, «Pronóstico de la generación eléctrica de sistemas fotovoltaicos. Un inicio en Cuba desde la universidad,» Universidad y Sociedad, vol. 13, nº 1, pp. 253-265, 2021.
Y. Li, Y. He, Y. Su y L. Shu, «Forecasting the daily power output of a grid-connected photovoltaic system based on multivariate adaptive regression splines,» Applied Energy, vol. 180, pp. 392-401, 2016.
A. Tesfaye, J. Zhang y D. Zheng, «Short-term photovoltaic solar power forecasting using hybrid Wavelet-PSO-SVM model based on SCADA and Meteorological information,» Renewable Energy, vol. 118, 2018.
I. Fraga, M. A. Gómez, J. R. Gómez y Z. García, «Predicción de la generación fotovoltaica usando Deep learning,» Revista Universidad y Sociedad, vol. 15, nº S1, pp. 266-275, 2023.
F. Succetti, A. Rosato, R. Araneo y M. Panella, «Deep learning networks for multivariate prediction of photovoltaic time series,» IEEEAccess. Special section on key enabling technologies for prosumer energy management, vol. 8, pp. 211490-21505, 2020.
N. H. A. Rahman, M. Z. Hussin, S. I. Sulaiman, M. A. Hairuddin y E. H. M. Saat, «Univariate and multivariate short-term solar power forecasting of 25MWac Pasir Gudang utility-sacle photovoltaic system using LSTM approach,» de 7th International Conference on Renewable Energy and Conservation, ICREC 2022, Paris, France, 2022.
M. S. Hossain y H. Mahmood, «Short-term photovoltaic power forecasting using an LSTM neural network and synthetic weather forecast,» IEEEAccess, vol. 8, pp. 172524-172533, 2020.
H. Zhen, D. Niu, K. Wang, Y. Shi, Z. Ji y X. Xu, «Photovoltaic power forecasting based on GA improved Bi-LSTM in microgrid without meteorological information,» Energy, vol. 231, p. 120908, 2021.
Z. Sen y A. D. Sahin, «Spatial interpolation and estimation of solar irradiation by cumulative semivariograms,» Solar Energy, vol. 71, nº 1, pp. 11-21, 2001.
M. C. Crespo, «Imputación de datos faltantes en redes de distribución de baja tensión: Aplicación a edificios de pública concurrencia,» 2018.
J. Adaro, J. Marchesi y J. Zizzias, «Tratamiento para Datos Faltantes en Series de Radiación Solar,» Revista Tecnología y Ciencia, vol. 12, nº 25, pp. 60-65, 2014.
M. Carbajal, C. Yarlequé, A. Posadas, E. Silvestre, A. mejía y R. Quiroz, «Reconstrucción de datos faltantes de precipitación pluvial diaria mediante la Transformada Wavelet,» REVISTA PERUANA GEO-ATMOSFÉRICA RPGA, vol. 2, pp. 76-88, 2010.
A. C. Arias, «Propuesta y evaluación de una estrategia para imputación múltiple y multivariada de valores faltantes en series de tiempo del campo meteorológico utilizando aprendizaje automático,» 2022.
Casta y Elkin, «Reconstrucción de datos de series de tiempo: una aplicación a la demanda horaria de la electricidad,» revista Colombiana de Estadística, vol. 30, nº 2, pp. 247-263, 2007.
J. Afriyie, S. Twumasi-Ankrah, K. Gyamfi, D. Arthur y W. Pels, «Evaluating the performance of unit root test in single time series processes.,» Mathematics and Statistic, vol. 8, nº 6, pp. 656-664, 2020.
M. Massaoudi, I. Chihi, L. Sidhom, M. Trabelsi, S. S. Refaat, H. Abu-Rub y F. S. Oueslati, «An effective hybrid NARX-LSTM model for point and interval PV power forecasting,» IEEEAccess, vol. 9, pp. 36571-36588, 2021.
