Development of a DNI Forecast Tool Based on Machine Learning for the Smart Operation of a Parabolic Trough Collector System with Thermal Energy Storage
Theory and Results
DOI:
https://doi.org/10.52825/solarpaces.v2i.802Keywords:
Direct Normal Irradiance (DNI) Forecasting, Long Short-Term Memory (LSTM), Recurrent Neural Network, RNN, Parabolic Trough Collector, PTC, Thermal Energy Storage, TESAbstract
In the research project Smart Solar System (S3), the Solar-Institut Jülich (SIJ) developed forecast tools to predict the direct normal irradiance (DNI) in hourly resolution for the current day. The aim of the daily DNI forecast is to use it as input to enable a smart operation of a parabolic trough collector (PTC) system with a concrete thermal energy storage (C-TES) located at the company KEAN Soft Drinks Ltd in Limassol, Cyprus. The main focus in this work is on the development of a DNI forecast tool based on long short-term memory (LSTM), which is a recurrent neural network (RNN), and its comparison with a DNI forecast tool based on analytical algorithms (non-machine learning). Only the non-machine learning DNI forecast tool with hourly update was tested in real-life PTC plant operation since end of 2022. The comparisons between the three DNI forecast tools show that the potential of using machine learning is very high. Different comparisons were made including an evaluation of the accuracy of the tool (i.e. comparison of the DNI forecast data with DNI measurement data). The DNI forecast based on the LSTM network proved to be more accurate than the non-machine learning DNI forecasts when considering errors greater than ±100 W/m2. The error of the LSTM network compared to DNI measurement data was as follows: 43.6 % of the data was within ±100 W/m2, 74.8 % was within ±200 W/m2, 89.8 % was within ±300 W/m2 and 95.8 % was within ±400 W/m2.
Downloads
References
[1] J. C. Sattler, S. Alexopoulos, R. A. Chico Caminos, J. Mitchell, V. Ruiz, S. Kalogirou, P. Ktistis, C. Teixeira Boura, U. Herrmann, “Dynamic simulation model of a parabolic trough collector system with concrete thermal energy storage for process steam generation”, AIP Conference Proceedings, vol. 2126, p. 150007, Jul., 2019, doi: https://doi.org/10.1063/1.5117663
[2] J. C. Sattler, R. A. Chico Caminos, N. Ürlings, S. Dutta, V. Ruiz, S. Kalogirou, P. Ktistis, R. Agathokleous, C. Jung, S. Alexopoulos, V. Atti, C. Teixeira Boura, U Herrmann, “Operational experience and behaviour of a parabolic trough collector system with concrete thermal energy storage for process steam generation in Cyprus”, AIP Conference Proceedings, vol. 2303, p. 140004, Dec., 2020, doi: https://doi.org/10.1063/5.0029278
[3] J. C. Sattler, R. A. Chico Caminos, V. Atti, N. Ürlings, S. Dutta, V. Ruiz, S. Kalogirou, P. Ktistis, R. Agathokleous, S. Alexopoulos, C. Teixeira Boura, U. Herrmann, “Dynamic simulation tool for a performance evaluation and sensitivity study of a parabolic trough collector system with concrete thermal energy storage”, AIP Conference Proceedings, vol. 2303, p. 160004, Dec., 2020, doi: https://doi.org/10.1063/5.0029277
[4] J. C. Sattler, V. Atti, S. Alexopoulos, C. Teixeira Boura, U. Herrmann, S. Dutta, and I. Kioutsioukis, “DNI Forecast Tool for the Smart Operation of a Parabolic Trough Collector System with Concrete Thermal Energy Storage”, in: SolarPACES 2022, 26 – 30 September 2022, Albuquerque, New Mexico, USA.
[5] The MathWorks® Inc., https://de.mathworks.com/help/deeplearning/ug/time-series-forecasting-using-deep-learning.html (last opened on 2 October 2022)
[6] Z. Bounoua, A. Mechaqrane, “Prediction of daily global horizontal solar irradiation using artificial neural networks and commonly measured meteorological parameters”, AIP Conf. Proc. 20 December 2018, 2056 (1): 020024, doi: https://doi.org/10.1063/1.5084997
[7] Z. Bounoua, A. Mechaqrane, “Estimation of Daily Direct Normal Solar Irradiation Using Machine-Learning Methods”, In: Bennani, S., Lakhrissi, Y., Khaissidi, G., Mansouri, A., Khamlichi, Y. (eds) WITS 2020, 2022, Lecture Notes in Electrical Engineering, vol 745. Springer, Singapore, doi: https://doi.org/10.1007/978-981-33-6893-4_58
[8] M. Khalifa Boutahir, Y. Farhaoui, M. Azrour, I. Zeroual, A. El Allaoui, "Effect of Feature Selection on the Prediction of Direct Normal Irradiance.", Big data mining and analytics, 5 (2022).:309-317, doi: https://doi.org/10.26599/bdma.2022.9020003
[9] B. Nouri, P. Kuhn, S. Wilbert, C. Prahl, R. Pitz-Paal, P. Blanc, T. Schmidt, Z. Yasser, L. Ramirez Santigosa, D. Heineman, “Nowcasting of DNI maps for the solar field based on voxel carving and individual 3D cloud objects from all sky images”, AIP Conf. Proc., 8 November 2018, 2033 (1): 190011, doi: https://doi.org/10.1063/1.5067196
[10] M. Deru, A. Ndiaye, “Deep Learning mit TensorFlow, Keras und TensorFlow.js”, 2., aktualisierte und erweiterte Auflage 2020, 1. Nachdruck 2023, Rheinwerk Verlag, Bonn 2020, Germany, Ch. 3, p. 71, ISBN: 978-3-8362-7425-8
Published
How to Cite
Conference Proceedings Volume
Section
License
Copyright (c) 2024 Johannes Christoph Sattler, Siddharth Dutta, Spiros Alexopoulos, Adnan Kawam, Cristiano Teixeira Boura, Ulf Herrmann, Ioannis Kioutsioukis
This work is licensed under a Creative Commons Attribution 4.0 International License.
Accepted 2024-04-09
Published 2024-10-15
