Ultra-short-term solar forecasting with reduced pre-acquired data considering optimal heuristic configurations of deep neural networks

Musaed Alrashidi; Musaed Alrashidi

doi:10.3934/math.2024603

AIMS Mathematics

2024, Volume 9, Issue 5: 12323-12356. doi: 10.3934/math.2024603

Previous Article Next Article

Research article Special Issues

Ultra-short-term solar forecasting with reduced pre-acquired data considering optimal heuristic configurations of deep neural networks

Musaed Alrashidi ^,

Department of Electrical Engineering, College of Engineering, Qassim University, Buraydah 52571, Saudi Arabia

Received: 03 February 2024 Revised: 20 March 2024 Accepted: 25 March 2024 Published: 28 March 2024
MSC : 62P30

Forecasting solar irradiance, particularly Global Horizontal Irradiance (GHI), has drawn much interest recently due to the rising demand for renewable energy sources. Many works have been proposed in the literature to forecast GHI by incorporating weather or environmental variables. Nevertheless, the expensive cost of the weather station hinders obtaining meteorological data, posing challenges in generating accurate forecasting models. Therefore, this work addresses this issue by developing a framework to reliably forecast the values of GHI even if meteorological data are unavailable or unreliable. It achieves this by leveraging lag observations of GHI values and applying feature extraction capabilities of the deep learning models. An ultra-short-term GHI forecast model is proposed using the Convolution Neural Network (CNN) algorithm, considering optimal heuristic configurations. In addition, to assess the efficacy of the proposed model, sensitivity analysis of different input variables of historical GHI observations is examined, and its performance is compared with other commonly used forecasting algorithm models over different forecasting horizons of 5, 15, and 30 minutes. A case study is carried out, and the model is trained and tested utilizing real GHI data from solar data located in Riyadh, Saudi Arabia. Results reveal the importance of employing historical GHI data in providing precise forecasting outcomes. The developed CNN-based model outperformed in ultra-short-term forecasting, showcasing average root mean square error results across different forecasting horizons: 2.262 W/m² (5min), 30.569 W/m² (15min), and 54.244 W/m² (30min) across varied day types. Finally, the findings of this research can permit GHI to be integrated into the power grid and encourage the development of sustainable energy systems.
- time series forecast,
- deep learning,
- historical GHI values,
- optimal heuristic configurations,
- energy systems
Citation: Musaed Alrashidi. Ultra-short-term solar forecasting with reduced pre-acquired data considering optimal heuristic configurations of deep neural networks[J]. AIMS Mathematics, 2024, 9(5): 12323-12356. doi: 10.3934/math.2024603

Related Papers:

Abstract

Forecasting solar irradiance, particularly Global Horizontal Irradiance (GHI), has drawn much interest recently due to the rising demand for renewable energy sources. Many works have been proposed in the literature to forecast GHI by incorporating weather or environmental variables. Nevertheless, the expensive cost of the weather station hinders obtaining meteorological data, posing challenges in generating accurate forecasting models. Therefore, this work addresses this issue by developing a framework to reliably forecast the values of GHI even if meteorological data are unavailable or unreliable. It achieves this by leveraging lag observations of GHI values and applying feature extraction capabilities of the deep learning models. An ultra-short-term GHI forecast model is proposed using the Convolution Neural Network (CNN) algorithm, considering optimal heuristic configurations. In addition, to assess the efficacy of the proposed model, sensitivity analysis of different input variables of historical GHI observations is examined, and its performance is compared with other commonly used forecasting algorithm models over different forecasting horizons of 5, 15, and 30 minutes. A case study is carried out, and the model is trained and tested utilizing real GHI data from solar data located in Riyadh, Saudi Arabia. Results reveal the importance of employing historical GHI data in providing precise forecasting outcomes. The developed CNN-based model outperformed in ultra-short-term forecasting, showcasing average root mean square error results across different forecasting horizons: 2.262 W/m² (5min), 30.569 W/m² (15min), and 54.244 W/m² (30min) across varied day types. Finally, the findings of this research can permit GHI to be integrated into the power grid and encourage the development of sustainable energy systems.

References

[1]	D. Gielen, F. Boshell, D. Saygin, M. D. Bazilian, N. Wagner, R. Gorini, The role of renewable energy in the global energy transformation, Energy Strateg. Rev., 24 (2019), 38–50. https://doi.org/10.1016/j.esr.2019.01.006 doi: 10.1016/j.esr.2019.01.006
[2]	N. Rahimi, S. Park, W. Choi, B. Oh, S. Kim, Y. Cho, A comprehensive review on ensemble solar power forecasting algorithms, J. Electr. Eng. Technol., 18 (2023), 719–733. https://doi.org/10.1007/s42835-023-01378-2 doi: 10.1007/s42835-023-01378-2
[3]	M. Perera, J. De Hoog, K. Bandara, S. Halgamuge, Multi-resolution, multi-horizon distributed solar PV power forecasting with forecast combinations, Expert Syst. Appl., 205 (2022), 117690. https://doi.org/10.1016/j.eswa.2022.117690
[4]	M. Alrashidi, S. Rahman, Short-term photovoltaic power production forecasting based on novel hybrid data-driven models, J. Big Data, 10 (2023). https://doi.org/10.1186/s40537-023-00706-7
[5]	R. Ahmed, V. Sreeram, Y. Mishra, D. Arif, A review and evaluation of the state-of-the-art in PV solar power forecasting: Techniques and optimization, Renew. Sust. Energ. Rev., 124 (2020), 109792. https://doi.org/10.1016/j.rser.2020.109792 doi: 10.1016/j.rser.2020.109792
[6]	D. V. Pombo, P. Bacher, C. Ziras, H. W. Bindner, S. V. Spataru, P. E. Sørensen, Benchmarking physics-informed machine learning-based short term PV-power forecasting tools, Energy Rep., 8 (2022), 6512–6520. https://doi.org/10.1016/j.egyr.2022.05.006 doi: 10.1016/j.egyr.2022.05.006
[7]	M. Alrashidi, M. Alrashidi, S. Rahman, Global solar radiation prediction: Application of novel hybrid data-driven model, Appl. Soft Comput., 112 (2021), 107768. https://doi.org/10.1016/j.asoc.2021.107768 doi: 10.1016/j.asoc.2021.107768
[8]	F. Pandžić, T. Capuder, Advances in short-term solar forecasting: A review and benchmark of machine learning methods and relevant data sources, Energies (Basel), 17 (2023), 97. https://doi.org/10.3390/en17010097 doi: 10.3390/en17010097
[9]	A. Dolara, S. Leva, G. Manzolini, Comparison of different physical models for PV power output prediction, Sol. Energy, 119 (2015), 83–99. https://doi.org/10.1016/j.solener.2015.06.017 doi: 10.1016/j.solener.2015.06.017
[10]	Ö. A. Karaman, Performance evaluation of seasonal solar irradiation models—case study: Karapınar town, Turkey, Case Stud. Therm. Eng., 49 (2023). https://doi.org/10.1016/j.csite.2023.103228
[11]	E. S. Solano, P. Dehghanian, C. M. Affonso, Solar radiation forecasting using machine learning and ensemble feature selection, Energies (Basel), 15 (2022), 19. https://doi.org/10.3390/en15197049 doi: 10.3390/en15197049
[12]	J. Lee, W. Wang, F. Harrou, Y. Sun, Reliable solar irradiance prediction using ensemble learning-based models: A comparative study, Energ. Convers Manage., 208 (2020), 112582. https://doi.org/10.1016/J.ENCONMAN.2020.112582 doi: 10.1016/J.ENCONMAN.2020.112582
[13]	X. Yang, Y. Ji, X. Wang, M. Niu, S. Long, J. Xie, Simplified method for predicting hourly global solar radiation using extraterrestrial radiation and limited weather forecast parameters, Energies (Basel), 16 (2023), 7. https://doi.org/10.3390/en16073215 doi: 10.3390/en16073215
[14]	M. Perera, J. De Hoog, K. Bandara, S. Halgamuge, Multi-resolution, multi-horizon distributed solar PV power forecasting with forecast combinations, Expert Syst. Appl., 205 (2022). https://doi.org/10.1016/j.eswa.2022.117690
[15]	F. Gurbuz, A. Mudireddy, R. Mantilla, S. Xiao, Using a physics-based hydrological model and storm transposition to investigate machine-learning algorithms for streamflow prediction, J. Hydrol., 628 (2024), 130504. https://doi.org/10.1016/J.JHYDROL.2023.130504 doi: 10.1016/J.JHYDROL.2023.130504
[16]	V. Narayan, S. Awasthi, N. Fatima, M. Faiz, S. Srivastava, Deep learning approaches for human gait recognition: A review, In: 2023 International Conference on Artificial Intelligence and Smart Communication, AISC, 2023,763–768. https://doi.org/10.1109/AISC56616.2023.10085665
[17]	F. Jiang, Y. Lu, Y. Chen, D. Cai, G. Li, Image recognition of four rice leaf diseases based on deep learning and support vector machine, Comput. Electron. Agr., 179 (2020), 105824. https://doi.org/10.1016/J.COMPAG.2020.105824 doi: 10.1016/J.COMPAG.2020.105824
[18]	S. Yang, Y. Wang, X. Chu, A survey of deep learning techniques for neural machine translation, arXiv preprint, 2020, 1–21.
[19]	S. P. Singh, A. Kumar, H. Darbari, L. Singh, A. Rastogi, S. Jain, Machine translation using deep learning: An overview, In: 2017 International Conference on Computer, Communications and Electronics, COMPTELIX, 2017,162–167. https://doi.org/10.1109/COMPTELIX.2017.8003957
[20]	S. Tajjour, S. S. Chandel, M. A. Alotaibi, H. Malik, F. P. G. Marquez, A. Afthanorhan, Short-term solar irradiance forecasting using deep learning techniques: A comprehensive case study, IEEE Access, 11 (2023), 119851–119861. https://doi.org/10.1109/ACCESS.2023.3325292 doi: 10.1109/ACCESS.2023.3325292
[21]	N. E. Michael, M. Mishra, S. Hasan, A. Al-Durra, Short-term solar power predicting model based on multi-step CNN stacked LSTM technique, Energies (Basel), 15 (2022), 6. https://doi.org/10.3390/en15062150
[22]	H. Kim, S. Park, H. J. Park, H. G. Son, S. Kim, Solar radiation forecasting based on the hybrid CNN-CatBoost model, IEEE Access, 11 (2023), 13492–13500. https://doi.org/10.1109/ACCESS.2023.3243252 doi: 10.1109/ACCESS.2023.3243252
[23]	V. Sansine, P. Ortega, D. Hissel, F. Ferrucci, Hybrid deep learning model for mean hourly irradiance probabilistic forecasting, Atmosphere (Basel), 14 (2023), 7. https://doi.org/10.3390/atmos14071192 doi: 10.3390/atmos14071192
[24]	A. Dairi, F. Harrou, Y. Sun, S. Khadraoui, Short-term forecasting of photovoltaic solar power production using variational auto-encoder driven deep learning approach, Appl. Sci. (Switzerland), 10 (2020), 1–20. https://doi.org/10.3390/app10238400 doi: 10.3390/app10238400
[25]	Y. Pang, M. Sun, X. Jiang, X. Li, Convolution in convolution for network in network, IEEE Trans. Neural Netw. Learn. Syst., 29 (2018), 1587–1597. https://doi.org/10.1109/TNNLS.2017.2676130 doi: 10.1109/TNNLS.2017.2676130
[26]	A. Johny, K. N. Madhusoodanan, Dynamic learning rate in deep CNN model for metastasis detection and classification of histopathology images, Comput. Math. Method. Med., 2021 (2021). https://doi.org/10.1155/2021/5557168 doi: 10.1155/2021/5557168
[27]	R. Chauhan, K. K. Ghanshala, R. Joshi, Convolutional neural network (CNN) for image detection and recognition, In: 2018 First International Conference on Secure Cyber Computing and Communication (ICSCCC), 2018,278–282.
[28]	S. Albawi, T. A. Mohammed, S. Al-Zawi, Understanding of a convolutional neural network, In: Proceedings of 2017 International Conference on Engineering and Technology, ICET 2017, 2017, 1–6. https://doi.org/10.1109/ICENGTECHNOL.2017.8308186
[29]	J. Wu, Introduction to convolutional neural networks, National Key Lab for Novel Software Technology. Nanjing University, China, 5 (2017), 495.
[30]	Z. Li, F. Liu, W. Yang, S. Peng, J. Zhou, A survey of convolutional neural networks: Analysis, applications, and prospects, IEEE Trans. Neural Netw. Learn. Syst., 2021, 1–21.
[31]	W. Lu, J. Li, Y. Li, A. Sun, J. Wang, A CNN-LSTM-based model to forecast stock prices, Complexity, 2020 (2020). https://doi.org/10.1155/2020/6622927 doi: 10.1155/2020/6622927
[32]	M. Marzouq, H. El Fadili, K. Zenkouar, Z. Lakhliai, M. Amouzg, Short term solar irradiance forecasting via a novel evolutionary multi-model framework and performance assessment for sites with no solar irradiance data, Renew. Energy, 157 (2020), 214–231. https://doi.org/10.1016/J.RENENE.2020.04.133 doi: 10.1016/J.RENENE.2020.04.133
[33]	A. P. Wibawa, A. B. P. Utama, H. Elmunsyah, U. Pujianto, F. A. Dwiyanto, L. Hernandez, Time-series analysis with smoothed convolutional neural network, J. Big Data, 9 (2022). https://doi.org/10.1186/s40537-022-00599-y
[34]	A. Alfadda, S. Rahman, M. Pipattanasomporn, Solar irradiance forecast using aerosols measurements: A data driven approach, Sol. Energy, 170 (2018), 924–939. https://doi.org/10.1016/j.solener.2018.05.089 doi: 10.1016/j.solener.2018.05.089
[35]	M. S. Hossain, H. Mahmood, Short-term photovoltaic power forecasting using an LSTM neural network and synthetic weather forecast, IEEE Access, 8 (2020), 172524–172533. https://doi.org/10.1109/ACCESS.2020.3024901 doi: 10.1109/ACCESS.2020.3024901
[36]	J. Zhang, A. Florita, B. M. Hodge, S. Lu, H. F. Hamann, V. Banunarayanan, et al., A suite of metrics for assessing the performance of solar power forecasting, Sol. Energy, 2015,157–175. Available from: https://www.sciencedirect.com/science/article/pii/S0038092X14005027.
[37]	A. Botchkarev, A new typology design of performance metrics to measure errors in machine learning regression algorithms, Interdiscip. J. Inform. Knowl. Manage., 14 (2019), 45–76. https://doi.org/10.28945/4184 doi: 10.28945/4184
[38]	A. Botchkarev, Performance metrics (error measures) in machine learning regression, forecasting and prognostics: Properties and typology, arXiv preprint, 1809 (2018), 03006. https://doi.org/https://doi.org/10.28945/4184
[39]	J. M. Bright, Solcast: Validation of a satellite-derived solar irradiance dataset, Sol. Energy, 189 (2019), 435–449. https://doi.org/10.1016/j.solener.2019.07.086 doi: 10.1016/j.solener.2019.07.086
[40]	Solar resource maps and GIS data Solargis, 2022. Available from: https://solargis.com/maps-and-gis-data/download/saudi-arabia.
[41]	X. Qing, Y. Niu, Hourly day-ahead solar irradiance prediction using weather forecasts by LSTM, Energy, 148 (2018), 461–468. https://doi.org/10.1016/J.ENERGY.2018.01.177 doi: 10.1016/J.ENERGY.2018.01.177
[42]	H. T. Yang, C. M. Huang, Y. C. Huang, Y. S. Pai, A weather-based hybrid method for 1-day ahead hourly forecasting of PV power output, IEEE Trans. Sustain. Energy, 5 (2014), 917–926. https://doi.org/10.1109/TSTE.2014.2313600 doi: 10.1109/TSTE.2014.2313600
[43]	S. E. Haupt, B. Kosovic, Big data and machine learning for applied weather forecasts: Forecasting solar power for utility operations, In: 2015 IEEE Symposium Series on Computational Intelligence, 2015,496–501. https://doi.org/10.1109/SSCI.2015.79
[44]	CAMS radiation service, Medium-Range Weather Forecasts (ECMWF), 2023. Available from: http://www.soda-pro.com/web-services/radiation/cams-radiation-service.
[45]	W. A. Beckman, J. A. Duffie, Solar engineering of thermal processes, 3 Eds., John Wiley & Sons, 2013.

Reader Comments

Your name:*

Email:*
© 2024 the Author(s), licensee AIMS Press. This is an open access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0)