Papers
Topics
Authors
Recent
Gemini 2.5 Flash
Gemini 2.5 Flash
126 tokens/sec
GPT-4o
28 tokens/sec
Gemini 2.5 Pro Pro
42 tokens/sec
o3 Pro
4 tokens/sec
GPT-4.1 Pro
38 tokens/sec
DeepSeek R1 via Azure Pro
28 tokens/sec
2000 character limit reached

Day-ahead regional solar power forecasting with hierarchical temporal convolutional neural networks using historical power generation and weather data (2403.01653v1)

Published 4 Mar 2024 in cs.LG

Abstract: Regional solar power forecasting, which involves predicting the total power generation from all rooftop photovoltaic systems in a region holds significant importance for various stakeholders in the energy sector. However, the vast amount of solar power generation and weather time series from geographically dispersed locations that need to be considered in the forecasting process makes accurate regional forecasting challenging. Therefore, previous work has limited the focus to either forecasting a single time series (i.e., aggregated time series) which is the addition of all solar generation time series in a region, disregarding the location-specific weather effects or forecasting solar generation time series of each PV site (i.e., individual time series) independently using location-specific weather data, resulting in a large number of forecasting models. In this work, we propose two deep-learning-based regional forecasting methods that can effectively leverage both types of time series (aggregated and individual) with weather data in a region. We propose two hierarchical temporal convolutional neural network architectures (HTCNN) and two strategies to adapt HTCNNs for regional solar power forecasting. At first, we explore generating a regional forecast using a single HTCNN. Next, we divide the region into multiple sub-regions based on weather information and train separate HTCNNs for each sub-region; the forecasts of each sub-region are then added to generate a regional forecast. The proposed work is evaluated using a large dataset collected over a year from 101 locations across Western Australia to provide a day ahead forecast. We compare our approaches with well-known alternative methods and show that the sub-region HTCNN requires fewer individual networks and achieves a forecast skill score of 40.2% reducing a statistically significant error by 6.5% compared to the best counterpart.

Definition Search Book Streamline Icon: https://streamlinehq.com
References (67)
  1. Short-term and regionalized photovoltaic power forecasting, enhanced by reference systems, on the example of Luxembourg, Renewable Energy 132 (2019) 455–470. URL: https://doi.org/10.1016/j.renene.2018.08.005. doi:10.1016/j.renene.2018.08.005.
  2. Improved satellite-derived PV power nowcasting using real-time power data from reference PV systems, Solar Energy 168 (2018) 118–139. URL: https://doi.org/10.1016/j.solener.2017.10.091. doi:10.1016/j.solener.2017.10.091.
  3. Operational solar forecasting for the real-time market, International Journal of Forecasting 35 (2019) 1499–1519. URL: https://doi.org/10.1016/j.ijforecast.2019.03.009. doi:10.1016/j.ijforecast.2019.03.009.
  4. Operational solar forecasting for grid integration: Standards, challenges, and outlook, Solar Energy 224 (2021) 930–937. URL: https://doi.org/10.1016/j.solener.2021.04.002. doi:10.1016/j.solener.2021.04.002.
  5. The value of day-ahead solar power forecasting improvement, Solar Energy 129 (2016) 192–203. doi:10.1016/j.solener.2016.01.049.
  6. Review of photovoltaic power forecasting, Solar Energy 136 (2016) 78–111. URL: http://dx.doi.org/10.1016/j.solener.2016.06.069. doi:10.1016/j.solener.2016.06.069.
  7. A short-term solar radiation forecasting system for the Iberian Peninsula. Part 2: Model blending approaches based on machine learning, Solar Energy 195 (2020) 685–696. URL: https://doi.org/10.1016/j.solener.2019.11.091. doi:10.1016/j.solener.2019.11.091.
  8. Regional forecasts and smoothing effect of photovoltaic power generation in Japan: An approach with principal component analysis, Renewable Energy (2014). doi:10.1016/j.renene.2014.02.018.
  9. Forecasting Regional Level Solar Power Generation Using Advanced Deep Learning Approach, Proceedings of the International Joint Conference on Neural Networks 2021-July (2021a). doi:10.1109/IJCNN52387.2021.9533458.
  10. Spatially Aggregated Photovoltaic Power Prediction Using Wavelet and Convolutional Neural Networks, in: 2021 International Joint Conference on Neural Networks (IJCNN), IEEE, 2021b, pp. 1–8. doi:10.1109/ijcnn52387.2021.9533513.
  11. A Data-driven Approach for Forecasting State Level Aggregated Solar Photovoltaic Power Production, Proceedings of the International Joint Conference on Neural Networks (2020). doi:10.1109/IJCNN48605.2020.9207594.
  12. Reconciling solar forecasts: Geographical hierarchy, Solar Energy (2017). doi:10.1016/j.solener.2017.02.010.
  13. Regional forecasts of photovoltaic power generation according to different data availability scenarios: A study of four methods, Progress in Photovoltaics: Research and Applications (2014). doi:10.1002/pip.2528.
  14. A Convolutional Neural Network for Regional Photovoltaic Generation Point Forecast, E3S Web of Conferences 185 (2020). doi:10.1051/e3sconf/202018501079.
  15. Improved convolutional neural network-based quantile regression for regional photovoltaic generation probabilistic forecast, IET Renewable Power Generation 14 (2020) 2712–2719. doi:10.1049/iet-rpg.2019.0949.
  16. Understanding forecast reconciliation, European Journal of Operational Research 294 (2021) 149–160. URL: https://doi.org/10.1016/j.ejor.2021.01.017. doi:10.1016/j.ejor.2021.01.017.
  17. A comparison of day-ahead photovoltaic power forecasting models based on deep learning neural network, Applied Energy 251 (2019) 113315. URL: https://doi.org/10.1016/j.apenergy.2019.113315. doi:10.1016/j.apenergy.2019.113315.
  18. Densely connected convolutional networks, in: Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017, volume 2017-Janua, 2017, pp. 2261–2269. doi:10.1109/CVPR.2017.243.
  19. Deep residual learning for image recognition, in: Proceedings of the IEEE conference on computer vision and pattern recognition, 2016, pp. 770–778.
  20. Multi-channel convolutional neural network for integration of meteorological and geographical features in solar power forecasting, Applied Energy 295 (2021) 117083. URL: https://doi.org/10.1016/j.apenergy.2021.117083. doi:10.1016/j.apenergy.2021.117083.
  21. Temporal convolutional networks for action segmentation and detection, Proceedings - 30th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2017 2017-Janua (2017) 1003–1012. doi:10.1109/CVPR.2017.113.
  22. WaveNet: A Generative Model for Raw Audio, arXiv preprint arXiv:1609.03499 (2016) 1–15. URL: http://arxiv.org/abs/1609.03499.
  23. An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling, arXiv preprint arXiv:1803.01271 (2018). URL: http://arxiv.org/abs/1803.01271.
  24. Temporal Convolutional Neural Networks for Solar Power Forecasting, Proceedings of the International Joint Conference on Neural Networks (2020). doi:10.1109/IJCNN48605.2020.9206991.
  25. History and trends in solar irradiance and PV power forecasting: A preliminary assessment and review using text mining, Solar Energy 168 (2018) 60–101. URL: https://doi.org/10.1016/j.solener.2017.11.023. doi:10.1016/j.solener.2017.11.023.
  26. Energy Forecasting: A Review and Outlook, IEEE Open Access Journal of Power and Energy 7 (2020) 376–388. doi:10.1109/OAJPE.2020.3029979.
  27. On recent advances in PV output power forecast, Solar Energy 136 (2016) 125–144. URL: http://dx.doi.org/10.1016/j.solener.2016.06.073. doi:10.1016/j.solener.2016.06.073.
  28. A review and evaluation of the state-of-the-art in PV solar power forecasting: Techniques and optimization, Renewable and Sustainable Energy Reviews 124 (2020) 109792. URL: https://doi.org/10.1016/j.rser.2020.109792. doi:10.1016/j.rser.2020.109792.
  29. M. Abdel-Nasser, K. Mahmoud, Accurate photovoltaic power forecasting models using deep LSTM-RNN, Neural Computing and Applications 31 (2019) 2727–2740. doi:10.1007/s00521-017-3225-z.
  30. X. Qing, Y. Niu, Hourly day-ahead solar irradiance prediction using weather forecasts by LSTM, Energy 148 (2018) 461–468. URL: https://doi.org/10.1016/j.energy.2018.01.177. doi:10.1016/j.energy.2018.01.177.
  31. Short-term solar power forecasting: Investigating the ability of deep learning models to capture low-level utility-scale Photovoltaic system behaviour, Applied Energy 285 (2021) 116395. URL: https://doi.org/10.1016/j.apenergy.2020.116395. doi:10.1016/j.apenergy.2020.116395.
  32. D. Korkmaz, SolarNet: A hybrid reliable model based on convolutional neural network and variational mode decomposition for hourly photovoltaic power forecasting, Applied Energy 300 (2021) 117410. URL: https://doi.org/10.1016/j.apenergy.2021.117410. doi:10.1016/j.apenergy.2021.117410.
  33. H. Acikgoz, A novel approach based on integration of convolutional neural networks and deep feature selection for short-term solar radiation forecasting, Applied Energy 305 (2022) 117912. URL: https://doi.org/10.1016/j.apenergy.2021.117912. doi:10.1016/j.apenergy.2021.117912.
  34. Digital elevation model-based convolutional neural network modeling for searching of high solar energy regions, Applied Energy 262 (2020) 114588. URL: https://doi.org/10.1016/j.apenergy.2020.114588. doi:10.1016/j.apenergy.2020.114588.
  35. Deep photovoltaic nowcasting, Solar Energy 176 (2018) 267–276. URL: https://doi.org/10.1016/j.solener.2018.10.024. doi:10.1016/j.solener.2018.10.024.
  36. Day-ahead photovoltaic power forecasting approach based on deep convolutional neural networks and meta learning, International Journal of Electrical Power and Energy Systems 118 (2020) 105790. URL: https://doi.org/10.1016/j.ijepes.2019.105790. doi:10.1016/j.ijepes.2019.105790.
  37. Photovoltaic power forecasting based LSTM-Convolutional Network, Energy 189 (2019) 116225. URL: https://doi.org/10.1016/j.energy.2019.116225. doi:10.1016/j.energy.2019.116225.
  38. P. Kumari, D. Toshniwal, Long short term memory–convolutional neural network based deep hybrid approach for solar irradiance forecasting, Applied Energy 295 (2021) 117061. URL: https://doi.org/10.1016/j.apenergy.2021.117061. doi:10.1016/j.apenergy.2021.117061.
  39. Assessing the performance of deep learning models for multivariate probabilistic energy forecasting, Applied Energy 285 (2021) 116405. URL: https://doi.org/10.1016/j.apenergy.2020.116405. doi:10.1016/j.apenergy.2020.116405.
  40. R. C. Deo, M. Şahin, Forecasting long-term global solar radiation with an ANN algorithm coupled with satellite-derived (MODIS) land surface temperature (LST) for regional locations in Queensland, Renewable and Sustainable Energy Reviews 72 (2017) 828–848. doi:10.1016/j.rser.2017.01.114.
  41. Regional PV power prediction for improved grid integration, Progress in Photovoltaics: Research and Applications (2011). doi:10.1002/pip.1033.
  42. A regional photovoltaic output prediction method based on hierarchical clustering and the mRMR criterion, Energies 12 (2019) 1–23. doi:10.3390/en12203817.
  43. An observational method for determining daily and regional photovoltaic solar energy statistics, Solar Energy 228 (2021) 12–26. URL: https://doi.org/10.1016/j.solener.2021.08.077. doi:10.1016/j.solener.2021.08.077.
  44. Bayesian parameterisation of a regional photovoltaic model – Application to forecasting, Solar Energy 188 (2019) 760–774. URL: https://doi.org/10.1016/j.solener.2019.06.053. doi:10.1016/j.solener.2019.06.053.
  45. Data-driven upscaling methods for regional photovoltaic power estimation and forecast using satellite and numerical weather prediction data, Solar Energy 158 (2017) 1026–1038. URL: https://doi.org/10.1016/j.solener.2017.09.068. doi:10.1016/j.solener.2017.09.068.
  46. T. Kim, J. Kim, A regional day-ahead rooftop photovoltaic generation forecasting model considering unauthorized photovoltaic installation, Energies 14 (2021). doi:10.3390/en14144256.
  47. Forecasting Solar Photovoltaic power production at the aggregated system level, in: 2014 North American Power Symposium, NAPS 2014, IEEE, 2014, pp. 1–6. doi:10.1109/NAPS.2014.6965389.
  48. A Solar Time Based Analog Ensemble Method for Regional Solar Power Forecasting, IEEE Transactions on Sustainable Energy 10 (2019) 268–279. doi:10.1109/TSTE.2018.2832634.
  49. A. E. M. Operator, Integrating utility-scale renewables and distributed energy resources in the swis, Australian Energy Market Operator: Melbourne, Australia (2019).
  50. 90–100% renewable electricity for the South West Interconnected System of Western Australia, Energy 122 (2017) 663–674. URL: http://dx.doi.org/10.1016/j.energy.2017.01.077. doi:10.1016/j.energy.2017.01.077.
  51. A. E. M. Operator, Renewable energy integration – swis update, Australian Energy Market Operator: Melbourne, Australia (2021).
  52. A review of behind-the-meter solar forecasting, Renewable and Sustainable Energy Reviews 160 (2022) 112224. URL: https://doi.org/10.1016/j.rser.2022.112224. doi:10.1016/j.rser.2022.112224.
  53. Short-term solar irradiation forecasting based on dynamic harmonic regression, Energy 84 (2015) 289–295. URL: http://dx.doi.org/10.1016/j.energy.2015.02.100. doi:10.1016/j.energy.2015.02.100.
  54. Very short-term photovoltaic power forecasting with cloud modeling: A review, Renewable and Sustainable Energy Reviews 75 (2017) 242–263. URL: http://dx.doi.org/10.1016/j.rser.2016.10.068. doi:10.1016/j.rser.2016.10.068.
  55. Recurrent Neural Networks for Time Series Forecasting: Current status and future directions, International Journal of Forecasting 37 (2021) 388–427. doi:10.1016/j.ijforecast.2020.06.008.
  56. Global models for time series forecasting: A simulation study, Pattern Recognition 124 (2022) 108441.
  57. P. Liashchynskyi, P. Liashchynskyi, Grid Search, Random Search, Genetic Algorithm: A Big Comparison for NAS, arXiv preprint arXiv:1912.06059 (2019) 1–11. URL: http://arxiv.org/abs/1912.06059.
  58. Automatic hourly solar forecasting using machine learning models, Renewable and Sustainable Energy Reviews 105 (2019) 487–498. URL: https://doi.org/10.1016/j.rser.2019.02.006. doi:10.1016/j.rser.2019.02.006.
  59. The cost of day-ahead solar forecasting errors in the united states, Solar Energy 231 (2022) 846–856.
  60. The value of solar forecasts and the cost of their errors: A review, Renewable and Sustainable Energy Reviews 189 (2024) 113915.
  61. Fedformer: Frequency enhanced decomposed transformer for long-term series forecasting, in: International Conference on Machine Learning, PMLR, 2022, pp. 27268–27286.
  62. A transformer-based multimodal-learning framework using sky images for ultra-short-term solar irradiance forecasting, Applied Energy 342 (2023) 121160.
  63. Are transformers effective for time series forecasting?, in: Proceedings of the AAAI conference on artificial intelligence, volume 37, 2023, pp. 11121–11128.
  64. Tsmixer: An all-mlp architecture for time series forecasting, arXiv preprint arXiv:2303.06053 (2023).
  65. Three types of incremental learning, Nature Machine Intelligence 4 (2022) 1185–1197.
  66. Identifying novelties and anomalies for incremental learning in streaming time series forecasting, Engineering Applications of Artificial Intelligence 123 (2023) 106326.
  67. A review of data-driven smart building-integrated photovoltaic systems: Challenges and objectives, Energy 263 (2023) 126082.
Citations (11)
View on Semantic Scholar

Summary

We haven't generated a summary for this paper yet.

Ai Sparkles Streamline Icon: https://streamlinehq.com Summarize Now