Benchmarking physics-informed machine learning-based short term PV-power forecasting tools

被引:33
作者
Pombo, Daniel Vazquez [1 ,2 ]
Bacher, Peder [3 ]
Ziras, Charalampos [1 ]
Bindner, Henrik W. [1 ]
Spataru, Sergiu, V [4 ]
Sorensen, Poul E. [5 ]
机构
[1] Tech Univ Denmark DTU, Dept Elect Engn, Frederikborsvej 399, DK-4000 Roskilde, Denmark
[2] Vattenfall AB, R&D Strateg Dev, Evenemangsgatan 13C, S-16956 Solna, Sweden
[3] Tech Univ Denmark DTU, Dept Comp Sci, Frederikborsvej 399, DK-4000 Roskilde, Denmark
[4] Tech Univ Denmark DTU, Photovolta Mat & Syst, Frederikborsvej 399, DK-4000 Roskilde, Denmark
[5] Tech Univ Denmark DTU, Dept Wind Energy, Frederikborsvej 399, DK-4000 Roskilde, Denmark
关键词
Machine learning; Physics informed; Solar PV; PV power forecasting; SOLAR; OPTIMIZATION;
D O I
10.1016/j.egyr.2022.05.006
中图分类号
TE [石油、天然气工业]; TK [能源与动力工程];
学科分类号
0807 ; 0820 ;
摘要
Uncertainty is one of the core challenges posed by renewable energy integration in power systems, especially for solar photovoltaic (PV), given its dependence on meteorological phenomena. This has motivated the development of numerous forecasting tools, recently focused on physics informed machine learning (ML). Virtually, every paper claims to provide better accuracy than the previous, yet the replicability of these studies is very low, motivating unfair, or erroneous comparisons. This paper reviews and compares the most relevant ML-methods identified in the literature (Random Forest, Support Vector Regression, Convolutional Neural Networks, Long-Short Term Memory and a Hybrid of the last two) with two statistical methods: persistence and an Semi-Parametric Auto-Regressive model. Furthermore, we propose a methodology to integrate a PV-performance model in ML models to forecast power several hours ahead with 5-min resolution. A basic dataset including power production and meteorological measurements is expanded with physics-informed features that capture the relationship between weather and PV operational state, while keeping strong correlation towards the intrinsic feature. This allows the models to learn about the physical interdependence of different features, potentially yielding a higher accuracy than conventional methods. Then, we also propose a physics-informed feature selection to tighten the search-space of the best performer. A case study of a PV array in Denmark is used for validation using both the original and expanded datasets. Results show how the best ML models consistently used physics-informed features in all cases. (C) 2022 The Author(s). Published by Elsevier Ltd.
引用
收藏
页码:6512 / 6520
页数:9
相关论文
共 50 条
[41]   A defect-based physics-informed machine learning framework for fatigue finite life prediction in additive manufacturing [J].
Salvati, Enrico ;
Tognan, Alessandro ;
Laurenti, Luca ;
Pelegatti, Marco ;
De Bona, Francesco .
MATERIALS & DESIGN, 2022, 222
[42]   Short-term rainfall forecasting using machine learning-based approaches of PSO-SVR, LSTM and CNN [J].
Adaryani, Fatemeh Rezaie ;
Mousavi, S. Jamshid ;
Jafari, Fatemeh .
JOURNAL OF HYDROLOGY, 2022, 614
[43]   Physics-informed machine learning based RANS turbulence modeling convection heat transfer of supercritical pressure fluid [J].
Cao, Yuli ;
Xu, Ruina ;
Jiang, Peixue .
INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER, 2023, 201
[44]   Physics-informed machine learning based RANS turbulence modeling convection heat transfer of supercritical pressure fluid [J].
Cao, Yuli ;
Xu, Ruina ;
Jiang, Peixue .
INTERNATIONAL JOURNAL OF HEAT AND MASS TRANSFER, 2023, 201
[45]   A Framework of Using Machine Learning Approaches for Short-Term Solar Power Forecasting [J].
Munawar, Usman ;
Wang, Zhanle .
JOURNAL OF ELECTRICAL ENGINEERING & TECHNOLOGY, 2020, 15 (02) :561-569
[46]   Evaluating machine learning in short-term forecasting time series of solar power [J].
Fraccanabbia, Naylene ;
Mariani, Viviana Cocco .
REVISTA BRASILEIRA DE COMPUTACAO APLICADA, 2021, 13 (02) :105-112
[47]   A Framework of Using Machine Learning Approaches for Short-Term Solar Power Forecasting [J].
Usman Munawar ;
Zhanle Wang .
Journal of Electrical Engineering & Technology, 2020, 15 :561-569
[48]   Machine-Learning based methods in short-term load forecasting [J].
Guo W. ;
Che L. ;
Shahidehpour M. ;
Wan X. .
Electricity Journal, 2021, 34 (01)
[49]   Hybrid model for microgrid short term load forecasting based on machine learning [J].
Khayat, Ahmed ;
Kissaoui, Mohammed ;
Bahatti, Lhoussaine ;
Raihani, Abdelhadi ;
Errakkas, Khalid ;
Atifi, Youness .
IFAC PAPERSONLINE, 2024, 58 (13) :527-532
[50]   Combined machine learning forecasting method for short-term power load based on the dynamic weight adjustment [J].
Hu, Xue ;
Tang, Xiafei ;
Zhang, Qichun ;
Chen, Zhongwei ;
Zhang, Yun .
ENERGY REPORTS, 2023, 9 :866-873