Hour-Ahead Photovoltaic Output Forecasting Using Wavelet-ANFIS
Abstract
:1. Introduction
- A new method of combined PV power forecasting based on the decomposition at different resolution levels to optimize the weight determination.
- A study of different combinations of wavelet mother functions is proposed to find the most suitable for PV power time series.
- The combination of wavelet decomposition and ANFIS would have a strong learning ability and handle non-linear sequences regarding the chaotic and high non-linearity PV power output.
- The use of 2 and 3 h of data each day to forecast 10 min to 60 min ahead to increase the forecasting accuracy and reduce the computation time.
- Up to 30 shuffles are conducted to have initial random weighting and capture their diversity for a reliable and robust proposed combination. Moreover, the reconstructed wavelet features are used to calculate the accuracy of the proposed method.
2. Wavelet Transform
3. Adaptive Neuro-Fuzzy Inference System
4. Case Study
5. Results
5.1. Forecasting Accuracy Evaluation
5.2. Wavelet Study Results
5.3. Forecasting Results
6. Discussion
7. Conclusions
Author Contributions
Funding
Institutional Review Board Statement
Informed Consent Statement
Data Availability Statement
Acknowledgments
Conflicts of Interest
References
- Nestle, D.; Ringelstein, J. Integration of DER into distribution grid operation and decentralized energy management. Smart Grids Eur. 2009, 19, 3–25. [Google Scholar]
- Allen, J.; Halberstadt, A.; Powers, J.; El-Farra, N.H. An Optimization-Based Supervisory Control and Coordination Approach for Solar-Load Balancing in Building Energy Management. Mathematics 2020, 8, 1215. [Google Scholar] [CrossRef]
- Elavarasan, R.M.; Selvamanohar, L.; Raju, K.; Vijayaraghavan, R.R.; Subburaj, R.; Nurunnabi, M.; Khan, I.A.; Afridhis, S.; Hariharan, A.; Pugazhendhi, R.; et al. A Holistic Review of the Present and Future Drivers of the Renewable Energy Mix in Maharashtra, State of India. Sustainability 2020, 12, 6596. [Google Scholar] [CrossRef]
- Ahmed, R.; Sreeram, V.; Mishra, Y.; Arif, M.D. A review and evaluation of the state-of-the-art in PV solar power forecasting: Techniques and optimization. Renew. Sustain. Energy Rev. 2020, 124, 109792. [Google Scholar] [CrossRef]
- Xie, T.; Zhang, G.; Liu, H.; Liu, F.; Du, P. A Hybrid Forecasting Method for Solar Output Power Based on Variational Mode Decomposition, Deep Belief Networks and Auto-Regressive Moving Average. Appl. Sci. 2018, 8, 1901. [Google Scholar] [CrossRef] [Green Version]
- Mo, J.Y.; Jeon, W. How does energy storage increase the efficiency of an electricity market with integrated wind and solar power generation?—A case study of korea. Sustainability 2017, 9, 1797. [Google Scholar] [CrossRef] [Green Version]
- Kosko, B. Neural Networks and Fuzzy Systems, a Dynamical Systems Approach; Prentice-Hall: Hoboken, NJ, USA, 1991. [Google Scholar]
- Nespoli, A.; Ogliari, E.; Leva, S.; Massi Pavan, A.; Mellit, A.; Lughi, V.; Dolara, A. Day-Ahead Photovoltaic Forecasting: A Comparison of the Most Effective Techniques. Energies 2019, 12, 1621. [Google Scholar] [CrossRef] [Green Version]
- Mellit, A.; Saǧlam, S.; Kalogirou, S.A. Artificial neural network-based model for estimating the produced power of a photovoltaic module. Renew. Energy 2013. [Google Scholar] [CrossRef]
- Ouedraogo, F.B.; Zhang, X.-Y.; Chen, C.-R.; Lee, C.-Y. Hour-Ahead Power Generating Forecasting of Photovoltaic Plants Using Artificial Neural Networks Days Tuning. In Proceedings of the 2020 IEEE International Conference on Systems, Man, and Cybernetics (SMC), Toronto, ON, Canada, 11–14 October 2020; pp. 3874–3879. [Google Scholar] [CrossRef]
- Huang, C.-M.; Chen, S.-J.; Yang, S.-P.; Kuo, C.-J. One-day-ahead hourly forecasting for photovoltaic power generation using an intelligent method with weather-based forecasting models. IET Gener. Transm. Distrib. 2015, 9, 1874–1882. [Google Scholar] [CrossRef]
- Ogliari, E.; Niccolai, A.; Leva, S.; Zich, R.E. Computational Intelligence Techniques Applied to the Day Ahead PV Output Power Forecast: PHANN, SNO and Mixed. Energies 2018, 11, 1487. [Google Scholar] [CrossRef] [Green Version]
- Li, G.; Wang, H.; Zhang, S.; Xin, J.; Liu, H. Recurrent Neural Networks Based Photovoltaic Power Forecasting Approach. Energies 2019, 12, 2538. [Google Scholar] [CrossRef] [Green Version]
- Mellit, A.; Pavan, A.M.; Lughi, V. Deep learning neural networks for short-term photovoltaic power forecasting. Renew. Energy 2021, 172, 276–288. [Google Scholar] [CrossRef]
- Yang, H.; Huang, C.; Huang, Y.; Pai, Y. A Weather-Based Hybrid Method for 1-Day Ahead Hourly Forecasting of PV Power Output. IEEE Trans. Sustain. Energy 2014, 5, 917–926. [Google Scholar] [CrossRef]
- Osório, G.J.; Gonçalves, J.N.D.L.; Lujano-Rojas, J.M.; Catalão, J.P.S. Enhanced Forecasting Approach for Electricity Market Prices and Wind Power Data Series in the Short-Term. Energies 2016, 9, 693. [Google Scholar] [CrossRef]
- Dong, W.; Yang, Q.; Fang, X. Multi-Step Ahead Wind Power Generation Prediction Based on Hybrid Machine Learning Techniques. Energies 2018, 11, 1975. [Google Scholar] [CrossRef] [Green Version]
- Fernandez-Jimenez, L.; Muñoz-Jimenez, A.; Falces, A.; Mendoza-Villena, M.; Garcia-Garrido, E.; Lara-Santillan, P.; Zorzano-Alba, E.; Zorzano-Santamaria, P. Short-term power forecasting system for photovoltaic plants. Renew. Energy 2012, 44, 311–317. [Google Scholar] [CrossRef]
- Singla, P.; Duhan, M.; Saroha, S. A comprehensive review and analysis of solar forecasting techniques. Front. Energy 2021. [Google Scholar] [CrossRef]
- Antonanzas, J.; Osorio, N.; Escobar, R.; Urraca, R.; de Pison, F.M.; Antonanzas-Torres, F. Review of photovoltaic power forecasting. Sol. Energy 2016, 136, 78–111. [Google Scholar] [CrossRef]
- Chung, Y.T.; Tse, C.H. The forecast of the electrical energy generated by photovoltaic systems using neural network method. Int. Conf. Electron. Control. Eng. 2011, 2758–2761. [Google Scholar] [CrossRef]
- Basurto, N.; Arroyo, Á.; Vega, R.; Quintián, H.; Calvo-Rolle, J.L.; Herrero, Á. A hybrid intelligent system to forecast solar energy production. Comput. Electr. Eng. 2019, 78, 373–387. [Google Scholar] [CrossRef]
- Takagi, T.; Sugeno, T. Fuzzy identification of system and its applications to modelling and control. IEEE Trans. Syst. Man Cybern. 1985, 15, 116–132. [Google Scholar] [CrossRef]
- Yona, A.; Senjyu, T.; Funabashi, T.; Kim, C.H. Determination method of insolation prediction with fuzzy and applying neural network for long-term ahead PV power output correction. IEEE Trans. Sustain. Energy 2013, 4, 527–533. [Google Scholar] [CrossRef]
- Jang, J. ANFIS: Adaptive-network-based fuzzy inference systems. IEEE Trans. Syst. Man Cybern. 1993, 23, 665–685. [Google Scholar] [CrossRef]
- Yaïci, W.; Entchev, E. Adaptive Neuro-Fuzzy Inference System modelling for performance prediction of solar thermal energy system. Renew. Energy 2016, 86, 302–315. [Google Scholar] [CrossRef]
- Solgi, A.; Nourani, V.; Pourhaghi, A. Forecasting daily precipitation using hybrid model of wavelet-artificial neural network and comparison with adaptive neurofuzzy inference system (case study: Verayneh station, Nahavand). Adv. Civ. Eng. 2014, 2014, 279368. [Google Scholar] [CrossRef] [Green Version]
- Al-Hilfi, H.A.H.; Abu-Siada, A.; Shahnia, F. Combined ANFIS–Wavelet Technique to Improve the Estimation Accuracy of the Power Output of Neighboring PV Systems during Cloud Events. Energies 2020, 13, 1613. [Google Scholar] [CrossRef] [Green Version]
- Osório, G.J.; Lotfi, M.; Shafie-khah, M.; Campos, V.M.A.; Catalão, J.P.S. Hybrid Forecasting Model for Short-Term Electricity Market Prices with Renewable Integration. Sustainability 2018, 11, 57. [Google Scholar] [CrossRef] [Green Version]
- Seo, Y.; Kim, S.; Kisi, O.; Singh, V.P. Daily water level forecasting using wavelet decomposition and artificial intelligence techniques. J. Hydrol. 2015, 520, 224–243. [Google Scholar] [CrossRef]
- Stefenon, S.F.; Ribeiro, M.H.D.M.; Nied, A.; Mariani, V.C.; Coelho, L.D.S.; da Rocha, D.F.M.; Grebogi, R.B.; Ruano, A.E.D.B. Wavelet group method of data handling for fault prediction in electrical power insulators. Int. J. Electr. Power Energy Syst. 2020, 123, 106269. [Google Scholar] [CrossRef]
- Wu, L.; Cheng, J.Z.; Li, S.; Lei, B.; Wang, T.; Ni, D. FUIQA: Fetal ultrasound image quality assessment with deep convolutional networks. IEEE Trans. Cybernet. 2017, 47, 1336–1349. [Google Scholar] [CrossRef]
- Lin, F.; Lu, K.; Ke, T. Probabilistic wavelet fuzzy neural network based reactive power control for grid-connected three-phase PV system during grid faults. Renew. Energy 2016, 92, 437–449. [Google Scholar] [CrossRef]
- Mallat, S.G. A Theory for Multiresolution Signal Decomposition: The Wavelet Representation. IEEE Trans. Pattern Anal. Mach. Intell. 1989, 11, 674–693. [Google Scholar] [CrossRef] [Green Version]
- Semero, Y.K.; Zhang, D.; Zhang, J. A PSO-ANFIS based Hybrid Approach for Short Term PV Power Prediction in Microgrids. Electr. Power Compon. Syst. 2018, 46, 95–103. [Google Scholar] [CrossRef]
- Gao, W.; Moayedi, H.; Shahsavar, A. The feasibility of genetic programming and ANFIS in prediction energetic performance of a building integrated photovoltaic thermal (BIPVT) system. Sol. Energy 2019, 183, 293–305. [Google Scholar] [CrossRef]
- De Jesús, D.A.R.; Mandal, P.; Velez-Reyes, M.; Chakraborty, S.; Senjyu, T. Data Fusion Based Hybrid Deep Neural Network Method for Solar PV Power Forecasting. In Proceedings of the 51st North American Power Symposium (NAPS), Wichita, KS, USA, 13–15 October 2019. [Google Scholar] [CrossRef]
Forward | Backward | |
---|---|---|
, , | Fixed | Gradient Descent |
, , | Least-squares estimator | Fixed |
Signals | Node outputs | Error signals |
12 Input Patterns | 18 Input Patterns | ||||||
---|---|---|---|---|---|---|---|
Forecasting Time | |||||||
RMSE (kWh) | 10 Min | 30 Min | 60 Min | 10 Min | 30 Min | 60 Min | |
Wavelet mother function | Haar | 3.5965 × 10−4 | 3.8168 × 10−4 | 3.9033 × 10−3 | 3.7218 × 10−3 | 4.3430 × 10−3 | 5.9257 × 10−3 |
db2 | 2.3432 × 10−4 | 2.5778 × 10−4 | 2.7985 × 10−4 | 1.6028 × 10−4 | 2.1587 × 10−4 | 3.1221 × 10−4 | |
db3 | 2.2701 × 10−4 | 2.2992 × 10−4 | 2.5655 × 10−4 | 1.5331 × 10−4 | 2.0677 × 10−4 | 2.1517 × 10−4 | |
db5 | 1.9905 × 10−4 | 2.1077 × 10−4 | 2.3834 × 10−4 | 1.7060 × 10−4 | 1.8702 × 10−4 | 2.5660 × 10−4 | |
db8 | 2.2789 × 10−4 | 2.8417 × 10−4 | 2.9929 × 10−4 | 1.5424 × 10−4 | 1.9433 × 10−4 | 2.9495 × 10−4 | |
coif1 | 2.1413 × 10−4 | 2.3070 × 10−4 | 2.3842 × 10−4 | 1.7586 × 10−4 | 2.0476 × 10−4 | 2.3773 × 10−4 | |
coif2 | 1.8734 × 10−4 | 2.1022 × 10−4 | 2.2619 × 10−4 | 2.1707 × 10−4 | 2.4265 × 10−4 | 2.7823 × 10−4 | |
coif3 | 2.1759 × 10−4 | 2.3643 × 10−4 | 2.4524 × 10−4 | 1.7536 × 10−4 | 2.3334 × 10−4 | 2.8807 × 10−4 | |
coif5 | 2.3467 × 10−4 | 2.8025 × 10−4 | 3.2647 × 10−4 | 2.1136 × 10−4 | 2.7625 × 10−4 | 3.3846 × 10−4 | |
sym4 | 2.2402 × 10−4 | 2.3770 × 10−4 | 2.7530 × 10−4 | 1.5365 × 10−4 | 1.8049 × 10−4 | 2.1273 × 10−4 | |
sym6 | 2.6657 × 10−4 | 2.8111 × 10−4 | 2.9734 × 10−4 | 1.4681 × 10−4 | 2.2755 × 10−4 | 2.5019 × 10−4 | |
sym8 | 2.0610 × 10−4 | 2.4349 × 10−4 | 3.1602 × 10−4 | 1.9454 × 10−4 | 2.1813 × 10−4 | 2.9410 × 10−4 |
12 Input Patterns | 18 Input Patterns | ||||||||||
---|---|---|---|---|---|---|---|---|---|---|---|
Inputs | Forecasting Time | nRMSE (%) | MAPE (%) | MAE (kWh) | RMSE (kWh) | STD (kWh) | nRMSE (%) | MAPE (%) | MAE (kWh) | RMSE (kWh) | STD (kWh) |
Wavelet-ANFIS | 10 min | 4.4604 × 10−3 | 2.4436 × 10−3 | 1.0263 × 10−4 | 1.8734 × 10−4 | 1.8718 × 10−4 | 3.4954 × 10−3 | 1.8751 × 10−3 | 7.8754 × 10−5 | 1.4681 × 10−4 | 1.4681 × 10−4 |
30 min | 5.0183 × 10−3 | 2.8624 × 10−3 | 1.2022 × 10−4 | 2.1077 × 10−4 | 2.1048 × 10−4 | 4.2973 × 10−3 | 2.2672 × 10−3 | 9.5222 × 10−5 | 1.8049 × 10−4 | 1.8049 × 10−4 | |
60 min | 5.3727 × 10−3 | 3.0646 × 10−3 | 1.2871 × 10−4 | 2.2565 × 10−4 | 2.2550 × 10−4 | 5.0651 × 10−3 | 2.6137 × 10−3 | 1.0977 × 10−4 | 2.1273 × 10−4 | 2.1273 × 10−4 | |
ANFIS | 10 min | 1.5059 × 10−2 | 8.7031 × 10−3 | 3.6553 × 10−4 | 6.3248 × 10−4 | 6.3196 × 10−4 | 1.3616 × 10−2 | 6.5940 × 10−3 | 2.7695 × 10−4 | 5.7185 × 10−4 | 5.7216 × 10−4 |
30 min | 2.0272 × 10−2 | 1.2034 × 10−2 | 5.0543 × 10−4 | 8.5144 × 10−4 | 8.5028 × 10−4 | 1.5735 × 10−2 | 7.3180 × 10−3 | 3.0736 × 10−4 | 6.6085 × 10−4 | 6.6011 × 10−4 | |
60 min | 2.5262 × 10−2 | 1.4211 × 10−2 | 5.9684 × 10−4 | 1.0610 × 10−3 | 1.0603 × 10−3 | 2.3642 × 10−2 | 1.1736 × 10−2 | 4.9290 × 10−4 | 9.9294 × 10−4 | 9.9184 × 10−4 | |
ANN | 10 min | 3.8131 × 10−2 | 2.9975 × 10−2 | 1.2589 × 10−3 | 1.6015 × 10−3 | 1.6065 × 10−3 | 0.49695 | 0.35041 | 1.4717 × 10−2 | 2.0872 × 10−2 | 1.8214 × 10−2 |
30 min | 4.6381 × 10−2 | 3.7486 × 10−2 | 1.5744 × 10−3 | 1.9480 × 10−3 | 1.9539 × 10−3 | 0.52487 | 0.36033 | 1.5134 × 10−2 | 2.2045 × 10−2 | 1.8964 × 10−2 | |
60 min | 5.4582 × 10−2 | 3.9186 × 10−2 | 1.6458 × 10−3 | 2.2924 × 10−3 | 2.2951 × 10−3 | 1.0199 | 0.71746 | 3.0133 × 10−2 | 4.2837 × 10−2 | 3.8449 × 10−2 |
Publisher’s Note: MDPI stays neutral with regard to jurisdictional claims in published maps and institutional affiliations. |
© 2021 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY) license (https://creativecommons.org/licenses/by/4.0/).
Share and Cite
Chen, C.-R.; Ouedraogo, F.B.; Chang, Y.-M.; Larasati, D.A.; Tan, S.-W. Hour-Ahead Photovoltaic Output Forecasting Using Wavelet-ANFIS. Mathematics 2021, 9, 2438. https://doi.org/10.3390/math9192438
Chen C-R, Ouedraogo FB, Chang Y-M, Larasati DA, Tan S-W. Hour-Ahead Photovoltaic Output Forecasting Using Wavelet-ANFIS. Mathematics. 2021; 9(19):2438. https://doi.org/10.3390/math9192438
Chicago/Turabian StyleChen, Chao-Rong, Faouzi Brice Ouedraogo, Yu-Ming Chang, Devita Ayu Larasati, and Shih-Wei Tan. 2021. "Hour-Ahead Photovoltaic Output Forecasting Using Wavelet-ANFIS" Mathematics 9, no. 19: 2438. https://doi.org/10.3390/math9192438
APA StyleChen, C.-R., Ouedraogo, F. B., Chang, Y.-M., Larasati, D. A., & Tan, S.-W. (2021). Hour-Ahead Photovoltaic Output Forecasting Using Wavelet-ANFIS. Mathematics, 9(19), 2438. https://doi.org/10.3390/math9192438