IDEAS home Printed from https://ideas.repec.org/a/gam/jeners/v16y2023i2p646-d1025957.html
   My bibliography  Save this article

Probabilistic Intraday PV Power Forecast Using Ensembles of Deep Gaussian Mixture Density Networks

Author

Listed:
  • Oliver Doelle

    (Siemens AG, Digital Industries—Data Visions, Siemenspromenade 1, 91058 Erlangen, Germany
    Faculty of Applied Computer Science, University of Augsburg, 86159 Augsburg, Germany)

  • Nico Klinkenberg

    (Faculty of Business Studies and Information Technology, Westphalian University of Applied Sciences, 46397 Bocholt, Germany)

  • Arvid Amthor

    (Siemens AG, Technology, Günther-Scharowsky-Str. 1, 91058 Erlangen, Germany)

  • Christoph Ament

    (Faculty of Applied Computer Science, University of Augsburg, 86159 Augsburg, Germany)

Abstract

There is a growing interest of estimating the inherent uncertainty of photovoltaic (PV) power forecasts with probability forecasting methods to mitigate accompanying risks for system operators. This study aims to advance the field of probabilistic PV power forecast by introducing and extending deep Gaussian mixture density networks (MDNs). Using the sum of the weighted negative log likelihood of multiple Gaussian distributions as a minimizing objective, MDNs can estimate flexible uncertainty distributions with nearly all neural network structures. Thus, the advantages of advances in machine learning, in this case deep neural networks, can be exploited. To account for the epistemic (e.g., model) uncertainty as well, this study applies two ensemble approaches to MDNs. This is particularly relevant for industrial applications, as there is often no extensive (manual) adjustment of the forecast model structure for each site, and only a limited amount of training data are available during commissioning. The results of this study suggest that already seven days of training data are sufficient to generate significant improvements of 23.9% in forecasting quality measured by normalized continuous ranked probability score (NCRPS) compared to the reference case. Furthermore, the use of multiple Gaussian distributions and ensembles increases the forecast quality relatively by up to 20.5% and 19.5%, respectively.

Suggested Citation

  • Oliver Doelle & Nico Klinkenberg & Arvid Amthor & Christoph Ament, 2023. "Probabilistic Intraday PV Power Forecast Using Ensembles of Deep Gaussian Mixture Density Networks," Energies, MDPI, vol. 16(2), pages 1-17, January.
  • Handle: RePEc:gam:jeners:v:16:y:2023:i:2:p:646-:d:1025957
    as

    Download full text from publisher

    File URL: https://www.mdpi.com/1996-1073/16/2/646/pdf
    Download Restriction: no

    File URL: https://www.mdpi.com/1996-1073/16/2/646/
    Download Restriction: no
    ---><---

    References listed on IDEAS

    as
    1. Philippe Lauret & Mathieu David & Hugo T. C. Pedro, 2017. "Probabilistic Solar Forecasting Using Quantile Regression Models," Energies, MDPI, vol. 10(10), pages 1-17, October.
    2. Hyndman, Rob J. & Koehler, Anne B. & Snyder, Ralph D. & Grose, Simone, 2002. "A state space framework for automatic forecasting using exponential smoothing methods," International Journal of Forecasting, Elsevier, vol. 18(3), pages 439-454.
    3. Alessandrini, S. & Delle Monache, L. & Sperati, S. & Cervone, G., 2015. "An analog ensemble for short-term probabilistic solar power forecast," Applied Energy, Elsevier, vol. 157(C), pages 95-110.
    4. David, Mathieu & Luis, Mazorra Aguiar & Lauret, Philippe, 2018. "Comparison of intraday probabilistic forecasting of solar irradiance using only endogenous data," International Journal of Forecasting, Elsevier, vol. 34(3), pages 529-547.
    5. Brusaferri, Alessandro & Matteucci, Matteo & Spinelli, Stefano & Vitali, Andrea, 2022. "Probabilistic electric load forecasting through Bayesian Mixture Density Networks," Applied Energy, Elsevier, vol. 309(C).
    6. van der Meer, D.W. & Widén, J. & Munkhammar, J., 2018. "Review on probabilistic forecasting of photovoltaic power production and electricity consumption," Renewable and Sustainable Energy Reviews, Elsevier, vol. 81(P1), pages 1484-1512.
    Full references (including those not matched with items on IDEAS)

    Citations

    Citations are extracted by the CitEc Project, subscribe to its RSS feed for this item.
    as


    Cited by:

    1. Sánchez-Balseca, Joseph & Pineiros, José Luis & Pérez-Foguet, Agustí, 2023. "Influence of environmental factors on the power produced by photovoltaic panels artificially weathered," Renewable and Sustainable Energy Reviews, Elsevier, vol. 188(C).

    Most related items

    These are the items that most often cite the same works as this one and are cited by the same works as this one.
    1. Le Gal La Salle, Josselin & Badosa, Jordi & David, Mathieu & Pinson, Pierre & Lauret, Philippe, 2020. "Added-value of ensemble prediction system on the quality of solar irradiance probabilistic forecasts," Renewable Energy, Elsevier, vol. 162(C), pages 1321-1339.
    2. Sun, Mucun & Feng, Cong & Zhang, Jie, 2020. "Probabilistic solar power forecasting based on weather scenario generation," Applied Energy, Elsevier, vol. 266(C).
    3. Luis Mazorra-Aguiar & Philippe Lauret & Mathieu David & Albert Oliver & Gustavo Montero, 2021. "Comparison of Two Solar Probabilistic Forecasting Methodologies for Microgrids Energy Efficiency," Energies, MDPI, vol. 14(6), pages 1-26, March.
    4. Elena Collino & Dario Ronzio, 2021. "Exploitation of a New Short-Term Multimodel Photovoltaic Power Forecasting Method in the Very Short-Term Horizon to Derive a Multi-Time Scale Forecasting System," Energies, MDPI, vol. 14(3), pages 1-30, February.
    5. Yang, Dazhi & van der Meer, Dennis, 2021. "Post-processing in solar forecasting: Ten overarching thinking tools," Renewable and Sustainable Energy Reviews, Elsevier, vol. 140(C).
    6. Alonso-Suárez, R. & David, M. & Branco, V. & Lauret, P., 2020. "Intra-day solar probabilistic forecasts including local short-term variability and satellite information," Renewable Energy, Elsevier, vol. 158(C), pages 554-573.
    7. Mitrentsis, Georgios & Lens, Hendrik, 2022. "An interpretable probabilistic model for short-term solar power forecasting using natural gradient boosting," Applied Energy, Elsevier, vol. 309(C).
    8. Gu, Bo & Shen, Huiqiang & Lei, Xiaohui & Hu, Hao & Liu, Xinyu, 2021. "Forecasting and uncertainty analysis of day-ahead photovoltaic power using a novel forecasting method," Applied Energy, Elsevier, vol. 299(C).
    9. John Boland & Sleiman Farah, 2021. "Probabilistic Forecasting of Wind and Solar Farm Output," Energies, MDPI, vol. 14(16), pages 1-15, August.
    10. Munkhammar, Joakim & van der Meer, Dennis & Widén, Joakim, 2021. "Very short term load forecasting of residential electricity consumption using the Markov-chain mixture distribution (MCM) model," Applied Energy, Elsevier, vol. 282(PA).
    11. Ricardo Aler & Javier Huertas-Tato & José M. Valls & Inés M. Galván, 2019. "Improving Prediction Intervals Using Measured Solar Power with a Multi-Objective Approach," Energies, MDPI, vol. 12(24), pages 1-19, December.
    12. Thomas Carrière & Rodrigo Amaro e Silva & Fuqiang Zhuang & Yves-Marie Saint-Drenan & Philippe Blanc, 2021. "A New Approach for Satellite-Based Probabilistic Solar Forecasting with Cloud Motion Vectors," Energies, MDPI, vol. 14(16), pages 1-19, August.
    13. Pierro, Marco & De Felice, Matteo & Maggioni, Enrico & Moser, David & Perotto, Alessandro & Spada, Francesco & Cornaro, Cristina, 2020. "Residual load probabilistic forecast for reserve assessment: A real case study," Renewable Energy, Elsevier, vol. 149(C), pages 508-522.
    14. Yang, Dazhi & Wu, Elynn & Kleissl, Jan, 2019. "Operational solar forecasting for the real-time market," International Journal of Forecasting, Elsevier, vol. 35(4), pages 1499-1519.
    15. Takahiro Takamatsu & Hideaki Ohtake & Takashi Oozeki & Tosiyuki Nakaegawa & Yuki Honda & Masahiro Kazumori, 2021. "Regional Solar Irradiance Forecast for Kanto Region by Support Vector Regression Using Forecast of Meso-Ensemble Prediction System," Energies, MDPI, vol. 14(11), pages 1-18, June.
    16. Hyndman, Rob J. & Ahmed, Roman A. & Athanasopoulos, George & Shang, Han Lin, 2011. "Optimal combination forecasts for hierarchical time series," Computational Statistics & Data Analysis, Elsevier, vol. 55(9), pages 2579-2589, September.
    17. Kourentzes, Nikolaos & Petropoulos, Fotios & Trapero, Juan R., 2014. "Improving forecasting by estimating time series structural components across multiple frequencies," International Journal of Forecasting, Elsevier, vol. 30(2), pages 291-302.
    18. Reza Fachrizal & Joakim Munkhammar, 2020. "Improved Photovoltaic Self-Consumption in Residential Buildings with Distributed and Centralized Smart Charging of Electric Vehicles," Energies, MDPI, vol. 13(5), pages 1-19, March.
    19. Alysha M De Livera, 2010. "Automatic forecasting with a modified exponential smoothing state space framework," Monash Econometrics and Business Statistics Working Papers 10/10, Monash University, Department of Econometrics and Business Statistics.
    20. Spiliotis, Evangelos & Makridakis, Spyros & Kaltsounis, Anastasios & Assimakopoulos, Vassilios, 2021. "Product sales probabilistic forecasting: An empirical evaluation using the M5 competition data," International Journal of Production Economics, Elsevier, vol. 240(C).

    Corrections

    All material on this site has been provided by the respective publishers and authors. You can help correct errors and omissions. When requesting a correction, please mention this item's handle: RePEc:gam:jeners:v:16:y:2023:i:2:p:646-:d:1025957. See general information about how to correct material in RePEc.

    If you have authored this item and are not yet registered with RePEc, we encourage you to do it here. This allows to link your profile to this item. It also allows you to accept potential citations to this item that we are uncertain about.

    If CitEc recognized a bibliographic reference but did not link an item in RePEc to it, you can help with this form .

    If you know of missing items citing this one, you can help us creating those links by adding the relevant references in the same way as above, for each refering item. If you are a registered author of this item, you may also want to check the "citations" tab in your RePEc Author Service profile, as there may be some citations waiting for confirmation.

    For technical questions regarding this item, or to correct its authors, title, abstract, bibliographic or download information, contact: MDPI Indexing Manager (email available below). General contact details of provider: https://www.mdpi.com .

    Please note that corrections may take a couple of weeks to filter through the various RePEc services.

    IDEAS is a RePEc service. RePEc uses bibliographic data supplied by the respective publishers.