• M. Linke, T. Meßmer, G. Micard, and G. Schubert, "Power grid operation in distribution grids with convolutional neural networks," Dec. 2024

    Abstract

    The efficient and reliable operation of power grids is of great importance for ensuring a stable and uninterrupted supply of electricity. Traditional grid operation techniques have faced challenges due to the increasing integration of renewable energy sources and fluctuating demand patterns caused by the electrification of the heat and mobility sector. This paper presents a novel application of convolutional neural networks in grid operation, utilising their capabilities to recognise fault patterns and finding solutions. Different input data arrangements were investigated to reflect the relationships between neighbouring nodes as imposed by the grid topology. As disturbances we consider voltage deviations exceeding 3% of the nominal voltage or transformer and line overloads. To counteract, we use tab position changes of the transformer stations as well as remote controllable switches installed in the grid. The algorithms are trained and tested on a virtual grid based on real measurement data. Our models show excellent results with test accuracy of up to 99.06% in detecting disturbances in the grid and suggest a suitable solution without performing time-consuming load flow calculations. The proposed approach holds significant potential to address the challenges associated with modern grid operation, paving the way for more efficient and sustainable energy systems.

    Contact

    Manuela Linke - manuela.linke@htwg-konstanz.de

    BibLaTex

     

    @article{LINKE2025100169,
    title = {Power grid operation in distribution grids with convolutional neural networks},
    journal = {Smart Energy},
    volume = {17},
    pages = {100169},
    year = {2025},
    issn = {2666-9552},
    doi = {https://doi.org/10.1016/j.segy.2024.100169},
    url = {https://www.sciencedirect.com/science/article/pii/S266695522400039X},
    author = {Manuela Linke and Tobias Meßmer and Gabriel Micard and Gunnar Schubert},
    keywords = {Power grid operation, Convolutional neural network, Artificial intelligence, Smart grids, Resilient energy system, Sector coupling},
    abstract = {The efficient and reliable operation of power grids is of great importance for ensuring a stable and uninterrupted supply of electricity. Traditional grid operation techniques have faced challenges due to the increasing integration of renewable energy sources and fluctuating demand patterns caused by the electrification of the heat and mobility sector. This paper presents a novel application of convolutional neural networks in grid operation, utilising their capabilities to recognise fault patterns and finding solutions. Different input data arrangements were investigated to reflect the relationships between neighbouring nodes as imposed by the grid topology. As disturbances we consider voltage deviations exceeding 3% of the nominal voltage or transformer and line overloads. To counteract, we use tab position changes of the transformer stations as well as remote controllable switches installed in the grid. The algorithms are trained and tested on a virtual grid based on real measurement data. Our models show excellent results with test accuracy of up to 99.06% in detecting disturbances in the grid and suggest a suitable solution without performing time-consuming load flow calculations. The proposed approach holds significant potential to address the challenges associated with modern grid operation, paving the way for more efficient and sustainable energy systems.}
    }
    
  • M. Arpogaus, J. Montalbano, M. Linke, and G. Schubert, "Probabilistic real-time grid operation management of future distribution grids with high penetration of renewable generators and electrical vehicles based on artificial intelligence," Jun. 2022

    Abstract

    In this paper, we propose a novel method for real-time control of electric distribution grids with a limited number of measurements. The method copes with the changing grid behaviour caused by the increasing number of renewable energies and electric vehicles. Three AI based models are used. Firstly, a probabilistic forecasting estimates possible scenarios at unobserved grid nodes. Secondly, a state estimation is used to detect grid congestion. Finally, a grid control suggests multiple possible solutions for the detected problem. The best countermeasures are then detected by evaluating the systems stability for the next time-step.

    Contact

    Marcel Arpogaus - marcel.arpogaus@htwg-konstanz.de

    BibLaTex

     

    @INPROCEEDINGS{9841826,
      author={Arpogaus, M. and Montalbano, J. and Linke, M. and Schubert, G.},
      booktitle={CIRED Porto Workshop 2022: E-mobility and power distribution systems}, 
      title={Probabilistic real-time grid operation management of future distribution grids with high penetration of renewable generators and electrical vehicles based on artificial intelligence}, 
      year={2022},
      volume={2022},
      pages={147-151},
      doi={10.1049/icp.2022.0681}}
    
  • M. Arpogaus, M. Voss, B. Sick, M. Nigge-Uricher, and O. Dürr, “Short-Term Density Forecasting of Low-Voltage Load using Bernstein-Polynomial Normalizing Flows,” Apr. 2022

    Preprint on arXiv

    Abstract

    The transition to a fully renewable energy grid requires better forecasting of demand at the low-voltage level to increase efficiency and ensure reliable control. However, high fluctuations and increasing electrification cause huge forecast variability, not reflected in traditional point estimates. Probabilistic load forecasts take future uncertainties into account and thus allow more informed decision-making for the planning and operation of low-carbon energy systems. We propose an approach for flexible conditional density forecasting of short-term load based on Bernstein polynomial normalizing flows, where a neural network controls the parameters of the flow. In an empirical study with 363 smart meter customers, our density predictions compare favorably against Gaussian and Gaussian mixture densities. Also, they outperform a non-parametric approach based on the pinball loss for 24h-ahead load forecasting for two different neural network architectures.

    Contact

    Marcel Arpogaus - marcel.arpogaus@htwg-konstanz.de

    BibLaTex

     

    @unpublished{Arpogaus2022a,
      title = {Short-{{Term Density Forecasting}} of {{Low-Voltage Load}} Using {{Bernstein-Polynomial Normalizing Flows}}},
      author = {Arpogaus, Marcel and Voss, Marcus and Sick, Beate and Nigge-Uricher, Mark and Dürr, Oliver},
      date = {2022-04-29},
      eprint = {2204.13939},
      eprinttype = {arxiv},
      primaryclass = {cs, stat},
      archiveprefix = {arXiv}
    }
  • M. Arpogaus, M. Voss, B. Sick, M. Nigge-Uricher, and O. Dürr, “Probabilistic Short-Term Low-Voltage Load Forecasting using Bernstein-Polynomial Normalizing Flows,” Jun. 2021

    Open Access

    Abstract

    The transition to a fully renewable energy grid requires better forecasting of demand at the low-voltage level. However, high fluctuations and increasing electrification cause huge forecast errors with traditional point estimates. Probabilistic load forecasts take future uncertainties into account and thus enables various applications in low-carbon energy systems. We propose an approach for flexible conditional density forecasting of short-term load based on Bernstein-Polynomial Normalizing Flows where a neural network controls the parameters of the flow. In an empirical study with 363 smart meter customers, our density predictions compare favorably against Gaussian and Gaussian mixture densities and also outperform a non-parametric approach based on the pinball loss for 24h-ahead load forecasting for two different neural network architectures.

    Contact

    Marcel Arpogaus - marcel.arpogaus@htwg-konstanz.de

    BibLaTex

     

    @inproceedings{Arpogaus2021,
      title = {Probabilistic {{Short-Term Low-Voltage Load Forecasting}} Using {{Bernstein-Polynomial Normalizing Flows}}},
      booktitle = {{{ICML}} 2021 Workshop on Tackling Climate Change with Machine Learning},
      author = {Arpogaus, Marcel and Voss, Marcus and Sick, Beate and Nigge-Uricher, Mark and Dürr, Oliver},
      date = {2021-06-23},
      url = {https://www.climatechange.ai/papers/icml2021/20}
    }
  • M. Linke, T. Messmer, G. Micard, A. Wenzel, G. Schubert, M. Kindl, A. Minde. „Artificial neural network based decision support system for the present power grid accounting for the successful integration of renewable energy sources such as pv systems” 2019

    Abstract

    We present an alternative approach to grid management in low voltage grids by the use of artificial intelligence. The developed decision support system is based on an artificial neural network (ANN). Due to the fast reaction time of our system, real time grid management will be possible. Remote controllable switches and tap changers in transformer stations are used to actively manage the grid infrastructure. The algorithm can support the distribution system operators to keep the grid in a safe state at any time. Its functionality is demonstrated by a case study using a virtual test grid. The ANN achieves a prediction rate of around 90% for the different grid management strategies. By considering the four most likely solutions proposed by the ANN, the prediction rate increases to 98.8%, with a 0.1 second increase in the running time of the model.

    Contact

    Manuela Linke - manuela.linke@htwg-konstanz.de

    BibLaTex

     

    @inproceedings{Linke2019,
        author = {M. Linke and T. Messmer and G. Micard and A. Wenzel and G. Schubert and M. Kindl and A. Minde},
        title = {Artificial neural network based decision support system for the present power grid accounting for the successful integration of renewable energy sources such as pv systems},
        booktitle = {Proceedings of the 36th European PV Solar Energy Conference and Exhibition},
        year = {2019}
    }
  • A. Wenzel, M. Linke, T. Meßmer, G. Micard, G. Schubert, A. Minde and M. Kindl. “Innovative grid optimization approach based on artificial neural networks". 2019.

    Abstract

    We present an innovative decision support system (DSS) for distribution system operators (DSO) based on an artificial neural network (ANN). A trained ANN has the ability to recognize problem patterns and to propose solutions that can be implemented directly in real time grid management. The principle functionality of this ANN based optimizer has been demonstrated by means of a simple virtual electrical grid. For this grid, the trained ANN predicted the solution minimizing the total line power dissipation in 98 percent of the cases considered. In 99 percent of the cases, a valid solution in compliance with the specified operating conditions was found. First ANN tests on a more realistic grid, calibrated with household load measurements, revealed a prediction rate between 88 and 90 percent depending on the optimization criteria. This approach promises a faster, more cost-efficient and potentially secure method to support distribution system operators in grid management.

    Contact

    Manuela Linke - manuela.linke@htwg-konstanz.de

    BibLaTex

     

    @inproceedings{Wenzel2019,
         author = {A. Wenzel and M. Linke and T. Meßmer and G. Micard and G. Schubert and A. Minde and M. Kindl},
         title = {Innovative grid optimization approach based on artificial neural networks},
         booktitle = {Proceedings of the 2019 IEEE PES Innovative Smart Grid Technologies Conference Europe (ISGT-Europe)},
         year = {2019}
    }
  • M. Linke, T. Messmer, G. Micard, A. Wenzel, G. Schubert, M. Kindl und A. Minde. „Netzoptimierungstool auf Basis künstlicher neuronaler Netze für den intelligenten Echtzeitbetrieb des Stromnetzes auf Verteilnetzebene“. 2019.

    Abstract

    Das hier vorgestellte Netzoptimierungstool kann dem Verteilnetzbetreiber bei einem Störfall im Netz in Echtzeit eine Lösung zur Steuerung seiner Betriebsmittel vorschlagen. Dadurch kann das bestehende Netz optimal genutzt werden und ein kostenintensiver Netzausbau im Mittel- und Niederspannungsnetz verringert oder sogar verhindert werden. Als Grundlage für den Netzoptimierer dient ein künstliches neuronales Netz (KNN). Zum Training des KNN wurden Störfälle generiert, die auf reellen Erzeugungs- und Lastprofilen aus dem CoSSMic-Projekt basieren. Für jeden Störfall wurde aus allen möglichen und sinnvollen Netzkonfigurationen eine optimierte Netztopologie anhand von Lastflussberechnungen ermittelt. Durch die Variation der Stufenschalter der Transformatoren und der Stellungen aller installierten Schalter im Netz wurde berechnet, wie der Stromfluss gelenkt werden muss, damit keines der Betriebsmittel die zulässigen Belastungsgrenzen mehr überschreitet. Für ein virtuelles Testnetz konnte mit einem trainierten KNN zu 90 Prozent die optimale Lösung des jeweiligen Störfalls erkannt werden. Durch die Anwendung der N-Best Methode konnte die Vorhersagewahrscheinlichkeit auf annähernd 99 Prozent erhöht werden.

    Contact

    Manuela Linke - manuela.linke@htwg-konstanz.de

    BibLaTex

     

    @inproceedings{Linke2019b,
         author = {M. Linke and T. Messmer and G. Micard and A. Wenzel and G. Schubert and M. Kindl and A. Minde},
         title = {Netzoptimierungstool auf Basis künstlicher neuronaler Netze für den intelligenten Echtzeitbetrieb des Stromnetzes auf Verteilnetzebene},
         booktitle = {Tagungsband des 26. REGWA Symposium (Nutzung Regenerativer Energiequellen und Wasserstofftechnik 2019, ISBN 978-3-9817740-4-7)},
         year = {2019}
    }