Fault localization method for power distribution systems based on gated graph neural networks


Fault localization is a key task on power systems operation and maintenance. When it comes to distribution networks, the problem is especially challenging due to the non-homogeneous characteristics and unique topology of each feeder. This paper presents a method based on gated graph neural network for automatic fault localization on distribution networks. The method aggregates problem data in a graph, where the feeder topology is represented by the graph links and nodes attributes can encapsulate any selected information such as operated devices, electrical characteristics and measurements at the point. The main advantage of the proposed solution is that it is immune to network reconfiguration and allows the use of a single trained model on multiple feeders. An experiment was conducted with faults simulated on 10 different feeders, all of them based on actual distribution feeders. The results shows that the model is able to generalize the correlations learned on training to correctly predict the fault region in most cases, even on a feeder it has not seen before.

This is a preview of subscription content, access via your institution.

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6


  1. 1.

    Aslan Y (2012) An alternative approach to fault location on power distribution feeders with embedded remote-end power generation using artificial neural networks. Electr Eng. https://doi.org/10.1007/s00202-011-0218-2

    Article  Google Scholar 

  2. 2.

    Cho K, van Merrienboer B, Gulcehre C, Bahdanau D, Bougares F, Schwenk H, Bengio Y (2014) Learning phrase representations using rnn encoder-decoder for statistical machine translation. arXiv:1406.1078

  3. 3.

    Das R, Madani V, Aminifar F, McDonald J, Venkata SS, Novosel D, Bose A, Shahidehpour M (2015) Distribution automation strategies: evolution of technologies and the business case. IEEE Trans Smart Grid 6(4):2166–2175. https://doi.org/10.1109/TSG.2014.2368393

    Article  Google Scholar 

  4. 4.

    Dashtdar M (2018) Fault location in distribution network based on fault current analysis using artificial neural. Network 1:18–32

    Google Scholar 

  5. 5.

    Dehghani F, Nezami H (2013) A new fault location technique on radial distribution systems using artificial neural network. In: 22nd international conference and exhibition on electricity distribution (CIRED 2013), pp 1–4. https://doi.org/10.1049/cp.2013.0697

  6. 6.

    Ekanayake J, Liyanage K, Wu J, Yokoyama A, Jenkins N (2012) Smart Grid: Technology and Applications, 1st edn. John Wiley & Sons Ltd. https://doi.org/10.1002/9781119968696

  7. 7.

    EPRI: The Open Distribution System Simulator (OpenDSS). https://www.epri.com/#/pages/sa/opendss?lang=en

  8. 8.

    Farias PE, de Morais AP, Rossini JP, Cardoso G (2018) Non-linear high impedance fault distance estimation in power distribution systems: a continually online-trained neural network approach. Electr Power Syst Res. https://doi.org/10.1016/j.epsr.2017.11.018

    Article  Google Scholar 

  9. 9.

    Gori M, Monfardini G, Scarselli F (2005) A new model for learning in graph domains. In: Proceedings of the international joint conference on neural networks. https://doi.org/10.1109/IJCNN.2005.1555942

  10. 10.

    Gururajapathy SS, Mokhlis H, Illias HA (2017) Fault location and detection techniques in power distribution systems with distributed generation: a review. https://doi.org/10.1016/j.rser.2017.03.021

    Article  Google Scholar 

  11. 11.

    IEEE (2014) IEEE Guide for Determining Fault Location on AC transmission and distribution lines IEEE power and energy society. In: IEEE Std C37.114-2014 (revision of IEEE Std C37.114-2004). https://doi.org/10.1109/IEEESTD.2015.7024095

  12. 12.

    Jarventausta P, Verho P, Partanen J (1994) Using fuzzy sets to model the uncertainty in the fault location process of distribution networks. IEEE Trans Power Deliv 10(1109/61):296278

    Google Scholar 

  13. 13.

    Kingma DP, Ba J (2014) Adam: a method for stochastic optimization. arXiv:1412.6980

  14. 14.

    Li Y, Tarlow D, Brockschmidt M, Zemel R (2015) Gated graph sequence neural networks. arXiv:1511.05493

  15. 15.

    Meira PCM (2018) DSS python. https://github.com/dss-extensions/dss_python

  16. 16.

    Northcote-Green J, Wilson R (2017). Control and automation of electrical power distribution systems. https://doi.org/10.1201/9781315221465

  17. 17.

    Pineda FJ (1987) Generalization of back-propagation to recurrent neural networks. Phys Rev Lett. https://doi.org/10.1103/PhysRevLett.59.2229

    MathSciNet  Article  Google Scholar 

  18. 18.

    Saha M, Das R, Verho P, Novosel D (2002) Review of fault location techniques for distribution systems. In: Power systems and communications infrastructures for the future. Beijing, China

  19. 19.

    Scarselli F, Gori M, Tsoi AC, Hagenbuchner M, Monfardini G (2009) The graph neural network model. IEEE Trans Neural Netw. https://doi.org/10.1109/TNN.2008.2005605

    Article  Google Scholar 

  20. 20.

    Yang F, Donde V, Li Zhao Wang Zhenyuan (2012) Advances in power distribution system management technology. In: Wang L (ed) Modeling and control of sustainable power systems: towards smarter and greener electric grids, pp 155–191. Springer, Berlin. https://doi.org/10.1007/978-3-642-22904-6_6

Download references


The authors would like to thank Pró-Reitoria de Pesquisa (PRPq) of the Universidade Federal de Minas Gerais (UFMG), FAPEMIG, CNPq, and CAPES; Companhia Energética de Minas Gerais (CEMIG) for providing data from its distribution system used in the experiments, and Eng. Ezequiel Campos Pereira of CEMIG’s Electrical System Planning and Expansion Engineering department, for the support with the simulations.

Author information



Corresponding author

Correspondence to Jonas Teixeira de Freitas.

Ethics declarations

Conflict of interest

The authors declare that they have no conflict of interest.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

de Freitas, J.T., Coelho, F.G.F. Fault localization method for power distribution systems based on gated graph neural networks. Electr Eng (2021). https://doi.org/10.1007/s00202-021-01223-7

Download citation


  • Fault location
  • Electric power systems
  • Distribution networks
  • Graph neural network (GNN)
  • Node prediction