Performance Comparison of Neural Network Training Algorithms for Load Forecasting in Smart Grids

Conference paper
Part of the Advances in Intelligent Systems and Computing book series (AISC, volume 656)

Abstract

Voltage control of distribution systems need load forecast. For improvement of efficiency and sustainability of the automated control of Smart Power Grids (SPG) the control system of the process should contain a subsystem of forecasting time series - load forecasting, as a parameter directly associated with voltage. The highest requirements applicate for the accuracy of short-term (day-week-month) and operational (within the current day) forecasts, as they determine the management of the current mode of operation of the SPG. The paper describes forecasting methods and concludes that using of artificial neural networks for this problem is preferable. It shows that for the complex real networks particle swarm method is faster and more accurate than traditional back propagation method. Finally the used model of short-term load forecasting (STLF) is described and tuning of all presented models is done. The paper concentrates of training methods of neural networks and uses “back propagation” algorithm and “particle swarm” algorithm for this purpose.

Keywords

Voltage control Load forecasting Artificial neural network 

References

  1. 1.
    Vanin, A., Aleshin, S., Nasirov, R., Novikov, D., Tulsky, V.: Investigation of voltage control at consumers connection points based on smart approach. MDPI Inf. 7(3), 42 (2016)Google Scholar
  2. 2.
    Rutkovskaja, D.: Neural Networks, Genetic Algorithms and Fuzzy Systems. Gorjachaja linija—Telekom, Moscow (2006)Google Scholar
  3. 3.
    Voronov, I.V.: Improving the efficiency of operation of the enterprises of power supply systems through the integrated use of smart grid and neural networks. Vestnik Kuzbasskogo gosudarstvennogo tehnicheskogo universiteta 2(90), 63–66 (2012)Google Scholar
  4. 4.
    Lis, R.A.: Application of data mining techniques to identify critical voltage control areas in power system. In: Proceedings of the Third International Conference on Computational Collective Intelligence: Technologies and Applications, ICCCI 2011, vol. I, pp. 152–162. ISBN 978-3-642-23934-2Google Scholar
  5. 5.
    Badar, I.: Comparison of conventional and modern load forecasting techniques based on artificial intelligence and expert systems. Int. J. Comput. Sci. 8(5), 504–513 (2011)Google Scholar
  6. 6.
    Roy, A.: Training artificial neural network using particle swarm optimization algorithm. Int. J. Adv. Res. Comput. Sci. Softw. Eng. 3(3), 430–434 (2013)Google Scholar
  7. 7.
    Haykin, S.: Neural Networks: A Comprehensive Course, 2nd edn. Viliams, Moscow (2008)MATHGoogle Scholar
  8. 8.
    Buhari, M.: Short-term load forecasting using artificial neural network. In: Proceedings of the International Multiconference of Engineers and Computer Scientists, vol. I, Hong Kong, 14–16 March 2012Google Scholar
  9. 9.
    Kennedy, J., Eberhart, R.: Particle swarm optimization. In: 2012 Proceedings of IEEE International Conference on Neural Networks, Perth, Australia. IEEE Service Center, Piscataway (1995)Google Scholar
  10. 10.
    Khothanzad, A., Hwang, R.C., Abaye, A., Maratukulam, D.: An adaptive modular artificial neural network hourly load forecaster and its implementation at electric utilities. IEEE Trans. Power Syst. 10(3), 1716–1722 (1995)CrossRefGoogle Scholar

Copyright information

© Springer International Publishing AG 2018

Authors and Affiliations

  • Robert Lis
    • 1
  • Artem Vanin
    • 2
  • Anastasiia Kotelnikova
    • 2
  1. 1.Wroclaw University of Science and TechnologyWroclawPoland
  2. 2.Moscow Power Engineering InstituteNational Research UniversityMoscowRussia

Personalised recommendations