An Optimized Nonlinear Grey Bernoulli Model for Forecasting the Electricity Consumption During COVID-19 Pandemic: A Case for Turkey


Konyalıoğlu A. K., Beldek T., Özcan T.

International Conference on Intelligent and Fuzzy Systems, INFUS 2021, İstanbul, Turkey, 24 - 26 August 2021, vol.307, pp.649-656 identifier

  • Publication Type: Conference Paper / Full Text
  • Volume: 307
  • Doi Number: 10.1007/978-3-030-85626-7_76
  • City: İstanbul
  • Country: Turkey
  • Page Numbers: pp.649-656
  • Keywords: COVID-19 pandemic, Electricity consumption, Forecasting, Genetic algorithm, Nonlinear grey Bernoulli model, Parameter optimization
  • Istanbul Technical University Affiliated: Yes

Abstract

© 2022, The Author(s), under exclusive license to Springer Nature Switzerland AG.Electricity is an unstorable and crucial resource, which is supposed to be planned and managed carefully. Accurate estimation of the electricity consumption can guide efficient energy policy making. On the other hand, to influence policy makers, prediction models need to be credible. It is a fact that electricity consumption has dramatically increased during COVID-19 (SARS-CoV-2) pandemic because of lockdown periods in Turkey on which directly affects electricity consumption. In this study, grey forecasting models are used to predict the monthly electricity consumption of Turkey during COVID-19 pandemic period. Furthermore, in this study, the Turkey’s electricity consumption monthly data for the period 2017–2020 are taken from Energy Market Regulatory Authority database. Firstly, an optimized NGBM is used to predict Turkey’s electricity consumption. In this model, the parameters of NGBM are optimized using genetic algorithm (GA). Then, GM (1,1), which is an optimized model and linear regression model (LRM) are used to evaluate forecasting performance of optimized NGBM model. Analysis results illustrate that by the aid of the optimized NGBM model, robust results can be obtained and the parameter optimization with rolling mechanism significantly increases the original NGBM’s performance.