Deseasonalisation in Electricity Load Forecasting

  • Authors

    • Maria Elena Binti Nor
    • Mohd Saifullah Rusiman
    • Suliadi Firdaus Sufahani
    • Mohd Asrul Affendi Abdullah
    • Sathwinee A/P Bataraja
    • Sabariah Saharan
  • Box-Jenkins, Deseasonalisation, Exponential Smoothing, Forecast Accuracy
  • Nowadays, there is an increasing demand for electricity however overproduction of electricity lead to wastage. Therefore, electricity load forecasting plays a crucial role in operation, planning and maintenance of power system. This study was designed to investigate the effect of deseasonalisation on electricity load data forecasting. The daily seasonality in electricity load data was removed and the forecast methods were employed on both the seasonal data and non-seasonal data. Holt Winters method and Seasonal-Autoregressive Integrated Moving Average (SARIMA) methods were used on the seasonal data. Meanwhile, Simple and Double Exponential Smoothing methods as well as Autoregressive Integrated Moving Average (ARIMA) methods were used on the non-seasonal data. The error measurement that were used to assess the forecast performance were mean absolute error (MAE) and mean absolute percentage error (MAPE). The results revealed that both Exponential Smoothing method and Box-Jenkins method produced better forecast for deseasonalised data. Besides, the study proved that Box-Jenkins method was better in forecasting electricity load data for both seasonal and non-seasonal data.

  • References

    1. [1] Kermanshahi B & Iwamiya H (2002), Up to year 2020 load forecasting using neural nets. International Journal of Electrical Power & Energy Systems 24(9), 789–797.

      [2] Makridakis SG, & Wheelwright SC (1973), Forecasting Metholds for Management, Wiley, 0–23.

      [3] Nelson M, Hill T, Remus W, & O’Connor M (1999), Time series forecasting using neural networks: should the data be deseasonalized first? Journal of Forecasting 18(5), 359–367.

      [4] Oracle (2006), The Bayesian Approach to Forecasting The Bayesian Approach to Forecasting, (September).

      [5] Pardo A, Meneu V, & Valor E (2002), Temperature and seasonality influences on Spanish electricity, Energy Economics 24(1), 55-70

      [6] Nor ME, Safuan HM, Shab NFM, Abdullah MAA, Mohamad NAI, Lee MH (2017), Neural Network Versus Classical Time Series Forecasting Models, AIP Conference Proceedings, 842 (1)

      [7] Ahmad NFY, Khalid K, Rusiman MS, Kamardan MG, Roslan R, Che-Him N (2018), Analysis Monthly Import of Palm Oil Products Using Box-Jenkins Model, Journal of Physics: Conference Series, 995(1)

      [8] Nor ME, Safuan HM, Shab NFM, Abdullah MAA, Mohamad NAI, Lee MH (2016), Malaysia Tourism Demand Forecasting by Using Time Series Approaches, The Social Sciences 11 (12), 2938-2945,

      [9] Song H, Li G, Witt SF & Athanasopoulos G (2011), Forecasting tourist arrivals using time-varying parameter structural time series models. International Journal of Forecasting 27(3), 855-869.

      [10] Smeral E & Wüger M (2005), Does Complexity Matter? Methods for Improving Forecasting Accuracy in Tourism: The Case of Austria. Journal of Travel Research 44(1), 100-110.

      [11] Paretkar PS (2008), Short-Term Forecasting of Power Flows over Major Pacific Northwestern Interties : Using Box and Jenkins ARIMA Methodology. Energy Economics 24(1), 55–70.

      [12] Weron R, (2014), Electricity price forecasting: A review of the state-of-the-art with a look into the future. International Journal of Forecasting 30(4), 1030–1081.

      [13] Widén J, Lundh M, Vassileva I, Dahlquist E, Ellegård K, & Wäckelgård E (2009), Constructing load profiles for household electricity and hot water from time-use data-Modelling approach and validation. Energy and Buildings 41(7), 753–768.

  • Downloads

  • How to Cite

    Nor, M. E. B., Rusiman, M. S., Sufahani, S. F., Abdullah, M. A. A., Bataraja, S. A., & Saharan, S. (2018). Deseasonalisation in Electricity Load Forecasting. International Journal of Engineering & Technology, 7(4.30), 448-450.