Improvement of performance of short term electricity demand model with meteorological parameters

  • Kamal Chapagain School of Information Communication Technology, SIIT, Thammasat University, PathumThani-12000, Thailand
  • Tomonori Soto Graduate School of Environmental Science, Hokkaido University, Sapporo, Japan
  • Somasak Kittipiyakul School of Information Communication Technology, SIIT, Thammasat University, PathumThani-12000, Thailand

Abstract

The accuracy of short term electricity demand forecasting is essential for operation and trading activities on energy market. This paper considers a parsimonious forecasting model to explain the importance of sophisticated weather parameters for hourly electricity demand forecasting. Temperature is the major factor that directly influence electricity demand, but what about the affect of other weather factors such as relative humidity, wind speed etc. on short term electricity demand forecasting, is the prime research question and this paper analyzed it quantitatively. We demonstrate three different multiple linear models including auto-regressive moving average ARMA (2,6) models with and without some exogenous weather variables to compare with performance for Hokkaido Prefecture, Japan. Since, Bayesian approach is used to estimate the weight of each variables with Gibbs sampling, it generates the weight of coefficients in terms of distribution as our interest. The performance of each models for complete one year out sample prediction shows that the average improvement of hourly forecast by 1 to 2 % can be achieve by including such weather factors.

Downloads

Download data is not yet available.
Abstract
1918
PDF
251

Author Biography

Kamal Chapagain, School of Information Communication Technology, SIIT, Thammasat University, PathumThani-12000, Thailand

Kamal Chapagain received B.E degree in Electronics and Communication Engineering from Pokhara Engineering College, Pokhara, Nepal in 2003 and M.Sc engineering from Pulchowk campus in 2013. He received excellent foreign scholarship (EFS) to conduct PhD at school of Information Communication Technology, Srindhron International Institute of Technology Thailand, and MEXT scholarship as long term exchange to Hokkaido University, Hokkaido, Japan. Currently, he is PhD candidate and his research is on short term electricity demand forecasting for Thailand.

Published
2018-12-14
How to Cite
Chapagain, K., Soto, T., & Kittipiyakul, S. (2018). Improvement of performance of short term electricity demand model with meteorological parameters. Kathford Journal of Engineering and Management, 1(1), 15-22. https://doi.org/10.3126/kjem.v1i1.22016
Section
Articles