In Japan, regional- hourly- electricity demand data per each sector (resi-dential, commercial, transport, etc) is not publicly available hence esti-mating it by bottom up approach is really important not only for electricity production and distribution companies, but also urban planners. One of the conventional methods to implement this is using the intensity method. However, the problem of this approach is in that intensity value may change drastically due to the wide diffusion of new technologies such as electric vehicles or photovoltaics. Hence the present paper proposes a sim-ple Bayesian updating method of the intensity by assimilating the infor-mation of smart meter data. Also, forecasting of the intensity is performed using several recently proposed statistical methods, and its performances are empirically compared. The results support the use of seasonal-ARIMA model, multivariate dynamic linear model, and tbats model proposed by De Livera et al. (2011).
|出版ステータス||Published - 2013|
|イベント||13th International Conference on Computers in Urban Planning and Urban Management, CUMPUM 2013 - Utrecht, Netherlands|
継続期間: 2013 7月 2 → 2013 7月 5
|Conference||13th International Conference on Computers in Urban Planning and Urban Management, CUMPUM 2013|
|Period||13/7/2 → 13/7/5|
ASJC Scopus subject areas