论文信息 - Stream-Based Short-Term Demand Forecasting Model Using ARIMA

Stream-Based Short-Term Demand Forecasting Model Using ARIMA

Short-term load forecasting for streaming load data is an important issue for power system planning, operation and control. Smart meters of Advanced Meter Infrastructure distributed around the distribution power grid produce streams of load at high-speed. The collected data can be characterized as a non-stationary continuous flow. A stream-based short-term demand forecasting model based on ARIMA is proposed. This method is used to forecast hourly electricity demand for next few days ahead. The performance of this methodology is validated with streaming data collected in real-time from the power grid.

Tao Liu | Jin Feng Yang | Rui Min Chen | Yong Xiao

[1] S. J. Kiartzis,et al. Short-term load forecasting using neural networks , 1995 .

[2] Robert I. Egbert,et al. Short term electric load forecasting via fuzzy neural collaboration , 2000 .

[3] P. McSharry,et al. A comparison of univariate methods for forecasting electricity demand up to a day ahead , 2006 .

[4] Ismet Erkmen,et al. A hybrid learning for neural networks applied to short term load forecasting , 2003, Neurocomputing.

[5] Chao-Ming Huang,et al. Analysis of an adaptive time-series autoregressive moving-average (ARMA) model for short-term load forecasting , 1995 .