Synthetic River Flow Time Series Generator for Dispatch and Spot Price Forecast

R.A. Flores (Sweden), J. Szczupak, and L. Pinto (Brazil)

Keywords

Modeling, planning, simulation, dispatch, spot price and forecast.

Abstract

This paper proposes a signal processing approach for generating new synthetic river flow time series as a support for planning and spot price forecasting. The proposed method offers an alternative synthetic representation for the original river flow time series, preserving correlations, basic statistics and seasonality. The method takes into consideration deterministic, periodic and non periodic components such as those due to El NiƱo Southern Oscillation phenomenon. The new synthetic time series presents high correlation factors with the original river flow time series, permitting its use as a possible practical replacement to the classical method of sorting historical river flow time series, as it is presently common in the planning and spot price expected value forecasting.

Important Links:



Go Back