Periodic ARMA models applied to weekly streamflow forecasts

Author

Maceira, M.E.P. ; Damazio, J.M. ; Ghirardi, A.O. ; Dantas, H.M.

Author_Institution

CEPEL, UERJ, Rio de Janeiro, Brazil

fYear

1999

Firstpage

86

Abstract

This paper presents a weekly streamflow forecasting model based on linear ARMA (p, q) models, considering both periodic and nonperiodic models. For each week, fifty possible models are automatically analyzed. The best modeling and parameter estimation are chosen based on the minimum square mean forecast error of the whole time series. The proposed model, which has been validated by the Brazilian Multi-Utility Hydrological Studies Working Group, is illustrated in case studies with several hydraulic plants of the Brazilian Southern, Southeastern, North and Northeastern systems.

Keywords

autoregressive moving average processes; hydroelectric power stations; hydrothermal power systems; parameter estimation; power system planning; time series; Brazilian Multi-Utility Hydrological Studies Working Group; hydraulic plants; hydrothermal system; inflow forecast; linear ARMA models; minimum square mean forecast error; nonperiodic models; operation planning; p, q models; parameter estimation; periodic ARMA models; stochastic models; time series; weekly streamflow forecasts; Costs; Parameter estimation; Predictive models; Production planning; Production systems; Reservoirs; Rivers; Stochastic processes; Topology; Uncertainty;

fLanguage

English

Publisher

ieee

Conference_Titel

Electric Power Engineering, 1999. PowerTech Budapest 99. International Conference on

Conference_Location

Budapest, Hungary

Print_ISBN

0-7803-5836-8

Type

conf

DOI

10.1109/PTC.1999.826517

Filename

826517