TY  - THES
T1  - State estimation techniques for monthly streamflow forecasting with application to the Angat Reservoir
A1  - Villa, Rhoel C.
LA  - English
UL  - https://ds.mainlib.upd.edu.ph/Record/UP-99796217604568903
AB  - The research study illustrates the use of state estimation techniques for monthly streamflow forecasting. Time series models of the ARMA and ARMAX types are cast within the state-space framework of the Kalman filter and used to forecast the one-month ahead inflow at the Angat Reservoir. The forecasts are two types, pure state estimation and combined state-parameter estimation. The former assumes that the model parameters are time invariant while the converse is assumed in the latter. The performance of the identified forecasting models are compared and evaluated through the use of the mean, standard deviation, correlation coefficient, standard forecast error and mean of peak forecast error of the forecasted series. The model selected for the system dynamics to be used within the Kalman filter framework is a combination of the AR (1) and AR (3) models.
NO  - Thesis (M.S. Civil Civil Engineering)--University of the Philippines Diliman.
NO  - Typescript.
CN  - LG 995 1987 E6 V54
KW  - Hydrological forecasting.
KW  - Angat Reservoir.
KW  - Stream measurements.
ER  -