TY  - THES
T1  - Real-time hourly forecasting with armax models with application for real-time flood operation at Angat Reservoir
A1  - Vergel, John Manuel B.
LA  - English
UL  - https://ds.mainlib.upd.edu.ph/Record/UP-99796217609940712
AB  - This paper presents a real-time forecasting model for hourly operation of the multi-purpose Angat Reservoir for real-time flood control operation. The operation of the surface water for flood control is essential and vital for minimizing the impact of flood during the real-time flood events. The forecasting model provides an anticipatory approach for 24-hours planning horizon and hourly forecasting for the next 24 hours at every hour. The forecasting models are formulated for each time steps ahead that lead to a total of 24 forecasting models. the techniques proposed here are the linear stochastic autoregressive model (AR). The forecasting models using the considered methods are assumed as a lumped system. Formal Box and Jenkins modeling procedures are presented such as model identification, parameter estimation, diagnostic check and forecast error properties to test the robustness of the forecasting model. AR model is a univariate streamflow series. ARMAX model are considered as multivariate model including inputs such as past inflow, rainfall for Anagt basin station, and synoptic station for Tayabas, Daet and Legaspi outside the basin. Simulation of the model based on forecasting are presented at Angat Reservoir for real-time flood operation. The results show that ARMAX model performed better compare to AR model. The ARMAX model gives quantitative information for the operation for the release of the spillway gates.
CN  - LG 995 2010 E63 V47
KW  - Hydrological forecasting.
KW  - Hydrological models.
ER  -