MULTITEMPORAL SCALE HYDROGRAPH PREDICTION USING ARTIFICIAL NEURAL NETWORKS

摘要

An artificial neural network (ANN) provides a mathematically flexible structure to identify complex nonlinear relationship between inputs and outputs. A multilayer perceptron ANN technique with an error back propagation algorithm was applied to a multitime-scale prediction of the stage of a hydro-logically closed lake, Devils Lake (DL), and discharge of the Red River of the North at Grand Forks station (RR-GF) in North Dakota. The modeling exercise used 1 year (2002), 5 years (1998-2002), and 27 years (1975-2002) of data for the daily, weekly, and monthly predictions, respectively. The hydrometeo-rological data (precipitations P_((t)), P_((t-1)), P_((t-2)_, P_((t-3)), antecedent runoff/lake stage R_((t-1)), and air temperature T_((t)) were partitioned for training and for testing to predict the current hydro-graph at the selected DL and RR-GF stations. Performance of ANN was evaluated using three combinations of daily datasets (Input Ⅰ = P_((t)), P_((t-1)), P_((t-1)), P_((t-3)), T_((t)) and R_((t-1)); Input Ⅱ = Input-Ⅰ less P_((t)), P_((t-1)), P_((t-2)), P_((t-3)); and Input Ⅲ = Input-Ⅱ less T_((t))). Comparison of the mode output using Input Ⅰ data with the observed values showed average testing prediction efficiency (E) of 86 percent for DL basin and 46 percent for RR-GF basin, and higher efficiency for the daily than monthly simulations.