Authors: WANG Tao, YANG Kai-lin, GUO Xin-lei, FU Hui

State key laboratory of simulation and key ulation of Water Cycle in River Basin
China Institute of Water Resources and Hydropower Research, Beijing 100038, China

Abstract: The Adaptive-Network-Based Fuzzy Inference System （ ANFIS ） has been applied to forecast the ice condition. By using a hybrid learning procedure, the ANFIS is employed to model the nonlinear functions, such as the forecast. By analyzing the ANFIS structure, the characteristic of the water temperature data and the factors related to forecast, and comparing the forecasted result based on changing their numbers and days, the type and number of membership functions and the forecast period are selected. The above study is applied to forecast the water temperature of the Shizuishan Hydrological Stations in the Yellow River for 4-year winter. The forecast results are in good agreement with the observed data.

Key words: ANFIS, ice condition, water temperature, Yellow River, membership function

Published in: Journal of Hydraulic Engineering, Vol. 43, No. 1, 2011

Article ID: 0559-9350 (2011) 01- 0112-06