Application of artificial neural networks to forecasting of river ice condition
Tao Wang · Journal of Hydraulic Engineering · 2005
The artificial neural networks(ANNs)theory is applied to establish the mathematical model for forecasting the ice condition of the Yellow River in section from Ningxia Autonomous Region to Inner Mongolia Autonomous Region.The Levenbeerg-Marquart algorithm is introduced to substitute the traditional BP algorithm to improve the computation of the model.The model can be used to forecast not only the date of ice run,freeze-up and break-up but also the temperature of water,ice flow density,ice jams as well as ice dam.The forecasted river ice condition in winter season of 2004 is in good agreement with the observation data.