Author(s): S. M. Mahbubur Rahman; Yonas B. Dibike
Linked Author(s):
Keywords: Numerical modelling techniques; Anrtificial neural networks; Computing resources
Abstract: Numerical modelling techniques are now well established methods for describing physical processes occurring in the aquatic environment. The development in information technology (IT) over the last decade has presented opportunities of extended computational ability together with improved data manipulation, storage and retrieval. As a result, the numerical models are now being used more extensively in the management, design and operation of water based assets[1]. However, despite their many successes, numerical modelling techniques are still to prove their practical usefulness in many areas of real time control and forecasting problems. The reason behind this is that in many areas of applications to complex flow systems the demand on computing time are of a magnitude that is far from acceptable. In recent years, new modelling techniques, which are mostly sub-symbolic, have demonstrated their potential in simulating physical processes based on measured or simulated data. This paper presents a case study on the Jamuna River in Bangladesh. The study attempts to show the ability of one of the sub-symbolic modelling techniques, namely Artificial Neural Network (ANN), in simulating hydrodynamic processes in the river with an accuracy comparable to physicallybased models and with an acceptable demand on computing time and other resources.
Year: 1999