Author(s): Ming Jin; Danny L. Fread
Linked Author(s): Ming Jin
Keywords: No Keywords
Abstract: Channel flood routing has long been of vital concern to man as he has sought to predict the characteristic features of a flood wave in his efforts to improve the transport of water through man-made or natural waterways and to determine necessary actions to protect life and property from the effects of flooding. Many channel routing models have been developed, and those based on the complete one-dimensional hydrodynamic equations (Saint-Venant) have found increasing use in the engineering community. Such dynamic channel routing models are based entirely on deterministic considerations. The outcome of a deterministic model is largely dependent on the accuracy of the model input, such as the specified hydraulic parameters within the mathematical equations used by the model, as well as boundary and initial conditions which need to be predetermined. Traditional deterministic methods cannot reflect the effects of possible inaccuracies in the equations, parameters, and boundary and initial conditions. When model results are applied to engineering practice, a margin of safety is often assigned to provide some degree of protection against the unknown effects. On the other hand, statistical models are receiving more attention because of their capability of reflecting the effects of uncertainties in the accuracy of the mathematical model, hydraulic parameters, and boundary and initial conditions. The Kalman filter is a statistical method that provides an updating technique to improve the simulation of unsteady flows for real-time river flood forecasting.
Year: 1993