A knowledge-based expert system for unsteady open channel flow

Abstract

Many engineering applications for flood control, flood prediction, etc. require the solution of the unsteady open channel flow equations. However, the mathematical modelling and the numerical simulation techniques of unsteady flow in a natural river network are very complicated, requiring highly specialized knowledge and experience. Hydraulic engineers face many difficulties in the design process, such as how to choose a suitable numerical model, how to determine the parameters, how to simulate the natural river boundary conditions, and so on. Recent developments in artificial intelligence and computational hydraulics have made it possible to develop an expert system on hydrodynamics to assist engineers in analyzing unsteady open channel flow in river or tidal networks. This paper describes the problem which arises within the specific field of unsteady open channel flow, and presents a knowledge-based expert system built to assist the engineers in their design practices.