AN EVOLUTIONARY COMPUTATION APPROACH TO SEDIMENT TRANSPORT MODELLING

摘要

The aim of this paper is to determine the capability of Evolutionary Computation (EC) method in predicting sediment suspension from the velocity field. EC is a robust search and optimisation technique based on the Darwinian principle of natural evolution, where the fittest will survive. There are different algorithms to simulate natural evolution in computer system. In this paper we use Genetic Programming (GP) methodology to explore a better correlation between the temporal pattern of fluid field and sediment transport. GP code developed by Langdon (1994) is modified for sediment transport modelling. Two datasets are used, one from numerical model results and other from field data (Sandy Duck), to evolve symbolic expression for sediment transport. The results from the numerical models are encouraging though GP found hard to find relationships from the field data.