Open Access Open Access  Restricted Access Subscription or Fee Access

Bridge Pier Scour Prediction by Multi-Objective Optimization Using the Genetic Algorithm

I. KIM, M. Y. FARD, A. CHATTOPADHYAY

Abstract


Local scour is a critical problem for the safety of bridges. In this study, a new bridge-pier-scour model is proposed by multi-objective optimization using genetic algorithm based on the traditional regression model and the inductive method. A model evaluation function has been established (HEC-18 equation, Froehlich equation, and GEP model). Two groups of field datasets; bridge scour management data base (358 sets) and FHWA (110 sets), are used to evaluate the models. Based on the model evaluation function and statistical evaluation, the new model is shown more efficient and less failed in predicting the scour depth at bridge piers.

Full Text:

PDF