摘要

Artificial neural network (ANN)s have been applied to solve a variety of problems related to the coastal and ocean areas over a period of last one and a half decades. Most of these studies have involved estimation or forecasting of environmental parameters, structural loads and responses. The aim in such studies were to carry out a cause-effect modeling, temporal or spatial mapping, and property integration and resolution. ANN models have been generally found to outperform the traditional empirical, statistical or numerical models to a smaller or larger extent. There are however certain grey areas where the modeling and its usefulness needs to be improved. This paper takes a stock of all such past works and summarizes author's experience in working with ANNs for the benefit of future researchers.

  • 出版日期2010-12