Abbass Yeganeh Bakhtiary - Department of Civil Engineering, Iran University of Science & Technology, Tehran, Narmak, Iran
Majid Zeinali - Department of Civil Engineering, Iran University of Science & Technology, Tehran, Narmak, Iran
Reza Valipour - Department of Civil Engineering, Iran University of Science & Technology, Tehran, Narmak, Iran
Tarkao Yamashita - RCDE, Disaster Prevention Research Institute (DPRI), Kyoto University, Kyoto, Japan

Design of port and harbor facilities highly depends on the nearshore hydrodynamics. Usually, the significant wave characteristics along with the most severe condition of the nearshore currents based on the field measurements is considered for the design purpose. On the other hand, optimal measurement cost and accurate numerical estimation depends on some key parameters of current velocity. The main objective of present paper is to describe an approach to more accurate and effective prediction of current velocity through key parameterization of observed data based on Root Mean Square (RMS). The procedure has significantly improved by using artificial neural networks due to ANN's capability in high functioning with rapid computation to solve the high nonlinearity and multi-variables systems.

Artificial Neural Network (ANN), Feed Forward (FF), Root Mean Square (RMS)