AN ARTIFICIAL NEURAL NETWORK STRATEGY TO IMPROVE WIND SPEED HINDCASTING

The veracious forecasts of wind waves are of great importance in ocean and coastal engineering applications as well as in managing oceanic resources. Accurate predictionsof wind waves with different lead times are necessary for a large scope of coastal and open ocean activities. Attempts to improve wave short-term forecasts based on artificial neural networks are reported by different researchers. Due to the lack of wind wave measured data, predicted wind waves data are used to determine wave climate in many regions. Nowadays spectral wind waves model are the most practical tool for wave prediction. Since the outputs of these models generally contain some errors, their results should modify based on the measured data. In this study, a new approach based on the error prediction of simulated wind data at the observing stations and distributing the errors in the computational domain was implemented for updating of the model outputs. To do so, wind simulation of meteorological model (WRF) was carried out over the Persian Gulf and the results were compared with the measured data. In this study, artificial neural networks are used to find out the relationship between the output of the WRF and measured data.