Significant wave height modelling using a hybrid Wavelet-genetic Programming approach

In this paper, Genetic Programming (GP) based wavelet transform (WGP) was developed to forecast Significant Wave Height (SWH) in different lead times. The hourly SWH values for two buoy stations located in the North Atlantic Ocean were applied to train and validate the WGP model. For this purpose, the SWH main time series was decomposed into some subseries using wavelet transform and then decomposed time series were imported to GP model to forecast the SWH. Furthermore, GP approach was independently used to the same data set for comparison purposes. Performance of the WGP model was evaluated using correlation coefficient (R), Root Mean Square Error (RMSE), index of agreement (Ia) and Mean Absolute Error (MAE). The analysis proved that the model accuracy is highly depended on the decomposition levels. The obtained results showed that WGP model is able to forecast the SWH with a high reliability.