Predicting sea wave height using Symbiotic Organisms Search (SOS) algorithm

In the present study, the Symbiotic Organisms Search (SOS) algorithm was used to predict the wave height in two time ranges, including hourly and daily; accordingly, the wave height data of the statistical years 2007-2011 and the data of February and March 2006 were used for daily and hourly predictions, respectively. Results of the SOS were compared with those of Particle Swarm Optimization (PSO) and Imperialist Competitive Algorithm (ICA) algorithms and intelligent methods including Support Vector Regression (SVR), Artificial Neural network (ANN) and Simulating Waves Nearshore (SWAN) dynamic model. The results indicated that the SOS had better performance in both hourly and daily time ranges, so that R2 (coefficient of determination), RMSE (Root Mean Square Error), d (Willmott's index of agreement), and MAE (Mean Absolute Error) were obtained equal to 0.9513, 0.0692, 0.9874, and 0.0472, respectively, for hourly prediction and 0.8607, 0.1707, 0.9615, and 0.1088, respectively, for daily prediction. Furthermore, the hybrid SWAN-SOS model was applied for the areas lacking enough observations and it was compared with the other methods. Comparing the obtained results indicated better performance of SOS and SWAN-SOS model in predicting the wave height for this region.