Prediction of wave transmission over submerged reef of tandem breakwater using PSO-SVM and PSO-ANN techniques

Abstract

ABSTRACT Protection of the damaged breakwater from the high-intensity wave action has become inevitable. Submerged reef can act as a protective structure in reducing the wave action. Further, placed the reef structures on the sea side of a conventional rubble mound breakwater will reduce the effects of wave action. The conventional breakwater and reef structure combination is a tandem breakwater. Keeping in mind the end goal to decrease the complexities associated in model scaling, time constraints and cost in conducting the experiments, an attempt is made to apply soft computing techniques such as an Artificial Neural Network (ANN) and Support Vector Machine (SVM) to model various problems of real case scenario, where mathematical modelling is also difficult. In the present study, Particle Swarm Optimization (PSO) optimizes various parameters of ANN and SVM model in predicting the wave transmission over a submerged reef of the tandem breakwater. The performance of proposed hybrid models such as PSO-ANN and PSO-SVM is evaluated using statistical indices. The results show that PSO-SVM tool performs better in predicting the wave transmission compared to PSO-ANN.