Parameters Optimization of Oyster Wave Energy Converter Devices Using Artificial Neural Network Model

A supervised machine learning model based on artificial neural network (ANN) is used for the optimization of various parameters of an Oyster device. Oyster is a surface-piercing flap type device which works in nearshore environment and extracts energy by pitching about a horizontal axis parallel to the seabed. In this study, the shape parameters associated with the Oyster's flap and foundation will be taken as input parameters to the ANN model and the efficiency of the Oyster device will be considered as output parameter. Initially, the problem will be solved using 3D boundary Element Method. For the training and testing database, Latin Hypercube Sampling technique will be used to obtain highly stratified input samples. The developed ANN model will be used to generate a large dataset consisting of combinations of various input parameters and associated output parameter. These large datasets will be used for the optimization purpose to achieve maximum efficiency from the Oyster device.