Abstract
The performance of an oscillating water column (OWC) wave energy converter depends on many factors, such as the wave conditions, the tidal level and the coupling between the chamber and the air turbine. So far most studies have focused on either the chamber or the turbine, and in some cases the influence of the tidal level has not been dealt with properly. In this work a novel approach is presented that takes into account all these factors. Its objective is to develop a virtual laboratory which enables to determine the pneumatic efficiency of a given OWC working under specific conditions of incident waves (wave height and period), tidal level and turbine damping. The pneumatic efficiency, or efficiency of the OWC chamber, is quantified by means of the capture factor, i.e. the ratio between the absorbed pneumatic power and the available wave energy. The approach is based on artificial intelligence—in particular, artificial neural networks (ANNs). The neural network architecture is chosen through a comparative study involving 18 options. The ANN model is trained and, eventually, validated based on an extensive campaign of physical model tests carried out under different wave conditions, tidal levels and values of the damping coefficient, representing turbines of different specifications. The results show excellent agreement between the ANN model and the experimental campaign. In conclusion, the new model constitutes a virtual laboratory that enables to determine the capture factor of an OWC under given wave conditions, tidal levels and values of turbine damping, at a lower cost and in less time than would be required for conventional laboratory tests.