Short-Term Prediction of an Artificial Neural Network in an Oscillating Water Column

An Oscillating Water Column (OWC) is a promising wave energy device due to its obvious advantages over many other wave energy converters: There are no moving components in the sea water. Though the bottom-fixed OWC have been quite successful in several practical applications, the projection of a massive wave energy production and the availability of wave energy resources have pushed the OWC applications from near shore to deeper water regions where the floating OWC are a better choice. In an OWC, the reciprocating air flow driving an air turbine to generate electricity is a random process. In such a working condition, a single design/operation point is nonexistent. To increase the energy extraction, and optimise the performance of the device, a system capable of controlling the air turbine rotation speed is desirable. For this purpose, this paper presents a short-term prediction of the random process using an artificial neural network (ANN), aiming to provide near-future information for the control system. In this research, the ANN is explored and tuned for a better prediction of the airflow and the device motions. It is found that, by carefully constructing the ANN platform and optimising the relevant parameters, the ANN is capable of predicting the random process a few steps ahead of the real time with good accuracy.