Model predictive control of wave energy converters

No wave energy converter (WEC) technology has yet reached the commercial stage due to its high levelised cost of energy (LCOE). To reduce the LCOE by increasing the energy extraction performance of WECs, most of the existing studies are focusing on the optimisation of the devices’ hydrodynamic performance. These designs may not be practically optimised under realistic irregular sea conditions. Therefore, the present study proposes a novel and realistic control strategy, which can be widely applied to improve the efficiency of various WEC devices under realistic random wave conditions. To achieve such objective, we developed a real-time controller method to maximise energy absorption by implementing a wave force prediction model. The novelty of the present work lies in the development of a real-time model predictive control (MPC) integrated with a long short-term memory recurrent neural network (LSTM) wave prediction model, which is applicable to control WECs under irregular waves in a real-time manner. Experimental tests are conducted in a wave tank to evaluate the prediction performance of the proposed LSTM algorithm. We concluded that the MPC controller with implementation of the novel LSTM algorithm could double the power absorption of a WEC model under realistic irregular wave conditions.