Latching control of wave energy converters using derivative-free optimization

We propose a control algorithm for wave energy converters (WECs) using derivative-free optimization (DFO), with relatively low computational burden and significant improvement in energy generation. The main idea is to consider the energy generation over a time horizon as a function of the latching time and formulate a suitable optimization problem. We demonstrate that the cost function can be discontinuous, which is the main reason for using DFO. A novel convergence proof is given for univariate discontinuous functions. We study two problem formulations, one that uses the past wave information to estimate the optimal latching time which can achieve at least 5 times more energy absorption than uncontrolled case. The other formulation obtains the optimal latching time by predicting the future wave information and results another 2-fold energy absorption over the first formulation.