%0 Conference Proceedings %T An Efficiency-Aware Model Predictive Control Strategy for a Heaving Buoy Wave Energy Converter %+ IFP Energies nouvelles (IFPEN) %A Tona, Paolino %A Nguyen, Hoaï-Nam %A Sabiron, Guillaume %A Creff, Yann %< avec comité de lecture %( 11th European Wave and Tidal Energy Conference %B 11th European Wave and Tidal Energy Conference - EWTEC 2015 %C Nantes, France %8 2015-09-06 %D 2015 %K wave energy converters %K model predictive control %K short-term wave prediction %K conversion efficiency %Z Computer Science [cs]/Automatic Control EngineeringConference papers %X In this paper we consider wave energy converters (WECs) of the heaving buoy type. We assume that the available power takeoff (PTO) mechanism is able to transform mechanical to electrical power (generator mode) and electrical to mechanical power (motor mode). In this context, a control system can be designed so as to take off as much energy as possible from the waves to the network, for a broad range of sea states. Among the implementable strategies surveyed in the literature, model predictive control (MPC) has the potential to lead to almost optimal performance, because of its ability to deal with many design objectives and performance criteria and to handle constraints on states and control inputs. However, a critical factor is often overlooked in the design and validation of MPC (as well as of most other WEC control strategies): the imperfect power conversion in the PTO. The MPC strategy presented here does take it into account, in the form of a conversion efficiency function in the criterion to be minimised. First, an optimal control problem is formulated and solved offline, which maximises the average net power output of the wave energy converter. Its performance is tested on five scenarios corresponding to different sea states and compared to that of a conventional control, based on proportional–integral (PI) velocity feedback. Then a nonlinear MPC strategy is introduced and its performance is tested on the same scenarios. It is shown that it yields roughly the same results as the optimal control, and can help harvesting up to 50% more energy than the reference PI control. %G English %2 https://ifp.hal.science/hal-01443855/document %2 https://ifp.hal.science/hal-01443855/file/EWTEC2015_Tona_et_al_postprint.pdf %L hal-01443855 %U https://ifp.hal.science/hal-01443855 %~ IFP