Influence of the excitation force estimator methodology within a predictive controller framework on the overall cost of energy minimisation of a wave energy converter

A large amount of energy is freely roaming around the world each day, without
us being able to exploit it: wave energy is a largely untapped source of renewable energy, which can have a substantial influence in the future energy mix. The reason behind the inability of using this free resource is linked to the cost of the energy (CoE) produced from the different wave energy converters (WEC). The CoE from the different WECs is not yet comparable with other energy resources, due to a relative low efficiency coupled with the high structural costs. Within the sector a large effort has been addressed to optimize the WEC efficiency by means of different control strategies. In several articles [1, 2], it has been shown that with simple modications of the control law, the absorbed energy can be doubled or quadrupled. Whilst the improvement of the efficiency will increase the revenue of the machine, the application of an advance control strategy will most probably increase the loads exerted on the structure, leading to an increment of the structural cost. Therefore, the problem of minimising the CoE produced by a WEC is at least a 2Dproblem. In a previous article [3], the minimisation problem has been investigated witha sequential approach, and the results have been reported for different control strategies.
The Model Predictive Controller (MPC) seemed to have superior performance in terms of energy maximisation and loads on the structure, leading to a minimal CoE. But as presented in [3] the MPC was implemeted with perfect knowledge of the future loadtime series, which is physically not achivable. This article is an extension of the work presented in [3] with a closer focus on the infuence of the excitation force prediction on the capability of the MPC architecture. Different estimator models of the excitation forcetime series are benchmarked, and validated with laboratory results.