Optimal operation of dielectric elastomer wave energy converters under harmonic and stochastic excitation

Dielectric elastomers are a promising technology for wave energy harvesting. An optimal system operation can allow maximizing the extracted energy and, simultaneously, reducing wear that would lead to a reduction in the wave harvester lifetime. We pursue a model-based optimization approach to identify optimal controls for wave energy harvesters based on dielectric elastomers. First, a direct method is used for time-discretization of the dielectric elastomer wave energy harvester in the optimal control problem. The two conflicting objectives are considered in a multiobjective optimization framework. Considering a periodic, sinusoidal wave excitation, the optimal solution shows turnpike properties for the optimal periodic mode of operation. However, since real wave motion is neither monochromatic nor predictable on longer time horizons, further extensions are pursued. First, we introduce a stochastic wave excitation. Second, an iterative model-predictive control scheme is designed. Due to multiple objectives, the control scheme has to include an automated adaption of the corresponding priorities. Here, we propose and evaluate a heuristic rule-based adaption in order to maintain the damage below target levels. The approach presented here might be used in the future to guarantee for autonomous operation of farms of wave energy harvesters.