Control-oriented wave surface elevation forecasting strategies: Experimental validation and comparison

• Published in: IFAC-PapersOnLine 2023-01, Vol.56 (2), p.10898-10903
• Main Authors: Papini, Guglielmo, Peña-Sanchez, Yerai, Pasta, Edoardo, Faedo, Nicolás
• Format: Article
• Language: English
• Subjects: AR model; Forecast; recursive least squares; wave energy; wave spectrum
• ISSN: 2405-8963; 2405-8963
• DOI: 10.1016/j.ifacol.2023.10.772

Optimal control strategies are a key development step towards commercialization of wave energy converters (WECs). Most of these rely on optimization routines to find a suitable control action to maximize WEC power production. Nevertheless, most of these solutions make use of device dynamical models, with the free-surface elevation as the external (uncontrollable) input, effectively representing the incoming wave field. Consequently, predictive strategies, such as model predictive control, strongly depend on the availability of future wave information, and hence suitable forecasters are commonly used to ‘restore’ the causality of the optimization problem. Motivated by the intrinsic requirement of suitable forecasting strategies within optimal WEC control, this study provides a validation and comparison of different algorithms, including adaptive and non-adaptive techniques, based on experimental data. The paper focuses on the adaptability of each algorithm, which must be capable to fit properly each wave surface elevation signal, thus not affecting the optimality condition by providing poor prediction results.