Official URL: https://doi.org/10.1016/j.conengprac.2023.105778

Model invalidation is the process of testing assumptions of a dynamical model by comparing simulated responses with experimental data, considering any discrepancies as evidence that the model may be invalid. In this study, a model invalidation methodology is presented, to obtain robust control oriented models for wave energy converters (WECs). In particular, this methodology can deal separately with dynamical uncertainty and external noise in experimental data sets. To this end, considering linear system theory, this study proposes a methodology for building input–output data sets for WEC systems, via a two-stage approach. Model invalidation results are analysed statistically, and the practical implications of considering dynamical uncertainty in WEC system models are discussed in terms of control performance, specifically absorbed energy. As indicated by the analysis and results presented in this study, failure to include dynamic uncertainty in the analysis can lead to performance overestimation. The importance of a good dynamical description for accurate estimation of experimental control performance is highlighted. Finally, this study emphasises the need for closed-loop controllers for WEC systems that can simultaneously maximise energy and guarantee robust stability, an area currently lacking within the WEC literature.