An adaptive machine learning framework is established for an implicit determination of the performance degradation of a ship due to marine growth, i.e., biofouling. The framework is applied in a case study considering telemetry data of a cruise ship operating predominantly in the Caribbean Sea. The dataset encompasses seven years including three dry-docking intervals and several in-water cleaning events. The COVID-19 period receives special focus due to the drastic change in the operational profile. A main outcome of the study is a comparison of the derived performance estimate to the corresponding results of the industry standard ISO 19030. Additional aspects of the present study include the use of special regularization techniques for incremental machine learning and the increase of transparency through the implementation of prediction intervals indicating model uncertainty. Overall, it is found that the developed machine learning framework shows good agreement with the industry standard underlining its plausibility.
In this paper, full-scale data for two ships have been used for the comparison of five different added resistance methods. The effect of using separate wave spectra for wind waves and swell on performance prediction has been explored. The importance of the peak enhancement factor(γ) in the JONSWAP spectrum for added resistance computation has been studied. Simulation model including calm water resistance, added resistance and wind resistance has been used. Ships have been simulated in the same weather conditions and propeller speed as in the case of full-scale ships using different methods for added resistance. The performance of these methods has been quantified by comparing speed and power predictions with the full-scale data. The paper also discusses the challenges involved in using full-scale data for such a comparison because of difficulty in isolating the effect of added resistance in full-scale data. It was observed that three out of five methods were able to predict added resistance even in high waveheights. Even though these methods showed significantly different RAOs, its effect on speed and power prediction was minor. Simulation results were not sensitive to the choice of peak enhancement factor(γ) in the JONSWAP spectrum. There was minor improvement in results by using separate wave spectra for wind waves and swell instead of single wave spectrum for combined wind waves and swell.