Statistical models for improving significant wave height predictions in offshore operations

Abstract

Installation and maintenance strategies regarding offshore wind farm operations involve extensive logistics. The main focus is the right temporal and spatial placement of personnel and equipment, while taking into account forecasted meteorological and ocean conditions. For these operations to be successful, weather windows characterized by certain permissive wave conditions are of enormous importance, whereas unforeseen events result in high cost and risk of safety. Numerical modelling of waves, water levels and current related variables has been used extensively to forecast ocean conditions. To account for the inherited model uncertainty, several error modelling techniques can be implemented for the numerical model forecasts to be corrected. In this study, various Bayesian Network (BN) models are incorporated, in order to enhance the accuracy of the significant wave height predictions and to be compared with other techniques, in conditions resembling the real-time nature of the application. The implemented BN models differ in terms of training and structure and provide overall the most satisfying performance. Supplementary, it is shown that the BN models illustrate significant advantages as both quantitative and conceptual tools, since they produce estimates for the underlying uncertainty of the phenomena, while providing information about the incorporated variables’ dependence relationships through their structure.