Ship powering performance modeling using machine learning technique based on full-scale operational data

Abstract

This dissertation presents an accurate and systematic method for estimating ship powering performance using machine learning algorithms based on full-scale operational data, and ship route optimization is conducted using the proposed models.The proposed models calculate the speed and engine power of a ship, which are the main factors of the ship operational efficiency, under various environmental and operating conditions.To design modeling structure of ship powering performance, graphical models are suggested in this study.Then, machine learning techniques are employed to estimate the powering performance of a full-scale ship by constructing regression models using the ship operational data.In order to minimize the risk of overfitting in the regression process, domain knowledge based on physical principles is combined into the regression models.Also, the uncertainty of the estimated performance is evaluated with consideration of the environmental uncertainties.To verify the validity of the proposed models, the proposed models are compared with the existing methods using actual measurement data.In addition, in this thesis, the proposed models are applied to the ship route optimization, which is representative application field of ship powering performance model, and the results of the routing are analyzed based on the actual forecast data.