모형시험 데이터베이스를 활용한 컨테이너선의 저항성능 예측

Abstract

This study suggests the prediction method for the residual resistance coefficient of the container carriers using KRISO model test results. The linear regression anaysis was adopted in order to obtain the prediction formula. In this regression process, the local hullform variables are defined and used. The machine learning approaches such as SVR (Support Vector Regression), RF (Random Forest), NN (Neural Network) are also tested for the regression. Some approaches gave an improved results comparing with the linear regression.