오사카 대학 Munehiko Minoura 교수님의 세미나를 아래와 같이 공지하오니 학생들의 많은 참석 바랍니다.
제목: Statistical Shaft Power Analysis by Vector Kernel Regression Model with On-board Monitoring Data
연사: Munehiko Minoura 교수 (오사카대학)
일시: 2017년 11월 14일(화), 4:30-5:30 P.M.
장소: 조선해양공학과 세미나실 (36동 204호)
A statistical model for evaluating ship service performance with on-board monitoring data is proposed. A shaft power function, which is representative of service performance, is discussed. A statistical model is based on the vector kernel regression model. Statistics data of 1200 seconds at the instantaneous value of 1Hz sampling rate, e.g. mean ship speed and standard deviation of pitch rate, are subject to analysis. Data measured for several consecutive months are used.
Added propulsion power caused by wave and wind resistances cannot be directly measured in actual seas. If the propulsion power is decomposed to such components, it would be useful to energy saving voyage, ship maintenance and performance design. When ship hull form, self-propulsion factors and engine characteristics are completely provided and the physical model is perfect, each power component can be estimated numerically. However, some of them are not always perfect. The statistical analysis enables to estimate ship service performance without touching these issues.
The dependent variables of the kernel regression model must be measureable; e.g. ship speed and motion. The actual on-board monitoring data and the simulated on-board monitoring data were provided to validate the regression model. When these data were transformed to be independent under the constraints based on the physical matters; e.g. zero-power at zero-speed, it became clear that the regression model is accurately identified and shaft power is rationally analyzed. This regression model enables to decompose the shaft power to each component.