Vessel Velocity Prediction Toward Model Demonstration And System Realization

In model demonstration period of a vessel, it calls for an overall assessment on the performances of the vessel. A vessel's performances include both technical performance and economic performance. On one hand, technical performances include buoyancy, stability, anti-sinking performance, powering performance, seakeeping performance, maneuverability, and so on. On the other hand, economic performances include single vessel economy, fleet economy and transport system economy.This thesis was to develop a program system. The program system focused on the assessment of a vessel's velocity prediction. When developing the program system, the author paid attention to compatibility with other performances and scalability, in order to pave the way for overall assessment on the other performances.This thesis attempted to use Visual Basic 6.0 language and database management software SQL Server 2000 to develope Vessel Velocity Prediction System.Firstly, the author established a database to store the various parameters effecting a vesell's velocity. The database collected as much datas as possible. All the datas could effect a vesell's velocity.Secondly, basing on this database, the author developed a set of program system with the help of multiple linear regression theory. The program system could search for data records meeting specific criteria the users set. What's more, the coefficients of the regression equation could be calculated so as to satisfy the need of vessel velocity prediction.Then, the author divided each characteristic parameters effecting a vessel's velocity into several intervals. In each interval,qualified records were searched for. The author applicated the program system, calculated a range of coefficients of the regression equation and made a set of coefficient tables. The tables were convenient for the users to applicate. What's more, a precision comparision with the Navy coefficient method was made.Finally, the author established a vessel velocity prediction model with the help of BP neural network toolbox in Matlab, and achieved the goal of vessel velocity prediction.