Study of semi-empirical methods for ship resistance calculations

In the early ship design process a quick overview of which shipdesign that could be the optimal choice for the intended usage needsto be investigated. Therefore the feasibility and accuracy ofinterpolating between measurement data from model resistance serieswhen estimating unknown hulls were conducted. A parametric study wasundertaken in order to investigate which parameters carry the mostimportance in regard to calm water resistance for semi-displacinghulls. In order to asses the whole estimation process one semidisplacing ship (FDS-5) and one bulk carrier with a bulbous bow (JBC)were estimated in regard to calm water resistance by using semiempirical methods and were later compared with CFD results. The CFDresults came from a in part parallel conducted work. The resultsshowed that it is possible to estimate the total resistance withsemi-empirical methods to an unknown hull by linear interpolationwith an accuracy of below 5% in the designed speed interval both forFDS-5 and JBC. The CFD simulations achieved a lower accuracy comparedto the semi-empirical approach, however by furhter calibrating themodels, the accuracy could potentially be improved. Linearinterpolation between two hulls in order to estimate an unknown hull,is only advised when the hulls are nearly identical. Meaning that thehulls must be of the same ship type and that only one parameter isallowed to differ compared to the unknown hull. The parametric studyresulted in parameter importance in falling order: Slenderness ratio,length-beam ratio, longitudinal prismatic coefficient, blockcoefficient and beam-draught ratio. Even though the CFD approach notyet is completely reliable, it could still be a useful complement tothe semi-empirical approach by calculating parameters such as adynamic wetted surface, resistance due to appendages or airresistance of the full-scale ship. Simply by incrementally increasingthe accuracy of individual resistance components an overallimprovement could potentially be achieved