New propeller optimisation process can analyse 10,000 designs a day!


A newly developed propeller optimiser makes it much faster to identify propeller-hull reactions and reach design decisions. Evert-Jan Foeth, e.j.foeth@marin.nl


I


n the EU project STREAMLINE the fuel efficiency of a chemical tanker was improved by optimising the hull form


and propeller. To this end, MARIN used its hull optimisation program PARNASSOS EXPLORER (see MARIN Report 109) and a newly developed propeller optimiser. The hull form itself was optimised for low resistance and for having a wake field as rotationally symmetric as possible, while simultaneously maintaining the displacement and sufficient space for the engine room. Meanwhile, the propeller was required to deliver the right thrust at the same ship speed and engine rate of revolutions.


Normally, the propeller designer starts when the hull lines have been set and the main propeller parameters have been decided upon. These parameters follow from series analysis - such as the Wageningen B-Series - plus hands-on experience and rules of


14 report


thumb to estimate the effect of propeller-hull interaction. For instance, the sector knows that the best propeller diameter in a wake field is smaller than given by the propeller series. Once the main dimensions are selec- ted, the designer continues to mould the geometry until the efficiency is high and the hindrance from cavitation is as low as possible. This is a time-consuming process that is particularly ill-suited to finding the best propeller when considering 500 hull form variations can be generated, as was the case here. It was clear that a tool that can quickly cover a wide range of design choices and return ballpark figures for efficiency, including the propeller-hull interaction effects, was required.


When a large number of propeller geometries are to be generated, the range should be as large as possible but certainly not by in- cluding every imaginable shape. Therefore,


the propeller blade geometry was built up from radial distributions for the pitch, chord length, skew, rake, camber and thickness and these distributions were described by newly developed parametric functions. In order to remain close to the shape of real propellers, MARIN analysed over 1,200 unique propeller designs in its database and used the results to train the parameters. Not only could the majority of propellers be well captured, the generation of a new blade by randomly changing the parameter values nearly always resulted in a plausible blade outline.


Avoids generation of unsuitable propellers Naturally, the proof of the pro- peller is in the computing. For STREAMLINE, all propeller geometries were analysed with MARIN’s Boundary-Element Method PROCAL. Although each propeller was unique, differing in blade shape as well as blade


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24