Page 215 - Practical Ship Design

180 Chapter 6

The total resistance coefficient Ct of a ship is subdivided into:

c,= C,(l +K> + c, + capp c, + ctr+ c, (6.42)
+

where
c, = frictional resistance coefficient to 1957 ITTC
(1 + K) = form factor
cw = wave-making resistance coefficient
= appendage resistance coefficient
CaPP
‘h = coefficient of the additional pressure resistance of a bulbous
bow near the surface
= coefficient of the additional pressure resistance of an immersed
Ctr
transom stern
= coefficient of model-ship correlation resistance.
‘a
This formula is very similar to that given as eq. (6.7) for the 1978 ITTC treatment
of resistance but there are a number of differences:
C, in the ITTC’78 formula has been subdivided into C,, C, and C,, whilst Cair
has been omitted and AC has been changed to C,.
The method provides regression analysis formulae for each of these resistance
components and goes on to provide further regression formulae for estimating the
propulsion factors of effective wake fraction, thrust deduction fraction and relative
rotative efficiency.
Further formulae for the prediction of the propeller open water efficiency
enables the calculation of the shaft horsepower to be completed in an expert system
type computer calculation.
In the 1984 paper the authors state that they had focused attention on improving
the power prediction of high block coefficient ships with low LIB ratios at one end
of the spectrum and of fine-lined slender naval ships at the other, so the method has
a wide application.
Due to a policy decision by Marin, not all the formula are given for the last
reference but those in general use are given in the 1984 paper and there seems no
point in repeating these here although as a sample of the great pains these authors
have taken to bring in every variable which may affect one of their factors and of
the complexity which results, a slightly modified version of the formula for (1 + K)
from the 1984 paper is given below.
The modification made to the formula consists of reducing the coefficients and
indices from five or six decimal places to three, which seems more appropriate to
the probable accuracy of a formula produced by regression analysis. This simpli-
fication has the advantage of enabling the formula to be fitted into a page more
easily and whilst it may have introduced some error this seems unlikely to be
significant.