Page 536 - Practical Design Ships and Floating Structures
P. 536

51 1

      For the inland ship the water depth was adjusted to ratios of the water depth to the draft h/T=1.5, 2.0,
      and 3.0 and for the subject container ship to h/T=2.0,2.6,  and 4.0 . The resistance tests were conducted
      by  free trim  and  sinkage. The measured total resistance and  sinkage for the inland ship model  are
      plotted in Fig. 2 as a function of the towing speed, e.g. the speed over ground. The towing speeds were
      all at the depth-Froude numbers  FA I 0.7.  In this subcritical speed range, the sinkage and the total
      resistance monotonously increase with the increased towing speed at the same water depth and with
      the decreased water depth at the same towing speed.
            40  I
                     I     1     I   I         70
                                               60
        -                                      50
        E                                      40
        E
        I                                      30
        N=
                                               20
                                                10
                                                0
              00    0.5    1 .a   1.5            0.0    0.5   1 .a   1.5
                        V,Im/sI                            V,Im/sl
         Figure 2: Characteristics of sinkage and measured resistance of the subject inland ship model
      3  ANALYSIS METHOD

      3.1  Introduction of a Mean Effective Speed Based on the Mean Sinkage

      Referring  to  the  definition  of  an  effective depth-Froude number  given  by  Graff (1963), a  mean
      effective speed VE based on the mean sinkage z,,  is introduced here



      where g denotes the acceleration due to gravity and V the ship speed. This effective velocity combines
      the blockage effect near the ship and the effective depth-effect under the ship. The former is important
      for  the  viscous effect  and  the  latter  for  the  wave  effect.  According  to  Horn  (1932), the  mean
      deformation of the water surface under the ship can be assumed to be equal to the mean sinkage z,, .
      Therefore,  the  mean  near-ship  water-velocity relative  to  the  ship  can  be  obtained  by  means  of
      Bernoulli’s equation:
                                    v, =Jv2+2gzv                              (2)
      For  a measured  sinkage zv, the velocity VB defined by  equation (2) combines both potential and
      viscous effects. Since the viscous effect on the mean sinkage is negligible small, the velocity V, can be
      called as a blockage speed near the ship. It  implies that the blockage velocity can be  estimated by
      means of potential calculations. However, the viscous effects on the ship are more associated with the
      blockage speed, since the local friction is a function of the local velocity relative to the ship’s surface,
      not necessarily of the ship speed over the ground.

      3.2  Model Resbtance as a Function of the Effective Speed
      Now by re-construction of the total resistance measured at different water-depths as a function of the
      effective speed, Fig. 3 and 4 show that the model resistance of the subject inland ship and the container
      ship, respectively, is almost a unit  function of the effective velocity and independent of the water-
   531   532   533   534   535   536   537   538   539   540   541