Chapter Vi Resistance Prediction

43

CHAPTER VI

RESISTANCE PREDICTION

6.1

Introduction

Nowadays there are many techniques which can be used in determining ship resistance. There are four methods which are: i)

Model Experiments

ii)

Standard Series Of Experiments

iii)

Statistical Methods

iv)

Diagrams

Model experiment method is the most widely used and applied among others since it uses models with similar characteristic of the ship and applicable to any kinds of ships. Meanwhile the other three methods can be used for prediction only because they have limitation and can be used only for a ship that has similar particulars to such group. In this project, only five methods are chosen. These methods are Van Ootmerssen’s Method, Holtrop’s & Mennen’s Method, Cedric Ridgely Nevitt’s Method, DJ Doust’s Method and Diagram method.

44 6.2

Van Oortmerssen’s Method

This method is useful for estimating the resistance of small ships such as trawlers and tugs. In this method, the derivation of formula by G. Van Ootmerssen is based on the resistance and propulsion of a ship as a function of the Froude number and Reynolds number. The constraint of this formula, also based on other general parameters for small ships such as trawlers and tugs that are collected from random tank data. The method was developed through a regression analysis of data from 93 models of tugs and trawlers obtained by the MARIN. Besides, few assumptions were made for predicting resistance and powering of small craft such as follows: 1. According to the Figure 6.1 there are positive and negative pressure peak distributions for the hull surface. For the ship hull scope, there are high pressure at the bow and stern, while in the middle it becomes a low pressure.

Figure 6.1: Pressure distribution around a ship hull

2. Small ship can be said to have a certain characteristics such as the absence of a parallel middle body, so the regions of low pressure and the wave system of fore and after shoulder coincide and consequently the pressure distribution is illustrated as in figure 6.2.

45

Figure 6.2: Wave system at fore and aft shoulder

3. The summation of viscous resistance and wave-making resistance representing the components of the total resistance.

The range of parameters for the coefficients of the basic expression is as follow: Table 6.1: Limitations for Van Oortmerssen’s Method

Parameter Length of water line, LWL Volume, ∇ Length/Breadth, L/B Breadth/Draft, B/T Prismatic coefficient, CP Midship coefficient, Cm Longitudinal center of buoyancy, LCB ½ entrance angle, ½ ie Speed/length, V/√L Froude number, Fn

Limitations 8 to 80 m 5 to 3000 m³ 3 to 6.2 1.9 to 4.0 0.50 to 0.73 0.70 to 0.97 -7% L to 2.8% L 10º to 46º 0 to 1.79 0 to 0.50

46

Van Oortmerssen’s suggested that the final form of the resistance equation is represented by the summation of viscous resistance and wave-making resistance as follow:

(

)

− ( 1 / 9 ) mFn + C2e − mFn + C3e − mFn sin Fn − 2 +  RT C1e = + ∆ C4e − mFn − 2 cos Fn − 2   0.075 ρSV 2   2   2( log Rn − 2 ) ∆  −2

−2

(

−2

)

where,

2

1.

2.

10 3 Ci = d i ,0 + d i ,1 LCB + d i , 2 LCB 2 + d i ,3C p + d i , 4 C p + d i ,5 LWL / B + d i , 6 ( LWL / B ) + d i ,7 CWL + d i ,8 ( CWL ) + d i ,9 B / T + d i ,10 ( B / T ) + d i ,11 + C m 2

m = b1 . − C p

( −b / 2 )

2

or for small ships this can be presented by: m = 0.14347 − C p

3.

2

( −2.1976 )

CWL is a parameter for angle of entrance of the load waterline, ie, where CWL = ie ( LWL / B )

4.

Approximation for wetted surface area is represented by: S = 3.223V 2 / 33 + 0.5402 LWLV 1 / 3

47 Table 6.2 below shows an allowance for frictional resistance and table 5.3 shows the values of regression coefficient given by Van Oortmerssen’s.

Table 6.2: Allowance for frictional resistance

Allowance for Roughness of hull Steering resistance Bilge Keel Resistance Air resistance

∆ CF 0.00035 0.00004 0.00004 0.00008

Table 6.3: Values of regression coefficient i di,0 di,1 di,2 di,3 di4 di,5 di,6 di,7 di,8 di,9 di,10 di,11

1 79.32134 -0.09287 -0.00209 -246.46896 187.13664 -1.42893 0.11898 0.15727 -0.00064 -2.52862 0.50619 1.62851

2 6714.88397 19.83000 2.66997 -19662.02400 14099.90400 137.33613 -13.36938 -4.49852 0.02100 216.44923 -35.07602 -128.72535

3 -908.44371 2.52704 -0.35794 755.186600 -48.93952 -9.86873 -0.77652 3.79020 -0.01879 -9.24399 1.28571 250.64910

4 3012.14549 2.71437 0.25521 -9198.80840 6886.60416 -159.92694 16.23621 -0.82014 0.00225 236.37970 -44.17820 207.25580

48

6.3

Holtrop’s & Mennen’s Method

This resistance prediction method is one of the techniques widely used in prediction of resistance of displacement and semi-displacement vessels. Like all methods, however, this technique is limited to a suitable range of hull form parameters. This algorithm is designed for predicting the resistance of tankers, general cargo ships, fishing vessels, tugs, container ships and frigates. The algorithms implements are based upon hydrodynamic theory with coefficients obtained from the regression analysis of the results of 334 ship model tests. In their approach to establishing their formulas, Holtrop and Mennen assumed that the non-dimensional coefficient represents the components of resistance for a hull form. It might be represented by appropriate geometrical parameters, thus enabling each component to be expressed as a non-dimensional function of the sealing and the hull form. The range of parameters for which the coefficients of the basic expressions are valid is shown as following:

Table 6.4: Limitation for Holtrop and Mennen’s method Ship type

Max Froude

Cp Min Max

L/B B/T Min Max Min Max

Tankers, bulk carries Trawlers, coasters, tugs Containerships, destroyer types

no. 0.24 0.38 0.45

0.73 0.55 0.55

0.85 0.65 0.67

5.1 3.9 6.0

7.1 6.3 9.5

2.4 2.1 3.0

3.2 3.0 4.0

Cargo liners RORO ships, car ferries

0.30 0.35

0.56 0.55

0.75 0.67

5.3 5.3

8.0 8.0

2.4 3.2

4.0 4.0

49

The step by step procedures are shown below to calculate resistance in order to predict the ship power. Calculate: 1.

Frictional Resistance, R F = (1 / 2 ) ρSV 2 ( C F (1 + k ) + Ca )

2.

Wetted Surface,

S = L( 2T + B ) CM

3.

( 0.5 )  0.4530 + 0.4425CB

− 0.2862CM   0.003467( B / T ) + 0.3696CWP

−  + 2.38 ABT / CB 

Form Factor

(1 + k ) = 0.93 + ( T / L ) 0.22284 ( B / LR ) 0.92497 ( 0.95 − CP ) − 0.52145(1 − CP ) + 0.0225 LCB 0.6906

where, LR = L(1 − C P ) + 0.06 4.

Correlation Factor, Ca = 0.0006( L + 100 ) −0.16 − 0.00205

5.

Frictional Resistance Coefficient, C F =

6.

Residuary resistance, R R = Ce ( m1Fn ) where,

−0.9

0.075

( LogRn − 2) 2

+ m2 cos ( λFn ) −2

50

(

(

) )

m1 = 0.0140407 L / T − 1.75254 ∇1 / 3 / L + ( − 4.79323B / L ) − 8.07981C P 2

+ 13.8673C P − 6.984388C P

3

2 2 m2 = 1.69385C P e − ( 0.1/ Fn )

λ = 1.446C P − 0.03L / B 7.

Therefore, total resistances are, RT = RF + RR

6.4

Cedric Ridgely Nevitt’s Method

This method is developed for a model test with trawler hull forms having large volumes for their length. In this method it covers a range of prismatic coefficients from 0.55 to 0.70 and displacement-length ratios from 200 to 500, their residuary resistance contours and wetted surface coefficients have been plotted in order to make resistance estimates possible at speed-length ratios from 0.7 to 1.5. The changing of beam-draft ratio also takes into account the effect on total resistance.

Table 6.5: Limitations for Cedric Ridgely Nevitt’s method Parameter Length/Breadth, L/B Breadth/Draft , B/T Vulome/length, ∇ / ( 0.01L ) 3 Prismatic coefficient, CP Block coefficient, CB Speed/length, V/√L Longitudinal center of buoyancy, LCB ½ entrance angle , ½ α˚e

Limitations 3.2 –5 .0 2.0 - 3.5 200 – 500 0.55 - 0.70 0.42 - 0.47 0.7 - 1.5 0.50% - 0.54% aft of FP 7.0˚ - 37.4˚

51 This method is applicable to fishing vessels, tugs, fireboats, icebreakers, oceanographic ships, yachts and other short and beamy ships falling outside the range of the Taylor Series or Series 60. Procedures of calculating ship resistances using this method are as follows: Calculate: i)

Parameter of V/L for every speed.

ii)

Parameter of ∆/(0.01L) ³ Residuary resistance coefficient CR then can be determined from the graph ∆ / (0.01L) ³ against prismatic CP at every V/L.

iii)

Parameter of B/T ratio The wetted surface coefficient, S / ∇L value can be determined from

the graph of wetted surface coefficient against prismatic coefficient Cp and calculated B/T ratio iv)

The wetted surface area from the wetted surface coefficient S / ∇L

v)

Reynolds Number, Rn =

vi)

Frictional resistance coefficient, C F =

vii)

Total resistance, RT = ( C R + C F )( ρ / 2) SV 2

ρUL µ 0.075

( LogRn − 2) 2

After all parameters are calculated, correction needs to be carried out with the total resistance according to the B/T ratio. The correction can be determined from the graph B/T correction factor against V/L. viii)

Calculate the real total resistance coefficient, RT ' = RT x correction factor

52 6.5

DJ. Doust’s Method

DJ. Doust’s Method is used for calculating resistance based on the resistance tests of about 130 trawler models carried out at the National Laboratory in Teddington, England. The results of the tests were transformed into a trawler standard length, between perpendiculars, of 61m (200ft). (Fyson J. 1985) There are sixs parameters used in the early stage of design. Those parameters are L/B, B/T, CM, CP, LCB and ½ α˚e.

Tables 6.6: Limitations for DJ Doust’s Method

Parameter Length/Breadth, L/B Breadth/Draft, B/T Midship coefficient, CM Prismatic coefficient, CP Longitudinal center of buoyancy, LCB ½ entrance angle, ½ α˚e

Limitations 4.4 – 5.8 2.0 – 2.6 0.81 – 0.91 0.6 – 0.7 0% - 6% aft of midship 5˚ - 30˚

This method is relevant to be used in predicting the resistance for fishing vessels. However, correction needs to be taken into consideration for the ships have different length compared to the standard ship length (200ft). Procedures of calculation for DJ Doust’s Method are as follows:

53 i)

Calculate the parameter required to determine factors used to calculate residuary resistance for the ship having standard length, 200 ft. These parameters are L/B, B/T and V/√L.

ii)

Determine three factors used to calculate residuary resistance using graph given. These three factors are F1 = f ( Cp, B / T ) ,

(

F2 = f ( Cp, LCB ) and F3 ' = f Cp,1 / 2α  , L / B iii)

)

Calculate F6 using F6 = 100a ( C m − 0.875). The parameter ‘a’ is a function V/√L and given by Table 6.7 '

iv)

Calculate residuary resistance, C R ( 200 ) = F1 + F2 + F3 + F6

v)

Calculate, S = 0.0935S / ∆ 1 2 / 3

vi)

Calculate, L' = 1.05V / L

vii)

Calculate Froude’s skin friction correction, SFC = SL'

−0.175

(0.0196 + 0.29L / 10 '

4

2

3

− 2.77 L' / 10 6 + 1.22 L' / 10 8

viii)

Calculate, δ 1 = (152.5 x SFC ) / ∆ ( 200 )

ix)

Calculate residuary resistance for new ship, C R ( new ) = C R ( 200 ) + δ 1

x)

Calculate total resistance, RT =

1/ 3

C R ( new ) ∆V 2 L

Tables 6.7: Values parameter ‘a ’ V/√L 0.8 0.9 1.0 1.1

a -0.045 -0.053 -0.031 -0.035

)

54

6.6

Diagram Method

This method represents the usage of a given chart to obtain the necessary power of the vessel. It’s also one of the method used to compare the result with the other method. Usually the values are taken directly from the chart and will be used as comparison with the other. This chart shows the curves for displacement and beam of a typical fishing vessel from 10 to 70m length (33 to 230 ft). (Appendix C) The solid line in this chart is for the optimum vessel and indicates how the necessary engine power can be reduced if the hull shape is favorably designed. (Fyson J. 1985) Dotted lines show the corresponding curve for average fishing vessels. Nevertheless, it also has limitations on the usage of this graph, they are:

Table 6.8: Limitation of Diagram method Parameter Length (m) Beam (m) Displacement (m³) Engine in SHP (hp) Speed (knots)

6.7

Limitation 10 – 70 m 0 – 10 m 0 – 3000 0 – 1500 7 – 13.5

Resistance prediction of trawler

The predictions on this trawler are done only by four methods. For all the techniques above, the method suitable for this trawler specifications are Van

55 Oortmerssen’s Method, Holtrop and Mennen’s Method, Cedric Ridgely Nevitt’s Method and Diagram method. The diagram and Cedric Ridgely Nevitt’s method are applied by calculation based on the graph while the Van Oortmerssen’s and Holtrop & Mennen’s method were calculated by Maxsurf program. The prediction of resistance will be compared with tank test result. The assumption on the coefficient is stated below.

Assumption of coefficient: 1. Gearing coefficient,η G

= 0.96-0.97 (0.97)

2. Shaft and bearing coefficient, η S .η B = 0.98 – machinery aft (0.98) = 0.97 – machinery amidships 3. Drive Coefficient, η D

= 0.5

Table 6.9: Resistance prediction using a various method

Speed

Cedric Ridgely

(knots)

Nevitt’s method

4 6 7 8 9 10 11 12 Speed 12.5 13 (knots)

11.98 18.09 27.75 Cedric Ridgely

Resistance Prediction, Rt (kN) Holtrop & Chart Chart (Average) 22.20 38.34 Power Chart (Average) -

(Optimum)

Mennen’s

1.32 2.88 4.01 5.56 7.87 13.32 10.33 24.22 PE (kW)13.80 Prediction, -Chart 23.79 Holtrop & 50.95 (Optimum) 78.14 Mennen’s

Nevitt’s

Van Oortmerssen’s Method 1.16 2.86 4.09 5.63 12.42 17.57 19.39 36.50 Van 51.22 67.95 Oortmerssen’s

Method

Method

3.655 11.935 19.36 30.7 48.89 71.26 104.68 196.92 439.32 700.735

3.19 11.855 19.765 31.095 77.11 121.18 147.15 302.17 441.65 609.37

method 4 6 7 8 9 10 11 12 12.5 13

82.61 137.28 229.74 -

153.12 290.94 -

91.88 183.75 -

56

Table 6.11: Summary of result (all)

Effective Power Against Ship Speed 800 700

Cedric Ridgely Nevitt’s

600 Chart (Average)

PE (HP)

500 Chart (Optimum) 400 Holtrop

300 200

Van Oortmerssen’s

100

Tank Test

0 0

2

4

6

8

10

12

14

Speed (knots)

Graph 6.1: Comparison of effective power between various methods and tank test

Differences (%) Total Resistance Against Ship Speed

Cedric

90

Holtrop

Speed 80 Towing

Ridgely

Chart

Chart

&

(knots) 70 tank (Hp)

Nevitt’s

(Average)

(Optimum)

Mennen’s

Rt (kN)

60

4 6 8 9 10 11 12 12.5 13

50 40 30 20 10 0

method 6.71 28.14 92.03 196.15 302.87 458.12 0 2 754.00 983.29 1217.24

42.62 36.96 4 35.9 -

Method 6.37 33.61 6 8 Speed (knots) -

36.18 15.61 10 -

12

14.61 10.73 29.81 47.56 50.52 51.93 14 45.1 6 21.11

Van

Cedric Ridgely Nevitt’s

Oortmerssen’s Chart (Average)

Method

Chart (Optimum)

0 11.34 Van Oortmerssen’s 28.95 Tank 17.29 test 15.96 32.42 15.93 5.5 5.33 Holtrop

57

Graph 6.2: Comparison of total resistance between various methods and tank test

6.8

Discussion on prediction result

The prediction of resistance for trawler hull form shows an acceptable value compared to the towing tank test result. All methods showed that resistance will increase gradually with increasing speed. In a certain speed, there is no result on that value. This condition occurs because parameter of the vessel can only measure at that certain speed. If the value is higher than the particular speed, the resistance cannot be measured, but its enough to see the comparison between the resistance test and prediction value. The comparison can be clearly seen on Graph 6.2. Differences on percentage are not to significant comparing to the tank test result and it still can be accepted.

6.8.1

Van Oortmerssen’s method

58 Usually, the Van Oortmerssen’s methods are useful for estimating the resistance of small ships such as trawlers and tugs. In general, Malaysian fishing vessels are short and beamy whilst their draught is relatively low. These kinds of vessel are normally located in shallow river estuaries. These factors will result in a relatively low breadth-draught ratio and block coefficient. According to the Graph 6.2, the results of the tank test are nearly close to this method. It’s also shows the lower percentage difference on the prediction value. Therefore Van Oortmerssen’s method is suitable because the hull characteristics are covered by this method.

6.8.2

Holtrop and Mennen’s method

Generally, Holtrop and Mennen’s method is suitable for a small vessel and this algorithm is designed for predicting the resistance of fishing vessels and tugs. However, with this method, there are still errors exist in the final result. Therefore, all the factors below are considered to determine the degree of uncertain parameter: i)

Increasing in Froude number which will create a greater residuary resistance (wave making resistance, eddy resistance, breaking waves and shoulder wave) is a common phenomenon in small ships. As a result, errors in total resistance increase. ii)

Small vessels are easily influenced by environmental condition such as wind and current during operational.

iii)

For smaller ship, the form size and ship type have a great difference.

This method is only limited to the Froude number below 0.5, (Fn < 0.5) and also valid for TF / LWL > 0.04. There is correlation allowance factor in model ship that will affect some 15% difference in the total resistance and the effective power. This

59 method is also limited to hull form resembling the average ship described by the main dimension and form coefficients used in the method. Graph 6.2 shows the difference is slightly large value but according to the Table 6.11, after 12 knots, the percentage will reduce. The critical range of speed between 8 to 11 knots shows the dissimilarity is because of the above reason.

6.8.3

Cedric Ridgely Nevitt’s Method

In this method, the resistances are base on the limitation of the parameter. The acceptable speed that can be used by these methods are from 7 to 13 knots. Referring to the calculation, this method also used the displacement-length ratio for measuring the residuary resistance from the chart. The error occurs when determining the residuary resistance and the correction factor of breadth-depth ratio from the graph given. The actual value cannot acquire because the prediction is only limited to a certain value. Therefore, an affected of residuary resistance also affect the total resistance of the ship.

6.8.4

Diagram method

60 This method can be used for preliminary design of new fishing vessel based on the requirement of that new vessel. In this analysis, the shaft horse power value can be used use for reference to design new ship. Based on the Graph 6.2, the tank test curve is in the middle of the optimum and average value. Therefore, it is on the range of selected engine power and can be accepted for the preliminary design.