See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/301289989

Analysis of Propulsion Performance of KVLCC2 in Waves

Conference Paper · July 2015

CITATIONS READS
4 1,014

2 authors:

Bhushan Taskar Sverre Steen

Technical University of Denmark Norwegian University of Science and Technology
26 PUBLICATIONS   159 CITATIONS    98 PUBLICATIONS   837 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

SINTEF Ocean Cavitation Tunnel View project

LEEDS - Low Energy and Emission Design of Ships View project

All content following this page was uploaded by Bhushan Taskar on 14 April 2016.

The user has requested enhancement of the downloaded file.

 Fourth International Symposium on Marine Propulsors
smp’15, Austin, Texas, USA, June 2015

Analysis of Propulsion Performance of KVLCC2 in Waves

Bhushan Taskar1, Sverre Steen2

1, 2 Department of Marine Technology, NTNU, Trondheim, Norway

ABSTRACT Therefore, previously considered assumptions and margins

In this paper, we have analyzed the propulsion performance should be revisited and updated by detailed knowledge of
of KVLCC2 in presence of waves. Different factors affecting propeller performance in waves.
the propulsion performance have been studied. Analysis of Along with the efficiency of the propeller, cavitation and
the extent of change in wake quality and its effect on the vibration characteristics should be studied in presence of
cavitation of propeller has been presented. Effect of wake waves as they depend on the wake distribution (Odabasi and
change alone was separately calculated to analyze its Fitzsimmons (1978) and Huse (1974)). Moreover, a change
importance in the design process, as wake data in waves is in wake distribution changes the angle of attack and the
usually not available. It was observed that wake change itself cavitation number of the propeller blades as shown by Albers
does not significantly affect the amount of cavitation hence; and Gent (1985). Chevalier and Kim (1995), Jessup and
cavitation margin should be considered only to handle Wang (1996) studied the cavitation of a propeller operating
increased load and relative stern motion. in waves by calculating wake velocities using potential flow
Keywords calculations. Drop in the cavitation inception speed of a
Cavitation Analysis in Seaway, Propeller in Waves, vessel was observed in waves.
Performance in Off-Design Conditions. The cavitation characteristics of propellers designed using
1 INTRODUCTION calm water wake data must be studied in order to validate
currently used cavitation margins, so that future propellers
Currently, propellers are designed using wake, thrust can be designed for low cavitation and noise along with
deduction and relative rotative efficiency obtained in calm acceptable performance even in rough weather.
water conditions. These factors vary when ship is subjected
to waves (Moor and Murdey 1970). Wake distribution also In this paper, we have evaluated the performance of the
changes due to waves and ship motion (Nakamura and Naito KVLCC2 propeller operating in waves. Time varying wake
1975). Similar results were obtained in the RANS simulation data in three different wavelengths provided by Sadat-
carried out by Guo, Steen et al. (2012) where the nominal Hosseini, Wu et al. (2013) have been used. The effect of
wake field was obtained in the presence of waves. In this waves on changes in the angle of attack and the cavitation
simulation, axial wake velocities increased up to 35% of ship number of propeller blade sections has been studied. The
speed in some regions. Such changes in the wake distribution effect of wake change and relative stern motion has been
of a ship travelling in waves were experimentally confirmed separately observed to decide the order of importance of each
by Wu (2013) using KVLCC2 ship model. PIV effect. The effect of this time varying wake on vibration and
measurements of wake field found strong variation in noise characteristics of the propeller has been calculated
presence of waves. using the BSRA wake criteria given by Odabasi and
Fitzsimmons (1978). Other possible factors causing changes
In view of this recently obtained data, which demonstrates in propulsion performance in waves have been noted.
significant effect of waves on wake, a possible drop in the
performance of the propeller should be calculated. Full-scale
2 METHODS AND VALIDATION
experiments performed by Kayano, Yabuki et al. (2013)
found a discrepancy between the calculated and obtained 2.1 Wake Data in Presence of Waves
performance of the ship. This can be due to inability of Experiments were performed by Sadat-Hosseini, Wu et al.
prediction methods to take into account the effect of waves (2013) to obtain wake data in three different wavelengths in
on the propulsion performance. Currently, off-design head sea condition at design speed. A model of KVLCC2 was
conditions are covered by simple sea margin, which may used for this purpose with model scale of 1:100. Ship
result in overdesign or failure in off design conditions. particulars are given in Table 1 (SIMMAN 2008). In these
experiments, PIV (Particle Image Velocimetry) was used to method, only viscous wake is scaled and a correction is
obtain time varying nominal wake field in the propeller applied to the potential component. However, in the absence
plane. CFD simulations were also performed and results were of potential wake data, we have contracted the whole wake
validated using existing data from PIV experiments. Since field towards the center plane by the ratio of viscous
the CFD data are smoother and less noisy, we have used them resistance coefficient between model and full scale. Hence,
in our calculations. These results were available for waves the difference between potential wake component of model
𝜆/𝐿 = 0.6, 1.1 and 1.6 at 8, 12 and 6 time intervals and ship has been neglected.
respectively in one wave period. Waveheight of these waves Potential wake is almost constant in a horizontal section in
correspond to the full-scale waveheight of 3m. the propeller plane as seen from the typical ship scale wake
presented in ITTC (2011). In such cases, the same full-scale
Table 1 Ship Particulars total wake would be obtained by scaling the total wake or just
the frictional component of the model-scale wake. The only
Length beween perpendiculars (m) 320.0
error would be due to the neglected correction in the potential
Length at water line (m) 325.5 wake.
Breadth at water line (m) 58.0
Depth (m) 30.0 2.4 Software Validation Using Existing Data
Draft (m) 20.8 After the initial wake assessment, a detailed study of
Displacement (m3) 312622 propeller operating in presence of waves was performed.
Block coefficient (CB) 0.8098 Existing KVLCC2 propeller design was analyzed in time
Design Speed (m/s) 7.97 varying wake. Details about propeller geometry can be seen
in Table 2 (more details can be obtained from SIMMAN
Table 2 Propeller Geometry (2008)). The open source program Openprop based on vortex
lattice lifting line theory (Epps 2010) has been used for this
Diameter (D) (m) 9.86 purpose.
No of blades 4 Openprop requires blade section details, corresponding
Hub diameter (m) 1.53 frictional drag coefficient, advance coefficient, axial and
Rotational speed (RPM) 76 tangential wake velocities and at each radial location for the
𝐴𝑒 / 𝐴0 0.431 analysis. Blade section details can be found in SIMMAN
(P/D)mean 0.690 (2008). Javafoil was used for the calculation of frictional
drag at each radial section for the given Reynolds number. It
Skew (°) 21.15
uses panel method to calculate velocity profile and pressure
Rake (°) 0 distribution over the foil section. Using these pressure and
velocity distributions, boundary layer calculations are
2.2 Wake Quality Assessment performed where drag is calculated using momentum loss in
In the preliminary investigation of the wake data in waves, the boundary layer (Hepperle).
the quality of wake was assessed and compared with the
quality of the calm water wake using the BSRA wake criteria
proposed by Odabasi and Fitzsimmons (1978). These criteria KT (Openprop) KT (Experiment)
are based on a large collection of wake distribution data and Efficiency (Openprop) Efficiency (Experiment)
noise and vibration characteristics of full-scale ships. Five
conditions are mentioned for assessing the wake. Although KQ (Openprop) KQ (Experiment)
satisfying these conditions does not guarantee good vibration 0.7 0.6
and noise characteristics, it is recommended to be extra 0.6 0.5
Efficiency and KT

careful when the conditions are not met. For our purpose, 0.5 0.4
these simple criteria are useful to assess the extent to which 0.4 0.3
KQ

waves can affect vibration and noise characteristics without

0.3 0.2
using any particular propeller geometry.
0.2 0.1
0.1 0
2.3 Wake Contraction Method 0 -0.1
For further investigation of the performance of the propeller 0 0.2 0.4 0.6 0.8
operating in waves, scaling of wake data from model scale to Advance Coefficient J
full scale was required. According to ITTC (2011), the wake
scaling procedure given by Sasajima, Tanaka et al. (1966) is Figure 1 Comparison of Openprop and open water data of
most commonly used and gives reasonable results. In this KVLCC2 propeller
 5

3 t/c = 0.2; f/c = 0.03

Angle of Attack 1 t/c = 0.1; f/c = 0.03

t/c = 0.06; f/c = 0.02
-1
brockett t/c = 0.2; f/c = 0.03
-3 brockett t/c = 0.1; f/c = 0.03

-5 brockett t/c = 0.06; f/c = 0.02

0.2 0.4 0.6 0.8 1 1.2 1.4 1.6
-Cpmin

Figure 2 Comparison of cavitation bucket diagrams obtained from Openprop with those calculated by Brockett (1966)

Openprop is based on steady lifting line theory. As we know, cavitation inception trends as seen in Figure 2. Even though
a propeller operating even in calm water condition faces time more complicated and accurate theories like lifting surface
varying inflow due to spatial variation of wake. Such cases theory and cavitating foil theory are available to predict exact
should ideally be analyzed with unsteady calculations. cavitation pattern, change in efficiency, thrust and torque of
Gaggero and Brizzolara (2009) have shown that a quasi- a cavitating propeller; we have used this simple theory since
steady approach also gives good results compared to fully we are interested in comparing the performance of a propeller
unsteady calculations. In their research, the quasi-steady in waves with that in calm water, rather than very accurately
approach was seen to correctly predict the change in thrust, predicting the performance in cavitating condition. Thus,
torque and efficiency between propeller and its modified correct prediction of trends would serve the purpose.
version. Hence, we have used quasi-steady approach for our While calculating the cavitation pattern, depth variation of
analysis. Openprop analyzes propeller in a steady flow with the propeller due to ship motion was also taken into account.
only radial wake variation, however, in reality there is Relative stern motion was calculated using the motion
angular as well as radial variation of wake. Hence, response of the ship. All the analysis was performed at
performance of the propeller with four blades facing different constant rpm. Hence, variation of rpm due to time varying
radial wake distribution was assumed to be the average torque was neglected in the analysis, which may cause some
performance of four hypothetical propellers, each facing the inaccuracies.
radial wake distribution faced by each blade.
Performance of Openprop with frictional drag obtained from 3 ANALYSIS
Javafoil was validated by comparing open water
3.1 Wake Assessment in Presence of Waves
characteristics with the experimental data. Thrust, torque and
efficiency in open water condition obtained using this Vibration and noise characteristics of a propeller depend on
approach match well with the experimental data as can be the wake field in which it operates. Odabasi and Fitzsimmons
seen from Figure 1. (1978) have listed certain criteria to be fulfilled by the wake
distribution for low noise and vibration. Time varying wake
Openprop has also been used to predict the cavitation on the in waves will now be compared with the calm water wake
propeller blades. In order to calculate the cavitation, pressure field considering four out of five BSRA wake criteria.
distribution over the foil has been calculated using linear foil
theory; possible effects of viscosity have been neglected. Criterion 1 –
Areas where pressure falls below the vapor pressure is The maximum wake measured inside the angular
assumed to cavitate. The cavitation bucket can be obtained interval 𝜃𝐵 = 10 + 360/𝑍 degrees and in the range 0.4–
by observing the angle of attack and cavitation number at 1.15𝑅 around the top dead center position of the propeller
which cavitation starts. Cavitation buckets were plotted for disc should satisfy the following:
foils with three different combinations of camber and
thickness. These plots were compared with those obtained by 𝑊𝑚𝑎𝑥 < 0.75 𝑜𝑟 𝑊𝑚𝑎𝑥 < 𝐶𝐵
Brockett (1966) where minimum pressure envelopes were whichever is smaller. Where 𝑍 is the number of
calculated for steady two dimensional flow with an empirical blades. 𝑊𝑚𝑎𝑥 has been obtained at given locations at
correction for the viscosity. There is discrepancy in the exact different times in one wave period to compare with the value
values of the angle of attack where cavitation inception is observed in calm water. Values greater than that in calm
predicted. However, Openprop correctly predicts the water can increase vibration and noise.
 w/L = 0.6 w/L = 1.1 w/L = 0.6 w/L = 1.1
w/L = 1.6 calm water w/L = 1.6 calm water

0.9 0.6

1.7 x mean wake - max wake at 0.7R

Maximum wake value

0.4
0.8
0.2

0
0.7
-0.2

0.6 -0.4
0 0.2 0.4 0.6 0.8 1 0 0.2 0.4 0.6 0.8 1
Time t/T Time t/T

Figure 3 Comparison of wake peak observed in waves and Figure 4 Comparison of wake with respect to criterion 2
in calm water
Criterion 3 –
Variation of 𝑊𝑚𝑎𝑥 in different waves can be seen in Figure Wake non-uniformity criterion is important to avoid
3. When the wave is shorter than the ship, the maximum unsteady cavitation and high levels of pressures on the hull.
value of wake is always smaller than in calm water. In the In this criterion, tip cavitation number is plotted against
longest wavelength, only few values are greater than that in average non-dimensional wake gradient. Tip cavitation
calm water. While, when wavelength is close to ship length, number is defined as-
𝑊𝑚𝑎𝑥 in wave is higher than that in calm water for almost
50% of the time as seen from Figure 3. Hence, this condition 𝐷
(9.903 – – 𝑍𝑝 + 𝑇𝐴 )
is not greatly affected due to waves except in case of 𝜎𝑛 = 2
wavelength close to ship length. 0.051(𝜋𝑛𝐷)2

Criterion 2 –
while average non-dimensional wake gradient is defined as
The maximum acceptable wake peak should satisfy the (𝛥𝑤/(1 − 𝑤)), where 𝐷 is the propeller diameter (m). 𝑍𝑝 is
following relationship with respect to the mean wake at 0.7R: the distance between the propeller shaft axis and the base line
(m). 𝑇𝐴 is the ship’s draught at the aft-perpendicular (m). 𝑛 is
̅0.7
𝑊𝑚𝑎𝑥 < 1.7𝑊 the propeller rotational speed (rev/s). 𝛥𝑤 is the wake
variation. Plotted point should lie above the dividing line of
Therefore, (1.7𝑊̅0.7 − 𝑊𝑚𝑎𝑥) was plotted in different Figure 5 to satisfy the criterion.
wavelengths, at different times in time varying wake and
compared with the calm water condition (Figure 4). Value of w/L = 0.6 w/L = 1.1 w/L = 1.6 calm water
this variable should be positive for the criterion to be
satisfied. 0.35

Figure 4 shows that this condition would be most stringent if

0.3
it is to be satisfied in wavelength 𝜆/𝐿 = 1.1. Since, all the
cavitation number

values of (1.7𝑊 ̅0.7 − 𝑊𝑚𝑎𝑥) are lower than that in calm

0.25
water and many of them are negative. Also in the
wavelength 𝜆/𝐿 = 1.6, values lower than that in calm water
0.2
are observed for almost 50% of the time period. Hence, in
case of designs where this condition is just satisfied in calm
0.15
water, its violation is highly probable in presence of waves.
This gives us an idea about the margin to be considered while
0.1
satisfying this condition in realistic sea when only calm water
0 0.5 1 1.5 2
wake data is available.
wake gradient

Figure 5 Wake non-uniformity criterion

 4.2
calm water at r/R = 0.7
3.8
r/R = 0.7; w/L = 0.6
3.4
Maximum Local wake gradient r/R = 0.7; w/L = 1.1
3 r/R = 0.7; w/L = 1.6
2.6 calm water r/R = 0.9

2.2 r/R = 0.9; w/L = 0.6

r/R = 0.9; w/L = 1.1
1.8
r/R = 0.9; w/L = 1.6
1.4
calm water at r/R = 1.1
1 r/R = 1.1; w/L = 0.6

0.6 r/R = 1.1; w/L = 1.1

0 0.2 0.4 0.6 0.8 1
r/R = 1.1; w/L = 1.6
Time t/T

Figure 6 Maximum wake gradient at different time intervals in three different wavelengths

Figure 5 shows the plot for this criterion in three different

w/L = 1.6; t/T = 0 w/L = 1.6; t/T = 0.33
head waves. Variation in cavitation number is due to change
in the submergence of the propeller due to ship motion while w/L = 1.6; t/T = 0.66 calm water
change in the horizontal axis variable (wake gradient) is due
3
to wake variations in waves. Wake gradient is becoming
Local Wake Gradient

2.5
favorable (i.e. less) in more cases than in those it is getting 2
worse than the calm water value. Some values are present in 1.5
the unacceptable region, which may cause intermittent 1
cavitation and vibration while ship is travelling in waves. 0.5
Figure 5 also provides the information about the extent to 0
which waves can worsen the cavitation and vibration 0 15 30 45 60 75 90
characteristics of the propeller. Therefore, appropriate Angular Location
cavitation margin can be considered for the calm water
design to have acceptable cavitation and vibration
Figure 7 Wake gradient at w/L=1.6 at different time intervals
performance in waves. as a function of angular position
Criterion 4 –
velocity was calculated (Figure 7) and maximum value was
For the propellers susceptible to the cavitation, that is near
obtained in the angular interval 𝜃𝐵 in the range 0.7R to
the grey area of Figure 5, the local wake gradient per unit
1.15R. This maximum value obtained at different time
axial velocity for radii inside the angular interval 𝜃𝐵 in the intervals of wave period is plotted in Figure 6.
range of 0.7–1.15𝑅 should be less than unity; that is,
As seen in Figure 6, from 0.7R to R, in wave 𝜆/𝐿 = 0.6
𝑑𝑤 values of wake gradient hardly exceed corresponding calm
1 ( ) water value. While in wavelengths 𝜆/𝐿 = 1.1 and 1.6 local
𝑑𝜃
𝑟 |(1 − 𝑤)| < 1.0 wake gradient is higher than calm water value for
( )
𝑅 approximately 66% and 33% of the time respectively. At r =
1.1R almost all the wake distributions in waves show higher
where 𝜃 is in radians. local gradients. Amount of exceedance, whenever it occurs
This criterion limits the wake gradient in order to reduce is considerable. Moreover, in this case, all the values
volume variations of the cavity. It is required only when the including those in calm water exceed the criterion limit, i.e.
relation between wake gradient and cavitation number lies in all values are greater than one. Therefore, points lying in grey
the grey area in Figure 5. However, here we are more or unacceptable region in Figure 5 are the cause of concern.
interested in comparing quality of wake in waves with calm Since, unstable cavitation in large wake gradient can cause
water wake. Hence, local wake gradient for unit axial significant amount of noise due to the cavity volume
 3 w/L = 0.6 3 w/L = 1.1

Net angle of attack (Deg)

Net angle of attack (Deg)

2 2

1 1

0 0

-1 -1

-2 -2
0.2 0.4 0.6 0.8 0.2 0.4 0.6 0.8
-Cpmin -Cpmin

3 w/L = 1.6
Net angle of attack (Deg)

Effect of wake variation and relative stern motion

2
Effect of Wake Variation
1 calm water
Cavitation Bucket
0

-1

-2
0.2 0.4 0.6 0.8
-Cpmin

Figure 8 Cavitation number and angle of attack faced by propeller blade section at 0.7R in calm water and in waves

variation. Radial section r/R = 1.1 is analyzed since there is considering the effect of wake change is tough. Wake in
chance of wake at that location coming in way of propeller waves can be obtained either experimentally or
due to propeller action (contraction of stream tube). computationally. Experimentally finding time varying wake
is not a common practice, it would require specialized
3.2 Propeller Analysis Using Openprop instruments like PIV. Moreover, multiple runs would be
required to find wake in different wavelengths.
After analyzing the wake quality, propeller geometry of Computationally finding wake variation in waves is also
KVLCC2 ship was examined in time varying wake using expensive. Therefore, it is important to know the extent to
Openprop. As noted earlier, significant change in the wake, which the wake change alone influences the propeller
observed in presence of waves is expected to affect the performance, especially due to significant changes in the
operation of wake-adapted propellers. Although wake wake field observed in presence of waves.
assessment gives some idea about possible cavitation, Therefore, propeller design was analyzed in time varying
examining the propeller geometry can reveal additional wake using the method based on the lifting line theory. These
details like changes in the type and the extent of cavitation, calculations were also used to predict the extent of cavitation
thrust and torque fluctuations. Therefore, performance of the on the propeller blade along with thrust and torque
propeller in waves was compared with that in calm water. fluctuations in different conditions.

3.2.1 Effect of Waves on Cavitation Cavitation Number and Blade Angle of Attack –
Propeller cavitation is affected by the following factors in the While designing the propeller, the knowledge of variation in
presence of waves: the angle of attack and cavitation number is important to
1. Relative stern motion causing change in the choose correct blade thickness. However, certain cavitation
margin has to be assumed for possible off design conditions
cavitation number
including the ship operation in rough sea, as wake in waves
2. Change in wake field leading to alteration of inflow
is rarely available. Therefore, since wake data in waves is
velocities and blade angle of attack
available in this case, the correctness of cavitation margin has
3. Added resistance causing increased propeller
been analyzed further. This would help propeller designers
loading
to estimate the change in the extent of cavitation in presence
Out of these three, relative stern motion and added resistance of waves as compared to the calm water condition.
of ship can be estimated at the design stage, while
 In Figure 8, influence of wake change and relative stern
motion was analyzed while effect of added resistance was not
considered. Out of these two, relative stern motion only
affects the range of –Cpmin and not angle of attack; since it
affects only cavitation number. While, wake change can
affect both the variables. Spread in the values of –Cpmin is
predominantly due to relative stern motion. In all 3 cases, the
effect of wake variation does not decrease the minimum
value of –Cpmin seen in calm water. Maximum of half a
degree increase in angle of attack can be seen due to wake
variation only in wavelength w/L = 1.6. In other two waves,
no significant change in maximum or minimum angle of
attack is observed. Similar trends were observed at other
blade sections as well.
Cavitation Due to Wake Variation –
Due to wake variation alone, there is no increase in the range
of cavitation numbers while angle of attack increases slightly
in some cases as compared to the calm water condition.
Influence of this slight increase in the angle of attack (only
due to wake variation) on the extent of cavitation can be seen
in Figure 9 where maximum amount of cavitation in each
Figure 9 Propeller cavitation in different conditions
condition has been plotted. No significant change in
cavitation is seen due to the effect of wake variation. This
Cavitation characteristics of the propeller were examined in
observation is in line with the earlier result of cavitation
presence of waves. Cavitation numbers and angle of attacks
bucket diagram. Since spread of operating points is similar to
faced by the blade section at 0.7R were plotted along with the
the one obtained in calm water wake, similarity in the extent
cavitation bucket of the blade section. Plotted points for the
and pattern of cavitation is expected.
calm water condition correspond to sixteen different angular
positions of the blade as it rotates in the calm water wake. These results obtained using quasi-steady approach were
While points in waves correspond to eight different angular validated using fully unsteady simulations with cavity
positions of the blade at ten different time intervals in one volume calculations. The unsteady panel method software
wave period. This can be seen in Figure 8. AKPA, developed by MARINTEK and University of St.
Petersburg, was used to simulate the propeller in calm water

In calm water In wave 𝜆/𝐿 = 1.6

Figure 10 Unsteady simulation results of propeller in calm water and in wave

wake and in case of 𝜆/𝐿 = 1.6 where maximum increase in
w/L = 0.6 w/L = 1.1
cavitation volume was observed in Figure 9. Effect of wake
change alone was considered in order to compare the results w/L = 1.6 calm water
with those in Figure 9. For the simulation in presence of
0.021
wave, wake at the time instance showing maximum
cavitation in Openprop simulation was chosen. Time period 0.02
of the propeller being much smaller than that of wake
variation, wake field was assumed constant in this unsteady

KQ
0.019
simulation. Cavitation pattern obtained from this analysis can
be seen in Figure 10. 0.018
Unsteady panel method (Figure 10) show significantly less 0.017
cavity volume as compared to Openprop (Figure 9). 0 0.2 0.4 0.6 0.8 1
However, maximum cavitation seen in presence of waves
hardly differs from the cavitation in calm water, as can be Time t/T
observed in Figure 10.
Figure 11 Variation of KQ in waves
Effect of Waves on Cavitation –
Therefore, the effect of wake change, excluding other factors, Fluctuations in KQ have been calculated assuming constant
on the cavitation is minor in spite of significant changes propeller rpm while in reality rpm is a function of torque and
observed in the wake field. This can be due to the large engine behavior.
induced velocities as compared to the wake velocities, and These variations in torque can cause transients in engine
that even though the change of wake pattern due to waves is operations, and might influence engine performance
quite significant, the critical features, like maximum wake negatively. Here it is important to observe the magnitude of
and wake gradient don’t worsen much. The effect of such torque fluctuations. Furthermore, engine simulation should
wake variation could be more pronounced in case of a lightly be carried out with this torque input in order to calculate
loaded propeller. It is important to note that this analysis has effect of torque fluctuation on engine operations, and its
been performed using wake data in regular waves of fixed efficiency and emissions. The changes in the propeller speed
due to engine response can be taken into account in the
waveheight. Therefore, influence of waves can increase in
propeller analysis, in order to “close the loop”.
case of higher waves. However, effects are expected to be
less severe in irregular waves with significant waveheight
equal to the height of the regular wave. 4 CONCLUSION
Increased load caused by added resistance increases the angle We now have an information about an extent to which the
of attack of blade sections making them susceptible to criteria required for good noise and vibration characteristics
backside sheet cavitation. In this case, since propeller is get affected due to waves. Hence, in future designs
already cavitating in calm water wake, increased load will appropriate margins can be considered for the similar type of
increase the extent of this cavitation. However, as noted vessels. However, such analysis should be performed for the
earlier, this effect can be easily taken into account while variety of ships for multiple propeller loadings in order to
designing the propeller, since the increased propeller load generalize the results.
can be calculated from the added resistance.
As per the analysis, it seems, presence of waves does not
significantly affect cavitation in spite of large changes
3.2.2 Thrust and torque fluctuations in waves observed in the wake field. Thus, a margin for cavitation
Along with the changes in cavitation and vibration would mainly be required for increased loading, relative stern
characteristics, wake variation also causes thrust and torque motion and not much for the wake change due to waves.
to fluctuate. The amount of these fluctuations should be Therefore, in practice the required margin can be estimated
examined to see if they affect the operation of the engine. using added resistance and relative stern motions.
Fluctuations of KQ at constant propeller rpm obtained using
Openprop can be seen in Figure 11 for three different Vibration and noise characteristics have been analyzed using
wavelengths. Maximum fluctuation is evident when the BSRA wake criteria. However, more advanced
wavelength is equal to the ship length. Change in mean value techniques should be used to quantify the pressure pulses in
of KQ as compared to the calm water value is due to the different wakes. Pressure pulses may increase since higher
increase in average inflow to the propeller caused by the wake gradients were observed in presence of waves.
pitching motion of ship. Present analysis being in regular waves gives conservative
estimate of the effect of waves on the propeller performance.
We expect the effects to be less severe in case of irregular
waves. It should also be mentioned that current wake field in Huse, E. (1974). Effect of afterbody forms and afterbody
waves is obtained at model scale corresponding to the actual fins on the wake distribution of single-screw ships, Ship
waveheight of 3m for a 340m long ship. Hence, there is a Research Inst. of Norway.
possibility of larger performance changes in presence of
ITTC (2011). Specialist committee on scaling of wake field.
higher waves. Thus, it would be of interest to perform a
Final report and recommendations to the 26th ITTC,
similar investigation, but for a significantly smaller ship.
ITTC. Volume 2.
Significant fluctuations observed in propeller torque in
waves should be analyzed further to calculate its effect on the Jessup, S. D. and H.-C. Wang (1996). Propeller Cavitation
engine operation. Coupled response of engine and propeller Prediction for a Ship in a Seaway, DTIC Document.
should be obtained to examine the effect of waves on whole Kayano, J., H. Yabuki, N. Sasaki and R. Hiwatashi (2013).
propulsion system. "A Study on the Propulsion Performance in the Actual
Sea by means of Full-scale Experiments." TransNav, the
ACKNOWLEDGMENTS International Journal on Marine Navigation and Safety
of Sea Transportation 7(4): 521-526.
Authors would like to thank Professor Frederick Stern
from the University of Iowa for providing the wake data in Moor, D. I. and D. C. Murdey (1970). "Motions and
waves used to analyze propeller in different conditions. We Propulsion of Single Screw Models in Head Seas, Part
also thank Professor Bjørnar Pettersen for helping us obtain II." The Royal Institution of Naval Architects Vol.
the wake data. This work is funded by the project ‘Low 112(No. 2).
Energy and Emission Design of Ships’ (LEEDS, NFR
216432/O70) where the Research Council of Norway is the Nakamura, S. and S. Naito (1975). "Propulsive performance
main sponsor. of a container ship in waves." J. Kansai Soc. N. A. Japan
No. 158.
REFERENCES Odabasi, A. Y. and P. A. Fitzsimmons (1978). "Alternative
Methods for Wake Quality Assessment." International
SIMMAN (2008). "SIMMAN 2008." from Shipbuilding Progress 25: 8 p.
http://www.simman2008.dk/KVLCC/KVLCC2/kvlcc2_
Sadat-Hosseini, H., P.-C. Wu, P. M. Carrica, H. Kim, Y.
geometry.html.
Toda and F. Stern (2013). "CFD verification and
Albers, A. B. and W. v. Gent (1985). Unsteady wake validation of added resistance and motions of KVLCC2
velocities due to waves and motions measured on a ship with fixed and free surge in short and long head waves."
model in head waves. 15th symposium on naval Ocean Engineering 59(0): 240-273.
hydrodynamics.
Sasajima, H., I. Tanaka and T. Suzuki (1966). "Wake
Brockett, T. (1966). Minimum pressure envelopes for Distribution of Full Ships." Journal of Zosen Kiokai
modified naca-66 sections wih naca a = 0.8 camber and 1966(120): 1-9.
buships type I and type II sections, David Taylor
Wu, P. C. (2013). A CFD Study on Added Resistance,
Model Basin.
Motions and phase averaged wake fields of full form
Chevalier, Y. and Y. H. Kim (1995). Propeller Operating in ship model in head waves, Osaka University.
a Seaway. PRADS'95. Seoul, Korea.
Epps, B. (2010). OpenProp v2.4 Theory Document.
Gaggero, S. and S. Brizzolara (2009). Parametric
Optimization Of fast Marine Propellers via CFD
Calculations. 10th International Conference on Fast Sea
Transportation. Athens, Greece.
Guo, B. J., S. Steen and G. B. Deng (2012). "Seakeeping
prediction of KVLCC2 in head waves with RANS."
Applied Ocean Research 35(0): 56-67.
Hepperle, M. "JavaFoil." from http://www.MH-
AeroTools.de/.
 DISCUSSION detail in this paper as wake data in waves was obtained for
the ship free to heave and pitch. The effect of variation in
Question from Tom van Terwisga propeller immersion has been compared with the effect of
Did you check the cavitation effect for the ship in waves wake variation in Figure 8.
considering an unsteady method and did you look at the
corresponding unsteady pressure variations? Question from Moustafa Abdel Maksoud

Author’s Reply Did you consider the effect of added resistance on the amount
of cavitation on the propeller surface? Do you think that the
Along with lifting line method, cavitation analysis was also following wave condition is more critical than the head
performed using unsteady panel method with cavity volume waves one?
calculations using the software AKPA as mentioned in the
paper. However, wake was assumed quasi steady i.e. Author’s Reply
calculations were performed for wake distribution at In this paper, added resistance has not been considered, since
different time intervals. Wake variation was assumed wake data was available for the design speed of ship and
constant in time in each calculation. We believe that quasi- since considering the added resistance would change the ship
steady wake assumption is reasonable since the frequency of speed as well as motion response, leading to significant
wake variation is much smaller than the frequency of changes in wake. Even if the speed could be kept the same
propeller rotation. by increasing the shaft power to compensate for the added
The effect of change in cavitation number due to relative resistance, this change in propeller operating point means
stern motion has been taken into account using hydrostatic that we would not be able to single-out the effect of the wake
approximation. Hence, the effect of dynamic pressure due to change. However, the effect of added resistance is planned to
wave has been ignored. However, we agree that it would have be included in future studies.
been interesting to include the effect of unsteady pressure Following wave condition can be more critical as waves
variations on the propeller performance. would be directly affecting the propeller. However, authors
are not aware of any measurement data or computations of
Question from Johan Bosschers wake in following waves. Limited availability of wake data
Can you say something about the influence of the change in is in general a limitation for analyzing propeller in waves.
transverse velocities on the results? Is the influence of ship
motions included? Question from Mehmet Atlar

Author’s Reply Interesting paper. The authors may also consider the BSRA
criteria of Odabasi and Fitzimmons in terms of propeller-
In some cases, transverse velocities show significant change. excited vibrations (PEV) since these two authors provided
We believe, change in transverse can affect the tip vortex diagram (i.e. criteria) for cavitation and PEV assessment in
inception, which has not been studied in this paper. their work (i.e. similar to the diagram in Figure 5 of the
Transverse velocities can affect the cavitation due to change paper) that would be interesting to compare with the
in the blade angle of attack. However, total induced velocities performance in waves.
being much larger compared to the transverse velocities, any
recognizable effect due to transverse velocities alone was not Author’s Reply
observed in the analysis. The paper mentions the Odabasi criterion. We believe, this is
Ship motions influences the propeller in two ways. Part of same as the BSRA criteria you mention. We have used the
wake variation is due to ship motion and cavitation number opportunity to update the paper so that it now also refers to
changes due to change in propeller immersion. Influence of this as the BSRA criterion.
wake variation on propeller operation has been studied in

View publication stats