Hydrodynamic Performance Analysis and Verification of Transverse Thrusters

YAO Zhen-qiu,YAN Zhou-guang

(School of Naval Architecture and Ocean Engineering,Jiangsu University of Science and Technology,Zhenjiang 212003,China)

1 Introduction

Transverse thrusters are widely used in ship and marine engineering especially in manoeuvring aids,but the difference from propeller is that there are no published design charts.As a standard component device of dynamic positioning system,the study on hydrodynamic performance of transverse thrusters is not sufficient.In the design and calculation,it mainly uses the rated thrust provided by manufacturers or modified design of ducted propeller.Thus,it can not reflect the true flow field,thrust structure,ship impact and can not compare the efficiency.In this paper,CFD technology is used to make the numeric calculation and flow simulation.Based on the thruster model experiments by Taniguchi[1],the effects of pitch ratio,expanded area ratio and boss ratio of the propeller,effects of tunnel length and wall inclination of the block are observed on the performance of the transverse thrusters.The expectation of this paper is to provide a reference for research and design of ship maneuvering and dynamic positioning system.

2 Performance characterization

The usual measure of propeller performance defined by trust coefficient KT,torque coefficient KQand open water efficiency η0are given by equation(1).However,at the transverse thrusters’ condition η0decreases to zero as the advance coefficient J tends to zero.Another measure of performance is needed in order to compare the efficiency of transverse thrusters.

The parameter most widely used in aeronautical applications is the static merit coefficient C which is usually used to characterize the performance of helicopter rotors and VTOL aircraft,and it is defined by equation(2)[1-2].The efficiency of transverse thrusters η comes from C,showed in equation(3)[1].

where,T is the total lateral thrust,SHP is the shaft horsepower,D is the tunnel diameter and ρ is the mass density of the fluid.

where,CTand CQare the usual thrust and torque coefficient,CFrepresents the force measured on the block.For the static condition,CFcan be valued by the surface force on the hull’s inlet side[3].CTFis the sum of CTand CF,representing the thrust coefficient of the whole thruster.

3 Mathematical model

3.1 Governing equation

By rotating reference frame transformation,the rotary movement of the blades can be regarded as three-dimensional incompressible steady fluid flow,the continuity and momentum equations are as follows[4]:

where,ui,ujare the time-averaged velocity components（i,j=1,2,3）,p is the mean pressure,ρ is the density of the fluid,μ is the dynamic viscosity,is the apparent stress.

3.2 RNG k-ε turbulence model

The turbulence model used in CFD calculation is RNG k-ε model,its transport equations are as follows[4]:

4 Performance analysis of thrusters with no ship speed

4.1 Calculation model

The four-leaf Kaplan type is chosen as controllable pitch propeller of the numerical model,and the basic geometric parameters of the model reference Taniguchi’s experiments are shown in Tab.1.The modeling transverse thruster used in the experiments is shown in Fig.1[1].

Tab.1 Basic geometric parameters of the thruster

Fig.1 Taniguchi’hull form arrangement

In order to solve the relative rotation between the propeller and the block,the Multiple Rotating Reference Frames model provided by Fluent is used.The propeller is placed in a small area slightly larger than the blade diameter as a dynamic field,with others a static field.Thus,propeller’s unsteady rotational motion is changed into a steady movement relative to a moving reference frame[5].Fig.2 shows the computational domain in Fluent.No-slip wall boundary conditions are applied on the surrounding wall,and pressure-inlet and pressure-outlet are set on the import and export.In Fig.3 and Fig.4 the surface grids on the transverse thruster and a blowup of the tunnel are shown respectively.

Mesh quality is the premise of the accuracy of numerical calculation.In order to improve the quality of mesh,a hybrid of structured grid and unstructured grid is used[6].The import and export of watershed V4 use structured grid,the watershed near the propeller and the block use unstructured grids.Blade surface mesh has a local refinement;surface mesh of the block uses size function provided by Gambit to control the density of the grid,making the grid denser near the tunnel.

Fig.2 Computational domain in Fluent

Fig.3 Surface grids on the block

Fig.4 Surface grids on the propeller

4.2 Effect of pitch ratio and expanded area ratio

This paper considers Kaplan blade design having pitch ratios 0.5,0.7,0.9,1.1 and 1.3,expanded area ratios 0.3,0.45 and 0.6 with a blade number of four.At the same speed 20rps,the trust coefficient,torque coefficient and the open water efficiency of the thrusters are calculated to study the influence of pitch ratio and expanded area ratio.The CFD simulation results are compared with experiment results by Taniguchi shown in Fig.5 and Fig.6,where the curve for the experimental values and point values for the CFD calculations.

Both of the results from the figures show that the trust and torque coefficients increase obviously with pitch ratio and slightly with expanded area ratio.The coefficients simulated and experimented are similar when pitch ratio is less than 1 and the deviation grows with pitch ratio continuing to increase.The main cause of the deviation is the trust of the block as is shown in Fig.5.In Fig.7,the results show that there is little difference among efficiency of different expanded area ratios and the efficiency becomes stable after the pitch ratio 1.

Tab.2shows the comparison of total thrust coefficient and torque coefficient by the CFD simulation and experiment.The errors show the deviation and several factors may cause this deviation:The gap between the propeller and tunnel which is not mentioned in reference 1;the measurement device used in the experiment could influence the result;the difference of the environment such as density of the fluid.However,the maximum error is around 7.5%and average error is less than 5%,which indicates that CFD method can accurately calculate the trust and torque of thrusters.Thus,a further calculation and analysis below makes sense.

Fig.5 Effect of Ac/Ad on trust

Fig.6 Effect of Ac/Ad on torque

Fig.7 Effect of Ac/Ad on efficiency

Tab.2 Comparison of CFD and experiments(Ac/Ad=0.45)

Fig.8 shows the pressure nephogram on section x=0 at the expanded area ratio of 0.45 and pitch ratio of 0.7.It can be observed that a low pressure forms near the import of the tunnel,resulting in the block wall surface of the inlet side produces thrust as the same direction as the propeller produces,as is shown in Fig.9.This is the reason why the block produces thrust.In Figs.10 and 11,pressure distribution on blade back and streamline near propeller are given.

Fig.8 Pressure nephogram on x=0

Fig.9 Wall pressure near inlet

Fig.10 Pressure distribution on blade back

Fig.11 Streamline near propeller

4.3 Effect of tunnel length

As practical used in vessels,the different installation positions of the thrusters lead to differences in tunnel length.So the thrusters at different tunnel lengths of 1D,2D,3D and 4D are considered,with pitch ratios at constant 0.75,expanded area ratio at constant 0.45,speed at 20 rps,the effect of tunnel length on hydrodynamic performance of thrusters is observed.

Fig.12shows the calculated results of CFD.The thrust of internal propeller does not change with the tunnel length while the torque and the thrust of block decrease slightly.The efficiency reaches a maximum at the tunnel length 2D.As compared with 2D the efficiency of thruster at tunnel length of 3D decreases by 4.6%.Therefore,in the process of designing thrusters,a full consideration should be given to the impact of tunnel length on the performance.Figs.13 and 16 show the wall pressure near inlet and pressure nephogram on section x=0 at the tunnel length of 2D.

Fig.12 Effect of tunnel length

Fig.13 Wall pressure near inlet(L=2D)

Fig.14 Pressure nephogram on x=0(L=2D)

4.4 Effect of wall inclination

In order to consider the effect of the angle between water line of bow and axis of the thruster,the model is simplified to observe the thrusters at the angle of wall inclination of 60°and 90°,with tunnel length at 2D,and other parameters invariant.The effect of wall inclination on hydrodynamic performance of thrusters is considered.Fig.15 shows the calculated results of CFD.

The thrust and torque of propeller are essentially the same while the thrust of block and efficiency reach a maximum at the angle 90°.Compared with the angle of 90°the thrust coefficient of the block and efficiency at the angle of 60°respectively decrease by 9.74%and 7.3%.So in the design,the tilt angle of the hull sides near the transverse thruster should not be too small.Figs.16 and 17 show the pressure nephogram on section y=0 at the angle of wall inclination of 60°and 90°,which show that α has an impact on the flow.

Fig.15 Effect of wall inclination

Fig.16 Pressure nephogram(α=60°)

Fig.17 Pressure nephogram(α=90°)

5 Performance analysis of thrusters with ship speed

5.1 Calculation model

In order to study the interactions between the ship and thrusters and accurately analyze the flow field around the thrusters,a bow equipped by a thruster designed by NAPA is used to calculate the performance of bow thrusters with ship speed.Fig.18 and Fig.19 show the 3D model of bow thruster and computational domain in Gambit.

Fig.18 3D model of bow thruster

Fig.19 Computational domain in Gambit

5.2 Effect of ship speed

With the rotation speed of propeller at constant 125 rad/s propeller thrust coefficient CT,torque coefficient CQand the outer surface of the bow thrust coefficient CFare calculated under different speeds.The results are listed in Tab.3.

Tab.3 Thrust and torque under different ship speed(m/s)

In this table,expressions of thrust coefficient,torque coefficient and efficiency are showed in equations(1)and(3).Thruster’s propeller thrust coefficient CTand torque coefficient CQincrease with the forward speed increased.The thrust coefficient of the surface force on the hull’s CFturns into negative and the thrust coefficient of the whole thruster CTFdecreases with the ship speed increased.The efficiency reaches maximum when speed is zero.With increase in speed,efficiency began to decline sharply,thruster becomes almost completely ineffective at the speed of 2 m/s.

5.3 Failure analysis of bow thrusters

The transverse thruster loses a significant amount of its effect with ship speeds,of the order of 1 to 2 m/s.The cause of this fall-off in net thrusting performance is due to the interaction between the fluid forming the jet issuing from the thruster tunnel and the flow over the hull surface.Fig.20 shows the effect in diagrammatic form.This interaction is conduce to low-pressure region to occur downstream of the tunnel on the jet efflux side,which can extend for a rather large area downstream.This induces a suction force on that side,which reduces the effect of the impeller thrust and alters the effective centre of action of the force system acting on the vessel.

Fig.20 Jet interactions with ship speed

Fig.21 Wall pressure near outlet

Fig.22 Wall pressure near inlet

Figs.21and 22 show the wall pressure near outlet and inlet,which can be observed that at the upstream region of inlet and downstream region of outlet a low-pressure area is formed.Low-pressure area near inlet is small and the produced suction force has the same direction as the thruster’s which is advantageous while the area near outlet is big and the produced suction force has the opposite direction which is disadvantageous.Hence,the differential pressure between the two sides is the most important reason lead up to the failure of bow thrusters.

6 Conclusions

In this paper,Fluent is used to simulate the performance of transverse thrusters.The following conclusions can be drawn from the above study.

(1)Use of CFD software for hydrodynamic performance analysis is feasible and can achieve a satisfactory result.The maximum error between the two is around 7.5%and average error is less than 5%.

(2)The coefficients simulated by CFD and experimented are similar when pitch ratio is less than 1 and the deviation grows with pitch ratio continuing to increase,and the main cause of the deviation is the trust of the block.

(3)When considering the internal propeller,expanded area ratio has little effect on the performance of thrusters;the thrust and torque coefficients increase obviously with pitch ratio.

(4)When considering the installation position of transverse thrusters,tunnel length has effect on the performance and the efficiency reaches a maximum at the tunnel length 2D;the thrust of block and efficiency reach a maximum at the wall inclination angle of 90°.It implies that a full consideration should be given to the impact of tunnel length and the tilt angle between water line of bow and axis of the thruster should not be too small.

(5)Transverse thrusters are at the most effective when the vessel is stationary.The thrusters tend to lose effectiveness as the vessel increases its ahead because of low-pressure region to occur downstream of the tunnel on the jet efflux side.Bow thrusters become almost completely ineffective at the speed of 2 m/s.

[1]Carlton J.Marine propellers and propulsion[M].Boston,MA:Elsevier,2007:333-342.

[2]Palmer A R.Analysis of the propulsion and manoeuvring characteristics of survey-style AUVs and the development of a multi-purpose AUV[D].University of Southampton,UK,2009.

[3]Beveridge J L,Naval S R A D.Design and performance of bow thrusters[M].Defense Technical Information Center,1971.

[4]Fluent Inc.FLUENT 6.3 User′s Guide[K].2006.

[5]Cai R Q,Chen F M,Feng X M.Calculation and analysis of the open water performance of propeller by CFD software Fluent[J].Journal of Ship Mechanics,2006,10(5):41-48.(in Chinese)

[6]Wang C,Huang S,Shan T B.Computations of the propeller’s hydrodynamic performance in abnormal working condition based on multi-block meshes[J].Journal of Ship Mechanics,2010,14(1-2):51-55.(in Chinese)