Second International Symposium on Fishing Vessel Energy Efficiency E-Fishing, Vigo, Spain, May 2012

Increased energy efficiency of the fishing fleet due to improved

hydrodynamic performance

Adrián Sarasquete1, Alejandro Caldas Collazo1, Sophie Coache1, Marcos Meis1, Valentina Ruiz1

1 Vicus Desarrollos Tecnológicos, Vigo, Spain.

Abstract— The main goal of the present article is to highlight the great potential that an improved hydrodynamic design could have on the energy efficiency of fishing vessels. A typical surface longliner was optimized employing Computational Fluid Dynamics; the results were validated by carrying out model testing on a towing tank.

Keywords- Energy efficiency, CFD, hull, propeller, rudder, longliner.

I. INTRODUCTION Surface longliner is a type of gear used for catching

different kinds of fish, like for instance swordfish and tuna. Swordfish surface longliners operate normally at speeds around 8 knots, while setting the line and during the hauling operation. Combining this profile with the transit speed while travelling to and from the fishing grounds, results in a quite high average speed. The improvement of the hydrodynamic performance of a typical surface longliner is feasible, and can be fully approached by using CFD methods. It must be borne in mind that some of the improvements can be applied to a new ship but are not economically feasible for an existing ship with the current oil prices. The optimization methodology was based in a segregated approach, as it was made in different stages which affect specific hydrodynamic parameters. The starting point was to carry out a systematic modification of the hull shape in order to reduce the resistance without losing sight of the hull-propeller interaction. After this, also propeller and rudder were modified for an improved efficiency.

II. ORIGINAL HULL AND RESTRICTIONS The original hull corresponds to a surface longliner with a

length of 36 m and a beam of 9 m. The main dimensions and non dimensional coefficients are presented in TABLE I. In Figure 1 a 3D representation of the hull is also shown. Since the project was based on a real ship, it is fair to define some restrictions during the improvement process if we want this to be considered as a real project. Major restrictions applied to the

main ship dimensions and coefficients are described below. The limits are defined as a percentage comparing the new project with the original one:

Displacement: ±5% Beam: ±5% Draft: ±5% Length waterline: ±5% Hull coefficients: ±5%

Lpp 36,0 m Lwl 39,0 m B 9,0 m

Tm 3,5 m ΔBH 662,2 m3 Sw 454,2 m2 Awl 286,1 m2 Lb 2,3 m

XCB 16,4 m CB 0,58 - CX 0,83 - Cwp 0,82 -

Lwl/B 4,33 - B/T 2,57 - fBT 0,1 -

TABLE I. MAIN DIMENSIONS.

Figure 1. Original hull geometry.

These values were chosen in order to give some freedom during the improvement process but without impairing the ship hull volumes and stability. These small percentages were not likely to have a significant impact on the ship new building price. After the redesign process, described in the present article, the variation on the main dimensions and coefficients are shown in Figure 2.

Figure 2. Deviations from the original design.

As it could be seen, there is no change in the main particulars (Lpp, Lwl, Beam, Draft) nor in non dimensional coefficients, since the changes are focused in hull shape. To achieve the main power reduction goal, the overall problem was decomposed in: Bare hull resistance reduction, propeller-hull interaction improvement, propeller improvement and propeller - rudder interaction improvement.

Figure 3. Modified hull geometry.

III. HULL OPTIMISATION

As shown in Figure 4 and 5, CFD methods were employed for redesigning the hull geometry. The improved characteristics using CFD were:

• Wave making resistance reduction. • Wake equalization and improvement of propeller -

hull interaction.

Figure 4. Hull-Propeller Interaction calculation.

Figure 5. Resistance calculation.

A. Resistance

Although the design process has entirely been carried out employing CFD methods, in the following lines towing tank results for the original and redesigned hull are presented. The towing tank testing was carried out in order to validate the CFD code, the results from the CFD matched very accurately the tank measurements. As stated before, the first stage aimed at reducing bare hull resistance. The effective power for 11,5 kts is around 356 kW and for 8 kts is around 68 kW. For 11,5 kts the Fn is around 0.3 and for 8 kts is around 0,2. The wave making resistance (Cw) for 11,5 kts is around 50% of the overall resistance while for 8 kts is around 10%; therefore, in this first stage, the main target for reducing power at 11.5 kts is a reduction in Cw. Taking advantage of CFD tools, a new hull geometry was developed. In addition to variations of the hull lines, the bulbous bow has been modified as shown in Figure 6. Its length was increased from 2.3 to 3.4 m and the sectional area was also increased.

Figure 6. Original and modified bulbous bow.

After this process, model tests were carried out at the towing tank facilities. Results of these tests are presented in Figure 8. The decrease in effective power for 11,5 kts is around 30%, as for this modified hull the delivered power is 265 kW. As expected, for low Fn the differences of power between the hulls are negligible.

Displacement

Wetted Surface Cb Cx Cwl fBT Lb/Lpp

% -2.7% 4.4% -2.7% 2.0% -2.3% 19.2% 47.8%

-4.0%

-2.0%

0.0%

2.0%

4.0%

%

Deviations from the original design

Figure 7. Towing tank test - Hull B.

Figure 8. Effective Power.

B. Wake

One of the indicatives that gives us a good measure of the hull-propeller interaction is the wake configuration. It is a fact that the propeller works inside a non-uniform velocity field influenced by the shapes of the hull upstream the propeller. It is also clear that this field should be as uniform as possible, since a non-uniform wake field will lead to vibration and cavitation problems and to a worse propeller performance due to a varying angle of attack at the blade profiles. In the next figure a representation of wake field measurements through the Taylor wake coefficient on the propeller plane is shown for the original hull.

Figure 9. Original wake field.

The Taylor wake coefficient (1) is defined as the difference between the local flow and the hull advance velocities divided by the hull velocity.

𝑤 = 1 −𝑉𝑋𝑉𝑆

(1)

It is obvious that values close to unity imply stagnation of the water flow, which happens for the original ship on the upper part of the propeller disc. Figure 10 shows local velocity divided by the ship advance velocity (Vx/Vs) against the angle for each radius. Again, a strong velocity change near 0º can be seen for all radii.

Figure 10. Original wake distribution.

This configuration was improved by numerical methods modifying the ship aft body. Not only the wake field but also important values as the suction are affected by the hull shapes. Because of this fact, both parameters have been taken into account when hull modifications were carried out.

Figure 11. Section 1 for hulls A and B.

Figure 11 shows section 1 for both ships, full line for Hull A and dashed line for Hull B. Taylor wake coefficient on the propeller plane for the new design is represented in Figure 12.

3

103

203

303

403

503

603

703

6 7 8 9 10 11 12 13 14

Pow

er [k

W]

Speed [kts]

Effective Power

Hull A. Rudder I. Hull B. Rudder I.

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0 100 200 300 Vx

/Vs

Ang [º]

Vx/Vs

R06 R08 R1

Figure 12. Modified wake field.

At first sight, it can be seen that the velocity field is more uniform and there is a significant reduction on the velocity peak in the upper part of the disk.

Figure 13. Modified wake distribution.

In Figure 14 a comparison of the angular variation of Vx/Vs for both ships on the 0.7 radii is shown. It is clear that the new design have a better wake uniformity and lower peaks.

Figure 14. R0.7 wake distribution.

C. Propeller hull interaction.

As discussed in the previous section, the wake field is one of the characteristics affecting the hull-propeller interaction; the other important factor is the thrust deduction factor. The thrust deduction is calculated by doing a self propulsion test / simulation and it represents the amount of thrust from the propeller used for pulling the ship forwards (i.e. not useful thrust). In Figure 15, the thrust deduction factor (2) is plotted against speed for the original hull shape.

𝑡 =𝑇 − 𝑅𝑇

(2)

Figure 15. Thrust deduction factor (original hull).

Taking into account ec. 2, it is evident that the higher the thrust deduction factor is, the higher the delivered power is. So, the thrust coefficient must be as small as possible. Figure 16 corresponds with the percent decrease of this coefficient for Hull B referred to the original case at different speeds.

Figure 16. Thrust deduction factor reduction.

It is clear from the above picture that the improvement is remarkable for all hull speeds. As shown in Figure 17 the delivered thrust is reduced due to a smaller resistance and lower thrust deduction factor, both due to an improvement on the hull shape. For 11 kts the reduction is around 30% (45 kN for the modified hull and 62 kN for the original). It could be seen that, as expected, improvements grow with the Froude number and the wave making resistance.

0.0

0.2

0.4

0.6

0.8

1.0

1.2

0 100 200 300

Vx/V

s

Angle [º]

Vx/Vs

R06 R08 R1

0

0.2

0.4

0.6

0.8

1

1.2

0 100 200 300

Vx/V

s

Angle [º]

Vx/Vs

R0,7 Hull B R0,7 Hull A

0 0.05

0.1 0.15

0.2 0.25

0.3 0.35

8 9 10 11 12 13 t

Speed [kts]

Thrust Deduction Factor

0% 5%

10% 15% 20% 25% 30% 35% 40% 45%

8.5 9 9.5 10 10.5 11 11.5 12 12.5 13 13.5

%

Speed [kts]

% t reduction

Figure 17. Delivered Thrust.

Another important magnitude for the quantification of the hull propeller interaction is the relative rotative efficiency. This value tries to capture the performance differences of the propeller when it works in open water conditions and in behind conditions. This value multiplied by the open water efficiency results in the behind efficiency which gives us a more real value of the propeller efficiency when it works behind the hull. Figure 18 compares this coefficient for the original and modified hull. As it could be seen the propeller works better behind the modified hull.

Figure 18. Behind efficiency.

Figure 19. Delivered Power.

In the previous figure a comparison of the power delivered by the engine is shown. As it could be clearly seen, the new design demands less power for the same velocity. For example, at 11 kts the power reduction is around 25% (from 479 kW to 360 kW).

Figure 20. Power Reduction.

IV. PROPELLER

The previous section has highlighted the importance of the hull - propeller interaction, but there are some other aspects to be improved: with a new design of wake adapted propeller, the propulsive efficiency can be increased. Indeed, with a uniform wake, the diameter can be increased since the pressure pulses and cavitation are no longer a limiting factor due to the better water flow; this increase in diameter implies a higher efficiency. The new diameter and reduced power lead to a smaller rotational speed of the propeller. It is also possible to reduce the area of the blade and the thickness and thickness / chord ratio, both leading to an increase in efficiency. All these different steps will allow designing a high efficient propeller adapted to the ship’s wake and thus achieving a higher relative rotative efficiency. The effect of propeller parameters on propeller operation and hull - propeller interaction has to be analysed. On the one hand, some statistical computations have been carried out modifying parameters such as blades number, diameter and RPM for the design speed, 11 kts. It has resulted that the efficiency was improved with an 8% increase of the diameter. Then, on the other hand, a potential based propeller panel method has been used. Panel methods should represent an improvement with respect to lifting surface methods, due to the possibility to properly model the effect of blade thickness, as well as the presence of other thick bodies (hub, duct, and portion of the hull). While the computation of the global forces (and moments) should have similar accuracy compared with lifting surface methods, panel methods should allow a better determination of the pressure distribution, especially at the leading edge, and this is a valuable improvement for predicting cavitation inception. At this stage, parameters such as profile, pitch and thickness have a huge influence on the propeller operation and efficiency. After several iterations, a new profile section had been selected to get a slightly higher efficiency. Finally, an improvement of 15% has been achieved with the new design. The characteristics of the two propellers are described in the table II.

0

20

40

60

80

100

120

140

8 9 10 11 12 13 14

Thru

st [k

N]

Speed [kts]

Delivered Thrust

Hull A. Rudder I. Hull B. Rudder I.

0.52

0.53

0.54

0.55

0.56

0.57

8.5 9 9.5 10 10.5 11 11.5 12 12.5 13 13.5

ƞ

Speed [kts]

ƞ Behind

Hull A. Rudder I. Hull B. Rudder I.

0 200 400 600 800

1000 1200 1400

8 9 10 11 12 13 14

Pow

er[k

W]

Speed [kts]

Delivered Power

Hull A. Rudder I. Hull B. Rudder I.

8.5 9 9.5 10 10.5 11 11.5 12 12.5 13 13.5 % -7% -11% -15% -20% -23% -25% -26% -26% -27% -28% -27%

-30%

-25%

-20%

-15%

-10%

-5%

0%

%

% Power Reduction

Characteristics Original New Diameter, m 2,486 2,7 Pitch at 0,7R/D 0,785 0,957 EAR 0,69 0,447 Blades number 5 4 Shaft power, kW 1015 850 Propeller RPM 230 190

TABLE II. PROPELLERS CHARACTERISTICS.

In the following figures, the diagrams of the two propellers are presented, the original propeller and the wake adapted one respectively. The maximum efficiency for the new design is quite higher than the original one and it is obtained for an advance coefficient around 0.7 and not around 0.6 as it is for the original propeller, due to the diameter and rpm difference.

Figure 21. Original Propeller Diagram.

Figure 22. Wake adapted propeller diagram.

V. RUDDER The last stage of the design process was to carry out a new rudder design employing two different criteria, the first one understanding the rudder as active manoeuvring device and the second one as energy recovery device working in the propeller slipstream. The new design has the same main

dimensions as the original, being the differences located in shape and the inclusion of a Costa Bulb. The original rudder is a NACA20 profile rudder, with constant chord length (1500 mm) and a height of 3000 mm as could be seen in Figure 23.

Figure 23. Original rudder geometry.

Again, CFD calculations were employed for the design of the new rudder geometry. Figure 24 is a representation of the pressure distribution on the propeller and on the original rudder profile.

Figure 24. Propeller rudder CFD calculation.

The same kind of pressure plot is presented on figure 25 but now for the new rudder design. As can be seen, the main differences on the tuned rudder are in the profile shapes and a modified propeller hub, leading to a Costa bulb.

Figure 25. Propeller rudder CFD calculation.

Again, after the design process, the new geometry was built for towing tank experiments in order to validate the calculations. In Figure 26 lift coefficients against the rudder angle are plotted. These results are measured at the towing

0

0.1

0.2

0.3

0.4

0.5

0.6

0 0.2 0.4 0.6 0.8 1

Kt, 1

0Kq,

Eta

J

KTM -Orix 10KQM -Orix η0m -Orix

0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

0 0.2 0.4 0.6 0.8 1 1.2

Kt, 1

0Kq,

Eta

J

KTM -Mod 10KQM -Mod η0m -Mod

tank facilities for fixed hull velocity and propeller rpm. As it could be seen the lift force is higher for the new rudder design, which in principle means that the manoeuvre capacity of the new rudder is also higher. Once again, these results matched very accurately the previous CFD predictions.

Figure 26. Rudder lift against angle.

As stated before, the second criteria is to take into account the energy recovery ability of the rudder. For this, we must employ two different viewpoints: the energy recovered by the rudder and the induced load into the propeller. These physic facts were taken into account when the CFD design was carried out. Beyond this, when analyzing towing tank data, both are recovered under the same magnitude: delivered power reduction.

Figure 27. Delivered power reduction due to the new rudder.

Figure 27 represents the power demand reduction for the case with the new rudder design referred to the same case with a conventional rudder. As it could be seen for the working point, 11 kts, the power demand reduction is around 12%.

VI. OVERALL RESULTS This section presents a summary of the improvement due to all proposals together. Figure 28 shows us the delivered power of the original long liner against the modified one for a range of operational speeds from 7,5 to 13 kts, with clean hull and no sea margin.

Figure 28. Delivered Power overall results.

It is clear that a correct hydrodynamic design will result in a drastic drop in the fuel bill along the lifetime of the ship. For 11 kts the reduction on the power demand is around 40%, maintaining similar values along the different operation speeds of the ship. In order to carry out an assessment of the fuel saving during a real fishing campaign, we have increased the power delivered by each ship by 25%, in order to account for sea margin and fouling. For clarity, the table below accounts only for a reduced number of speeds.

Figure 29. Delivered Power with 25% sea margin.

The overall propulsion energy consumption is calculated for one year, this means approximately 300 working days a year. The typical operational conditions of the surface longliner have been studied, i.e. Sail, Set, Haul, Search and Stopped, correlating them with each speed. For each condition and speed, a figure for the yearly propulsion energy consumption is calculated; this leads to yearly savings in MWh. This figure is translated into money with a value of 0.2 € / kWh, deriving from 230 gr/kWh and 0.75 € per litre of fuel. The estimated results in an overall amount of near 140.000 € per year, just with the savings on the propulsion and without taking into account aspects like the better manoeuvrability, which also influences the hauling process, making it more efficient. Another aspect we should not forget is the comfort onboard, the improved wake and propeller which lead to a reduced cavitation and transmitted pressure pulses to the hull; this will reduce significantly the vibration and noise onboard, making life and work easier. If we combine savings derived from a hybrid propulsion system as the one shown in [10], together with some improvements on the freezing plant, the energy efficiency of a fishing vessel with these features could be improved up to 50% compared to a conventional vessel.

-0.2

-0.1

0

0.1

0.2

-40 -20 0 20 40

Cl

Angle [º]

Lift VS Angle

Rudder I Rudder II

9.5 10 10.5 11 11.5 12 12.5 13 % -5% -10% -12% -12% -11% -10% -6% -2%

-14% -12% -10%

-8% -6% -4% -2% 0%

%

% Delivered Power Reduction due to new rudder

7.5 9.0 10.0 11.0 12.0 13.0 Hull A 108.00 183.00 293.00 479.00 753.00 1084.00

Hull B 68.95 126.49 186.00 286.00 431.00 640.00

% 36% 31% 37% 40% 43% 41%

0 200 400 600 800

1000 1200

Pow

er [k

W]

Power Vs Speed

7.5 9.0 10.0 11.0 12.0 13.0 Hull A. Rudder I 135 229 366 599 941 1355

Hull B. Rudder II 96 172 259 406 619 911

% 29% 25% 29% 32% 34% 33%

0 200 400 600 800

1000 1200 1400 1600

Pow

er [k

W]

Power Vs Speed

Condition Sail Set Haul Search Stop Av. Speed (kts) 11 9 VAR. 7,5 0 %Time 35% 15% 31% 10% 9% Hours 2520 1080 2232 720 648

Av. Pow. Orig. (kW) 599 220 99 130 0 Energy (MWh) 1509 237,6 222 93,6 0 Av. Pow. Mod. (kW) 382 163 76 91 0 Energy (MWh) 963 176 169 66 0 %Saving 32% 25% 24% 29% 0% Sav. Energy (MWh) 546 62 53 28 0 € Saved 110.849 12.493 10.667 5.699 0

Total Saved 139.707 €

TABLE III. ECONOMIC STUDY RESULTS.

VII. CONCLUSIONS

• A correct hydrodynamic design is of key importance during the concept design of a ship; it is very important to pay attention to this during the early design stages or any improvement might become a headache if the project is mature.

• It is important not only under the scope of a reduced resistance but also for an improvement on the hull - propeller interaction.

• A good propeller should be always designed adapted to the real ship wake and operational profile.

• Wake adapted rudder can provide huge savings at a reduced cost.

• All the proposed modifications are technically feasible for a new project, while others are feasible for retrofitting an existing ship as well (like rudder and propeller)

• The proposed improvements can be combined with other measures in order to reach savings up to 50% for a new project compared to a conventional one.

ACKNOWLEDGMENT

The presented results have been obtained in cooperation with Baliño SA on the BAIP 2020 research project, partly sponsored by CDTI agency under the Spanish Ministry of Industry.

REFERENCES [1] Bertram, V (2000). Practical Ship Hydrodynamics. 2nd ed. Woburn:

Butterworth-Heinemann

[2] Caldas, A., Meis, M., Sarasquete, A. (2010) CFD validation of different

propeller ducts on Open Water condition. 13th Numerical Towing Tank Symposium, Germany.

[3] Caldas, A., Sarasquete, A.. (2009) Energy efficiency improvement via

rudder geometry modification (Alejandro Caldas Collazo, Adrián Sarasquete Fernández). 12th Numerical Towing Tank Symposium. Cortona, Italy.

[4] Caldas, A., Sarasquete, A., Meis, M., (2011) Numerical analysis of

rudder effects upon ducted propeller units. Symposium on Marine Propulsors SMP 2011. Hamburg, Germany.

[5] Ferziger, J.H. & Peric, M. (2002). Computational Methods for Fluid

Dynamics. 3rd ed. Springer-Verlag Berlin Heidelberg.

[6] Hoekstra, M. (2006), A RANS-based analysis tool for ducted propeller

systems in open water conditions, International Shipbuilding Progress, Vol. 53, No. 3, Maritime Research Institute Netherlands

[7] Priego, M., González, L., Izquierdo, J., Sarasquete, A., Pérez, L., (2007) RANSE with free surface computations around fixed DTMB 5415 model and other Baliño’s fishing vessels. 9th International Conference on Numerical Ship Hydrodynamics, Michigan, USA

[8] Reichel, M. (2009) Influence of rudder location on propulsive

characteristics of a single screw container ship. First International Symposium on Marine Propulsors, Trondheim, Norway.

[9] Sánchez-Caja,A., Sipilä, T.P. & Pylkkänen, J.V. (2009) Simulation of

viscous flow around a ducted propeller with rudder using different RANS-based approaches.

[10] Sarasquete, A., Rojas, D., Solla M., Doval, J., Iturralde, M.. (2010)

Potential of hybrid systems with permanent magnet motors for propulsion improvement on surface longliners. First international Symposium on Fishing Vessel Energy Efficiency. Vigo, Spain.

[11] Sarasquete, A., Caldas, A . (2009) Investigación y desarrollo con CFD

en aplicaciones marinas. Marine Simulation Workshop. Madrid, Spain.

[12] Sarasquete, A., Caldas, A. (2009) Investigación sobre la eficiencia energética en timones. Líneas de investigación en Vicus Desarrollos Tecnológicos S. L. 48º Congreso de Ingeniería Naval e Industria Marítima. Vigo, Spain.

[13] Sarasquete, A., Juandó, A., Caldas, A., Zapata, F. (2011) Energy

efficiency audits on ships: Hydrodynamic aspects for energy efficiency improvements. 2nd International Ship Design and Naval Engineering Congress. Cartagena de Indias, Colombia.

[14] First International Symposium on Marine Propulsors, Trondheim,

Norway. [15] STAR CCM + User Guide (Version 4.0.6) (2009).

[16] ITTC Symbols & Terminology (2008).