Experimental Study of Air Layer Drag Reduction with Bottom Cavity for A Bulk Carrier Ship Model

Air lubrication by means of a bottom cavity is a promising method for ship drag reduction. The characteristics of the bottom cavity are sensitive to the flow field around the ship hull and the effect of drag reduction, especially the depth of the bottom cavity. In this study, a ship model experiment of a bulk carrier is conducted in a towing tank using the method of air layer drag reduction (ALDR) with different bottom cavity depths. The shape of the air layer is observed, and the changes in resistance are measured. The model experiments produce results of approximately 20% for the total drag reduction at the ship design speed for a 25-mm cavity continuously supplied with air at Cq = 0.224 in calm water, and the air layer covers the whole cavity when the air flow rate is suitable. In a regular head wave, the air layer is easily broken and reduces the drag reduction rate in short waves, particularly when λ/Lw1 is close to one; however, it still has a good drag reduction effect in the long waves.


Introduction
A brief review of the technologies of friction drag reduction was published by Ahmadzadehtalatapeh and Mousavi (2015). Ceccio (2010) presented different applications and a detailed explanation of the distribution of air under a ship's hull during air injection. Air layer drag reduction (ALDR) by means of a bottom cavity has been proven to be an effective method of ship drag reduction.
The theoretical model of ALDR can be found in the paper published by Butuzov (1966), and then he conducted an experiment with a full-scale ship in 1988 (Butuzov et al., 1988). The first full-scale ship test was carried out by Basin et al. (1969). Drag reduction rates up to 20% have been obtained for inland vessels (Butuzov et al., 1990). Subsequently, researchers conducted a large number of experimental studies in the laboratory. In most ALDR experiments, the air is injected through continuous slots, which are either open or filled with porous material, which ensure a uniform air layer under the test surface.
To obtain a uniform and stable air layer under the bottom hull, Zverkhovskyi (2014) designed a device called a 'cavitator' to impede the flow of water on the underside of the vessel, with a small slotted opening behind the cavitator delivering the air supply. This method was found to be particularly successful because it significantly reduces the air flow rate required to create the air layer. The effect of air layer thickness on drag reduction was verified experimentally by Mizokami et al. (2010). The result shows that the energy consumed by the blower depends on the thickness of the bottom air layer. Following the fundamental experimental study on a flat plate with a cavity conducted in the Emerson Cavitation Tunnel of Newcastle University by Slyozkin et al. (Slyozkin, 2011;Slyozkin et al., 2014), Butterworth et al. (2015) conducted an experimental test on an existing container ship model with a cavity of 0.0387 m 2 at the bottom of the ship model. The model experiment results show that the effect of ALDR can reach 4%−16% and the air escape from the bottom hull was affected by the characteristics of the bottom cavity.
According to conclusions of Slyozkin and Butterworth, the design of the bottom cavity plays an important role in the evaluation of the ALDR. Mäkiharju et al. (2010Mäkiharju et al. ( , 2012 conducted a series of tests with a flat plate with a device called a 'BFS', which helps the air layer to form easily while reducing the air supply and pressure. Experimental test on a flat plate and a ship model with air lubrication system were successively conducted by Jang et al. (2014). From the tests, it is considered that the air lubrication at the bottom of the ship has great potential for energy saving if there is no significant deterioration under the real sea condition. Matveev (2003Matveev ( , 2007 solved an important problem with a simplified model, which is the interaction between the waves and the cavity. The result shows the main parameters that affect air layer in the ship bottom cavity are the shape of cavity and the velocity of the ship model. The same author has developed a platform to test different types of Air Cavity Ships (ACS) using the necessary instrumentation to improve and optimize this technique (Matveev, 2015). The interaction between the cavity and the boundary of ship was also analyzed by Amromin (2016).
The above studies on ALDR mainly focus on the drag reduction rate of the model in calm water. It is very rare that the research about the shape of the bottom cavity on drag reduction for low-speed vessels, especially, there are few studies on the movement of the ship and the shape of the bottom air layer in the waves. Within this framework, we propose an experimental study with an aim to investigate the effect of the characteristics of the bottom cavity, especially the depth of the bottom cavity on the ship drag reduction and the air layer shape. In this study, a model ship based on a bulk carrier was tested in the towing tank using an ALDR technique with bottom cavity. Section 2 of the paper describes the experiments description, which includes the test model, test facility and test setup. In Section 3 the experiment results in the calm water and regular wave are presented and discussed, which include the ship resistance, the air layer shape and the ship motion. Finally, Section 4 presents the conclusions obtained from the study.

Experimental model
A flat large ship was selected as the test hull form and a ship model was manufactured with a scale ratio of 38.0 to perform the tests. The bottom of the ship has a broad flat structure with a long parallel body section, which is suitable for the application of ALDR technology. The principal particulars of the ship are shown in Table 1.
The flat bottom area, where reduction in the frictional resistance is expected to occur, is approximately 40% of the total wetted surface area under the design displacement condition. Fig. 1 shows the test model with the bottom cavity. As the bottom of the ship is an arc-shaped flat structure, an arc-shaped cavity is used under the bottom of ship and a slope structure is set at the tail of the cavity. The dimensions of the cavity are 4.32 m×0.82 m. The area of the bot-tom cavity is approximately 84% of the large flat structure.

Experimental setup
Two types of air injection devices are used to cover the bottom of the cavity with an air layer: arc air injection device and hole-shaped air injection device, as shown in Fig. 2. Fig. 3 shows the installation schematics of two air injection devices for the ship model. The arc air injection device is installed at the fore-stepped position of the cavity, which forms an integral part of the cavity. The hole-shaped air injection device consists of eleven through-holes with 10 mm in diameters, which are mounted on the rear of the cavity's front step, along the edge of the cavity. To avoid an undesirable increase in ship resistance, the air injection devices were mounted flush with the bottom surface of the ship hull.
The main purpose of this study is to investigate the influence of the bottom cavity depth. Therefore, it is necessary to set different depths for the bottom cavity. In the experiment of ALDR, the method of adding a basic plank to change the cavity depth is adopted, where the thickness of the basic plank is 5 mm. First, the depth of the bottom cavity is set to 25 mm, and then it is changed in increments of five basic planks to 20, 15, and 10 mm.
Air was injected through an array of holes on the base plate of each air injection device according to the air supply chain, as shown in Fig. 4. Owing to the sensitivity of the measuring instrument on the carriage, it was not possible to keep the air compressor running throughout the test as the vibrations generated were detected within the results. Therefore, compressed air generated by an air compressor was stored in an air cylinder on the tank trailer, and the flow rate of air supplied to the air injection devices was adjusted by a personal computer and measured using several air flow meters.
The model ship for tests in the towing tank is shown in Fig. 5. To avoid interference between the pipe and ship model, the air injection pipes are made of plastic hose and they prevent contact between the pipes and the ship model.

Experimental results
To further reveal the effect of air flow rate and flow velocity on the air layer shape, the non-dimensional air flow rate coefficient, C q , is defined as follows: where Q is the air flow rate; V is the inflow velocity; B is the transverse width of the air injection entrance; and δ is the thickness of the boundary layer of the air injection entrance without air injection, which is given by Eq. (2): where Re is the Reynolds number of the air injection entrance, which is given by where x is the distance between the air injection and bow of the ship model, and ν is the water viscosity coefficient, where ν = 1.003×10 −6 at normal temperature. The relative drag reduction rate and absolute drag reduction rate of the resistance of the model ship is defined as follows: where is the resistance of the ship model with a cavity, but without air; is the resistance of the ship model with both a cavity and air; and is the resistance of the ship model without a cavity or air.

Baseline results
A speed curve of the datum hull form was generated as a basis for comparison. Fig. 6 shows the resistance curve of the ship model for the Fr number ranging from 0.107 to 0.182. It can be seen that the resistance of the ship model increases with the presence of the cavity, and the added value  It can be seen that the trim and the heave exhibit similar tendencies as Fr increases. With the increase in speed, the ship model trims by the head and the center of gravity of the ship model sinks; this phenomenon becomes more obvious as the speed increases. In addition, the value of trim and heave are larger for the 25-mm cavity.

Impact of the air injection mode
Firstly, the impacts of air injection mode from different injection units at the design speed of Fr = 0.155 and 25-mm cavity were investigated. Fig. 9 shows the air injection mode of the two types of devices. Table 2 shows the eight air injection test conditions at the air flow rate of 10 m 3 /h. Fig. 10 presents the results of the ship model resistance under eight different air injection modes for an Fr number ranging from 0.139 to 0.161. It can be seen that the resistance of the ship model is significantly reduced with air injection. The value of resistance is very close under the eight different air injection modes at the flow rate of 10 m 3 /h, which indicates that the effect of the ALDR is not related to the air injection mode. It is shown that the air layer in the cavity has a strong adaptability to the air injection mode, which is very favorable in practical engineering applications.

Effect on resistance
Figs. 11−14 show the variation curves of the relative drag reduction rate of the ship model with the dimensionless air flow coefficient under different cavity depths at Fr = 0.107, 0.129, 0.155, and 0.161. The influence of the cavity depth on the relative drag reduction rate of ship model is great. The drag reduction rate under different speeds and dimensionless air flow coefficients is lower when the cavity depth is 10 mm. The ship speed, air flow, and bottom cav-      Fig. 13, thus, the drag reduction effect of each speed at this cavity depth is analyzed. Fig. 15 shows the resistance curve of the ship model with the 25-mm cavity at different speeds and air C q values. It can be seen that the total resistance decreases greatly when the air is ejected to the bottom of the ship model's cavity. However, the drag of the ship model increases as C q increases. There is a critical value of C q that changes with the speed. When Fr < 0.129, the critical value is approximately 0.26; when Fr > 0.129, the critical value is larger than 0.5. Fig. 16 is a graphical representation of the percentage of ship mode resistance reduction with an Fr number ranging from 0.107 to 0.161. Based on the experimental results, it is clear that nearly 20% of the total resistance can be reduced at the ship design speed (Fr = 0.155) if the air cavity can be continuously supplied with air at C q =0.224. The lower speeds provide larger potential for reducing the resistance of the ship model. Fig. 17 shows the air layer shape under different cavity depths. It can be seen that it is difficult to cover the whole cavity when C q is small and the 10-mm cavity, and the air      bubble is released from the side of ship model. The cover area of the cavity increases with the increasing C q at the same depth cavity and Fr. The air cavity is fully covered when C q = 0.224, Fr = 0.155, and the air bubble is evenly spilt from both sides of the stern.

Experimental result in head waves
The seakeeping experiment of the ship model was carried out at Fr = 0.155 as described in Table 3. The ship model is set to free pitch and heave motions. Time histories of the total resistance R, heave motion z A , and pitch angle θ A are obtained from each test run. The acceleration was measured at different positions of the ship model; a F , a G , and a A represent the first, middle, and tail acceleration, respectively, of the ship model. η ∆ η Rw The added resistance rate and absolute drag reduction rate of the resistance of the model ship in the wave is defined as follows:

R aw
R calm where is the mean resistance in the wave and is the resistance in calm water.

Effect on resistance
Figs. 18−20 show the change in the relative drag reduction rate of the ship model with the wave length under different cavity depths for C q = 0.112, 0.168, and 0.224 and Fr = 0.155. It can be seen that the change in relative drag reduction rate with wavelength has the same tendency at different C q values. In general, the relative drag reduction rate is larger in the 25-mm cavity and in the long wave. The relative drag reduction rate of the ship model can reach more than 30% in the long wave at C q = 0.224, which is smaller than the value for calm water.

Effect on motion
Figs. 21 and 22 show the experimental results of the ship model pitch and heave motions with the wavelength under different cavity depths when C q = 0.224 and Fr = 0.155. It can be seen that the cavity depth has little effect on the pitch and heave motions of the ship model in the short Fig. 18. Variation in the relative drag reduction rate with the wavelength for C q = 0.112, Fig. 19. Variation in the relative drag reduction rate with the wavelength when C q = 0.168.    wave; however, in the long wave, the motion amplitude of the ship model is small in the 25-mm cavity. When the bottom cavity of the vessel is large, the air flow is large for the continuous air injection, which can relieve the movement of the ship model in the long wave. Fig. 23 shows the experimental result of the ship model local acceleration with the wavelength under different cav-ity depths when C q = 0.224 and Fr = 0.155. It can be seen that the change in local acceleration with the wavelength has the same tendency at different air layer thicknesses. The local acceleration increases firstly and then decreases with the change in wavelength, and the cavity depth has little effect on the local acceleration of the ship model. Fig. 24 shows the air layer shape under the bottom of the ship model in the regular wave at Fr = 0.155 and C q = 0.224. It can be seen that the air layer breaks in the regular waves. In addition, the rupture phenomenon in the short wave is more obvious than that in the long wave. The phenomenon of air layer break is most serious when is close to one. The air layer is easily broken when the cavity depth is 10 mm; however, the air layer covers most of the bottom cavity at the 25-mm depth, which shows that a larger cavity depth is conducive to improve the stability of the air layer in waves.

Discussion and conclusions
Experiments were conducted on the ALDR on the lower surface of the ship model in calm water and in regular waves. It can be confirmed from the experiment that the air layer generated by air injection at the bottom of the hull can be used as an effective means to reduce the frictional resistance of the ship. Accordingly, the following conclusions can be drawn from the experimental analysis: (1) The resistance of ship model increases when the bottom cavity is set, but has little effect on the trim and heave of the ship model.
(2) The air layer in the bottom cavity has a strong adaptability to the air injection mode, and the ALDR is not related to the air injection mode in this experiment.
(3) The absolute drag reduction is nearly 20% at the ship design speed if the 25-mm cavity is continuously supplied with air at C q = 0.224 in the calm water. In addition, the drag reduction is good under the same condition in the long wave.
(4) The depth of cavity has little effect on the pitch and heave motions of the ship model in the short wave; however, in the long wave, the motion amplitude of the ship model is small in the 25-mm cavity.
(5) The air layers of the ship bottom in waves and calm water obviously differ in shape. The air layer breaks in regular waves. Moreover, the rupture phenomenon in the short wave is more obvious than in the long wave. A larger cavity depth is conducive to improve the stability of the air layer in the waves.