Online monitoring method of hydrate agglomeration in natural gas pipelines based on acoustic active excitation

Measurement 92 (2016) 11–18

Contents lists available at ScienceDirect

Measurement journal homepage: www.elsevier.com/locate/measurement

Online monitoring method of hydrate agglomeration in natural gas pipelines based on acoustic active excitation Zhigang Qu a,⇑, Huayang Wang a, Yang An a, Huanhuan Yue a, Jian Li b, Yu Zhang b, Yanfen Wang a, Tingting Yue c, Weibin Zhou a a b c

College of Electronic Information and Automation, Tianjin University of Science & Technology, No. 1038 Dagu Nanlu, Hexi District, Tianjin 300222, PR China State Key Laboratory of Precision Measuring Technology & Instruments, Tianjin University, Tianjin 300072, PR China Beijing Electric Power Economic Research Institute, No. 15, Guanganmen Station West Street, Xicheng District, Beijing 100055, PR China

a r t i c l e

i n f o

Article history: Received 19 March 2016 Received in revised form 18 May 2016 Accepted 25 May 2016 Available online 26 May 2016 Keywords: Hydrate Acoustic Natural gas pipeline Online monitoring

a b s t r a c t Natural gas hydrate, which forms and agglomerates under the condition of low temperature and high pressure, can easily affect production or even lead to serious accidents. In order to prevent hydrate agglomeration in a pipeline, the industries generally add seriously excess inhibitor with experience, which causes serious environmental pollution and huge costs. However, due to the uncontrollable production and environmental conditions, hydrate plugging in a pipeline still occurs. According to the issues above, the method, for hydrate agglomeration online monitoring in a natural gas pipeline based on acoustic active excitation, has been presented in this paper. The method is to monitor and locate the hydrate agglomeration in a pipeline in order to take measures in time. Both the modeling and the experimental results show that the method can monitor and locate the hydrate agglomeration online at different locations with good accuracy. Ó 2016 Elsevier Ltd. All rights reserved.

1. Introduction With the continuous deterioration of the environment and excessive consumption of the resources, natural gas, a clean and environment friendly energy source, has been seen as a promising alternative energy source in the future [1]. Compared to other fossil fuels, natural gas has relatively lower carbon intensity and higher fuel efficiency in power production [2]. The growth rate of annual global demand of natural gas is expected to over 10% from 2007 to 2035 [3]. Natural gas will exceed coal to be the world’s largest energy source after 2030. Flow assurance, which ensures safe transportation through a pipeline, is of crucial importance. However, how to keep the pipeline unobstructed is a considerable puzzle in the area of natural gas industry. A main reason of blocking the pipeline is the natural gas hydrate which can form under normal transportation environment [4]. Natural gas hydrates (also known as flammable ice) usually are ice-like crystalline substances that are composed by methane and water, and can burn in air to form water and carbon dioxide [5,6]. Hydrates can form in natural gas pipelines under suitable condi-

⇑ Corresponding author. E-mail address: [email protected] (Z. Qu). http://dx.doi.org/10.1016/j.measurement.2016.05.084 0263-2241/Ó 2016 Elsevier Ltd. All rights reserved.

tions (usually high pressure and low temperature) with little hindrance and often lead to serious obstructions in pipelines. To ensure the flow assurance, different measures have been employed to prevent the hydrate formation in pipelines, which either reduce the water amount in the pipeline or destroy one of the essential conditions for hydrate formation. Currently, in the oil and gas industry, the most applied measure to prevent hydrate blockage is to inject hydrate inhibitors into the pipelines. Most popular hydrate inhibitors are: (1) Thermodynamic inhibitors (THIs) [7]; (2) Low-dosage hydrate inhibitors (LDHIs) [5]; (3) The combination of THIs and LDHIs. However inhibitors can lead to huge cost and environment pollution [8]. Even so, accidents still happen led by hydrate blockage due to the complex environment changes along pipelines, which can cause serious safety and financial loss [9,10]. An hydrate early warning system has been developed by HeriotWatt University [11,12], in which a test sample need to be taken from an well or transport line, then sound velocity and electrical conductivity in the sample are measured and input to a trained artificial neural network (ANN). The output of the ANN can estimate whether hydrate is forming or not. However, the system can only work offline and cannot detect leakages in gas pipelines. University of North Dakota reported a transient analysis method to locate and characterize the plugs in gas wells [13]. In

12

Z. Qu et al. / Measurement 92 (2016) 11–18

Fig. 1. Measurement principle.

principle, this method can estimate the location and the size of a blockage, however, concrete methods don’t come forward and cannot detect gas leakages as well. Also the localization accuracy depends on the dimensions of the deposit. Southwest Petroleum University provided a partial blockage detection solution for the natural gas pipelines [14]. This method employed the transient flow model and the relevant analytical solution previously proposed for partially plugged trunk natural gas pipelines. Nevertheless, it’s only a theoretical method based on the Tikhonov regularization without actual experiments. The Research center of Norway national oil company [15] used an ultrasonic technique for on-line monitoring of solid deposition in radial directions of a pipe. However, this method is suitable for a very short range and cannot be used for a long pipeline.

An acoustic pipeline blockage detection method has been reported [16,17]. However, the research for hydrate monitoring has not been reported. In order to detect and locate hydrate formation inside a pipeline, this paper presents a method based on acoustic active excitation, which is able to monitor and locate hydrate blockages at multiple locations online. Furthermore the method can be applied to monitor leakages in a gas pipeline, which is not covered by this paper. 2. Measurement principle and experimental facilities 2.1. Measurement Principle Fig. 1 shows the schematic configuration of the measurement system. A loudspeaker placed at the start point of the pipeline is

Fig. 2. (a) Test pipe schematic diagram. (b) Pipe with the pump. (c) Details of a bend.

Fig. 3. Description of acoustic wave incident with angles.

Z. Qu et al. / Measurement 92 (2016) 11–18

13

Fig. 4. Reflection signal for one deposit at 18.05 m.

used to generate the incident sound wave which is launched into the pipeline. When the incident sound wave comes across a abnormal event such as a hydrate or leakage, the reflected wave will propagate back and be detected by microphone which is close to the loudspeaker. The reflection signal is then acquired by a Data Acquisition (DAQ) card and processed in the personal computer (PC). In addition to the data acquisition, the DAQ card is also used to generate the driving signal which is amplified by a power amplifier to drive the loudspeaker. Meanwhile the system can detect a leakage of a gas pipeline, which is not within the scope of this paper. The position of the hydrate in pipeline X can be calculated by the following equation:

X¼

ct ; 2

ð1Þ

where c is the propagation velocity of sound in the pipeline, t is the time that the reflected wave takes traveling back to the microphone. The reflected time t can be worked out by cross correlation algorithm. 2.2. Experimental facilities The laboratory experiments have been done using a pipe of 21.58 m, which is made in carbon steel (United States Standard ASTMA 539-90a) with the internal diameter of 100 mm. The pipe is sealed away from the atmosphere and the external temperature is about 25 °C (298.15 K). The experimental pipe is shown in Fig. 2 below.

Fig. 5. Reflection signals for two hydrate deposits at 15 m and 18.05 m.

Fig. 6. (a) Reflection signals for cases with the same length but different diameters. (b) Reflection signals for cases with the same diameter but different lengths.

14

Z. Qu et al. / Measurement 92 (2016) 11–18

Fig. 7. (a) Experiment setup. (b) Details of the start point of the pipe. (c) NI-6366 DAQ card. (d) Markaudio Alpair7. (e) Superlux ECM888B.

3. Hydrate deposit detection models

r2 ¼ divðgradpÞ ¼

3.1. Sound propagation in pipelines The equation of one dimensional acoustic wave is derived from kinetic equation, equation of continuity and equation of state. The equation of three-dimensional acoustic wave is illustrated as follows [18]:

r2 p ¼

2

1 @ p ; c20 @t 2

ð2Þ

where r2 is the Laplacian and is illustrated as follows in Cartesian coordinates:

@2 @2 @2 þ þ : @x2 @y2 @z2

ð3Þ

Acoustic waves can be divided into three classes: the spherical wave, the cylindrical wave and the plane wave. The sound pressure of plane wave which propagates only in one direction can be illustrated below:

p ¼ AejðxtkxÞ ;

ð4Þ

where A is the pressure amplitude of incident wave p, k ¼ x=c is the wave number of acoustics, x ¼ 2pf is angular frequency, f is the frequency of acoustic wave. Acoustic wave propagation depends on the shape, side and materials of the cylindrical pipe. As a result, the relations are shown as follows:

15

Z. Qu et al. / Measurement 92 (2016) 11–18


 1 @ @p 1 @2p @2p 1 @2p r þ 2 2þ 2¼ 2 2: r @r @r r @h @z c0 @t

ð5Þ

Sound signal energy attenuates during the propagation along a pipeline. The Navier–Stokes equations can be used to form the basic equations for acoustic wave damping effect (including viscous absorption and heat conduction).

@p þ r  qV ¼ 0: @t   @V g0 q þ ðV rÞV ¼ rp þ g0 rV þ ðg00 þ Þrðr  VÞ; @t 3

ð6Þ ð7Þ

where q and V are the density and volume of the medium, respectively. Then, the propagation equation of acoustic waves in the damping medium is obtained as

p ¼ A0 eax ;

ð8Þ

where A0 is the amplitude of acoustic waves at the initial moment, Pa. And a is the damping absorption coefficient (also known as the attenuation coefficient) of medium which describes the attenuation speed of the amplitude with the distance. The attenuation coefficient a of a plane wave in a pipe [19] is represented by

a ¼ ag0 þ ag00 þ av 1 ¼ rc

sffiffiffiffiffiffiffiffiffi






g 0x x2 x2 1 1 :  g00 þ v  þ Cv Cp 2q0 2q0 c3 2q0 c3

ð9Þ

As a result, the amplitude and the longest propagation distance of acoustic waves after attenuation can be calculated if all the following parameters of the pipelines are known: x ¼ 2pf is the angular frequency, f is the frequency of acoustic wave, Hz; r is the inner radius of pipeline, m; q0 is the density of medium,kg=m3 ; c is the speed of acoustic wave in the pipeline,m=s; g0 is the shear viscous coefficient of medium, Pa  s; g00 is the volume-change viscous coefficient of medium, Pa  s; v is the heat transfer coefficient, W=ðm  kÞ; C v is the constant volume specific heat and C p is the constant-pressure specific heat, kJ=ðkg  KÞ.For ideal gas, the state equation of ideal gas based on Clapeyron equation is shown in Eq. (10) below.

PV ¼

c¼

M RT; m

sffiffiffiffiffiffi cP

q

¼

ð10Þ rffiffiffiffiffiffiffiffiffi cRT ; m

ð11Þ

where P, V, M are the pressure, volume, mass of the gas, respectively. And R is the universal gas constant, T is the temperature of the gas (K), m is the molecular weight of the gas (kg/mol). The speed of sound can be estimated by Eq. (11), where c ¼ C P =C V is the ratio of specific heat. When an acoustic wave travels from one medium to another, an oblique incident analysis must be considered. During the propagation, a sound wave consists of a number of plane waves with angles. As shown in Fig. 3, there are six kinds of waves: the incident pressure wave, the incident shear wave, the reflected pressure wave, the reflected shear wave, the transmitted pressure wave and the transmitted shear wave [20]. The relation between these waves is described as Eq. (12) below:

0

10

1 Asr CB B kp1 cos hpr ks2 sin hst kp2 cos hpt CB Apr C B ks1 sin hsr C CB B C B q sin 2hsr q1 cos 2hsr q2 sin 2hst q2 cos 2hst C A@ Ast A @ 1 l1 q1 l2 q2 q1 cos 2hsr  k1 þ2l sin 2hpr q2 cos 2hst  k2 þ2l sin 2h Apt 1 2 1 0 ks1 cos hsi kp1 sin hpi C
 B kp1 cos hpi C Asi B ks1 sin hsi C ¼B ; B q sin 2hsi q1 cos 2hsi C A Api @ 1 l1 q1 q1 cos 2hsi  k1 þ2l sin 2hpi ks1 cos hsr

kp1 sin hpr

ks2 cos hst

kp2 sin hpt

1

ð12Þ where Asi , Api are the incident vectors of the shear and pressure waves, respectively. Asr and Apr are the reflection vectors of the shear and pressure waves, respectively. Ast and Apt are the transmission vectors of the shear and pressure waves. hsi , hpi , hsr , hpr , hst and hpt are the incident, reflection and transmission angles of the shear and pressure waves, respectively. q1 and q2 are the densities of medium A and medium B, respectively. cs1 , cs2 , cp1 and cp1 are the speeds of sound of the shear and pressure waves in each medium. ks1 , ks2 , kp1 and kp1 are the wave numbers of the shear and pressure wave in each medium and k ¼ x=c. k1 , l1 , k2 and l2 are the Young modulus and shear modulus of medium A and medium B, respectively. The sound pressure can be indicated as below,

pi ¼ Api ejðxtkp1 xÞ :

ð13Þ

pr ¼ Apr ejðxtþkp1 xÞ :

ð14Þ

pt ¼ Apt ejðxtkp2 xÞ :

ð15Þ

pi ¼ pr þ pt :

ð16Þ

Table 1 Comparison of related properties of ice and SI and SII hydrates.

Fig. 8. Deposit in the experimental pipe.

Property

Ice

SI

SII

Isothermal Young’s modulus at 268 K, 109 Pa Poisson’s ratio Bulk modulus (GPa) Shear modulus (GPa) Compressional velocity, VP (m/s) Shear velocity, VS (m/s) Velocity ratio (compression/shear) Adiabatic bulk compression at 273 K (GPa) Refractive index (632.8 nm, 3 °C) Density (g/cm3)

9.5 0.3301 9.097 3.488 3870.1 1949 1.99 12 1.3802 0.91

8.4est 0.31403 8.762 3.574 3778 1963.6 1.92 14est 1.346 0.94

8.2est 0.31119 8.482 3.6663 3821.8 2001.14 1.91 14est 1.350 1.291

Note: est = estimated

16

Z. Qu et al. / Measurement 92 (2016) 11–18

Fig. 9. Reflection signals for the two cases: with and without single hydrate deposit within the pipe.

Fig. 10. Reflection signals for the two cases: with two hydrates and without hydrate deposits within the pipe.

Table 2 Comparison between the modeling (Fig. 5) and the experimental results (Fig. 9).

Hydrate A Hydrate B

Location in modeling

Location in experiment

14.956 m 18.005 m

14.936 m 18.087 m

Api is the complex pressure amplitude of incident wave, Apr is that of the reflected wave and Apt is that of the transmitted wave. Since the sound speed is different, the wave numbers kp1 in medium A and kp2 in medium B are different. The reflection coefficient, R, led by the acoustic impedance [21] changes, is given by Eq. (17).

R¼


2 rA  rB : rA þ rB

r ¼ qc;

ð17Þ ð18Þ

where rA is the acoustic impedance of medium A, r B is the other one acoustic impedance of medium B, q is the density of the medium and c is the sound speed in medium.

the start point of the pipe. The reflection signal is shown in Fig. 4 and an obvious reflection can be found at 18.096 m. Similarly, when two hydrates are placed within the pipe at 15 m and 18.05 m respectively from the start point, two reflection signals can be found as shown in Fig. 5 below, which indicates that the method is able to locate hydrate deposits at multiple locations. The following two results show the reflection signals for the cases with different diameters and lengths hydrate deposits in the pipeline. Fig. 6(a) shows the reflection signal of the hydrate deposits with the same length but different diameters (from 10 mm to 50 mm). The model shows that there is a link between the reflection signals and the geometry of the hydrate deposits. If the diameter of a hydrate deposit is bigger, then the amplitude of the reflection signals increases (or vice versa) accordingly. Similarly, Fig. 6(b) shows the reflection signals of four hydrate deposits with the same diameter but different lengths (5 cm, 9 cm, 15 cm, and 19 cm). In this case if the length of a hydrate increases then the length of the reflection signal increases accordingly. 4. Experiments and results

3.2. Hydrate deposit detection models

4.1. Experiment setup

A series of 2D models, based on the theories in Section 3.1 and the experiment pipe, have been built to understand the whole process and verified by the laboratory experiments. All the models have been built under the normal situation (298.15 K, sealed away from the atmosphere and air inside the pipe). In the following model, one hydrate deposit (50 mm diameter  100 mm long) is placed inside the pipe at about 18.05 m from

A loudspeaker (Markaudio Alpair7, 30 W 6 X) is placed at the start point of the pipe (see Fig. 2(a)), is used to generate the incident sound wave. A microphone (Superlux ECM888B, 94dBSPL, Omni-directional) is fixed closely to the loudspeaker to detect the reflection signals caused by a abnormal event like a hydrate or leakage. An NI-6366 DAQ card is used for not only acquiring the reflection signals but also generating the driving signal. Then

17

Z. Qu et al. / Measurement 92 (2016) 11–18

Fig. 11. (a) Reflection signals with five different diameters of hydrate in the pipe. (b) Enlarged reflection signals.

Table 3 Comparison of the amplitude between the modeling and the experimental results (Normalized value).

D = 20 mm D = 30 mm D = 40 mm D = 50 mm

Simulation result

Measurement result

0.040 0.058 0.077 0.110

0.023 0.047 0.086 0.130

the driving signal is amplified by a power amplifier to drive the loudspeaker. The experiment setup is shown in Fig. 7. 800 Hz is proven to be a good driving signal frequency balancing between both the propagation attenuation and the localization resolution by the modeling result. In the experiments, in order to simplify the procedures, ice, which forms toward the center of the cross-section, is employed (as shown in Fig. 8) rather than a real hydrate deposit as their physical properties, especially acoustic properties, are very similar [5,6], which are shown in Table 1 below.

5. Results and discussions An ice cylinder (50 mm diameter  100 mm long) is placed inside the pipe (as shown in Fig. 8) at about 18.05 m from the start point and then one cycle 800 Hz sine wave was transmitted into the pipe. Fig. 9 shows the experimental result, in which the location of the deposit is at 18.115 m. For this case in the simulation result from the model, the deposit location is at 18.096 m, which is shown in Fig. 4 and close to the experimental result. Also the complex internal structure within the pipe, like bends, defects etc., can lead to small reflection signals, which can be found in Fig. 9 as well.

Table 4 Comparison of the position of the rear end between the modeling and the experimental results.

L = 5 cm L = 9 cm L = 15 cm L = 19 cm

Simulation result (m)

Measurement result (m)

14.922 14.960 14.983 15.039

14.926 14.966 15.018 15.046

Another ice cylinder (40 mm diameter  100 mm long) is placed at 14.9 m from the start point of the pipe. For this case two reflection signals can be found in Fig. 10 according to the two hydrate deposits. Table 2 below shows the comparison between the modeling results (Fig. 5) and the experimental results (Fig. 10). From the comparison it can be concluded that the method is able locate multiple hydrate deposits within a pipe and the modeling results are close to the experimental results. Fig. 11 shows the reflection signals for five cases with different cross-sectional areas of hydrate (diameter range from 0 mm to 50 mm). The experimental results indicate that the amplitude of a reflection signal increases when its cross-sectional area increases. Table 3 shows the comparison of the amplitude between the modeling results (Fig. 6(a)) and the experimental results (Fig. 11 (a)), which shows that the modeling results are close to the experimental results. Fig. 12 illustrates the reflection signals for four cases with different lengths of hydrate (lengths range from 20 mm to 50 mm). The experimental results show that the length of the reflection signal increases when the length of the hydrate increases. Table 4 below shows the comparison for the position of the rear end between the modeling results (Fig. 6(b)) and the experimental results (Fig. 12(a)), which shows that the modeling and the

Fig. 12. (a) Reflection signals with four different lengths of hydrate in the pipe. (b) Enlarged reflection signals.

18

Z. Qu et al. / Measurement 92 (2016) 11–18

experimental results are close to each other and the method could be used to estimate the length of a hydrate deposit. 6. Conclusions As is demonstrated in both the modeling and experiments, the method based on acoustic active excitation, is able to monitor and locate hydrate agglomeration at multiple locations online. The measurement principle of the method is based on detecting and analyzing the reflection signals from a hydrate deposit. One cycle 800 Hz sine wave is used as the excitation signal in the experiments. To simplify the experimental procedures, ice is employed in the experiments rather than a real hydrate as their physical properties, especially acoustic properties, are very similar. Both the modeling and experimental results show that the method can locate hydrates with good accuracy. Furthermore, the amplitude and length of a reflection signal are related with the geometry of a hydrate deposit, which indicates that the method is able to roughly estimate the profile of the hydrate. Acknowledgments This work was funded by the Tianjin Research Program of Application Foundation and Advanced Technology (No. 15JCZDJC39200), The Initial Research Founding for oversea talents of Tianjin University of Science and Technology (No. 10216), Foundation of State Key Laboratory of Precision Measuring Technology and Instruments (Tianjin University) (No. PIL1506), and also partly funded by the Young academic team construction projects of the ’Twelve Five’ integrated investment planning in Tianjin University of Science and Technology. The authors offer sincere gratitude to Professor Daxian Yan for helpful discussion. References [1] S. Balitskiy, Y. Bilan, W. Strielkowski, D. treimikien, Energy efficiency and natural gas consumption in the context of economic development in the European Union, Renew. Sust. Energ. Rev. 55 (2016) 156–168.

[2] Ö. Dilaver, Z. Dilaver, L.C. Hunt, What drives natural gas consumption in Europe? Analysis and projections, J. Nat. Gas. Sci. Eng. 19 (2014) 125–136. [3] F. Furuoka, Naturalgasconsumption and economic development in China and Japan: an empirical examination of the Asian context, Renew. Sust. Energy Rev. 56 (2016) 100–115. [4] Y.F. Makogon, Hydrates of Hydrocarbons, PennWell Publishing Company, Oklahoma, 1997. [5] E.D. Sloan, Clathrate Hydrates of Natural Gases, Clathrate hydrates of natural gases, third ed., M. Dekker, 2008. [6] E.D. Sloan, Hydrate Engineering, Society of Petroleum Engineers Inc., Texas, 2000. 63–61. [7] E. Heidaryan, A. Salarabadi, J. Moghadasi, A. Dourbash, A new high performance gas hydrate inhibitor, J. Nat. Gas Chem. 19 (3) (2010) 323–326. [8] S.V. Joshi, G.A. Grasso, P.G. Lafond, I. Rao, E. Webb, Experimental flowloop investigations of gas hydrate formation in high water cut systems, Chem. Eng. Sci. 97 (7) (2013) 198–209. [9] J.L. Creek, S. Subramanian, D.A. Estanga, New method for managing hydrates in deepwater tiebacks, in: Offshore Technology Conference, May2-5, Houston, Texas, USA, 2011. [10] S.L. Scott, L.A. Saterwhite, Evaluation of the backpressure technique for blockage detection in gas flowlines, J. Energy Resour. – ASME 120 (1) (1998) 27–31. [11] J. Yang, A. Chapoy, S. Mazloum, B. Tohidi, Development of a hydrate inhibition monitoring system by integration of acoustic velocity and electrical conductivity measurements, Explor. Prod. Oil Gas Rev. 9 (1) (2011) 36–40. [12] J. Yang, A. Chapoy, S. Mazloum, B. Tohidi, A novel technique for monitoring hydrate safety margin, SPE Prod. Facil. 27 (4) (2011) 376–381. [13] A.R. Hasan, G.E. Kouba, X. Wang, Transient analysis to locate and characterize plugs in gas wells, in: SPE Annual Technical Conference and Exhibition, October 6–9, Denver, Colorado, 1996, pp. 653–662. [14] Z. Yuan, Z. Deng, M. Jiang, Extended partial blockage detection in a gas pipeline based on Tikhonov regularization, J. Nat. Gas Sci. Eng. 27 (2015) 130–137. [15] M.I. Andersen, Isaksen, O. Urdahl, Ultrasonic instrumentation for on-line monitoring of solid deposition in pipes, in: SPE Production Operations Symposium, Oklahoma City, Oklahoma, March 9–11, 1997. [16] X. Wang, K.M. Lewis, K.A. Papadopoulou, B. Lennox, J.T. Turner, Detection of hydrate and other blockages in gas pipelines using acoustic reflectometry, Proc. Inst. Mech. Eng., J. Mech. Eng. Sci. 226 (7) (2011) 1800–1810. [17] X. Wang, B. Lennox, G. Short, J. Turner, K. Lewis, Z. Ding, K. Dawson, C. Lewis, Detecting blockages and valve status in natural gas pipelines, Adv. Energy Syst. Div. 3 (2010) 531–537. [18] M. Bruneau, Fundamentals of Acoustics, first ed., Wiley, London, 2006. [19] L.E. Kinsler, A.R. Frey, A.B. Coppens, J.V. Sanders, Fundamentals of Acoustics, fourth ed., John Wiley and Sons Inc., New York, NY, 2000, p. 126. [20] A.I. Gunawan, N. Hozumi, S. Yoshida, Y. Saijo, K. Kobayashi, Numerical analysis of ultrasound propagation and reflection intensity for biological acoustic impedance microscope, Ultrasonics 61 (2015) 79–87. [21] T.H. Melling, The acoustic impedance of perforates at medium and high sound pressure levels, J. Sound Vib. 29 (1) (1973) 1–65.