当前位置:首页 >> 教育学/心理学 >>

AJK2011-31002-ASME论文


Proceedings of ASME-JSME-KSME Joint Fluids Engineering Conference 2011 AJK2011-FED July 24-29, 2011, Hamamatsu, Shizuoka, JAPAN

AJK2011-31002
NUMERICAL INVESTIGATION OF NEW DESIGN OF THE ETOILE FLOW STRAIGHTENER
Boualem LARIBI Industrial Fluids Laboratory, Measurements and Applications University of Khemis Miliana, Algeria E-mail : boualemlaribi@yahoo.com Abdelkader YOUCEFI Mechanical department University of sciences and technology of Oran. Algeria E-mail : youcefi_a@yahoo.fr Djelloul BELKACEMI Industrial Fluids Laboratory, Measurements and Applications University of Khemis Miliana, Algeria E-mail: belkacemidjelloul@gmail.com

ABSTRACT Flow metering of industrial fluids remains the concern of several researchers and exporting countries of gas and oil in the world. It is in this context that a vast numerical investigation is done in our laboratory of Industrial Fluids Measurements and Applications (FIMA). This article examines numerically a new design of the Etoile flow straightener which is described by the standard ISO 5167. This new design consists in removing the central part of the Etoile which, according to researchers, leads to a very high level of turbulence. Our intervention relates to the development and the establishment of the flow parameters downstream the Etoile with and without central part. The flow is produced by air in a 100mm pipe diameter and 40D of length with a Reynolds number of 2.5x105. The disturbance is a valve maintained 100%, 50% and 30% open. The flow parameters examined are velocity profile, turbulence intensity profile, and the fluid gyration angle. The code CFD Fluent is used for this simulation. The results obtained are compared according to directives of the standard ISO 5167. The results obtained show that for the valve settings 30% and 50% open, upstream the Etoile, we have a high turbulence level and a velocity profile with recirculation zones more significant for the valve 30% open than for the valve 50 % open. It is also noted that the valve develop very high fluid gyration angle apart from the standard values. The flow behavior downstream the central part of Etoile described by the ISO 5167 is well simulated with the valve open at 100%, with a deficit of flow and a very high degree of turbulence. At this stage for the two designs, the noticed results seem so identical beyond a certain stations downstream the Etoile.

INTRODUCTION Orifice plates have been used for flow measurement for many years for process and fiscal purposes. The ability to accurately measure the gas flow rate in a duct is of major concern and vital importance when large volumes are handled. The Algerian petroleum company recorded receipts of 56x109 m3 of gas over one the year. The quality of gas measurement, receipt and major delivery points disturbed through 13000 km on pipe line. Errors in flow measurement can have large cost and efficiency implications. The majority of the meters must be calibrated and this is done in fully developed pipe flow, axisymmetric, free from swirl and pulsation. Standard such as ISO 5167[1] and AGA-3[2] define a satisfactory flow as one which has a swirl angle less than two degrees and for which the ratio of the axial velocity to the maximum axial velocity is within ±5% of the corresponding ratio in fully developed flow measured in the same pipe after 100 pipe diameters of development length. While high accuracy about 0,3% mass flow measurement is required, disturbances in the flow introduced by contractions, bends, valves and other components introduce errors of accuracy to 3% [3 ]. Given that most industrial installations include bends, valves, expanders and reducers, which are sources of swirl, asymmetries and turbulence distortions, insuring that fully developed flow in terms of mean flow and turbulence structure approaches the meter is difficult to achieve in practical situations. For best accuracy, a flow meter needs to be presented with an axisymmetric, fully developed velocity profile with zero swirl. Either very long lengths of straight pipe work upstream of the meter must be provided as recommended by ISO 5167 [1] and these may need to be of the order of 80 to 100 pipe diameter, which will give a higher installation cost and greater space requirement. Alternatively, disturbances can be attenuated

1

Copyright ? 2011 by ASME

by using flow straightener and/or flow conditioner to control the quality of the flow approaching the metering device. Recent research work, Morrison and nd al [4], Morrow and Park [5], Rans and al [6], Ahmadi [7], Laws and Ouazzane [8], Darin and Bowles [9], Aichouni and al [10], [1 Laribi and al [11,12] have reported a number of experimental and computational studies of installation n effects on orifice meter performances. . Most of these studies investigated the effect of flow conditioner locations with respect the orifice meter on the discharge coefficient. In this paper, a numerical experimentation of the effect of entrance flow velocity ocity profile, generated by a valve in different settings, 100%, 50% and 30% open is examined. The efficiency of the modified Etoile flow straightener, , without central part to remove flow distortions and produce the fully developed conditions are investigated and compared to the Etoile described actually by standard ISO 5167. NOMENCLATURE A Swirl Angle D Pipe inner diameter I Turbulence intensity k Turbulence kinetic energy R Pipe inner radius Rey Reynolds number t Time U Mean axial velocity Uavg Bulk average velocity Umax Maximal velocity V Mean radial velocity y Radial coordinate z Axial coordinate ? Dissipation rate of k EXPERIMENTAL SIMULATION onduit Conduit The flow in the duct, Fig. 1, is produced by air with a Reynolds number Rey=2.5x105. The air leads eads from the entrance on the length 22D for good flow conditions ns at the t entrance of the valve. The measure stations are 0.5D upstream (-0.5D), then 0.5D, 4.5D, 5.5D downstream of the unit The entrance of the Etoile flow traightener is 3D downstream of the th valve with three settings 100%, 50% and 30% open.

The Modified Etoile Flow Straightener The Etoile flow straightener used in this study and described by standard is shown in Fig. (2-a). In this study we examine the removing of 0.5D of the central part of the Etoile as shown in Fig. (2-b,c), the length of the two units is 2D. The Etoile straightener is composed of four plates of thickness 0.01D arranged as in Fig. 2.

(a)

(b)

(c)

FIGURE 2.

ETOILE AND ITS ALVER ALVERNATIVE

TURBULENT MODELE IN FLUENT Basic Equations e fluid is considered as incompressible, three The dimensional and stationary. . The general equation used in CFD and by the code Fluent t is given by Eq. (1).

? ??? ?t

? ? ? div ??? U ??

div ??? . grad ? ? ? S ? (1)

Where ? is the general dependent variable which can be the mean velocity, the turbulen turbulent kinetic energy k or the rate of dissipation ? of the turbulent turbulen kinetic energy. S? is the term source ource of the variable ? ?? is the coefficient of diffusion of ? Model Used in the Simulation imulation The code Fluent presents present several models of turbulence, and for the lightening of the text we will not reproduce the equations in this article. The reader can co consult literature of code Fluent uent for more details [13]. Here we only present a summary of the model (k-ε) used. The k-ε Model. The k-ε model is the simplest model known as two equations model. model This model assumes that the turbulence regime is fully developed through all the section of the pipe and the effects of molecular viscosity are negligible compared to the turbulent viscosity (far wall). It is based on the Boussinesq assumption. It t is a semi empirical model. Two transport equations are used, used one for the turbulent kinetic energy k and the other for its dissipation rate ε.

FIGURE 1.

CONDUIT

2

Copyright ? 2011 by ASME

Reference Data Examined Velocity profile. The velocity profiles of U/Umax are compared to the profile obtained by the universal power law with ±5% tolerance as recommended by ISO 5167 [1]. It is done by the following Eq. (2), [14,15]:

Ui ?y ? ?? i ? U max ? R ?

1

7

(2)

Turbulence intensity profile. For the turbulence intensity, expressed in percentage terms, I(%), the reference profile is determined by the profile obtained at station 100D where the flow is supposed fully developed as suggested by ISO 5167 [1]. We can obtain the turbulence intensity by Eq. (3) as bellow [16]:

Figure 6 shows the development of the turbulence intensity profile at the same stations for the Etoile without central part. We can remark that the peak developed previously became a wake in this case with a level about 3.5% who disappears rapidly at the same stations. Figures 7 and 8 show the swirl angle A(°) obtained by using the two components U and V of the mean velocity at different locations in the conduit with valve fully open. All the results same to reach the values adopted by the standard. We can note some deviation in Fig. 7 of the swirl angle at station z/D=0.5 downstream the unit without any time to leave the limits of 2°. Flow Parameters Development With Valve 50% Open Figure 9 shows the time mean velocity profiles measured at different stations downstream and upstream of the standard Etoile with valve 50% open. All profiles are far from the theoretical profile ±5% according standard. We see also the redistribution of the asymmetry velocity profile, generated by the valve in this setting, to reach the standard profile from station 0.5D to 5.5D. The deficit of flow is clearly visible in the axis of the conduit at station z/D=0.5 and for 1>y/R>0.5, due to the central solid part of the unit. Figure 10 shows the time mean velocity profiles measured at different stations downstream and upstream of the Etoile modified with valve 50% open. It is clear that there is no deficit in flow but with the stations z/D=4.5 and 5.5 the velocity profile seems to follow the same development to reach the theoretical profiles. At this stage we have the same results with the two units. Figure 11 shows the development of the turbulence intensity profile at the same stations for the standard Etoile. We note the same remark that for the velocity profile. All the profiles at different stations same to reach the developed intensity profile except at station z/D=0.5 where the profile develop a very high turbulent level with a value about 35% who decay rapidly to mean value about 5% downstream. The same development is visible with the modified unit in Fig. 12. Here also there is no difference in the results of the units. Figures 13 and 14 show the swirl angle A(°) obtained by using the two components U and V of the mean velocity at different locations in the conduit with the valve 50% open. All the results same to reach the values adopted by the standard. We can note the high degrees of swirl introduced by the valve 50% open about ±40° at station z/D=-0.5 for the two units. This value disappears rapidly downstream the two units. In reality, a detailed examination of Fig. 12 and Fig. 13 on a small scale in ordinate showed that there was no difference between the results of the swirl angle A(°) downstream the two units to reach the standard values of ±2°.

I (%) ?

2 k 3 U avg

(3)

Fluid swirl. The swirl angle A(°) was obtained by using the two components U and V of the mean velocity obtained at different locations in the conduit with Eq. (4), [17]:

?V ? A(?) ? arctg ? ? ?U ?
RESULTS AND DISCUSSION

(4)

Flow Parameters Development With Valve 100% Open Figure 3 shows the time mean velocity profiles measured at different stations downstream and upstream of the standard Etoile with valve fully open (100% open). All profiles same to reach the theoretical profile ±5% according standard except the profile downstream the unit at station z/D=0.5 where the velocity profile develop a wake, deficit in flow for 1.5>y/R>0.5, due to the solid central part. This wake decay rapidly at station z/D=4.5 and 5.5. We can note a large turbulent mixture at these stations which make disappear this wake. On the other hand, in Fig. 4, for Etoile modified, all velocity profile have reached fully developed velocity profiles and established flow, which shows at first sight the performances of Etoile without the central solid part. Figure 5 shows the development of the turbulence intensity profile at the same stations for the standard Etoile. We note the same remark that for the velocity profile. All the profiles at different stations same to reach the developed intensity profile except at station z/D=0.5 where the profile develop a peak and the value reaches some 4.5%, who disappears rapidly by station z/D=4.5 and downstream to reach the fully developed turbulence intensity profile.

3

Copyright ? 2011 by ASME

Flow Parameters Development With Valve 100% Open
1,2 1,0 0,8
1,2 1,0 0,8

U/Umax

U/Umax

0,6 0,4 0,2 0,0 0,0 0,5

-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED ±5%

0,6 0,4 0,2 0,0

-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED ±5%

1,0

1,5

2,0

0,0

0,5

1,0

1,5

2,0

y/R

y/R

FIGURE 3. VELOCITY DEVELOPMENT ETOILE ISO

FIGURE 4. VELOCITY DEVELOPMENT ETOILE MODIFIED
10 8 6 4 2 0

10 8

-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED

I(%)

4 2 0 0,0 0,5 1,0 1,5 2,0

I(%)

6

-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED

0,0

0,5

1,0

1,5

2,0

y/R

y/R

FIGURE 5. INTENSITY DEVELOPMENT ETOILE ISO
10 8 6 4

FIGURE 6. INTENSITY DEVELOPMENT ETOILE MODIFIED

-0.5D +0.5D +4.5D +5.5D ISO 5167

10 8 6 4

A(°)

0 -2 -4 -6 -8 -10 0,0 0,5 1,0 1,5 2,0

A(°)

2

-0.5D +0.5D +4.5D +5.5D ISO 5167

2 0 -2 -4 -6 -8 -10 0,0 0,5 1,0 1,5 2,0

y/R

y/R

FIGURE 7. SWIRL VALUES ETOILE ISO

FIGURE 8. SWIRL VALUES ETOILE MODIFIED

4

Copyright ? 2011 by ASME

Flow Parameters Development With Valve 50% Open
1,2 1,0 0,8 0,6 0,4 0,2 0,0 0,0 0,5 1,0
-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED ±5%

1,2 1,0 0,8

U/Umax

U/Umax

0,6 0,4 0,2 0,0
-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED ±5%

1,5

2,0

0,0

0,5

1,0

1,5

2,0

y/R

y/R

FIGURE 9. VELOCITY DEVELOPMENT ETOILE ISO

FIGURE 10. VELOCITY DEVELOPMENT ETOILE MODIFIED

40 35 30

I(%)

I(%)

25 20 15 10 5 0 0,0 0,5 1,0

-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED

40 35 30 25 20 15 10 5 0 0,0 0,5 1,0

-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED

1,5

2,0

1,5

2,0

y/R

y/R

FIGURE 11. INTENSITY DEVELOPMENT ETOILE ISO

FIGURE 12. INTENSITY DEVELOPMENT ETOILE MODIFIED

40 30 20

40
-0.5D +0.5D +4.5D +5.5D ISO 5167

30 20

A(°)

A(°)

10 0 -10 -20 -30 -40 0,0 0,5 1,0 1,5

10 0 -10 -20 -30 -40 0,0 0,5 1,0

-0.5D +0.5D +4.5D +5.5D ISO 5167

2,0

1,5

2,0

y/R

y/R

FIGURE 13. SWIRL VALUES ETOILE ISO

FIGURE 14. SWIRL VALUES ETOILE MODIFIED

5

Copyright ? 2011 by ASME

Flow Parameters Development With Valve 30% Open
1,2 1,0 0,8 1,2 1,0 0,8

U/Umax

U/Umax
-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED ±5%

0,6 0,4 0,2 0,0 0,0 0,5 1,0

0,6 0,4 0,2 0,0 0,0 0,5 1,0
-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED ±5%

1,5

2,0

1,5

2,0

y/R

y/R

FIGURE 15. VELOCITY DEVELOPMENT ETOILE ISO

FIGURE 16. VELOCITY DEVELOPMENT ETOILE MODIFIED

60 50 40

-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED

60 50 40

-0.5D +0.5D +4.5D +5.5D FULLY DEVELOPED

I(%)

I(%)

30 20 10 0 0,0 0,5 1,0 1,5 2,0

30 20 10 0 0,0 0,5 1,0 1,5 2,0

y/R

y/R

FIGURE 17. INTENSITY DEVELOPMENT ETOILE ISO

FIGURE 18. INTENSITY DEVELOPMENT ETOILE MODIFIED

80 60 40

A(°)

A(°)

20 0 -20 -40 -60 -80 0,0 0,5 1,0 1,5

-0.5D +0.5D +4.5D +5.5D ISO 5167

80 60 40 20 0 -20 -40 -60 -80 0,0 0,5 1,0 1,5

-0.5D +0.5D +4.5D +5.5D ISO 5167

2,0

2,0

y/R

y/R

FIGURE 19. SWIRL VALUES ETOILE ISO

FIGURE 20. SWIRL VALUES ETOILE MODIFIED

6

Copyright ? 2011 by ASME

Flow Parameters Development With Valve 30% Open Figure 15 shows the time mean velocity profiles measured at different stations downstream and upstream of the standard Etoile with valve 30% open. All profiles are far from the theoretical profile ±5% according standard. We see also the redistribution of the asymmetry velocity profile, generated by the valve in this setting, to reach the standard profile from station 0.5D to 5.5D. The deficit of flow is clearly visible in the axis of the conduit at station z/D=0.5 and for 0.7>y/R>0.5, due to the central solid part of the unit like previously. Figure 16 shows the time mean velocity profiles measured at different stations downstream and upstream of the Etoile modified with valve 30% open. It is clear that there is no deficit in flow but with the stations z/D=4.5 and 5.5 the velocity profile seems to follow the same development to reach the theoretical profiles. There is no difference between the results with the two units. Figure 17 shows the development of the turbulence intensity profile at the same stations for the standard Etoile. We note the same remark that for the velocity profile. All the profiles at different stations same to reach the developed intensity profile except at station z/D=0.5 where the profile develop a very high turbulent level with a value about 50% who decay gradually to reach fully developed profile downstream the unit. The same development is visible with the modified unit in Fig. 18. Here also there is no difference in the results of the units. Figures 19 and 20 show the swirl angle A(°) obtained by using the two components U and V of the mean velocity at different locations in the conduit with the valve 30% open. All the results same to reach the values adopted by the standard. We can note the high degrees of swirl introduced by the valve 30% open about ±60° at station z/D=-0.5 for the two units. These values persist around ±40° for the first unit at the station z/D=0.5 for the radial station 0.75>y/R>0.25. For the modified unit the values of the swirl angle at the station z/D=0.5 are in good agreement with the standard values. CONCLUSION The purpose of this work is to study the effectiveness of a new modified Etoile flow straightener with 2D length described by the standard ISO 5167. The modification consists in removing the central part of Etoile. The numerical analysis with CFD code Fluent enabled us to analyze the development of flow downstream a valve in various settings, 100%, 50% and 30% open. The analyzed parameters are the velocity, the intensity of turbulence profiles and the swirl angle of the fluid. The results obtained with this numerical experimentation showed that the two units have the same performances for the development and establishment towards standard profiles for the different positions of the valve, except for the position of the valve 30% open where the results show better performances of the

modified Etoile which eliminates almost all the gyration of fluid and gives values within the limits of the standard. We think that a future experimental study will make it possible to well discuss the results of the current standard. ACKNOWLEDGEMENTS This work is part of the research project no. J0303920080007 approved by the Ministry of Higher Education and Scientific Research in Algeria. 2009. Contribution to conceive a new flow conditioner. REFERENCES [1] ISO 5167, 2003, Measurement of fluid flow by means of orifice plates nozzles and ventury tubes inserted in circular cross section conduits running fuel,. http://www.iso.org/iso/ [2] AGA-3, 1980, ‘Orifice Metering of Natural Gas and the Related Hydrocarbone Fuels”. http://www.aga.org [3] Yeh, T.T and Mattingly, G., 1996. E, “Flow meter installation effects due to a generic header”. NIST Technical note 1419.. http://catalogue.nla.gov.au/ [4] Morrison G. L. and al, 1997, “Flow development downstream of a standard tube Bundle and three different porous plate Flow conditioners”. Flow Meas. Instrum., Vol. 8, N°2, pp. 61-76. [5] Morrow, T.B., Park, J.T. and Mckee, R. J., 1991. “Determination of Installation Effects for a 100 mm Meter using a Sliding Vane Technique”. Flow Measurement and Instrumentation Vol .2, N°1, pp 14-20. [6] Rans, R. and al , 2008. “Flow Conditioning and Effects on Accuracy for Fluid measurement”. 7th East Asia Hydrocarbon Flow Measurement Workshop 5th – 7th, Malisia, March http://www.flowconditioner.com [7] Ahmadi A., 2009. “Experimental Study of a New Flow Conditioner on Disturbed Flow in Orifice Plate Metering”. J. Fluids Eng., May, Volume 131, Issue 5 http://www.asme.org [8] Laws E.M. and A.E.K. Ouazzane, 1995. “A preliminary study into the effect of Length on the performance of the Etoile flow straightener”. Flow Meas. Instrum., Vol. 6, N°3, pp. 225 233,. [9] Darin L. & Bowles E. B., 2008. “Conditioning On Gas Measurement”. Pipeline & Gas Journal, PP 59-62 February. [10] Aichouni M. and Laribi B. ,2000. “Computational Study of the Aerodynamic Behaviour of the Laws Vaned Plate Flow Conditioner”. ASME Fluids Engineering Division Summer Meeting, Boston, USA, june 11-15. http://www.asme.org [11] Laribi B., Wauters P. and Aichouni M., 2000. “Experimental Study of the Decay of Swirling Turbulent Pipe Flow and its Effect on Orifice Flow

7

Copyright ? 2011 by ASME

Meter Performance”. ASME Fluids Engineering Division Summer Meeting, FEDSM'01, May 28. http://www.asme.org [12] Laribi B. and al, 2002, “Experimental Study of Aerodynamic Behaviour Downstream of Three Flow Conditioners”. ASME Fluids Engineering Division Summer Meeting, Montréal, Canada, july 14-18. http://www.asme.org [13] Fluent v6.3, 2006, Fluent incorporated, Journal Centerra source park, USA. http://www.fluent.com [14] Laribi B., Wauters P. and Aichouni M., 2003. “Comparative study of the aerodynamic behavior of three flow conditioners”. European Journal of Mechanical and Environmental Engineering, Vol. 48, No. 4, pp. 21-30, March. [15] Laribi B., Wauters P. and Aichouni M., 2003. “Further analysis of the aerodynamic behavior of flow conditioner”. European Journal of Mechanical and Environmental Engineering, Vol. 48, No. 3, pp. 167-176. [16] Morrison G. L. and al, 1997. “Flow development downstream of a standard tube Bundle and three different porous plate Flow conditioners”. Flow Meas. Instrum., Vol. 8, N°2, pp. 61-76. [17] Laribi B. and al, 2010. “Numerical investigation of contribution of three flow conditioners in the development and establishment of turbulent flows”. 2010 Asme Fluids Engineering Division Summer Meeting and 8th International Conference on Nanochannels, Microchannels and Minichannels”. Montreal, Canada, August. http://www.asme.org

8

Copyright ? 2011 by ASME


赞助商链接
相关文章:
更多相关标签: