A model for estimating flow assurance of hydrate slurry in pipelines A model for estimating flow assurance of hydrate slurry in pipelines

A model for estimating flow assurance of hydrate slurry in pipelines

  • 期刊名字:天然气化学(英文版)
  • 文件大小:745kb
  • 论文作者:Wuchang Wang,Shuanshi Fan,Deqi
  • 作者单位:Department of Storage and Transportation Engineering,South China university of Technology,Guangzhou Institute of Energy
  • 更新时间:2020-11-03
  • 下载次数:
论文简介

Available online at www.sciencedirect.com| Joumal ofScienceDirect7I Natural GasChemistryEL SEVIERJournal of Natural Gas Chemistry 19(2010)380 ,384www.elsevier.com/locate/jngcA model for estimating flow assurance of hydrate slurry in pipelinesWuchang Wangl*,Yuxing Lil1. Department of Storage and Transportation Engineering, China University of Petroleum, Qingdao 266555, Shandong, China;2. South China university of Technology, Key Laboratory of Enhanced Heat Transfer and Energy Conservation, MOE, Guangzhou 510640, Guangdong, China;3. Guangzhou Institute of Energy Conversion, Chinese Academy of Sciences, Guangzhou 5 10640, Guangdong, ChinaI Manuscript received February 6, 2010; revised April 21, 2010 ]AbstractThe problem of hydrate blockage of pipelines in offshore production is becoming ever-increasing severe because oil fields in ever-increasingunusual environments have been brought in production. HCFC- 141b and THF were selected as the substitutes to study the flow assurance ofthe hydrates in pipelines. There are critical hydrate volume concentrations for these two slurries. Hydrate slurries behave like Bingham fluidsand have high agglomerating tendency when the hydr + volume conqentrations arf larger than the critical ones. Based on rheological behaviorsof these two hydrates, a non-dimensional parameter proposed rotgh studyin, the driving forces of agglomeration among hydrate particles,which shows the agglomerating probability of hydrate particles ip pipeline and ct be used to judge the safety of the pipeline. Moreover, a safemodel to judge the safely flow hydrate surries was pressdhtedand Hr夫AAA中he experimental data, which demonstrates that the model |effective to judge whether the pipeline can be ryp say o not.Key wordshydrate slurry; flow assurance; rheological behavior; afe model1. IntroductionIn this paper, HCFC- 141b and THF were chosen as thesubstitutes, since HCFC- 141b can not unite with water whileOver the past decade, hydrates have become the numberTHF can mix with water very well, to study the flow assur-one issue in flow assurance, especial in offshore condensateance of hydrate slurry on the flow loop. And both of them canpipeline running at high pressure and relative low temperatureform hydrate of structure II at atmosphere pressure. The de-[1,2. However, the limitation of traditional methods by pre-tailed flow behaviors of HCFC- 141b and THF were reportedventing the formation of hydrates is obviously ever-increasingin Refs. [14,15], respectively. In this work the flow assuin costs [3,4]. So there is an interest in developing technologyance was presented in detail and a model to judge the fl(that hydrates can be transported as a slury, while avoidingassurance based on the flow behaviors of hydrate slurries wasdeduced.plugs [5-7].Understanding the formation and flow characters of hy-drates in pipeline is necessary to prevent the formation of2. Experimentalplugs and let pipeline systems work safely [8]. Unfortunately,litle is known about the phenomena involved in the hydrateExperiments on flow characters of hydrate slurry wereplug formation [9]. While many works were devoted to theperformed on the flow loop in Guangzhou Institute of Energystudy of hydrate structures as well as rheological behaviorsConversion (Giec, China). This flow loop (Figure 1), which is .of some kinds of slurries in pipeline and several authors have a two pass loop consisting of a 42.0 mm diameter pipe, 30.0 mstudied the formation of pipeline hydrates and hydrates plugslong, was specially built to perform experiments on hydratewith low-pressure flow system, such as CH3CCl2F (HCFC-slurry in low pressure (no more than 1.5 MPa). The flow loop141b) hydrae, tetrahydrofuran (THF) hydrate, TBAB hydrateis enclosed in a temperature chamber (4.83x 3.30x2.55 m),as well as R11 hydrate, there are short of systematic re-which can keep a stable and constant temperature environmentsearches on hydrates morphology and blockage in pipelinewith a temperature ranging from - 40°C to 80 °C. More de-[10-13]. .tails about the flow loop can be found in Ref. [14].中国煤化工' Corresponding author. Tel: 0532-86981223-091; E-mail: wangwuchangupc@ 126.comMYHCNM HGCopyrightO2010, Dalian Institute of Chemical PChinese Academy olSciences. All rights reserved.doi: 10.1016/S 10039953(09)60094-3w.st,Journal of Natural Gas Chemistry Vol. 19 No.4 2010381_DP-_B240 rpm and the pressure drop rise began at 18% hydrates at520 rpm. Moreover, the trend of increasing flow loop pres-sure drop with hydrate volume fraction is qualitatively similar4to the increase in viscosity observed in the laboratory rheolog-ical studies [12]. .All these above mentioneQfacts suggest that agglomer-ation may be the cause of the-suddenly and/or steadily in-crease at the critical volume concentration zone of hydrate.Hydrate particles begin to agglomerate and slush-like hydratesare formed when the solid volume concentration reaches thegure 1. Schematic of the flow loop. 1- _Ta 2- _Mg etic centrifu-criticalcentration, which may be responsible forpump; 3- - Flowmeter; 4--Buffer tankelines in offshore production.o-Drain valve; 7- Pressure sensor;8- T hpe atures sor; A- Sideline,! B- - Sideline for flowmeter, C- Dip part,1 ewpe,E- Venrcalpart3.2. eounve aviors of hydrate slurries in pipelineCommercial THF with a certit edpurity pf 99.9 % andHCFC-141b of 99.5 wt% certified purity (4 pjiang 9 nhuanstand the agglomerating characters of hy-Chemical Co., Zhejiang, China) were used in these t以peri-drate particles when the hydrate volume concentrations arements. More details about the experimental protocols of theselarger than the critical ones, rheological behaviors of thesetwo materials can be found in Refs. [14,15].two kinds of hydrates were analyzed with the assistance ofexperimental data on their flow behaviors. The following sec-3. Results and discussionstion aims at the correlation of the measured data (Q, OP) withthe classical rheological parameters: the shear rate (7) andthe shear stress (τ). In the case of a horizontal cylindrical3.1. Flow behaviors of hydrate slurries in pipelinepipe, the integration of this equation gives a simple relationHCFC- 141b are unsolvable with water what at the bottd mbetween the linear pressure drop△P/L and the shear stress:T = (DintOP/4L)(r/R)= Tw(r/ R), where Tw = (DintOP/4L) isof the pipeline for its higher density when they are flowing inthe wall shear stress. The shear rate is the opposite of thepipeline while THF solves with water very well. Howeverlocal gradient of velocity, γ=-du2/dr in our case. It is re-changes of the morphologies of the two refrigerants hydratelated to the volume flow rate by the Rabinovith Eq. (1).slurries are similar. And the other common phenomenon ofIts Tw derivate gives the expression of the wall shear ratethe two hydrate slurries is that there is a critical hydrate vol-Eq. (2)[11]:ume concentration zone, which depends on the velocity ofQthe hydrate slurries in pipeline. Pressure drops begin to in-品=一"r2ndr(1)crease substantially when the solid volume concentrations arelarger than the critical ones while pressure drops are almostγ= 32Q (3n+1 )/4nπDm withindependent of the solid volume concentration when theyare less than the critical ones especially for turbulent flow.dIn( DmOPThe critical hydrate volume concentrations for HCFC-141bdln(Tw)hydrate slurry and THF hydrate slurry are 28.5%~37.5%n= -8w(2)and 39.4%~50.4%, respectively, with a mean velocity fromIn (In(5)Dimt,0.5 m/s to 3.5 m/s in pipeline. More details about the flow be-haviors on these two materials can be found in Refs. [14,15].Consequently, in the case of a laminar flow in a horizon-The relation of pressure drop versus solid volume con-tal cylindrical pipe, the expressions of the shear rate and thecentraghm so has been reported by many researchers. In thesher stress are functions of measured data (Q, AP). Theof oil-based and water-based hydrate slurries re-cufes representing AP versus Q are approximately straightportel. [12,17], it was sen that even at the high con-lInidH which, however, do not intercept the origin. Conse-centitraXAH3% ), the frictional pressure drop of the slurry wasquently, it seems that these two kinds of hydrate slurries in-equal to that of the pure carrying fluids, provided the flgwduce yield pp s lure drops at the hydrate content higher thanwas turbulent. This implied that no additional pressure内the critical ones. This leads us to assume that both HCFC-could be initiated when hydrate slurries were transporte141b and THF hydrate slurries with hydrate volume con-pipelines, as compared to pipe transport of pure oil or Wadcentrations higher than the critical ones behave like Bing-ter. However, the critical solid volume concentration washam fluids and are characterized bv an extrapolated yieldnot found in these experiments due to the limitation of theirshear stress TO an中国煤化工This assump-mass flow meter used in the experiments being limited by thetion of Bingham ,MH. CNMHGP we plot thelarger hydrate concentration. While in the flow experimentscurve of ln[(DimO101-J Li\unxHihn)] as we ob-of THF hydrate/Conroe oil slurries in [12,17], the pressure .tain straight lines with slopes very close toTlthat implies thatC382Wuchang Wang et alL./ Journal of Natural Gas Chemistry Vol. 19 No.42010eating Tw = (DimtOP/4L) versus γ= (8w/Dim) is sufcient in Table1. The data presented in Table 1 are the yield stessesRMlerive the apparent viscosity and the yield shear stress, re-of HCFC-141b and THF hydrate slurries with their hydratespectively, from the slope and the intercept point if straightvolume concentrations larger than the critical ones.lines are obtained [11]. This is done and the results are listedTable 1. Yield srerses of HCFC-141b and thF hydrate surriesHCFC-141bTHFHydrate volume concentration (%)Yield sresses (Pa)Yield stresses (Pa)37.539.40.8041.64.4444.51.9947.98.9650.62.7153.210.5655.812.044062.115.3461.318.9268.018.5665.230.20 .By now there are no well approbated models to calcu-late the yield stress of Bingham fluid [13], so here polynomialgravity should not be the main driving force to initiate the ag-equations were regressed with experimental data for the twoglomeration of hydrate particles. The energy to separate thehydrate slurries with hydrate volume concentrations largerhydrate particles in the zone defined by the two dashed linesthan the critical ones. The results calculated by Eq.(3) are forwith a length of L in Figure 3 can be described as follows [18]:E= SDryzylrate slurries, respectively. Both the experimental data andSihe calculated curves are shown in Figure 2. It is clearly illus-in which, E is the energy, S is the wetted perimeter of thetrated that the two equations can be well used to calculate thepipeline, D is the diameter of the pipeline, and Ty is the sumyield stresses of the two slurries.of agglomerating forces among hydrate particles.TB=-24.31 + 9.92中- 26.03中2 37.5%< 中< 68% .(3)TB= 92.91 - 444.85中n + 536.39中n2 39.4% <中n < 5.2%比4)Figure 3. Schematic diagram of hydrate particles in pipeline21 5●Experinental results (HCFC-141b)If the superficial surface and the volume of the particlesExperimnental results (THF)are described as Sp and Vp respectively, the energy to separate. Calculated results (HCFC-141b)---- Caleulated results (THF)the hydrate particles into a unit volume can be deduced from30Eq. (5) as follows:En = DOnTy(Sp/Vp)(6意2(in which En is the energy to separate the hydrate particles ina unit volume.10n the deduction, the hydrate particles are supposed asspheres with a same diameter of dp. Now a new non-Lodimensional parameter of Cn was defined as the ratio of ki-netic energy and separating energy of the hydrate particles intoHydrate volume concentraion (%)a unit volume, which is shown as follows:Figure 2. Yield stresses of HCFC- 141b and THF hydrate slurries with solidvolume concentrations larger than the critical onesPmw2/2(7)Ch= 6DOnTy/dpwhere Pm is the density of the mixture of hydrates and wa-3.3. Safe model of hydrate surries in pipelineter in pipeline, which depends on the volume concentration ofhydrate slurries. Since the densities of HCFC- 141b and THFBased on the results of THF and HCFC-141b hydrateshydrates are very close to water densitv. 0m is replaced withas well as the results reported by other researchers, it is ob-the density of wat中国煤化工vious that hydrates are inclined to agglomerate with a largeIn the Eq. (7).MHCN M H Gm of agglomer-dying prce inducq fr4OthemselveA 94She other hfpd,5 ( Dating forcep ag?l4t9y19 v心diult be-0.70thtre eitte dfference tEtween the derstcs of these kmnds'U cause there dr? More than one Kinmdt force among hehydrateHvdr ate vol une concentrai on( %)Journal of Natural Gas Chemistry VoL. 19 No.42010383?sticles, including the force from the liquid bridge amongMYrate particles, atractive force among hydrate particles anderating force among hydrate particles; secondly, yield stress isso on. By far there are short of systemic related research onjust the minimal value of the agglomerating force among hy-the forces among hydrate particles. Fortunately both the twodrate particles. Both the two factors explain the above resultkinds of hydrates slurries are Bingham fluids based on thethat the values of agglomerating force used in calculation wasanalysis of the experiments, and the yield stress of the Bing-less than the values in experiments, which also can be usedham fluid, which is determined by the characteristic parame- to explain that the calculated critical values are a lttle largerters of中n, dp, should be considered as the least force amongthan the experimental critical values.the particles [11,19,20]. So the yield stresses (TB) of the twohydrates slurries are used to replace the sum of agglomerating70forces among hydrate particles (τy). And the Eq. (7) can beCalculated results (THF)transformed as follows:Experimental results (THFCalculated results (HCFC-141b)rw2 /260Experimental results (HCFC-141b)Ch = 6D中nrB/dpAs shown in Eq. (8), if Cn has a value larger than 1.0,E 50which means the force to separate the particles is larger thanthe force to agglomerate the particles, the hydrate particleswill not be agglomerated. On the contrary, if Cn has a value40larger than 1.0, hydrate particles will be agglomerated and thepipeline can not be run safely. So the non-dimensional pa-rameter of Ch can be used to judge whether the pipeline runs300.人safely or not.w (m/s)Figure 4. Application of safe model of hydrate slurry in pipeline装3.4. Validation on the model of hydrate slurries in pipelineMoreover, HCFC-141b hydrate particles have a higher50 In order to judge the applicability of the non-dimensional agglomerating tendency than THF hydrate particles inparameter of Ch, a test with the data from the experimentspipeline as shown in Figure 3 since HCFC- 141b hydrates havewas carried into execution. The basic parameters of the twolarger critical volume concentrations than THF hydrates withekinds of hydrates are listed in Table 2, in which the diametersa mean velocities less than 3.5 m/s. The agglomerating ten-of the hydrate particles are the average values measured in thedency also can be seen from the experimental critical hydrateexperiments.volume concentrations of 28.5%~37.5% and 39.4%~50.4%for HCFC-141b hydrate slurry and THF hydrate slurry, re-Table 2. Basic parameters of the two kinds of hydratesspectively. And the higher agglomerating tendency of HCFC-ParameterHCFC-141bTHF141b can be explained with its higher yield stress values thanDiametter (mm)442THF, as shown in Figure 2.Diameter of hydrate particles (mm)0.30.45In a word, the model supposed above can be used to cal-Liquid density (kg-m-3 )998.5culate the critical volume concentration of a kind of hydratesflowing in pipeline at a velocity. And when the actual hy-40.The critical lines calculated by supposing the value of Chdrate volume concentration is lower than the calculated one,equal to 1.0 with the yield stresses calculated with Eq. (3)the pipeline will be free of hydrate blockage; on the contrary,for HCFC-141b hydrate and those with Eq. (4) for THP hy-pipeline will be easy to be blocked. The proposed model, ofdrate respectively, are shown in Figure 4. Figure 4 presentscourse, should be further improved in order to correctly judgethe corresponding critical hydrate volume concentrations forwhether the pipeline can be run safely or not with natural gasevery flow velocity of hydrate in pipeline. Meanwhile bothhydrate slurries.the experimental critical points obtained by calculating the in-crease of pressure drop and the observing ones through the4. Conclusionsview point on the experimental flow loop are also shown inFigure 4.Risk management on hydrates in pipeline has been ac-According to Figure 4, the calculated critical line fits wellcepted more and more widely around the world. HCFC-141bwith the experimental critical points especially at the normaland THF were selected as the substitutes to study the flows{ow velocity in pipeline. While the ltte difrence betweene calculated values and the experimental ones can be ex-temic researches (中国煤化工hydrate slurriesplained as follows: at first, the hydrate particles in the safewere conducted arYH.| CNMH Gollows:mopel me supposep) to5have equal 1dia netespsinsead可ac-2 1Turbulent fofhydrates. Aggtomeration of hydrate particles accleratesI gu qyiulaie thefongnation4.0tuarones dstribued over a relative tortidrube eten.PAndW(m/s)384Wuchang Wang et alL./ Journal of Natural Gas Chemistry Vol. 19 No.42010quickly when the solid volume concentration is larger than the[5] Sloan E D. In: The Fifth International Conference on Gas Hy-critical one, which leads the pipeline to a dangerous situation.drates, 2005, 4: 14312) Hydrate suries with solid volume concentration larger[6] Tohidi B. UK DTI IOR Seminar, Aberdeen, Scotland, UK, 2003than the critical ones behave like Bingham fluids and the cal-[7] Ramesh A K. In: The Fifth International Conference on GasHydrates, 2005, 4: 1215culating equations of yield shear stress (7o) and apparent vis-cosity (μo) for HCFC- 141b and THF hydrate slurries were re-[8] Mehta A P, Klomp U C. In: The Fifth International Conferenceon Gas Hydrates, 2005, 4: 1089gressed.[9] Haghighi H. SPE 107335, 20073) A model based on a new and non-dimensional param-[10] Balakin B V, Pedersen H, Kilinc Z, Hoffmann A C, Kosinski P,eter of Ch, which is defined as the ratio of kinetic energy andHoiland s. J Pet Sci Technol, 2010, 70: 177separating energy of the hydrate particles in pipeline, can be [11] Darbouret M, Coumnil M, HeriJM. IntJ Refrig, 2005, 28: 663 .used to calculate the critical volume concentrationof a kind of [12] Derek M K [MS Disertation]. Colorado: Colorado School ofhydrates flowing in pipeline at a velocity. The new model canMines, 2005be used to judge whether the pipeline can be run safely or not [13] Fidel-Dufour A, Cruy F, Heri J M. Chem Eng Sci, 2006. 61:and can give some introductions to further study of the flow05assurance on natural gas hydrate slurries.[14] Wang W C, FanSS, LiangD Q, Yang X Y. Int J Refrig, 2008,31: 371[15] Wang W C, FanS S, LiangD Q, Li Y X. J Natur Gas Chem,References2010, 19: 318[16] Marinhas s, Delahaye A, Fournaison L. Int J Refrigeration,[1] Emmanuel D, Giorgio G, Lissett B, Douglas E, Ramon D. In:2007, 30: 758The Sixth International Conference on Gas Hydrates, 2008. [17] Turner D J, Larry T. In: The Sixth Intermnational Conference on5650Gas Hydrates, 2008. 54342] Douglas A E, Jefferson C, Sivakumar S, Ramesh A K. In: The [18] Hirochi T, Maeda Y, Yamada S, Shirakashi M, Hattori M, SaitoSixth International Conference on Gas Hydrates, 2008. 5658A.J Fluids Eng, 2004, 126: 4363] Yousif M H, Dunayevsky VA. SPE 30641[19] Kitanovski A, Vuarnoz D, Ata-Caesar D, Egolf P W, Hansen T[4] Roghieh A, Antonin C, Ross A, Bahman T In: The Sixth Inter-M, Doetsch C. ImtJ Refrig, 2005, 28: 37national Conference on Gas Hydrates, 2008. 5332[20] Niezgoda-Zelasko B, Zalewski w. Int J Refrig, 2006, 29: 418中国煤化工MHCNM HG

论文截图
版权:如无特殊注明,文章转载自网络,侵权请联系cnmhg168#163.com删除!文件均为网友上传,仅供研究和学习使用,务必24小时内删除。