煤炭气化气流床气化炉的数学模拟

第24卷第4期煤炭转化Vol.24 No.42001年10月COAL CONV ERSIONOct.2001煤炭气化气流床气化炉的数学模拟步学朋)彭万旺’ 徐 振刚2)商要简要介绍了煤炭气流床气化的原理,总结了到目前为止煤炭气化气流床气化炉数学模拟情况，包括简单平衡模型和动力学模型(一维或多维),给出了这些数学模型模拟的主要内容(对气化过程流体力学、热力学、化学反应和质量、能量及动量平衡考虑情况)和模型的主要结论，以及典型气流床气化炉的模拟煤气组成和煤炭转化率数值与实验值或实际操作值的比较情况,结果显示主要组分模拟误差较小.关键词煤炭气化,气流床气化， 数学模拟中图分类号TQ529造,而我国第一座IGCC示范电厂也已立项,其气化0引言岛也将采用气流床气化工艺.因此应重视对气流床气化工艺的研究及开发工作.煤炭气化是煤洁净利用的关键技术之一,它以对气化炉气化过程数学模拟的研究不仅有利于提高碳转化率、冷煤气效率,降低气化过程的氧耗及深入了解气化过程规律,而且可以用于指导气化炉煤耗为目标,并向加压及液态排渣、大型化等方向发设计及生产过程优化控制.对气流床气化炉的数学展,从而达到改善环境及降低产品成本的目的.加压模拟主要始于20世纪70年代，国内外对此进行了气流床气化技术是国内外优先发展方向之,气流床大量研究,取得了许多成果.气化技术有如下特点:①能够气化任何变质程度的煤及煤的加氢残渣、石油焦等;②气化强度很高，碳1气流床气化原理转化率高;③产品煤气中不含焦油和酚类物质,环境友好;④其缺点是氧耗高、需设置磨粉、显热回收1.1 气流床气化过程描述及除尘等较庞大的辅助装置.已工业化或正在示范从气流床气化工艺分析看.煤粉(或水煤浆)与的加压气流床气化技术包括湿法加料的Texaco,气化剂(O2/H2O)经喷嘴喷入气化炉的燃烧区,由于Destec和干法加料的Shell, Prenflo 及GSP等,并气化炉该处温度高达1 500 C~2 000 C,因此煤粉在世界范围内得到广泛应用.受热升温速度很快(>10+C/s),可认为煤粉中的我国20世纪70年代先后进行过熔渣炉气化及.残余水分快速(瞬间)蒸发,同时由于热分解反应速K-T气化研究,也建立过工业化装置,后因耐火材度大大高于煤粉的燃烧及气化反应速度,所以细小料等问题而停止.20世纪80年代后在鲁南化肥厂、的煤粉颗粒开始发生快速热分解，即脱挥发分,生成上海焦化有限公司、渭河化肥厂、淮南化肥厂等引进半 焦和气体产物.在富含氧气及高温条件下,挥发分.了Texaco气化炉用于生产合成氨及甲醇,通过多 中的活性可燃成分如CO,H2,CH,及焦油与O2发年的实践积累了丰富的经验,已达到了长周期、高负生气相燃烧反应,生成CO2和H2O.并放出大量热荷、安全和稳定运行的要求,且在喷嘴、耐火砖及水量 供煤粉继续热解及气化反应的进行.由于气相燃煤浆泵等设备国产化方面取得了突破性进展.目前，烧反应速度很快，因此-般认为在氧气存在的情况利用干法加料的Shell气化工艺正用于某化肥厂改下,上述气相燃烧反应达到完全,亦即在氧气存在8煤炭转化2001年时,气相中不含CO,H,CH,及焦油.煤中的挥发分反应平衡时的煤气组成及平衡温度.项友谦[]用能析出后,发生半焦燃烧及气化反应,与水蒸气及CO2量最小原理,建立了加压气化平衡模型,并用4种方反应.如此时仍有氧气存在，则在气相中仍发生CO法对微分方程求解.模拟结果显示平衡模型对气流和H2燃烧反应.气化炉中的氧气反应完后,半焦与床的模拟效果要好于固定床. Watkinson[1°]提出平水蒸气、CO2和H2等继续发生气化反应,同时气相衡模型,通过质量和能量平衡及反应平衡方程式关中还有水煤气变换反应和甲烷裂解反应等.对气流联,可以得到产品煤气组成、产率和最佳适宜温度，床气化,一般将变换反应和甲烷化反应视为平衡反并对9种气化炉型的工业化和半工业化气化数据进行了比较.对产品煤气中的CO和H2含量误差在土从如上分析可以认为在对气流床模拟时,考虑0.1%之内,H2S和COS浓度可以准确的预测，但气化炉中气固相的组成分别为气相:O,H2O,CO,，CO2预测值的准确性要差一些.结果还表明该模型CO, H2,CH,N2(或NHs)，H2S(+ COS), Tar;固对气流床气化炉模拟效果最好,流化床次之，而固定相:C,H,O,N, s, Ash, Moisture.另外考虑气固相床由于一些不确定因素如挥发分含量组成等的存温度Tg, T..数学模拟的目的就是求出气化炉内每在,模拟误差较大.Ruprecht等DI建立了平衡模型,并用于评价试验数据分析.平衡模型可用于对出口-质点处的上述参数的值.组成及温度进行简单预测.由于其假定的条件较理1.2 气化反应动力学想，如假定所有反应都达到平衡,在实际过程是不可煤粉热解反应速度很快,主要根据经验式对热能的，因此其用途受到-定限制,对过程控制及气化解速度和产物进行估算,大致有Wen等[1+4给出的炉设计参考价值较少.几种表达形式.气固非均相反应,由于气化炉内反应2.2考虑气化过程动力学的模型温度很高，最高温度可达2000 C以上,所以半焦与为了更准确的用数学模型对气化及燃烧过程进O2,H2O,COz的反应速度由气膜扩散及灰层扩散控行模拟，需考虑气化过程三传一反及动力学行为.许制,而半焦与H2的反应较慢,它由化学反应控制.多研究者对此进行了研究,建立了一维、二维及多维从非均相反应动力学模型看,分为未反应核收缩模模型. Field等对1967年以前的工作进行了归纳,模型及灰层剥离模型,文献[5]认为灰层在反应过程中型属活塞流. Ubhayakar等$忽略了表面反应,但考剥离,因此反应受气膜扩散和化学反应动力学控制;虑了轴向混合、挥发分的气相反应及热解反应.而文献[1,6]认为,由于气流床气化炉中煤粉颗粒所Smith等[8,123 先后建立了一维及二维煤粉气流床燃占体积很小(<1%),且停留时间只有2 s~3 s,因烧及气化模型,模型中考虑了煤粉颗粒大小分布对此煤粉颗粒碰撞几率较小，在煤粉颗粒表面因反应反应速度的影响. Wen等[给出一维模型来描述形成的灰层可认为仍保留而不剥离,所以反应速度Texaco气化炉中的混合情况,每-微元被处理成气可由未反应核收缩模型得到反应速度.其它学者也相是完全混合,固相以活塞流通过整个反应器,固相研究了气化反应动力学问题,基本一致的观点是要粒子的速度用Stokes方程近似描述.模拟了中试装同时考虑化学反应控制及扩散控制.使用比较多的置使用煤及加氢残渣气化情况,操作参数的变化对如对加氢气化过程模拟时,采用了Johnosn]的加煤气组成的影响等. Govind等明对上述模型进行了氢热解动力学模型;Wen等[1]3提出的未反应核收缩修正,加入了动量平衡.考查了煤气组成与进料速模型也使用较多，Smith等0]也采用了独特的动力度、氧/煤比及汽/煤比的关系. Sprouse等13.14] 模拟学表达式.加氢气化反应,前者使用了经验关联式,但不适合次烟煤,后者尽管较通用和完善,但因其复杂性导致其2气流床气化数学模型综述应用困难.Brown等0研究了4种煤沿气化炉的轴向及径向气相组成分布,讨论了氧碳比及蒸汽量等对模拟结果的影响. Vamvukal[5]建立了一维稳态模2.1平 衡模型型,发现气固相最大温度位于氧气耗尽处,考察了关第4期步学朋等煤炭气化气流床气化炉的数学模拟数学模拟的攻关课题,取得了丰硕成果.现将最近的任何模型模拟的目的是能反映气化炉内物料运研究成果及文献[6,12]对以前的煤粉气流床气化及动及化学反应情况,并获得出口处煤气组成及温度、燃烧模拟情况的总结一并列于表1中.碳转化率、气化效率等参数.气流床气化炉数学模拟与实际(或实验)结果的对比情况见表2,可见对几.3模拟煤气组成与实际(或实验)结果种不同炉型煤气中的主要组分,模拟误差小，说明模拟基本上是成功的.对比表1气流床 气化数学模拟汇总Table 1 Reviews of models for coal entrained-bed gasifiersAuther(s)ReferencesYearMain Contents/DescriptionMajor ResultsMehta1976Phenomenological model combination of plug flow Describes the outlet conditions for aand perfectly-stirred calculations, gas phase chemi- three -stage entrained flow gasifier.cal equilibrium. overall char gasification with Ar-rhenius- type rates.Sprouse* ,131979One- dimensional hydrogasification; free-stream e Extensive study of boundary layerquilibrium chemistry; one-step devolatilization and around the particles of Rockwell Inthydrogasification model.FHP gasifier, compared well with out-let measurements for coal gasifierFinsonet al1978Steady, one-dimensional model，with gaseous ki- Predictions compared with laboratorynetics; kinetics from data for coal pyrolysis; het- gasification measurements of axial tem-erogeneous oxidation; considered the char struc- perature and gas composition.ture and reactivity.Ubhayakar,￥,31977One-dimensional steady-state ;two-steple- Applied to gasification combustion andStickler &. Cannonvolatilization; combined diffusion &. surface reachydropyrolysis with good agreement..ntion rates; no radiation; equilibrium gas phase re-actions.Blake et al1977-1979 Three- dimensional，transient， mixed finite-ele- Study the effect of particle size， thement, finite- difference scheme with separate Eule- conservation of mass, momentum andrian particle equations; two equation gaseous tur- energy, studied the locus or velocitybulence; separate turbulent kinetics energy for par- vectorsticles.Barnhart et alCombination of plug-ments; gas phase in equilibrium with two equation composition， temperature; reportedkinetics for pyrolysis. char oxidation and CO oxi- fair to good comparison between exper-dation.iment 8. theory for cyclone gasifier.Smith &. Smoo￥.81979- 1981 One dimensional steady-state, two-step de- Good agreement with one- dimensionalvolatilization, combined diffusion &. surface reac- combustor data。 poor agreement withtion rates. one- dimensional zonal radiation， equi- gasifier measurements.librium gas phase reactions.Smith, Fletcher, * .121979- 1981 Two dimensional, axisymmetric， with recircula-Gas -phase components reported andSmoottion, k-ε turbulence model. Lagrangian particles. compared with experiment data.two-step devolailization, diffusion and kinetic het-erogeneous rates, four-flux radiation with scatter-ing.Wen & Chuang兴,10Divided gasifier into three zones; one -step de- Predieted temperature and gas compo-volatilization, gas phase combustion completely or sition data for Texaco gasifier com-equilibrium gas phase reactions, consider diffusion pared with pilot plant gasified coal andand kinetic rates for char reaction.H-coal residue, with good agreement.Goyal141980One-dimensional steady-state model， Johnson's Modeling bench- scale hydrogasifier ofhydropyrolysis model. continuity equations for Citys Service R &. DCo.both phases, mixture momentum balance and ther-mal energy balance considered.Govind and Shan61984Refined the Wen &. Chuang model, considered the The gas composition and carbon con-momentum balance,version depends on three. essential pa-rameters: the fuel rate. the oxygen tofuel ratio and the steam fuel ratio; theoptimum steam-fuel ratio is between0.8-0.9 to achieve 98%-99% conver-sinnt the: steam fu.l ratinsicnifirantly10煤炭转化2001年.续表1Auther(s)ReferencesYearMain Contents/ DescriptionMajor ResultsGong Sunling et al11987Image the gasifier being composed of several paral- Axial gas/solid temperature and conlel reactors of which each has respective. homoge- version profiles were got. the initial re-neous particle size and also its own material and action rate very fast, and then slowheat balances ; one-dimensional equations similar to down significantly; the wall tempera-Smith's; the finite element method was used to cal-ture has influence on reaction.culate. the wall temperature. distribution and heat should be taken as an important param-loss.eter. :Brownet al51988Refined the Smith two dimensional model, gave Modeling the radial and axial gas phaseout a heat loss calculation method.profiles of four coals : investigated theeffect of full/ partial equilibrium. coalvolatiles。stoichiometric coefficients.heterogeneous. reaction rates. and heatloss on prediction results; the effect ofO/C and steam on gas composition andgasification efficiency.Ruprecht et alConsidered only three reaction, the gas composi- The agreement between prediction andtion inelude CO.H2.CO2. no CH4; equilibrium gas measured data is very good. the dis-phase reactions, carbon conversion controlled by crepancy normally amounts to 1% orO/C; mass and energy balance.less; it has been used for the evaluationof tests or process data.Vamvuka et al1:1995One- dimensional steady-state. plug-flow. ash did The boundary between combustionnot remain on the reacting particle surface. com- zone and gasification zones distinct, itsbined diffusion and chemical reaction kinetics, used location depend: on the gasifier pres-TGA data, equilibrium gas phase reactions, the sure; peak temperature of gas and solidmomentum balance neglected.near the boundary; coal conversion is. Lhutnot sensitive to the pyrolysis rate，sensitive to pressure; the effect of op-eration conditions was discussed.Ni &. Williams :A multivariable model was set up on the basis of e- The oxygen to coal ratio is the mostquilibrium. mass balance and energy balance. the important control variable for the. gasi-Shell gasifier as a typical model; the gas composi- fier operation; the cold gas efficieneytion are good agreement with realistic data.increases with the increasing tempera-ture of inlet flow. steam to coal ratiohas influence on the efficiency.LiuG Set al181999 !One-dimensional steady -state. plug- flow，the gas The predicted carbon conversion agreeturbulence is negleeted, kinetic data from PTGA with those measured in the gasifier of 8tests; considered the pore structure and reactivity coals; the better prediction results canof char; the heat balance between the wall, gas, be got by considered char structure;and particles.modeled the CSIRO atmosphere. gasifi-er and 2.4 MPa Texaco gasifier.Bu Xuepeng,￥2000One-dimensional steady- state, divided gasifier into The 48 t/d and 2 600 t/d Prenflo gasi-Peng Wanwang,three zones; the devolatilization composition by fiers were modeled， axial gas-solidXu Zhengangcalculated, gas phase combustion completely, oth- temperature and composition were got，er gas reactions equilibrium; considered diffusion main gas composition agree with realand kinetic rates for gas-char reactions.data; the effect of oxygen to coal ratio,steam- coal ratio and gasification inten-sity on gasifier. operation results wdiscussed.Yu Zunhong,Based on cold model experiment, divided gasifier Based on Texaco gasifier realistic oper-Wang Fuchen.into three zones. ; the reactions different in each ation data, predicted the feature. tem-Gong Xin et aulzone; the mixed time and reaction time are consid- perature and gas composition at threeered as main variables, chemical reactions equilibri- different load, they are agreement withumrealistic data.Li Zheng,Divided gasifier into lots of zones along axial direc- Predicted axial gas-solid temperatureHan Zhiming ,tion, in order to consider the effect of gasifier and composition of Texaco gasifier;Wang Tianjiao et alstructure and size; gas phase and solid phase reac- and the effect of pressure, water slurrytion kinetics; retention time involved.composition, coal kinds on operationresults.Gao Juzhong.Equilibrium model, only thermodynamics was con- Predicted the gas composition and gasi-Wang Ningbo,sidered; took therelation between carbon conver- fication indexes of two coals gasified inZhang Yarusion and O/C ratio, reaction extent as boundary Shell gasifiers in pilot plant and demon-第4期步学朋等煤炭气化气流 床气化炉的数学模拟I1表2模拟结果与实际结果的对比Table 2 1 Comparison of computational results from the models with experimental/ operation resultsAuthor(s)/Gas composition/%1SourceCarbon conversion/%GasifiersHCOCO2CH; H2S+COS N2+Ar H2OWenC YI1Experiment35.79 50.71 13. 140.090.030.2492.70 .TexacoModel37.30 47.86 14. 450.050.070.2694.87Watkinson[1024.30 47.10 13.20 0. 092.20.0.4012.723.70 47.20 13.30 0. 33:2.210.3013.0Watkinson30.60 61.501.600.001.304.70Shell30.30 61.501.400. 081.33; 5.3834.60 55. 407.00 0.001.951.01K-T34.90 55. 406. 700.001.931.03Qizhi Ni[17]Aetual)29.80 65. 401.70 0.000.941.9099.70 .29.7165.651.360.012.33:99. 70.Yu ZunbongOperation !37.47 45.84 16. 290.100.41Texcao38.31 44.84 15.010.110.910.82896. 20Li ZhengDocument29.80 41.05 10.20 0. 301. 100.8017.10Texeao30.55 40. 8010. 270.131.080.9616.20Gao JuzhongOperation25.60 65.100. 800.478.0399.0025.89 63. 790.00.357. 981.1999. 10Bu Xuepeng26.00 59. 593.73.0.729.96Prenflo25.93 60.212.660. 020.7710.4199.05Note:1) Percent of volume;2) Include NH;3) Others 0. 21.础上,建立二维及多维动态模型,并充分考虑颗粒大4结束语小分布、气化炉内气固相流体力学行为等.另外为适应气化炉使用中国煤种的操作，需要研究适宜中国国外对气流床气化炉的数学模拟研究起步较煤粉的高温高压快速热解反应动力学以确定其热解早,经过最近几年的努力,我国对气流床气化炉的数速度及热解煤气组成，研究适宜煤种的气化反应动学模拟也取得了初步成功，但正如专家们指出的那力学，为数学模拟提供基础数据.最后对建立的数学样，目前离真正应用还有-定距离.根据在模型建立模型还要到生产实际中进行检验，对某些参数进行及求解过程的经验，笔者认为下一步应继续进行气修正,以求能较准确反应气化炉内气化行为并用于流床气化炉的数学模拟工作.在目前-维稳态的基指导气化炉设计及生产操作.参考文献[1] WenC Y, Chuang T 2. Entrained bed Coal Gasification Modeling Interim Report. US Department of Energy Report. FE-2274-T1.1979. 1-41[2] Anthony D B. Howard J B. Coal Devolatilization and Hydrogasification, AIChE J. 1976.22<4):625-656[3] Ubhayakar S K, Stichler D B, Ganon R E. Modelling of Entrained Bed Pulverized Coal Gasifiers. Fuel, 1977 ,56(2):281-286[4] Fu Weibiao, Zhang Yanping, Han Hongqiao et al. A General Model of Pulverized Coal Devolatilization. Fuel, 1989.68<5);505- 509[5] Brown B w. Smoot L D, Smith P J et al. Measurement and Prediction of Entrained- flow Gasification Processes. AIChE J,1988.34(3):435-446[6] Govind R, Shah J. Modeling and Simulation of an Entrained Flow Coal Gasifier. AlChE J.1984.130(1):79-92[7] Johnson J L. Kinetics of Coal Gasification. New York ;John Willy 8. Sons.1979. 1-100[8] Smith P J, Smoot L D. Prediction and Measurement of Optimum Operating Conditions for Entrained Coal Gasification Processes. DE82015610.1981. 1-40T顶右:排12煤炭转化2001年[11] Ruprecht P. Schafer W. Wallace P. A Computer Model of Entrained Coal Gasification. Fuel, 1988.67(6):739-743[12] Smith P J. Smoot L D. Prediction and Measurement of Optimum Operating Conditions for Entrained Coal Gasification Pro-cesses. DE82015611.1981. 1-30[13] Spouse K M. Modeling Pulverized Coal Conversion in Entrained Flows. AIChE J,1980.126(6):964-975C 14] Goyal A. Mathematical Modeling of Entrained Flow Coal Gasification Reactors :[Ph D Thesis J. Chicago: llinoiss Inst Tech-nol, 1980. 1- 100[15] Vamvuka D, Woodburn E T, Senior P R. Modeing of An Entrained Flow Coal Gasifier. Fuel, 1995，74(10):1452- 1465[16] Gong Sunling. Xu Zhihong. Improved Simulation System of Pulverized Coal Gasifier.J of Chemical Industry and Engineer-ing (China),1997. 12(2):245-254[17] Ni Qizi. Williams A. A Simulation Study on the Performance of An Entrained flow Coal Gasifier. Fuel. 1995. 74(1):102-110[18] LiuG S, RezaeiH R, Tate A G et al. Modeling of a Pressurised Entrained Flow Coal Gasifier. In: Prospects for Coal Sciencein the 21st Century. Taiyuan : Shanxi Science & Technology Press. 1999. 579-582MATHEMATICAL MODELING OF COAL ENTRAINED-BED GASIFIERSBu Xuepeng Peng W anwang and Xu Zhengang(Beijing Research Institute of Coal Chemistry, China Coal Research Institute. 100013 Beijing)ABSTRACT The coal entrained bed gasification principle was briefly introduced in this pa-per. Based on relative documents, the mathematical modeling of coal entrained-bed gasifiers werereviewed. These include simple equilibrium models and kinetic models (one- dimensional or two/three- dimensional). The main contents-about hydromechanics, thermodynamics, chemical reac-tion kinetics, and mass balance, energy balance, momentum balance, of the models were givenout. The major results of models were also given. Finally. the comparison of computational re-sults from the models with experimental or operational results were given. It can be seen that theerrors of the main gas composition are very low.KEY WORDS coal gasification ,entrained-bed gasification , mathematical modeling(上接第6页)DEVELOPMENT OF COAL STRUCTURECheng Jun Zhou Anning and Li Jianwei(Department of Material Engineering, Xi' an Unirversity ofScience and Technology, 710054,Xi' an)ABSTRACT The coal aggregative structure is discussed, including the origin, structure andswelling behavior of several macerals. Coal chemical structure models, physical structure modelsand synthesized models are analyzed and discussed. The development of coal structure research israised. At last, the utilization of coal structure research in new material field is summarized,which reveals the importance of coal structure research in further.