四墙切圆煤粉燃烧室内湍流气粒两相流动的数值模拟研究
|
摘要
煤粉燃烧室内煤粉流动与燃烧的研究对于控制煤燃烧产生的污染问题有着至关重要的意义。由于炉内换热过程的复杂性，单纯依靠实验室试验台进行研究是很困难的。随着计算机技术的不断发展，人们开始寻求用实验室试验结合数值模拟的方法来研究炉膛内的流动、传热及燃烧过程。目前国内外有不少研究者针对煤粉燃烧室进行了数值模拟，但是很多机理性问题和模拟算法尚待进一步解决与完善。所以本课题研究的主要内容为首先建立关于燃烧的合理的数学模型，然后用数值模拟方法分析煤粉燃烧室内湍流流场、温度场、浓度场，并与试验数据作对照以证明模型的合理性。
本文以一台1MW的四墙切圆煤粉燃烧室为研究对象，对炉内的速度场以及颗粒的运动轨迹进行了计算分析。本文作者根据上下炉膛的结构特点和流动特点，采用了将网格系统扭转一定角度的方法来对燃烧器区域进行网格划分，并采用块结构化网格方法和网格自适应技术来实现整个炉膛的网格划分和加密。这种网格划分的方法可以有效的减小数值模拟中的伪扩散，大大减轻了网格生成的难度，使网格数量不至于过大，有效地节省了计算所花费的时间。
对于气相流场，本文作者针对k-ε双方程模型，可实现性k-ε双方程模型以及雷诺应力模型进行了模拟比较。发现在相同初始条件和边界条件下、可实现性k-ε双方程模型和雷诺应力模型预报结果优于k-ε模型，而雷诺应力模型预报结果与试验结果更为一致。可实现性k-ε双方程模型因其计算量小，耗费机时少，所以在一般的计算条件下用于炉内模拟更为合理。此外，对流项的离散格式分别采用一阶迎风格式、二阶迎风格式及QUICK格式进行计算，通过对比分析确定二阶迎风格式作为对流项离散格式。
在模拟两相流中的颗粒轨迹时采用了颗粒轨道模型。此模型在不过分耗费计算机存储量及机时的条件下，能够模拟有复杂经历的颗粒相，而且颗粒相用拉各朗日方法可以减小伪扩散。颗粒的湍流扩散是通过随机轨道模型来模拟的。文中对采用单向耦合和双向耦合计算方法的计算结果进行了比较，结果表明炉内离散相在流场中的质量及动量承载率较低时，单向耦合计算方法适用于研究对象的冷态流场计算。
The research on the pulverized coal flow and combustion in the pulverized coal furnaces is one of the most significant studies of the control of pollution resulting from the coal combustion. Due to the complicated heat exchange in the chamber, it is difficult to deal with the research in the laboratory solely. With the progress of the computer technology, people are beginning to apply the approach that includes the test in laboratory and the numerical simulation to study the process of the flow, the heat transfer and combustion in the chamber. Though some people study the pulverized coal furnaces with numerical simulation method, there is some mechanism problem and the numerical simulation methods to be resolved and make perfect. Therefore, the main text of this paper is, firstly to construct a reasonable mathematics model for the combustion, And then to analyze the turbulence flow velocity field, the temperature field and the density field with numerical simulation method, lastly to compare the test data an
d to prove the rationality of the model.
This paper is to take the test model of a 1MW pulverized coal furnaces of four wall-tangentially-fired as the research subject. According to the structure characteristic of the chamber and the flow characteristic, the author uses two approaches: one is an approach of rotating the mesh system to divide the area of burning; the other is an approach of the block-structured grid method and the adaptive grid technology to divide the mesh and refine the grid. This method can reduce false diffusion in the numerical simulation, ease the difficulty of the mesh generation, and control the numbers of the mesh, which can save the computing time efficiently.
As to the gas phase flow field, author chooses the standard k - model, the realizable k - model and the Reynolds stress model to make simulation and comparison. In the same case of the initial conditions and boundary conditions the expectation result of the realizable k - model and the Reynolds stress model is better to the standard k - model, and moreover the expectation result of the Reynolds stress model is close to the test result. The realizable k - model is more reasonable in the chamber simulation on the common condition for its little computation amount. In addition, the
paper uses first-order upwind scheme and second-order upwind scheme and QUICK scheme to compute convection. Through comparison and analysis, the paper chooses second-order upwind scheme as discrete scheme of convection.
This paper adopts the Particle-Trace model to simulate the track of particle. This model can reduce the computer memory and save time, simulate the complex process of the particle phase and use Lagrange method to reduce the false diffusion. The Random Particle-Trace model Simulate the turbulence diffusion. This paper compares the computing results of the one-way coupling method and two-way coupling method, and the result indicates that the one-way coupling method is applicable when the mass and momentum ratio of the discrete phase in the turbulence flow velocity field is small in the chamber.
本文以一台1MW的四墙切圆煤粉燃烧室为研究对象，对炉内的速度场以及颗粒的运动轨迹进行了计算分析。本文作者根据上下炉膛的结构特点和流动特点，采用了将网格系统扭转一定角度的方法来对燃烧器区域进行网格划分，并采用块结构化网格方法和网格自适应技术来实现整个炉膛的网格划分和加密。这种网格划分的方法可以有效的减小数值模拟中的伪扩散，大大减轻了网格生成的难度，使网格数量不至于过大，有效地节省了计算所花费的时间。
对于气相流场，本文作者针对k-ε双方程模型，可实现性k-ε双方程模型以及雷诺应力模型进行了模拟比较。发现在相同初始条件和边界条件下、可实现性k-ε双方程模型和雷诺应力模型预报结果优于k-ε模型，而雷诺应力模型预报结果与试验结果更为一致。可实现性k-ε双方程模型因其计算量小，耗费机时少，所以在一般的计算条件下用于炉内模拟更为合理。此外，对流项的离散格式分别采用一阶迎风格式、二阶迎风格式及QUICK格式进行计算，通过对比分析确定二阶迎风格式作为对流项离散格式。
在模拟两相流中的颗粒轨迹时采用了颗粒轨道模型。此模型在不过分耗费计算机存储量及机时的条件下，能够模拟有复杂经历的颗粒相，而且颗粒相用拉各朗日方法可以减小伪扩散。颗粒的湍流扩散是通过随机轨道模型来模拟的。文中对采用单向耦合和双向耦合计算方法的计算结果进行了比较，结果表明炉内离散相在流场中的质量及动量承载率较低时，单向耦合计算方法适用于研究对象的冷态流场计算。
The research on the pulverized coal flow and combustion in the pulverized coal furnaces is one of the most significant studies of the control of pollution resulting from the coal combustion. Due to the complicated heat exchange in the chamber, it is difficult to deal with the research in the laboratory solely. With the progress of the computer technology, people are beginning to apply the approach that includes the test in laboratory and the numerical simulation to study the process of the flow, the heat transfer and combustion in the chamber. Though some people study the pulverized coal furnaces with numerical simulation method, there is some mechanism problem and the numerical simulation methods to be resolved and make perfect. Therefore, the main text of this paper is, firstly to construct a reasonable mathematics model for the combustion, And then to analyze the turbulence flow velocity field, the temperature field and the density field with numerical simulation method, lastly to compare the test data an
d to prove the rationality of the model.
This paper is to take the test model of a 1MW pulverized coal furnaces of four wall-tangentially-fired as the research subject. According to the structure characteristic of the chamber and the flow characteristic, the author uses two approaches: one is an approach of rotating the mesh system to divide the area of burning; the other is an approach of the block-structured grid method and the adaptive grid technology to divide the mesh and refine the grid. This method can reduce false diffusion in the numerical simulation, ease the difficulty of the mesh generation, and control the numbers of the mesh, which can save the computing time efficiently.
As to the gas phase flow field, author chooses the standard k - model, the realizable k - model and the Reynolds stress model to make simulation and comparison. In the same case of the initial conditions and boundary conditions the expectation result of the realizable k - model and the Reynolds stress model is better to the standard k - model, and moreover the expectation result of the Reynolds stress model is close to the test result. The realizable k - model is more reasonable in the chamber simulation on the common condition for its little computation amount. In addition, the
paper uses first-order upwind scheme and second-order upwind scheme and QUICK scheme to compute convection. Through comparison and analysis, the paper chooses second-order upwind scheme as discrete scheme of convection.
This paper adopts the Particle-Trace model to simulate the track of particle. This model can reduce the computer memory and save time, simulate the complex process of the particle phase and use Lagrange method to reduce the false diffusion. The Random Particle-Trace model Simulate the turbulence diffusion. This paper compares the computing results of the one-way coupling method and two-way coupling method, and the result indicates that the one-way coupling method is applicable when the mass and momentum ratio of the discrete phase in the turbulence flow velocity field is small in the chamber.
引文
[1] 杨起：中国煤炭资源的展望与建议，科学报海外版，1995，11，25
[2] 岑可法，池涌：洁净煤技术的研究和进展，动力工程，No．1，1997
[3] 谭厚章：四墙切圆水平浓淡燃烧方式试验研究与数值模拟，西安交通大学博士论文，1998
[4] B. E. Launder, D. B. Spalding: Mathematical Modlels of Turbulence, Academic Press, London, 1972
[5] R. K. Boyd, J.H. Kent: Three-Dimensional Furnace Computer Modeling., 21st Symp. (int.) on Combusion, the Combusion Institute, Pittsburgh, 265～274, 1986
[6] K. Corner, W. Zinser: Prediction of Three-Dimensional Flows in Utility Boiler Furnace and Comparison with Experiment. Comb. Sci. and Tech, 58: 43～47, 1988
[7] 岑可法，樊建人：工程气固多相流动的理论和计算，浙江大学出版社，1990
[8] 岑可法，樊建人：数值试验(CAT)在大型电站锅炉设计及调试中的应用前景，浙江大学学报，1992(1)
[9] 魏小林：浓淡煤粉燃烧的实验研究与数值模拟，西安交通大学博士论文，1995
[10] 还博文，王振宇：强旋湍流气固两相流动和煤粉燃烧数值模拟，动力工程，6(3)：43～48，1998
[11] 李力，周力行等：空气动力学学报，2001，(1)
[12] 窦国仁：湍流力学(上册)，北京：高等教育出版社，1985．99
[13] 周力行：湍流两相流动与燃烧的数值模拟，北京：清华大学出版社，1991
[14] 周力行：湍流气粒多相流数值模拟模拟理论的最近进展，燃烧科学与技术，19(1)，1995
[15] 凯斯W．M，克拉福特M．E著，陈熙等译：对流传热与传质，北京：科学出版社，1986
[16] 刘彤，姚文达：锅炉炉内流动的数学模型AS-F模型的应用，工程热物理学报，10(2)：209～211，1989
[17] Shih T H, Liou W W, Shabbir A, Yang Z G, Zhu J: A new k-ε eddy viscosity model for high Reynolds number turbulent flows, Comput Fluids, 24(3): 227～238, 1995
[18] 帕坦卡 S．V．著，张政译：传热与流体流动的数值计算，科学出版社，1984
[19] 陶文铨：数值传热学，西安：西安交通大学出版社，2001
[20] 蔡树棠，刘宇陆编著：湍流理论，上海：上海交通大学出版社，152，145，1993
[21] 陈云：旋流燃烧器内气固两相流动数值试验的研究，浙江大学硕士学位论文，2002
[22] Daly B L, Harlow F H: Transport equation of turbulence, Phys Fluids, 13: 2634～
2639，1970
[23] 吴乘康，卫景彬等：应用三维LDA对同轴射流煤粉预燃室流场的研究，工程热物理学报，10(1)：81～87，1989
[24] Farhanieh B, Davidson L, Sunden B: Employment of second moment closure for calculation of turbulent recirculating flows in complex geometries with collocated variable arrangement, Int J Number Methods Fluids, 16: 525～544, 1993
[25] 陶文铨：计算传热学的近代发展，科学出版社，2000
[26] 王良壁：复杂截面及扭转流道中紊流流动与换热的实验与数值研究，西安交通大学博士论文，1996
[27] Huang Huaxiong, Prosperetti: Effect of grid orthorgonality on the solution accuracy of the two-dimensional convection-diffusion equation, Number Heat Transfer, Part B, 26: 1～20, 1994
[28] Coelho. P.J, Darvalho. M.G: Application of a domain decomposition technique to the mathematical modeling of a utility boiler, [J]. Int J For Numerical Methods in Eng, (36): 3401～3419, 1993
[29] 苏铭德，黄素逸：计算流体力学基础，北京：清华大学出版社，1997
[30] 陈景仁：流体力学及传热学，国防工业出版社，1984
[31] Patankar. S.V, Sparrow. E.M, Ivanovic. M. Thermal interaction among the confining walls of a turbulent recirculating flow. Int J Heat Mass Transfer. 24: 269～274 1978
[32] C. Lockwood, C. Papadopoulos, A.S. Abbas: Prediction of a Corner-Fired power Station Combustor, Combust. Sci. and Tech, 58: 5～23, 1988
[33] 岑可法，樊建人：燃烧流体力学，水利电力出版社，1991
[34] F. Odar: Verification of proposed equation for calculation of forces on a sphere accelerating in a vicous fluid, J. Fluid Mech, 25(3): 591～592, 1966
[35] W. E. Ranz and W. R. Marshall: Jr. Evaporation from Drops, Part Ⅱ. Chen. Eng. Prog, 48(4): 173～180, 1952
[36] 郝吉明，马广大：大气污染控制工程，北京：高等教育出版社，2002
[37] 徐进，葛满初：气固两相流动的数值计算，工程热物理学报，3(2)，1998
[2] 岑可法，池涌：洁净煤技术的研究和进展，动力工程，No．1，1997
[3] 谭厚章：四墙切圆水平浓淡燃烧方式试验研究与数值模拟，西安交通大学博士论文，1998
[4] B. E. Launder, D. B. Spalding: Mathematical Modlels of Turbulence, Academic Press, London, 1972
[5] R. K. Boyd, J.H. Kent: Three-Dimensional Furnace Computer Modeling., 21st Symp. (int.) on Combusion, the Combusion Institute, Pittsburgh, 265～274, 1986
[6] K. Corner, W. Zinser: Prediction of Three-Dimensional Flows in Utility Boiler Furnace and Comparison with Experiment. Comb. Sci. and Tech, 58: 43～47, 1988
[7] 岑可法，樊建人：工程气固多相流动的理论和计算，浙江大学出版社，1990
[8] 岑可法，樊建人：数值试验(CAT)在大型电站锅炉设计及调试中的应用前景，浙江大学学报，1992(1)
[9] 魏小林：浓淡煤粉燃烧的实验研究与数值模拟，西安交通大学博士论文，1995
[10] 还博文，王振宇：强旋湍流气固两相流动和煤粉燃烧数值模拟，动力工程，6(3)：43～48，1998
[11] 李力，周力行等：空气动力学学报，2001，(1)
[12] 窦国仁：湍流力学(上册)，北京：高等教育出版社，1985．99
[13] 周力行：湍流两相流动与燃烧的数值模拟，北京：清华大学出版社，1991
[14] 周力行：湍流气粒多相流数值模拟模拟理论的最近进展，燃烧科学与技术，19(1)，1995
[15] 凯斯W．M，克拉福特M．E著，陈熙等译：对流传热与传质，北京：科学出版社，1986
[16] 刘彤，姚文达：锅炉炉内流动的数学模型AS-F模型的应用，工程热物理学报，10(2)：209～211，1989
[17] Shih T H, Liou W W, Shabbir A, Yang Z G, Zhu J: A new k-ε eddy viscosity model for high Reynolds number turbulent flows, Comput Fluids, 24(3): 227～238, 1995
[18] 帕坦卡 S．V．著，张政译：传热与流体流动的数值计算，科学出版社，1984
[19] 陶文铨：数值传热学，西安：西安交通大学出版社，2001
[20] 蔡树棠，刘宇陆编著：湍流理论，上海：上海交通大学出版社，152，145，1993
[21] 陈云：旋流燃烧器内气固两相流动数值试验的研究，浙江大学硕士学位论文，2002
[22] Daly B L, Harlow F H: Transport equation of turbulence, Phys Fluids, 13: 2634～
2639，1970
[23] 吴乘康，卫景彬等：应用三维LDA对同轴射流煤粉预燃室流场的研究，工程热物理学报，10(1)：81～87，1989
[24] Farhanieh B, Davidson L, Sunden B: Employment of second moment closure for calculation of turbulent recirculating flows in complex geometries with collocated variable arrangement, Int J Number Methods Fluids, 16: 525～544, 1993
[25] 陶文铨：计算传热学的近代发展，科学出版社，2000
[26] 王良壁：复杂截面及扭转流道中紊流流动与换热的实验与数值研究，西安交通大学博士论文，1996
[27] Huang Huaxiong, Prosperetti: Effect of grid orthorgonality on the solution accuracy of the two-dimensional convection-diffusion equation, Number Heat Transfer, Part B, 26: 1～20, 1994
[28] Coelho. P.J, Darvalho. M.G: Application of a domain decomposition technique to the mathematical modeling of a utility boiler, [J]. Int J For Numerical Methods in Eng, (36): 3401～3419, 1993
[29] 苏铭德，黄素逸：计算流体力学基础，北京：清华大学出版社，1997
[30] 陈景仁：流体力学及传热学，国防工业出版社，1984
[31] Patankar. S.V, Sparrow. E.M, Ivanovic. M. Thermal interaction among the confining walls of a turbulent recirculating flow. Int J Heat Mass Transfer. 24: 269～274 1978
[32] C. Lockwood, C. Papadopoulos, A.S. Abbas: Prediction of a Corner-Fired power Station Combustor, Combust. Sci. and Tech, 58: 5～23, 1988
[33] 岑可法，樊建人：燃烧流体力学，水利电力出版社，1991
[34] F. Odar: Verification of proposed equation for calculation of forces on a sphere accelerating in a vicous fluid, J. Fluid Mech, 25(3): 591～592, 1966
[35] W. E. Ranz and W. R. Marshall: Jr. Evaporation from Drops, Part Ⅱ. Chen. Eng. Prog, 48(4): 173～180, 1952
[36] 郝吉明，马广大：大气污染控制工程，北京：高等教育出版社，2002
[37] 徐进，葛满初：气固两相流动的数值计算，工程热物理学报，3(2)，1998