节点文献

循环流化床回路流动及颗粒分层特性研究

Study on Hydrodynamic and Particles Segregation in the Loop of Circulating Fluidized Bed

【作者】 姚宣

【导师】 杨海瑞;

【作者基本信息】 清华大学 , 动力工程及工程热物理, 2010, 硕士

【摘要】 循环流化床具有良好的气固混合、传质、传热特性,是目前商业化程度最好的清洁煤燃烧技术。目前的研究和工程实践表明,对于循环流化床,还存在很多的问题亟待解决:比如在工业锅炉中物料循环流率的测定、立管中气体流动对于系统物料平衡的影响以及宽筛分颗粒的分层特性等等。本文通过搭建冷态循环流化床试验台,基于循环流化床系统回路的角度,系统研究了窄筛分颗粒床料条件下,操作参数对于各部件的压降特性影响。根据冷态测量结果,得出了各部件压降和循环流率之间的变化关系,绘制了提升管床压降-循环流率图谱,给出了保证快速流态化的床压降区间,用以指导提升管内的流态重构;发现旋风分离器的压降呈现出非线性的变化规律,随着循环流率增大先下降后上升,压降-循环流率的转捩点与风速、颗粒物性相关;返料阀水平段的阻力随循环流率增加上升,竖直段阻力随着循环流率增加而下降,阀总压降呈现下降趋势,表征输送物料的能力增加;同时,采用压力梯度、气体示踪和光导纤维等测量方式,直接测量了立管中的气体流动规律,并与移动床理论计算结果进行了对比,验证了立管内气固流动处于黏性滑移流动状态;立管通过调节气固滑移速度来得到不同压差梯度,实现低料封-高压头的特性,从而吸收其它部件的压力波动,维持整个循环流化床系统的压力平衡。在单组元颗粒回路特性的基础上,本文研究了双组元体系在循环流化床内的颗粒分层特性,以及物料在提升管和立管之间的分配规律。实验结果表明,提升管内的颗粒分层随着流化风速、循环流率的增大而减弱,在较高的风速(6.0 m/s)和循环流率情况下,颗粒分层消失,系统达到完全混合的状态;根据文献中的实验结果,提出了端头结构对于系统颗粒分层的影响,并在冷态实验中得到了验证。由于端头强约束形式的存在,会使得较多粗颗粒在炉膛出口发生返混,更多的粗颗粒停留在提升管侧,直接影响了双组元颗粒在提升管、立管之间的分配。根据物料平衡的基本理论,本文还建立了循环流化床内宽筛分颗粒的流动和分层模型,并针对冷态实验值中的实验工况进行了计算,计算值较好的吻合了实验结果,验证了模型的可靠性。

【Abstract】 Circulating fluidized bed (CFB), as one of the best commercialized clean-coal technologies, has good advantages in gas-solid mixing, mass and heat transfer in the boiler. However, current researches and engineering experience show that lots of issues remained unsolved in the area, such as the measurement of solid circulating rate (Gs) in industry boiler, the impact of gas flow in standpipe on material balance in whole CFB system, particle segregation in the furnace and so on.In this paper, a cold test rig of CFB was constructed to study the gas-solid flow in the CFB loop. In the view of system loop, the impacts of operation parameters on the pressure drops of each component with narrow size range of particles were investigated. From the experimental results, the influence of Gs on each component’s pressure drop was obtained. The relationship between riser pressure drop and Gs was drawn. The relationship helps to choose the pressure drop in the riser to ensure the fast bed flow regime in the riser, which can be used to guide the reconstruction of flow state. The pressure drop of the cyclone firstly decreases and then increases with solid concentration increasing. The transition point of the variation was influenced by the gas velocity and particle prosperity. The pressure drop of the horizonal section part of the Loop seal increases with Gs, while that of the vertical section part decreases. The overall pressure drop of Loop seal that indicates the ability of transporting particles decreases with Gs. Combination with the pressure gradient measurement, lacer optic probe and direct gas trace method, the flow state in the standpipe was also studied. The packed flow state in the standpipe was verified by the comparison of experimental result and the theoretical calculation. By adjusting the relative gas-solid slip velocity, different pressure gradient of gas-solid flow is achieved. The special feature with low seal height to provide high pressure head helps to maintain the mass and pressure balance of the whole loop system.Based on the above results, this paper also studied particles segregation with binary partices with different sizes in the CFB. Experimental results showed that segregation intensity in the riser becomes weaker with increaseing gas velocity and Gs. In the case of higher gas velocity (6.0 m/s) and higher Gs, segregation disappeares and the system reaches completed mixing state. With other published literatures, this paper proposed the exit geometry may have impact on segregation in the system and verified it through the experiment. Because of the inner separation of exit geometry, more coarse particles are easily separated and flow down back into the lower riser, comparing with fine particles. Therefore, the distribution of particle between the riser and standpipe was changed, which lead to more coarse particles in the riser than in the standpipe. At last, a CFB mass balance model with wide size distribution particles, including the segregation model and pressure balance mocel, was developed. The model predicting results were in good agreement with data in present study and and literatures, which verified the reliability of the model.

  • 【网络出版投稿人】 清华大学
  • 【网络出版年期】2012年 05期
节点文献中: