【作者】 胡和敏；

【导师】 杜小泽；

【作者基本信息】 华北电力大学 ， 热能工程， 2014， 博士

【摘要】 空冷技术具有明显的节水优势,是富煤贫水地区火力发电的重要选择。火电厂空冷机组的流动换热性能不仅取决于翅片的换热效率,还受到翅片管束、空冷凝汽器单元、空冷岛、电厂主要构筑物、气象和地形地貌条件等多个尺度层次的制约。最大尺度和最小尺度之间的跨度为106-107m,是典型的系统多尺度问题。火电厂空冷系统中流动传热规律的研究所采用的实验和数值方法需要消耗巨大的计算时间和计算资源。在同时考虑所有尺度层次相互作用机制的情况下,数值计算中所采用的多重网格法建立的模型容易受到大小尺度区域之间界面上网格耦合的限制,伺时庞大的模型很容易超出现有计算机的计算能力。在自然环境风影响下,引入本征正交分解(POD)方法对二维空冷凝汽器单元的流场和温度场等变量建立了低维模型,快速预测了新工况下的变量分布,并对其流动传热规律进行了分析。其中低维模型基本模态的权系数采用三次样条插值和通量守恒两种方法求得,并分析了各自的优缺点。用优化的本征正交分解补集(PODc)法得到了适合整个参数研究范围的鲁棒低维模型。用本征正交分解方法的低维模型得到的凝汽器单元变量场,在保持与数值模拟结果相近精度的情况下,计算效率得到了明显提高,大大节省了计算时间和计算资源。引入的本征正交分解方法对于处理火电厂空冷系统中凝汽器结构和自然环境风综合作用下所表现出的复杂流动传热问题是非常有效的。该方法为空冷机组流动换热规律的研究、空冷系统的结构优化以及运行的实时操作控制等提供了很好的手段。针对三维凝汽器单元中流动换热性能对环境风向有着高度依赖性的特点,提出本征正交分解备用(PODs)法与本征正交分解补集(PODc)法相结合的PODs-c方法来处理环境风向角较小工况下凝汽器单元中由于大变量梯度产生的各种湍流涡引起的强烈非线性问题,计算精度得到了显著提高。提出的PODs-c方法有效拓展了用本征正交分解的低维模型处理空冷凝汽器单元中流动换热问题的参数研究范围。针对火电厂空冷系统中多尺度流动传热问题,提出本征正交分解与计算流体力学相结合的跨尺度模拟方法。该方法中采用正交分解法对小尺度区域的各变量建立低维模型并快速准确得预测其流动换热状态,在大小尺度区域的跨尺度界面上,将低维解保留的小尺度传输信息作为边界条件,实现对大尺度区域流动和传输特性的数值模拟。采用提出的跨尺度模拟方法来研究包含2×30个小尺度凝汽器单元的空冷岛和大尺度的电厂主要构筑物以及地形地貌影响下的流动换热特性。揭示出了自然环境风影响下空冷岛的热风回流以及背风侧空冷凝汽器的传热恶化等特征。提出的跨尺度模拟方法能够实现动力与运动的相互关联,并实现空冷系统中空气流动传热特性的跨尺度关联,从而揭示出不同尺度湍流涡的相互作用机制。相较于多重网格数值模拟技术,显著得节省了计算时间和计算资源,却能得到了与之相近精度的模拟结果。该方法也可用于解决其他领域存在的多尺度问题。更多还原

【Abstract】 Air-cooled condensers with obvious water conservation benefit can be an important alternative for power plant near coal mines where water source is of shortage over the past few decades. The flow and heat transfer performance not only depends on heat transfer efficiency among fins, it is also restricted by finned tube bundles, air cooled condensers (ACC), air cooled island, main structures of power plant, and the meteorological and the geographic conditions. The smallest and largest length scale in this typical multi-scale system have bridged106-107m order of magnitude. The experiment and numerical simulation methods adopted by flow and heat transfer research of air cooling system in power plant always consume gigantic calculating time and resource. Moreover, considered simultaneously interacting mechanism among all length scale levels, the established models used multi-grid method in numerical simulation were easily restricted by the mesh interconnection on the interfaces between large and small scale regions. And the gigantic models were easy to exceed the calculating ability of the existing computer.Influenced by the natural environmental wind, the reduced order models based proper orthogonal decomposition (POD) method were established about air side velocity and temperature fields of two-dimensional air cooled condenser. In order to analyze the flow and heat transfer laws, the variable fields at new cases were predicted quickly and accurately. The weight coefficients for POD modes were obtained by cubic spline interpolation and flux matching procedure, and analyzed the respective possessing advantages and disadvantages. In order to obtain the robust reduced order models over the whole researching range of parameter, the improved complementary POD(PODc) approach was adopted in the present research.The calculating efficiencies of variable fields have been improved obviously in air cooled condenser, and the calculating time and resources were significantly saved through the reduced order models based POD method with similar precision to mutlti-grid CFD simulating results. The introduced POD method is very effective to deal with the complex flow and heat transfer problem in air cooled condenser of power plant under the comprehensive influence of ACC structure and ambient wind. The present investigation may provide a rapid and reliable approach for the research of flow-thermo rules, structure optimization, and real-time operation control of air cooling system in power generating unit.Due to the high dependence of flow and heat transfer performance of three-dimensional ACC onto the natural wind angle, combined procedure (PODs-c) of spare POD (PODs) with complementary POD (PODc) was reformulated to treat the high nonlinear issue with large variable gradients generating kinds of vortices in ACC at the relative small natural direction angle cases. And the calculating precisions have been obviously improved. The reformulated PODs-c effectively extends the researching parameter range of flow and heat transfer for ACC utilized reduced order models based POD.In order to deal with the multi-scale flow and heat transfer problem of air cooling system in power plant, a cross scale modeling methodology combined POD and CFD was proposed in the present paper. Reduced order models based proper orthogonal decomposition were established about variables in small-scale region, and the flow and heat transfer status were predicted quickly and accurately. On the cross scale interfaces between large and small scale regions, numerical simulation of flow and transport characteristics on large-scale region were realized, with the boundary condition of the reserving transport information from reduced order solution in small scale region.The proposed cross scale modeling methodology was adopted to research the flow and heat transfer characteristic of air cooling system crossing two neighbor characteristic lengths, including that of the air cooling island region containing2x30air cooled condensers with small length scale, and the large-scale one consisted of power plant buildings and the geomorphology. The hot plume recirculation of the air cooling island caused by environmental natural wind, as well as the heat transfer deterioration of downwind ACC cells were revealed.The cross scale relationship of air flow and heat transfer characteristic, as well as the interdependent relationship between kinematic and dynamic in air cooling system were realized adopted the present cross scale modeling methodology. And interacting mechanism among different length scale turbulent vortex were revealed. The calculating time and resources were significantly saved through the present cross scale modeling methodology with similar precision to multi-grid CFD simulating technique. The cross scale modeling methodology can be used to solve the existing multi-scale problem in other fields, and the researched achievement exerts positive contribution to the research of multi-scale science.更多还原