THE 1993 IGTI SCHOLAR LECTURE - LOSS MECHANISMS IN TURBOMACHINES
THE 1993 IGTI SCHOLAR LECTURE - LOSS MECHANISMS IN TURBOMACHINES
复制标题
DOI：
10.1115/1.2929299
复制
发表时间：
1993-10-01
影响因子：
1.7
通讯作者：
DENTON, JD
中科院分区：
文献类型：
--
作者：
DENTON, JD
讨论了叶轮机械损失的来源和影响，重点是试图了解损失的物理原因，而不是回顾现有的预测方法。用熵增定义了损耗，推导并讨论了它与更常见的损耗系数之间的关系。通常，熵的来源是：边界层中的粘性效应、混合过程中的粘性效应、激波和跨越温差的热传递。首先对这些问题进行了一般性的讨论，然后将结果应用于叶轮机械流动。要理解热传递造成的损失，需要对循环热力学进行一些讨论。各节专门讨论了叶片边界层和尾缘损失、叶尖泄漏损失、端壁损失、换热影响和其他损失。边界层分离造成的损失特别难以量化。大部分的讨论都是基于轴流机械，但有一个单独的章节专门讨论径向流动机械的特殊问题。在某些情况下，例如附加的叶片边界层，损失机制被很好地理解，但即使如此，损失也很少能被很准确地预测。在许多其他情况下，例如端壁损失，损失机制仍然不清楚，预测方法仍然非常依赖相关性。文章强调，相关性的使用不应取代试图理解损失来源的尝试，并建议对后者的良好物理理解可能比定量预测更有价值。
The origins and effects of loss in turbomachines are discussed with the emphasis on trying to understand the physical origins of loss rather than on reviewing the available prediction method . Loss is defined in terms of entropy increase and the relationship of this to the more familiar loss coefficients is derived and discussed. The sources of entropy are, in general: viscous effects in boundary layers, viscous effects in mixing processes, shock waves, and heat transfer across temperature differences. These-are first discussed in general and then the results are applied to turbomachinery flows. Understanding of the loss due to heat transfer requires some discussion of cycle thermodynamics. Sections are devoted to discussing blade boundary layer and trailing edge loss, tip leakage loss, endwall loss, effects of heat transfer, and miscellaneous losses. The loss arising from boundary layer separation is particularly difficult to quantify. Most of the discussion is based on axial flow machines, but a separate section is devoted to the special problems of radial flow machines. In some cases, e.g., attached blade boundary layers, the loss mechanisms are well understood, but even so the loss can seldom be predicted with great accuracy. In many other cases, e.g., endwall loss, the loss mechanisms are still not clearly understood and prediction methods remain very dependent on correlations. The paper emphasizes that the use of correlations should not be a substitute for trying to understand the origins of loss, and suggests that a good physical understanding of the latter may be more valuable than a quantitative prediction.