Prediction of heat exchangers performance is primordial in the world of machine design especially when the industrial environment becomes demanding in terms of size reducing or price lowering. The most common adopted procedure for this kind of investigation is defining the exchanger's overall heat transfer coefficient with a purely empiric way using interpolations and mathematical operations missing physical senses in the modeling. The present work offers a first step in theoretically modeling the heat transfer coefficient and presents a semi-empirical method which allows expressing the latter in function of the Reynolds' numbers of the fluids' flows when the variations of the Prandtl numbers of the two fluids are small. The method consists in an iterative technique based on a single test approach and allows the prediction of the heat exchanger performance for a large range of flowrates. Experimental validation of the method is performed for a cross-flow heat exchanger with water and air as working fluids, showing a relative error less than 1.6%. The relative error of the mathematical methods applied is less than 1.7%.

换热器性能的预测在机器设计领域是最重要的，尤其是当工业环境在尺寸减小或价格降低方面变得苛刻时。此类调查最常用的程序是使用插值和数学运算，在建模中使用缺少物理意义的纯经验方法来定义换热器的整体传热系数。目前的工作提供了对传热系数进行理论建模的第一步，并提出了一种半经验方法，当两种流体的普朗特数的变化为小的。该方法包括基于单一测试方法的迭代技术，并允许预测大范围流量的换热器性能。该方法在以水和空气为工作流体的错流换热器上进行了实验验证，相对误差小于1.6%。所应用的数学方法的相对误差小于1.7%。