In order to enhance thermal efficiency, most process industries are fitted with heat exchangers. Without using external power, we can increase the heat transfer rate by providing different strips in tubes. This paper deals with the passive method of elliptic leaf insertion in a double pipe heat exchanger with opposite orientation and same direction and the heat transfer rate and pressure drop are found. The double elliptical leaf used in experiment has major to minor axes ratios as 2:1 and distance of 50 mm between two leaves are arranged at different angular orientations from 0⁰ to 180⁰ with 10⁰ intervals. Experimental and numerical simulations are carried and the thermal performance is evaluated at various Reynolds number. Nusselt number increases with increases of Reynolds number and among all elliptical leaf angles, 60⁰ leaf angles gives more heat transfer rate because of high surface area and secondary flows near by the walls of the tube. Friction factor decreases with increased Reynolds number. From the obtained experimental results, a statistical tool generalized regression neural network is considered where the output values are obtained by giving known input parameters without conducting any experimentation.

为了提高热效率，大多数过程工业都装有热交换器。在不使用外部电源的情况下，我们可以通过在管中设置不同的条带来提高传热速率。本文研究了一种在方向相反，方向​​相同的双管换热器中椭圆叶片插入的被动方法，发现了传热速率和压降。实验中使用的双椭圆形叶片的长轴与短轴之比为2：1，两片叶片之间50 mm的距离以从0 ⁰到180 ^0的不同角度方向排列，其中10 ⁰间隔。进行了实验和数值模拟，并在各种雷诺数下评估了热性能。努塞尔数随雷诺数的增加而增加，并且在所有椭圆形叶片角中，由于高表面积和在管壁附近的二次流动，60 ⁰叶片角可提供更高的传热率。摩擦系数随雷诺数的增加而减小。根据获得的实验结果，考虑使用统计工具广义回归神经网络，其中通过给出已知的输入参数而无需进行任何实验即可获得输出值。