The helical micro fin tubes (HFT) are commonly used in various double pipe heat exchangers because of the excellent processing and anti-fouling performance. It is of great significance to further improve the overall efficiency of the HFT so as to diminish energy consumption. In this work, the heat transfer and flow characteristics of the HFT are studied by numerical simulation. The results show that the heat transfer enhancement factors of the HFT are the secondary flow generated near the wall and the increase of the heat exchange area. In addition, the effects of the geometrical parameters on thermal–hydraulic performance are studied at Re = 36,636. It is found that the micro fin height (e), the helical angle (φ), and the number of starts (Ns) have a significant impact on the overall performance, and there is a strong mutual coupling between them. According to the parametric analysis, the HFT with a low micro fin height and a large number of starts is considered to be a better geometrical type. Finally, in order to select (or design) the HFT quickly under the specific working conditions, based on the exergy destruction minimization principle, the geometrical parameters are optimized by using the artificial neural network and genetic algorithm. An optimal solution (e = 0.23 mm, φ = 36.1°, and Ns = 66) is selected from the Pareto front by the TOPSIS method. The results indicate that the optimal solution has a sensible balance between the exergy destruction caused by heat transfer and fluid flow. Besides, it has a better thermal–hydraulic performance as well (PEC = 1.73). This work fills the gap of heat transfer and the geometrical optimization study of HFT based on the second law of thermodynamics and provides strong evidence that the exergy destruction minimization principle is still applicable in the case of the periodic model and fully developed turbulence. We hope that it will be contributed to the structural design of the HFT.