The application of machine learning based on neural networks NNs and genetic algorithm GA in multi- objective optimization of heat exchangers is studied Taking the tube fin heat exchanger TFHE as the

In summary, the study demonstrates the effectiveness of integrating machine learning and genetic algorithm in multi-objective optimization of heat exchangers. The proposed approach can provide a more efficient and accurate way to obtain the optimal design parameters for heat exchangers. The results also show that the inlet air velocity and ellipticity of tubes are critical parameters to consider in the optimization process. The optimized TFHE can achieve a significant reduction in pressure drop while maintaining a high level of heat transfer performance.