GMDH and RSM models for prediction of heat transfer parameters in an ultrasonic vibrating fin-and-tube heat exchanger

This study employed the Group Method of Data Handling (GMDH) and Response Surface Methodology (RSM) to predict key heat transfer parameters, including the Nusselt number (Nu), outlet temperature, heat flow, and convective heat transfer coefficient in a fin-and-tube heat exchanger (FTHX) subjected to ultrasonic vibrations. Experimental investigations were conducted with varying inlet temperatures (10–140 °C), flow rates (2–6 l/min), air velocities (0.1–4 m/s), and ultrasonic power levels (0 or 50 W). The models’ accuracy was validated against experimental data, showing high correlation coefficients exceeding 0.98. The GMDH model slightly outperformed the RSM model. The maximum absolute average relative error (AARE) was 0.0633, demonstrating the models’ precision. These findings provide valuable insights for optimizing thermal systems and enhancing heat transfer efficiency in heat exchangers through ultrasonic vibration.