Efficient thermal management of high temperature synthesis gas produced during underground coal gasification is essential for improving system performance and ensuring safe operation of gas processing units. Membrane helical coil heat exchangers have demonstrated superior heat transfer capabilities compared with conventional straight and serpentine tube designs due to the formation of secondary flow structures within curved tubes. However, predicting thermo hydraulic performance under varying operating conditions remains a complex task because of nonlinear interactions between flow parameters and geometric characteristics.

This study proposes a machine learning-based predictive framework for estimating heat transfer and pressure drop characteristics in membrane helical coil heat exchangers used for high pressure syngas cooling. Computational Fluid Dynamics (CFD) simulations were first performed to generate a comprehensive dataset covering a wide range of Reynolds numbers, coil diameters, pitch ratios, and operating pressures. The generated dataset was then used to train machine learning models including Artificial Neural Networks (ANN), Random Forest (RF), and Support Vector Regression (SVR).

The performance of the developed models was evaluated based on prediction accuracy for Nusselt number and friction factor. Results show that the ANN model achieved the highest prediction accuracy with a coefficient of determination (R²) greater than 0.97. The machine learning models were capable of predicting thermo-hydraulic performance significantly faster than traditional CFD simulations.

The proposed framework enables rapid evaluation and optimization of heat exchanger designs for high temperature syngas cooling applications. The integration of machine learning with CFD provides a powerful approach for accelerating the design of advanced heat exchangers used in underground coal gasification systems and other high-temperature energy technologies.