Prediction model of heat transfer coefficient for high pressure water descaling based on water flux

While existing mathematical formulas have provided empirical references, significant discrepancies were observed in the calculated heat transfer coefficient (HTC) values when these formulations were applied to high-pressure water descaling processes. Consequently, the analysis of the descaling temperature field was rendered inaccurate. This study undertook a systematic investigation of the impacts of various operational variables, including water flow rate, nozzle-to-billet standoff distance, nozzle geometric parameters, and installation configuration, which were subsequently converted into water flux. Following a parametric analysis of the effects of water flux and surface temperature, comprehensive computational fluid dynamics (CFD) simulations were performed to quantitatively assess the HTC in high-pressure water descaling processes. To improve practical applicability, a mathematical model for the HTC in high-pressure water descaling was developed through nonlinear regression analysis, thereby establishing a predictive relationship between process parameters and heat transfer characteristics. The HTC under operational conditions was quantitatively assessed by employing a regression-based mathematical model formulated in this study, which delineated explicit relationships between process variables and thermal transport mechanisms in high-pressure water descaling systems. The HTC values obtained were subsequently utilized in finite element analysis (FEA) to perform numerical simulations of thermal profiles during the descaling process. To validate the thermal behavior predicted by the FEA, real-time infrared thermography measurements of billet surface temperature were conducted during industrial-scale descaling operations, thereby establishing a comparative validation framework between numerical simulations and field-acquired thermal data. The results indicated that the maximum discrepancy between the simulated and measured temperatures was 28 °C, while the minimum discrepancy was recorded at 1 °C. A quantitative correlation was established between the numerical predictions and experimental measurements, thereby affirming the predictive accuracy of the regression-derived HTC model developed in this study.