Acoustic emission sensing for non-destructive characterization of orientation-dependent flow condensation

Orientation of gravity strongly influences flow condensation heat transfer, yet conventional diagnostics for quantifying local heat transfer remain intrusive and difficult to implement in different orientations. This work introduces a non-destructive method, acoustic emission (AE) sensing, to differentiate flow condensation orientation and characterize local heat transfer under different orientations. AE sensing is applied to a tube-in-tube external flow condensation facility with varying flow rates. Steady-state condensation experiments are conducted in a vertical downward orientation and integrated with prior data obtained in a horizontal orientation. An unsupervised machine learning model, K-means clustering, successfully separates vertical and horizontal condensation from AE features, with the Counts feature providing the dominant contribution and achieving adjusted Rand indices above 0.99 for all cases. In addition, orientation-independent AE features are identified to explicitly map AE features into condensation heat transfer of both orientations by exhaustive AE-polynomial regression. Heat fluxes and heat transfer coefficients at various axial locations are accurately predicted by orientation-independent AE features with mean absolute percentage errors (MAPEs) of 2.61% and 5.47%, respectively. These results demonstrate that AE sensing provides a robust diagnostic for both flow orientation identification and quantitative heat transfer prediction in flow condensation across different orientations. The proposed orientation-independent acoustic descriptors offer a practical alternative to conventional thermal measurements and enable real-time monitoring of flow condensation systems across varied orientations.