Visual data-based diagnostics of regime transitions in nucleate pool boiling for next-generation digital twins

We introduce a novel metric, the Index of Visual Similarity (IVS), to qualitatively characterize the change in extent of heat transfer within nucleate pool boiling regime using visual data. IVS is defined by combining morphological similarity (using Scale Invariant featuresbased feature matching) along with physical similarity (using vapor area estimation by Mask R-CNN). We have demonstrated the feasibility of the approach through pool boiling of water on two distinct surfaces, polished copper and porous foam. An additional validation is performed using Silicon surface. IVS has shown to capture critical changes in bubble size and morphology that correspond to transitions in domination of heat transfer mechanisms as pointed out in multiple studies. For further validation, we have also compared IVS with an equivalent metric, ϕ, derived from measured heat transfer coefficients (HTC). Results have shown a strong correlation and reliability in detecting boiling regime transitions, including the onset of nucleate boiling and proximity to critical heat flux (CHF). Hence IVS could be a critical development towards non-intrusive diagnostic tool for realtime boiling systems as well as a digital twin for estimation of boiling crisis failure in the systems utilizing boiling phenomena.