Solving 3D inverse heat transfer problems in living insects using small sample data

Researching bio-heat transfer in living insects presents significant challenges, primarily driven by the intricate 3D body geometries, uncertain bio-thermal properties, dynamic metabolic processes, and complex environmental interactions, etc. Furthermore, cost and instrumental limitations typically restrict data accuracy, as well as spatiotemporal resolution. In this work, temperature measurement data for the living insect species under investigation, i.e. mosquitoes, are available for only three sites—head, thorax, and abdomen—at five-second intervals. Because mosquitoes excrete and expel droplets from the abdomen during blood feeding, the transient heat flux in this area is complex and impractical to measure directly. To estimate this unknown boundary condition, we develop a numerical framework that solves a 3D transient inverse heat transfer problem (IHTP) at the whole-body level of a living mosquito, based on the limited measurement data. The approach integrates an improved Tikhonov regularization method within a high-throughput computation architecture. To ensure transferability, an idealized, physically and biologically informed model is constructed to account for the combined effects of complex structures, uncertain thermal parameters, and dynamic variations. To validate the efficacy of this method, a simulated ring-shape boiling heat flux data set is employed as a supplementary case study within the engineering context. The high-throughput computing design reduces computation time by at least an order of magnitude compared to previous serial algorithms. To our knowledge, this work represents the first application of such a novel computational strategy for solving 3D IHTP in living insects at the whole-body scale. It is anticipated that this study will facilitate future research in thermal biology and the assessment of climate change impacts on organisms by leveraging infrared-based non-destructive measurement techniques.