Prof. Hongwei Wu

Head of Energy and Sustainable Manufacturing, University of Hertfordshire, UK

E-mail: h.wu6@herts.ac.uk

Bio

Prof. Hongwei Wu is a Professor of Thermofluids and Head of Energy and Sustainable Manufacturing Group in the School of Physics, Engineering and Computer Science at the University of Hertfordshire, UK. He has been elected Fellow of the Institution of Mechanical Engineers (FIMechE), the Royal Aeronautical Society (FRAeS), the Energy Institute (FEI), and the World Society of Sustainable Energy Technologies (FWSSET). His work places particular emphasis on two-phase and multiphase flows, conjugate heat transfer, advanced cooling technology, and battery thermal management system. Professor Wu leads major UK and European research programmes funded by Horizon Europe, EPSRC, Royal Academy of Engineering, Royal Society, Innovate UK, and industry partners. He is a Panel Member for EPSRC Engineering Prioritisation, EPSRC Open Fellowships and Open Plus Fellowships, EPSRC New Horizons, UKRI Future Leaders Fellowships (FLF) Sift and Interview Panel, Royal Society, and British Council International Science Partnerships Fund (ISPF). He has authored more than 280 publications, including over 200 peer-reviewed journal articles. He is frequently invited as a plenary and keynote speaker and serving as General Chair/Co-Chair, Session Chair or Co-Chair, Organising Committee Member (OCM) and Technical Program Committee (TPC) members at many major International Conferences.

Title

A Quantum-Inspired Tensor-Train-Assisted Simulator for Conjugate Heat Transfer in Channels with Shaped Pin-Fin Elements

Abstract

A quantum-inspired Tensor-Train-assisted simulator is developed for conjugate heat transfer in a shallow cooling channel containing directly heated shaped pin-fin elements. The study addresses an optimisation-stage problem: shaped fins can improve heat removal by modifying fluid-solid interaction, but they can also obstruct flow and increase hydraulic demand. A depth-averaged finite-volume model is used to evaluate forced-convection thermal-fluid response under one-way thermal coupling, while Tensor-Train/Matrix-Product-State diagnostics characterise compact representations of the resulting thermo-fluid and geometry/source maps. For a reference shaped-fin arrangement investigated at Re=51, 102 and 151, increasing through-flow reduces the maximum solid temperature from 26.888 to 24.729,^∘ C and the thermal-resistance metric from 8.611 to 5.912 K∙W^(-1), while increasing the simulated inlet-to-outlet pressure difference from 46.947 to 140.682 Pa. The TT/MPS diagnostic maps require maximum retained ranks no greater than 30 with relative reconstruction errors below 5.72×10^(-5). These results demonstrate a physics-based and tensor-compatible workflow for future shaped-fin placement and geometry optimisation, with higher-fidelity connected-base validation reserved for selected candidate designs.