Cova-PINN improves heat transfer accuracy in complex heat exchangers
Cova-PINN: Cross-Domain Conservation Physics-Informed Neural Network for Fluid-Solid Conjugate Heat Transfer in Complex Geometries
Machine Learning
Summary
Heating and cooling devices rely on accurately predicting temperature changes where fluids and solids meet. The authors identify that existing methods sometimes guess temperatures well but get the total heat transfer wrong. They introduce Cova-PINN, a new neural network approach that better models how heat moves through these combined fluid-solid spaces. Tested on advanced heat exchanger shapes, it reduces errors significantly, leading to more reliable temperature and energy predictions.
What this means in practice
- •For thermal engineers: Simulate heat exchanger performance more accurately in complex geometries to design more efficient fluid-solid thermal systems.
- •For energy system designers: Improve predictions of outlet temperatures and energy transfer in devices like heat exchangers for better system optimization.
Authors
Weizheng Zhang, Xunjie Xie, Hao Pan, Lin Lu
Abstract
Multi-domain physics-informed neural networks (PINNs) flexibly model medium-specific representations to solve fluid--solid conjugate heat transfer (CHT). However, standard multi-domain PINNs enforce governing equations and interface conditions on separately sampled domain supports, which can yield plausible temperature fields but inaccurate end-to-end energy transfer and outlet temperatures. We propose Cova-PINN, a multi-domain PINN framework that aligns conservation support with thermal interaction paths in complex geometries. Cova-PINN jointly optimizes cross-domain composite control-volume balances at the local scale and paired-wall closure at the global exchanger scale. We evaluate Cova-PINN on four triply periodic minimal surface (TPMS) heat exchangers and a geometrically distinct DualMS design against CHT-specific, optimization-oriented, and complex-geometry PINN baselines under a common protocol. Relative to the closest baseline, MUSA-PINN-CHT, Cova-PINN reduces average outlet-temperature and device-level closure errors across the four TPMS topologies by $37.7\%$ and $60.2\%$, respectively, while also improving full-field and heat-duty accuracy, with consistent gains on DualMS.