Physics-Informed Modeling for Wood Thermal Analysis and Prediction
2026-06-22 • Machine Learning
Machine Learning
AI summaryⓘ
The authors studied how wood's thermal behavior varies across its surface and is hard to predict using usual methods that treat wood as uniform. They created new deep learning models that incorporate physics laws about heat to better predict temperature patterns from wood images. They tested two methods: one that adds physics as a penalty during training and another that builds physics directly into the model's structure. Using real wood samples, their physics-aware models predicted temperatures more accurately and gave understandable physical insights, compared to purely data-driven models.
thermal propertiesdeep learningpartial differential equationsheat transferphysics-informed neural networksconvolutional neural networkswood heterogeneitysteady-state heat equationphysics-based modeling
Authors
Jingren Xie, Alex John Buckthal, Ryan Anthony O'Connor, Isak Worre Foged, Dim P. Papadopoulos
Abstract
Wood materials exhibit complex, spatially varying thermal properties that challenge traditional architectural assumptions of material homogeneity. Although data-driven approaches can directly map wood RGB images to their corresponding thermal responses, they operate as uninterpretable black boxes that prioritize statistical correlation and may absorb experimental noise rather than thermodynamic plausibility. To address these limitations, we present physics-informed deep learning frameworks that integrate partial differential equations (PDEs) to predict pixel-level thermal responses of spatially heterogeneous wood materials using wood RGB images and testbed temperature maps. Specifically, we investigate two distinct approaches to enforcing a normalized 2D steady-state heat transfer equation derived from the general heat transfer equation: Physics-Informed Convolutional Neural Networks (PICNNs), which embed physics as a soft penalty term in the loss function, and Physics-Integrated Convolutional Neural Networks (PInteCNNs), which hard-code an analytical approximator-predictor-corrector solver directly into convolutional neural networks. To validate our proposed approaches, we collect three real-world multimodal datasets of Poplar, Grandis Cross-Cut (Grandis-CC), and Grandis Radial-Cut (Grandis-RC) wood samples. We further demonstrate that embedding physical inductive biases successfully balances predictive accuracy, physical interpretability, and intra-species diversity, outperforming data-driven approaches in handling complex wood material heterogeneity and enabling the extraction of interpretable physical parameters. Project: https://zekifayes.github.io/pim