A deep learning method based on partition modeling for reconstructing temperature field

Abstract

Physical field reconstruction is highly desirable for the measurement and control of engineering systems. The reconstruction of the temperature field from limited observation plays a crucial role in thermal management for electronic equipment. Deep learning has been employed in physical field reconstruction, whereas the accurate estimation for the regions with large gradients is still difficult. To solve the problem, we propose a novel deep learning method based on partition modeling to accurately reconstruct the temperature field of electronic equipment from limited observation. Firstly, the temperature field reconstruction (TFR) task of electronic equipment is modeled mathematically and transformed as an image-to-image regression problem. Then a partition modeling framework consisting of an adaptive UNet and a shallow multilayer perceptron (MLP) is developed to establish the mapping from the observation to the temperature field. The adaptive UNet is utilized to reconstruct the whole temperature field, while the MLP is designed to predict the patches with large temperature gradients. Numerical case studies employing finite element simulation data are conducted to demonstrate the accuracy of the proposed method. Furthermore, the generalization is evaluated by investigating cases under different heat source layouts, power intensities, and observation point locations. The maximum absolute errors of the reconstructed temperature field are less than 1 K under the partition modeling approach.