TFRD benchmark

Welcome to Temperature Field Reconstruction Dataset （TFRD） benchmark.

Temperature field reconstruction of heat-source systems

Temperature field reconstruction benchmark services for thermal management of electronic devices. It tries to learn the temperature information of Point-of-interests (PoIs) from limited monitoring points and finally realizes real-time health detection of the system.

One instance one task

• Point-based modelling

  learning maps from the coordinates of point to the corresponding temperature value

  

One family one task

• Vector-based modelling

  learning maps from the temperature information of monitoring points to that of point-of-interests (PoIs)

  

• Image-based modelling

  learning maps from the temperature information of monitoring points to that of PoIs（domain discretization）

  

• Graph-based modelling

  learning maps from the temperature information of monitoring points to that of PoIs（building graph correlation）

  

Citing

@article{gong_tfrd,
    Author = {Xiaoqian Chen and Zhiqiang Gong and Xiaoyu Zhao and Weien Zhou and Wen Yao},
    Title = {A Machine Learning Modelling Benchmark for Temperature Field Reconstruction of Heat-Source Systems},
    Journal = {arXiv preprint arXiv:2108.08298},
    Year = {2021}
} 
        

@article{gong_pirl,
    Author = {Zhiqiang Gong and Weien Zhou and Jun Zhang and Wei Peng and Wen Yao},
    Title = {Physics-Informed Deep Reversible Regression Model for Temperature Field Reconstruction of Heat-Source Systems},
    Journal = {arXiv preprint arXiv:2106.11929},
    Year = {2021}
}