PyTorch implement of Boiler Furnace Temperature Field Measurement and Reconstruction Error Elimination Based on Temperature Field Residual Correction Network
TRCN: Temperature Field Residual Correction Network
ATFRSD(-W): Acoustic Temperature Field Reconstruction Simulation Dataset (with Water-cooled Walls)
*TRCN is described in section II-B&C and shown in Fig.3-5 of the paper.
**ATFRSD(-W) is described in section III-A.
Ubuntu 18.04.5 LTS
python = 3.8.0
PyTorch = 1.11.0
numpy = 1.22.3
ATFRSD and ATFRSD-W are located in the folder data and are named separately. The temperature fields in T.mat or TT.mat is to be measured, is also called target fields. The temperature fields in TR.mat or TTR.mat is reconstructed by traditional method, and its details are described in the paper.
*You can also generate datasets tailored to your needs based on the details provided in our paper, as long as abiding by academic norms.
You can use the default parameters run. The following configurations have been defined as default parameters, and you do not need to configure or change them.
batchSize = 32
epochs = 50
milestone = 30
lr = 1e-3
num_of_layers is the total number of TRCN layers, and the default setting in the code is the same as in the paper, which is 17.
num_of_layers = 17
All results involving TRCN and ATFRSD(-W) have been disclosed in the paper.
If this work has provided you with a reference, please cite our article.
@ARTICLE{10399821, author={Duan, Yixin and Chen, Liwei and Zhou, Xinzhi and Shi, Youan and Wu, Nan}, journal={IEEE Transactions on Instrumentation and Measurement}, title={Boiler Furnace Temperature Field Measurement and Reconstruction Error Elimination Based on Temperature Field Residual Correction Network}, year={2024}, volume={73}, number={}, pages={1-15}, keywords={Temperature measurement;Acoustic measurements;Pollution measurement;Measurement uncertainty;Boilers;Temperature distribution;Reconstruction algorithms;Acoustic temperature measurement;convolutional neural network;error elimination;ill-conditioned problem;temperature field construction}, doi={10.1109/TIM.2024.3353873}}
If you have any questions about TRCN and ATFRSD, please contact the authors by email.