1
Return

Deep-learning reconstruction of surface temperature and heat flux in microchannel heat sinks from infrared foil thermography

delete2026-08-11
delete0
PRE
AI
K
Kumar Nishant Ranjan Sinha
H
Hadee Muhamed
M
Md Motiur Rahaman
S
Sarit K. Das
A
Arvind Pattamatta *
DOI：10.1016/j.icheatmasstransfer.2026.112257delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• A residual-context RED-Net reconstructs hidden thermal fields from foil-side infrared thermography. • Surface temperature and local heat flux are predicted directly from measured foil-temperature maps. • Case-level data splitting prevents patch-level leakage during model evaluation. • RED-Net inference reduces repeated reconstruction time from hours to less than one second per case. • Geometry-aware inputs are identified as essential for broader extrapolation to unseen microchannel designs.
Keywords:
Microchannel heat sink
Infrared thermography
Surface temperature reconstruction
Heat flux reconstruction
RED-Net
Deep learning

Journal

International Communications in Heat and Mass Transfer cover
International Communications in Heat and Mass Transfer
IF:
6.4
Papers:
1.0W
Citations:
2.5W

Organization

P
Purdue University
Scholars:
2.6W
Papers: 2.0W
Citations: 147
I
indian institute of technology madras
Scholars:
720
Papers: 288
Citations: 0
Cited Papers

Cited Papers

Citing Papers

Citing Papers