A physics-informed neural operator trained on 20 simulated thermoelectric materials infers thermal conductivity and Seebeck coefficient for 60 unseen materials from six sparse measurements, with test R-squared above 0.97.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cond-mat.mtrl-sci 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Physics-Informed Neural Operators for Generalizable and Label-Free Inference of Temperature-Dependent Thermoelectric Properties
A physics-informed neural operator trained on 20 simulated thermoelectric materials infers thermal conductivity and Seebeck coefficient for 60 unseen materials from six sparse measurements, with test R-squared above 0.97.