1
Return

Physics-informed deep learning for cross-fuel spray evolution prediction and blend screening

delete2026-08-12
delete0
PRE
AI
X
Xinrui Tao
J
Jinhong Fu
S
Shangning Wang
Z
Zhiyin Ma
X
Xuesong Li *
DOI：10.1016/j.applthermaleng.2026.132775delete
deleteOriginal
deleteOriginal request for help
deleteShare
deleteSave
Abstract

Abstract

En 中文
• Physics-informed deep learning predicts cross-fuel sprays and screens blends. • Composite physics loss and temporal tracking ensure fluid dynamic plausibility. • The model generalizes reliably to unseen alternative fuels under flash boiling. • Data-driven screening identifies optimal mixture ratios, 3D CFD validated.
Keywords:
Alternative fuels
Flash boiling spray
Physics-informed deep learning
Surrogate model
Fuel blend screening

Journal

Applied Thermal Engineering cover
Applied Thermal Engineering
IF:
6.9
Papers:
2.6W
Citations:
10.6W

Organization

S
shanghai jiao tong university
Scholars:
15.1W
Papers: 11.5W
Citations: 159
T
tongji university
Scholars:
7.5W
Papers: 5.8W
Citations: 98
Cited Papers

Cited Papers

Citing Papers

Citing Papers