Haoze Zhang

Thermodynamics-Informed Input Reparameterization for Neural Prediction of Real-Fluid Thermodynamic Properties in Supercritical Combustion

Haoze Zhang, Han Li, Ke Xiao, Yangchen Xu, Runze Mao, Zhi X. Chen

Real-fluid thermodynamic property evaluation is a major computational cost in supercritical combustion simulations. In the enthalpy-based pressure-correction formulation, the closure evaluates temperature T, density $ρ$, and compressibility coefficient $ψ$ from the solver state (…Read more