TY - JOUR
T1 - Physics-informed neural networks for dynamic process operations with limited physical knowledge and data
AU - Velioglu, Mehmet
AU - Zhai, Song
AU - Rupprecht, Sophia
AU - Mitsos, Alexander
AU - Jupke, Andreas
AU - Dahmen, Manuel
PY - 2024
Y1 - 2024
N2 - In chemical engineering, process data are expensive to acquire, and complex phenomena are difficult to fully model. We explore the use of physics-informed neural networks (PINNs) for modeling dynamic processes with incomplete mechanistic semi-explicit differential–algebraic equation systems and scarce process data. In particular, we focus on estimating states for which neither direct observational data nor constitutive equations are available. We propose an easy-to-apply heuristic to assess whether estimation of such states may be possible. As numerical examples, we consider a continuously stirred tank reactor and a liquid–liquid separator. We find that PINNs can infer immeasurable states with reasonable accuracy, even if respective constitutive equations are unknown. We thus show that PINNs are capable of modeling processes when relatively few experimental data and only partially known mechanistic descriptions are available, and conclude that they constitute a promising avenue that warrants further investigation.
AB - In chemical engineering, process data are expensive to acquire, and complex phenomena are difficult to fully model. We explore the use of physics-informed neural networks (PINNs) for modeling dynamic processes with incomplete mechanistic semi-explicit differential–algebraic equation systems and scarce process data. In particular, we focus on estimating states for which neither direct observational data nor constitutive equations are available. We propose an easy-to-apply heuristic to assess whether estimation of such states may be possible. As numerical examples, we consider a continuously stirred tank reactor and a liquid–liquid separator. We find that PINNs can infer immeasurable states with reasonable accuracy, even if respective constitutive equations are unknown. We thus show that PINNs are capable of modeling processes when relatively few experimental data and only partially known mechanistic descriptions are available, and conclude that they constitute a promising avenue that warrants further investigation.
KW - Chemical engineering
KW - Dynamic process modeling
KW - Liquid–liquid separator
KW - Physics-informed neural networks
KW - State estimation
KW - Van de Vusse reaction
UR - http://www.scopus.com/inward/record.url?scp=85208365306&partnerID=8YFLogxK
U2 - 10.1016/j.compchemeng.2024.108899
DO - 10.1016/j.compchemeng.2024.108899
M3 - Article
AN - SCOPUS:85208365306
SN - 0098-1354
VL - 192
JO - Computers and Chemical Engineering
JF - Computers and Chemical Engineering
M1 - 108899
ER -