TY - JOUR
T1 - The Potential of Hybrid Mechanistic/Data-Driven Approaches for Reduced Dynamic Modeling
T2 - Application to Distillation Columns
AU - Schäfer, Pascal
AU - Caspari, Adrian
AU - Schweidtmann, Artur M.
AU - Vaupel, Yannic
AU - Mhamdi, Adel
AU - Mitsos, Alexander
PY - 2020
Y1 - 2020
N2 - Extensive literature has considered reduced, but still highly accurate, nonlinear dynamic process models, particularly for distillation columns. Nevertheless, there is a need for continuing research in this field. Herein, opportunities from the integration of machine learning into existing reduction approaches are discussed. First, key concepts for dynamic model reduction and their limitations are briefly reviewed. Afterwards, promising model structures for reduced hybrid mechanistic/data-driven models are outlined. Finally, crucial future challenges as well as promising research perspectives are presented.
AB - Extensive literature has considered reduced, but still highly accurate, nonlinear dynamic process models, particularly for distillation columns. Nevertheless, there is a need for continuing research in this field. Herein, opportunities from the integration of machine learning into existing reduction approaches are discussed. First, key concepts for dynamic model reduction and their limitations are briefly reviewed. Afterwards, promising model structures for reduced hybrid mechanistic/data-driven models are outlined. Finally, crucial future challenges as well as promising research perspectives are presented.
KW - Distillation columns
KW - Dynamic model reduction
KW - Hybrid modeling
KW - Machine learning
KW - Surrogate models
UR - http://www.scopus.com/inward/record.url?scp=85092078832&partnerID=8YFLogxK
U2 - 10.1002/cite.202000048
DO - 10.1002/cite.202000048
M3 - Review article
AN - SCOPUS:85092078832
SN - 0009-286X
VL - 92
SP - 1910
EP - 1920
JO - Chemie-Ingenieur-Technik
JF - Chemie-Ingenieur-Technik
IS - 12
ER -