Pyomo.DOE: An open‐source package for model‐based design of experiments in Python

Predictive mathematical models are a cornerstone of science and engineering. Yet selecting, calibrating, and validating said science‐based models often remains an art in practice. Model‐based design of experiments (MBDoE) provides a systematic framework to maximize information gain from experiments...

Celý popis

Uložené v:
Podrobná bibliografia
Vydané v:AIChE journal Ročník 68; číslo 12
Hlavní autori: Wang, Jialu, Dowling, Alexander W.
Médium: Journal Article
Jazyk:English
Vydavateľské údaje: Hoboken, USA John Wiley & Sons, Inc 01.12.2022
American Institute of Chemical Engineers
Wiley Blackwell (John Wiley & Sons)
Predmet:
applied statistics
Computer applications
Data science
Design of experiments
Differential equations
Fisher information
Fisher information matrix
Mathematical analysis
Mathematical models
Optimization
Sensitivity analysis
Statistical analysis
Stochastic programming
ISSN:0001-1541, 1547-5905
On-line prístup:Získať plný text
Tagy: Pridať tag
Žiadne tagy, Buďte prvý, kto otaguje tento záznam!
Popis
Shrnutí:Predictive mathematical models are a cornerstone of science and engineering. Yet selecting, calibrating, and validating said science‐based models often remains an art in practice. Model‐based design of experiments (MBDoE) provides a systematic framework to maximize information gain from experiments while minimizing time and resource costs. But MBDoE remains limited to niche application areas, in part because practitioners must integrate expertise in statistics, computational optimization, and modeling. To help reduce this barrier, we introduce Pyomo.DOE, an open‐source package for MBDoE. Pyomo.DOE uses a nonlinear sensitivity analysis code k_aug to quickly approximate the Fisher information matrix and leverages a new stochastic programming ion. We demonstrate Pyomo.DOE with the first application of MBDoE to fixed‐bed breakthrough experiments, which highlights the power of Pyomo.DOE to quantify the value of experimental modifications a priori for large‐scale partial differential‐algebraic equation (PDAE) models. We also provide a mathematical primer on MBDoE targeted at general chemical engineers.
Bibliografia:Funding information
U.S. DOE Office of Fossil Energy and Carbon Management; Carbon Capture Simulation for Industry Impact (CCSI
)
2
ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
USDOE
ISSN:0001-1541
1547-5905
DOI:10.1002/aic.17813