A tutorial review of neural network modeling approaches for model predictive control

Yi Ming Ren; Mohammed S. Alhajeri; Junwei Luo; Scarlett Chen; Fahim Abdullah; Zhe Wu; Panagiotis D. Christofides

doi:10.1016/j.compchemeng.2022.107956

A tutorial review of neural network modeling approaches for model predictive control

Yi Ming Ren, Mohammed S. Alhajeri, Junwei Luo, Scarlett Chen, Fahim Abdullah, Zhe Wu, Panagiotis D. Christofides

Multiscale Materials

Research output: Contribution to journal › Review article › peer-review

80 Scopus citations

Abstract

An overview of the recent developments of time-series neural network modeling is presented along with its use in model predictive control (MPC). A tutorial on the construction of a neural network-based model is provided and key practical implementation issues are discussed. A nonlinear process example is introduced to demonstrate the application of different neural network-based modeling approaches and evaluate their performance in terms of closed-loop stability and prediction accuracy. Finally, the paper concludes with a brief discussion of future research directions on neural network modeling and its integration with MPC.

Original language	English
Article number	107956
Journal	Computers and Chemical Engineering
Volume	165
DOIs	https://doi.org/10.1016/j.compchemeng.2022.107956
State	Published - Sep 2022

Funding

Financial support from the National Science Foundation and the Department of Energy is gratefully acknowledged.

Keywords

Chemical processes
Encoder–decoder architecture
Feed-forward neural networks
Model predictive control
Nonlinear systems
Recurrent neural networks
Time-series forecasting

Access to Document

10.1016/j.compchemeng.2022.107956

Cite this

@article{5ced4997e9024660bdbad4586d88834b,

title = "A tutorial review of neural network modeling approaches for model predictive control",

abstract = "An overview of the recent developments of time-series neural network modeling is presented along with its use in model predictive control (MPC). A tutorial on the construction of a neural network-based model is provided and key practical implementation issues are discussed. A nonlinear process example is introduced to demonstrate the application of different neural network-based modeling approaches and evaluate their performance in terms of closed-loop stability and prediction accuracy. Finally, the paper concludes with a brief discussion of future research directions on neural network modeling and its integration with MPC.",

keywords = "Chemical processes, Encoder–decoder architecture, Feed-forward neural networks, Model predictive control, Nonlinear systems, Recurrent neural networks, Time-series forecasting",

author = "Ren, {Yi Ming} and Alhajeri, {Mohammed S.} and Junwei Luo and Scarlett Chen and Fahim Abdullah and Zhe Wu and Christofides, {Panagiotis D.}",

note = "Publisher Copyright: {\textcopyright} 2022 Elsevier Ltd",

year = "2022",

month = sep,

doi = "10.1016/j.compchemeng.2022.107956",

language = "English",

volume = "165",

journal = "Computers and Chemical Engineering",

issn = "0098-1354",

publisher = "Elsevier",

}

TY - JOUR

T1 - A tutorial review of neural network modeling approaches for model predictive control

AU - Ren, Yi Ming

AU - Alhajeri, Mohammed S.

AU - Luo, Junwei

AU - Chen, Scarlett

AU - Abdullah, Fahim

AU - Wu, Zhe

AU - Christofides, Panagiotis D.

PY - 2022/9

Y1 - 2022/9

N2 - An overview of the recent developments of time-series neural network modeling is presented along with its use in model predictive control (MPC). A tutorial on the construction of a neural network-based model is provided and key practical implementation issues are discussed. A nonlinear process example is introduced to demonstrate the application of different neural network-based modeling approaches and evaluate their performance in terms of closed-loop stability and prediction accuracy. Finally, the paper concludes with a brief discussion of future research directions on neural network modeling and its integration with MPC.

AB - An overview of the recent developments of time-series neural network modeling is presented along with its use in model predictive control (MPC). A tutorial on the construction of a neural network-based model is provided and key practical implementation issues are discussed. A nonlinear process example is introduced to demonstrate the application of different neural network-based modeling approaches and evaluate their performance in terms of closed-loop stability and prediction accuracy. Finally, the paper concludes with a brief discussion of future research directions on neural network modeling and its integration with MPC.

KW - Chemical processes

KW - Encoder–decoder architecture

KW - Feed-forward neural networks

KW - Model predictive control

KW - Nonlinear systems

KW - Recurrent neural networks

KW - Time-series forecasting

UR - http://www.scopus.com/inward/record.url?scp=85136478556&partnerID=8YFLogxK

U2 - 10.1016/j.compchemeng.2022.107956

DO - 10.1016/j.compchemeng.2022.107956

M3 - Review article

AN - SCOPUS:85136478556

SN - 0098-1354

VL - 165

JO - Computers and Chemical Engineering

JF - Computers and Chemical Engineering

M1 - 107956

ER -

A tutorial review of neural network modeling approaches for model predictive control

Abstract

Funding

Keywords

Access to Document

Other files and links

Fingerprint

Cite this