Machine‐learning‐based construction of barrier functions and models for safe model predictive control

Chen, Scarlett; Wu, Zhe; Christofides, Panagiotis D.

doi:10.1002/aic.17456

Machine‐learning‐based construction of barrier functions and models for safe model predictive control

Journal Article · Wed Oct 06 00:00:00 EDT 2021 · AIChE Journal

DOI:https://doi.org/10.1002/aic.17456· OSTI ID:1824904

Chen, Scarlett ^[1]; Wu, Zhe ^[1]; ^[2]

Department of Chemical and Biomolecular Engineering University of California Los Angeles California USA
Department of Chemical and Biomolecular Engineering University of California Los Angeles California USA, Department of Electrical and Computer Engineering University of California Los Angeles California USA

Abstract

In this paper, we propose a control Lyapunov‐barrier function‐based model predictive control method utilizing a feed‐forward neural network specified control barrier function (CBF) and a recurrent neural network (RNN) predictive model to stabilize nonlinear processes with input constraints, and to guarantee that safety requirements are met for all times. The nonlinear system is first modeled using RNN techniques, and a CBF is characterized by constructing a feed‐forward neural network (FNN) model with unique structures and properties. The FNN model for the CBF is trained based on data samples collected from safe and unsafe operating regions, and the resulting FNN model is verified to demonstrate that the safety properties of the CBF are satisfied. Given sufficiently small bounded modeling errors for both the FNN and the RNN models, the proposed control system is able to guarantee closed‐loop stability while preventing the closed‐loop states from entering unsafe regions in state‐space under sample‐and‐hold control action implementation. We provide the theoretical analysis for bounded unsafe sets in state‐space, and demonstrate the effectiveness of the proposed control strategy using a nonlinear chemical process example with a bounded unsafe region.

View Accepted Manuscript (Publisher)

Sponsoring Organization:: USDOE

OSTI ID:: 1824904

Journal Information:: AIChE Journal, Journal Name: AIChE Journal Journal Issue: 6 Vol. 68; ISSN 0001-1541

Publisher:: Wiley Blackwell (John Wiley & Sons)Copyright Statement

Country of Publication:: United States

Language:: English

References (21)

Machine learning‐based predictive control of nonlinear processes. Part I: Theory Wu, Zhe; Tran, Anh; Rincon, David AIChE Journal, Vol. 65, Issue 11 https://doi.org/10.1002/aic.16729	journal	August 2019
A ‘universal’ construction of Artstein's theorem on nonlinear stabilization Sontag, Eduardo D. Systems & Control Letters, Vol. 13, Issue 2 https://doi.org/10.1016/0167-6911(89)90028-5	journal	August 1989
A universal formula for stabilization with bounded controls Lin, Yuandan; Sontag, Eduardo D. Systems & Control Letters, Vol. 16, Issue 6 https://doi.org/10.1016/0167-6911(91)90111-Q	journal	June 1991
Barrier Lyapunov Functions for the control of output-constrained nonlinear systems Tee, Keng Peng; Ge, Shuzhi Sam; Tay, Eng Hock Automatica, Vol. 45, Issue 4 https://doi.org/10.1016/j.automatica.2008.11.017	journal	April 2009
Stabilization with guaranteed safety using Control Lyapunov–Barrier Function Romdlony, Muhammad Zakiyullah; Jayawardhana, Bayu Automatica, Vol. 66 https://doi.org/10.1016/j.automatica.2015.12.011	journal	April 2016
Control Lyapunov-Barrier function-based model predictive control of nonlinear systems Wu, Zhe; Albalawi, Fahad; Zhang, Zhihao Automatica, Vol. 109 https://doi.org/10.1016/j.automatica.2019.108508	journal	November 2019
Handling bounded and unbounded unsafe sets in Control Lyapunov-Barrier function-based model predictive control of nonlinear processes Wu, Zhe; Christofides, Panagiotis D. Chemical Engineering Research and Design, Vol. 143 https://doi.org/10.1016/j.cherd.2019.01.002	journal	March 2019
Process operational safety using model predictive control based on a process Safeness Index Albalawi, Fahad; Durand, Helen; Christofides, Panagiotis D. Computers & Chemical Engineering, Vol. 104 https://doi.org/10.1016/j.compchemeng.2017.04.010	journal	September 2017
Control Lyapunov-Barrier function-based predictive control of nonlinear processes using machine learning modeling Wu, Zhe; Christofides, Panagiotis D. Computers & Chemical Engineering, Vol. 134 https://doi.org/10.1016/j.compchemeng.2019.106706	journal	March 2020
Robustness of Control Barrier Functions for Safety Critical Control**This work is partially supported by the National Science Foundation Grants 1239055, 1239037 and 1239085. Xu, Xiangru; Tabuada, Paulo; Grizzle, Jessy W. IFAC-PapersOnLine, Vol. 48, Issue 27 https://doi.org/10.1016/j.ifacol.2015.11.152	journal	January 2015
Stabilization of nonlinear systems with state and control constraints using Lyapunov-based predictive control Mhaskar, Prashant; El-Farra, Nael H.; Christofides, Panagiotis D. Systems & Control Letters, Vol. 55, Issue 8 https://doi.org/10.1016/j.sysconle.2005.09.014	journal	August 2006
Barrier Lyapunov functions for the output tracking control of constrained nonlinear switched systems Niu, Ben; Zhao, Jun Systems & Control Letters, Vol. 62, Issue 10 https://doi.org/10.1016/j.sysconle.2013.07.003	journal	October 2013
Model predictive control using neural networks Draeger, A.; Engell, S.; Ranke, H. IEEE Control Systems, Vol. 15, Issue 5, p. 61-66 https://doi.org/10.1109/37.466261	journal	October 1995
High-order neural network structures for identification of dynamical systems Kosmatopoulos, E. B.; Polycarpou, M. M.; Christodoulou, M. A. IEEE Transactions on Neural Networks, Vol. 6, Issue 2 https://doi.org/10.1109/72.363477	journal	March 1995
Control barrier function based quadratic programs with application to adaptive cruise control Ames, Aaron D.; Grizzle, Jessy W.; Tabuada, Paulo 2014 IEEE 53rd Annual Conference on Decision and Control (CDC), 53rd IEEE Conference on Decision and Control https://doi.org/10.1109/CDC.2014.7040372	conference	December 2014
A sampling approach to finding Lyapunov functions for nonlinear discrete-time systems Bobiti, Ruxandra; Lazar, Mircea 2016 European Control Conference (ECC) https://doi.org/10.1109/ECC.2016.7810344	conference	June 2016
Lyapunov-Based Model Predictive Control of Nonlinear Systems Subject to Data Losses Munoz de la Pena, David; Christofides, Panagiotis D. IEEE Transactions on Automatic Control, Vol. 53, Issue 9 https://doi.org/10.1109/TAC.2008.929401	journal	October 2008
Learning-Based Model Predictive Control: Toward Safe Learning in Control Hewing, Lukas; Wabersich, Kim P.; Menner, Marcel Annual Review of Control, Robotics, and Autonomous Systems, Vol. 3, Issue 1 https://doi.org/10.1146/annurev-control-090419-075625	journal	May 2020
Combining Control Lyapunov and Barrier Functions for constrained stabilization of nonlinear systems Jankovic, Mrdjan 2017 American Control Conference (ACC) https://doi.org/10.23919/ACC.2017.7963232	conference	May 2017
Constructive Safety Using Control Barrier Functions Wieland, Peter; Allgöwer, Frank IFAC Proceedings Volumes, Vol. 40, Issue 12 https://doi.org/10.3182/20070822-3-ZA-2920.00076	journal	January 2007
Recurrent Neural Network-Based Model Predictive Control for Continuous Pharmaceutical Manufacturing Wong, Wee; Chee, Ewan; Li, Jiali Mathematics, Vol. 6, Issue 11 https://doi.org/10.3390/math6110242	journal	November 2018

Similar Records

Machine Learning-Based Model Predictive Control of Two-Time-Scale Systems

Journal Article · Wed Sep 06 00:00:00 EDT 2023 · Mathematics · OSTI ID:2418521

Dynamic process modeling with recurrent neural networks

Journal Article · Fri Oct 01 00:00:00 EDT 1993 · AIChE Journal (American Institute of Chemical Engineers); (United States) · OSTI ID:5996396

Nonlinear Multiple Models Adaptive Secondary Voltage Control of Microgrids

Journal Article · Fri Sep 10 00:00:00 EDT 2021 · IEEE Transactions on Smart Grid · OSTI ID:1765350

Machine‐learning‐based construction of barrier functions and models for safe model predictive control

Citation Formats

References (21)

Similar Records

Related Subjects