A Zonotope-based Big Data-driven Predictive Control Approach for Nonlinear Processes

Shuangyu Han; Yitao Yan; Jie Bao; Biao Huang

doi:10.1016/j.ifacol.2025.07.117

A Zonotope-based Big Data-driven Predictive Control Approach for Nonlinear Processes

Shuangyu Han^*
, Yitao Yan^*
, Jie Bao^*
, Biao Huang

^*Corresponding author for this work

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

1 Citation (Scopus)

Abstract

A zonotope-based big data-driven predictive control (BDPC) approach is developed to partition the nonlinear process behaviour (represented by an input-output trajectory set) into multiple linear sub-behaviours using a two-step hierarchical clustering: Euclidean distance-based clustering and linear subspace distance-based clustering. By approximating every linear sub-behaviour as a zonotope, a data-driven interpolation is developed based on the convex combination of zonotopes. During online control, a BDPC controller is designed by determining an interpolated zonotope where its centre trajectory is closest to the online trajectory and computing the control action subject to an optimisation problem. The proposed BDPC approach is illustrated using a case study on controlling an aluminium smelting process.

Original language	English
Pages (from-to)	31-36
Number of pages	6
Journal	IFAC-PapersOnLine
Volume	59
Issue number	6
DOIs	https://doi.org/10.1016/j.ifacol.2025.07.117
Publication status	Published - 1 Jun 2025
Externally published	Yes
Event	14th IFAC Symposium on Dynamics and Control of Process Systems, including Biosystems, DYCOPS 2025 - Bratislava, Slovakia Duration: 16 Jun 2025 → 19 Jun 2025

Keywords

Behavioural systems theory
Big data
Clustering
Data-driven predictive control
Zonotope

Access to Document

10.1016/j.ifacol.2025.07.117

Cite this

@article{29913fd86051405cb2126034dc27f19d,

title = "A Zonotope-based Big Data-driven Predictive Control Approach for Nonlinear Processes",

abstract = "A zonotope-based big data-driven predictive control (BDPC) approach is developed to partition the nonlinear process behaviour (represented by an input-output trajectory set) into multiple linear sub-behaviours using a two-step hierarchical clustering: Euclidean distance-based clustering and linear subspace distance-based clustering. By approximating every linear sub-behaviour as a zonotope, a data-driven interpolation is developed based on the convex combination of zonotopes. During online control, a BDPC controller is designed by determining an interpolated zonotope where its centre trajectory is closest to the online trajectory and computing the control action subject to an optimisation problem. The proposed BDPC approach is illustrated using a case study on controlling an aluminium smelting process.",

keywords = "Behavioural systems theory, Big data, Clustering, Data-driven predictive control, Zonotope",

author = "Shuangyu Han and Yitao Yan and Jie Bao and Biao Huang",

note = "Publisher Copyright: Copyright {\textcopyright} 2025 The Authors.; 14th IFAC Symposium on Dynamics and Control of Process Systems, including Biosystems, DYCOPS 2025 ; Conference date: 16-06-2025 Through 19-06-2025",

year = "2025",

month = jun,

day = "1",

doi = "10.1016/j.ifacol.2025.07.117",

language = "English",

volume = "59",

pages = "31--36",

journal = "IFAC-PapersOnLine",

issn = "2405-8971",

number = "6",

}

TY - JOUR

T1 - A Zonotope-based Big Data-driven Predictive Control Approach for Nonlinear Processes

AU - Han, Shuangyu

AU - Yan, Yitao

AU - Bao, Jie

AU - Huang, Biao

PY - 2025/6/1

Y1 - 2025/6/1

N2 - A zonotope-based big data-driven predictive control (BDPC) approach is developed to partition the nonlinear process behaviour (represented by an input-output trajectory set) into multiple linear sub-behaviours using a two-step hierarchical clustering: Euclidean distance-based clustering and linear subspace distance-based clustering. By approximating every linear sub-behaviour as a zonotope, a data-driven interpolation is developed based on the convex combination of zonotopes. During online control, a BDPC controller is designed by determining an interpolated zonotope where its centre trajectory is closest to the online trajectory and computing the control action subject to an optimisation problem. The proposed BDPC approach is illustrated using a case study on controlling an aluminium smelting process.

AB - A zonotope-based big data-driven predictive control (BDPC) approach is developed to partition the nonlinear process behaviour (represented by an input-output trajectory set) into multiple linear sub-behaviours using a two-step hierarchical clustering: Euclidean distance-based clustering and linear subspace distance-based clustering. By approximating every linear sub-behaviour as a zonotope, a data-driven interpolation is developed based on the convex combination of zonotopes. During online control, a BDPC controller is designed by determining an interpolated zonotope where its centre trajectory is closest to the online trajectory and computing the control action subject to an optimisation problem. The proposed BDPC approach is illustrated using a case study on controlling an aluminium smelting process.

KW - Behavioural systems theory

KW - Big data

KW - Clustering

KW - Data-driven predictive control

KW - Zonotope

UR - https://www.scopus.com/pages/publications/105013966221

U2 - 10.1016/j.ifacol.2025.07.117

DO - 10.1016/j.ifacol.2025.07.117

M3 - Conference article

AN - SCOPUS:105013966221

SN - 2405-8971

VL - 59

SP - 31

EP - 36

JO - IFAC-PapersOnLine

JF - IFAC-PapersOnLine

IS - 6

T2 - 14th IFAC Symposium on Dynamics and Control of Process Systems, including Biosystems, DYCOPS 2025

Y2 - 16 June 2025 through 19 June 2025

ER -

A Zonotope-based Big Data-driven Predictive Control Approach for Nonlinear Processes

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this