Distributed Model Predictive Control of Linear Systems with Coupled Constraints Based on Collective Neurodynamic Optimization

被引：1

作者：

Yan, Zheng ^{[1
]}

Lu, Jie ^{[1
]}

Zhang, Guangquan ^{[1
]}

机构：

[1] Univ Technol Sydney, Fac Engn & IT, Ctr Artificial Intelligence, Ultimo, NSW 2007, Australia

来源：

AI 2018: ADVANCES IN ARTIFICIAL INTELLIGENCE | 2018年 / 11320卷

基金：

澳大利亚研究理事会;

关键词：

Collective neurodynamic optimization; Recurrent neural networks; Distributed optimization; Model predictive control; RECURRENT NEURAL-NETWORK; LIMITING ACTIVATION FUNCTION;

D O I：

10.1007/978-3-030-03991-2_31

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Distributed model predictive control explores an array of local predictive controllers that synthesize the control of subsystems independently yet they communicate to efficiently cooperate in achieving the closed-loop control performance. Distributed model predictive control problems naturally result in sequential distributed optimization problems that require real-time solution. This paper presents a collective neurodynamic approach to design and implement the distributed model predictive control of linear systems in the presence of globally coupled constraints. For each subsystem, a neurodynamic model minimizes its cost function using local information only. According to the communication topology of the network, neurodynamic models share information to their neighbours to reach consensus on the optimal control actions to be carried out. The collective neurodynamic models are proven to guarantee the global optimality of the model predictive control system.

引用

页码：318 / 328

页数：11

共 22 条

[11] Model Predictive Control of an Asymmetric Flying Capacitor Converter [J].

Lezana, Pablo ;

Aguilera, Ricardo ;

Quevedo, Daniel E. .

IEEE TRANSACTIONS ON INDUSTRIAL ELECTRONICS, 2009, 56 (06) :1839-1846

[12] A one-layer recurrent neural network for constrained nonsmooth invex optimization [J].

Li, Guocheng ;

Yan, Zheng ;

Wang, Jun .

NEURAL NETWORKS, 2014, 50 :79-89

[13] A one-layer recurrent neural network with a discontinuous hard-limiting activation function for quadratic programming [J].

Liu, Qingshan ;

Wang, Jun .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2008, 19 (04) :558-570

[14] A One-Layer Recurrent Neural Network for Constrained Nonsmooth Optimization [J].

Liu, Qingshan ;

Wang, Jun .

IEEE TRANSACTIONS ON SYSTEMS MAN AND CYBERNETICS PART B-CYBERNETICS, 2011, 41 (05) :1323-1333

[15] Finite-Time Convergent Recurrent Neural Network with a Hard-Limiting Activation Function for Constrained Optimization with Piecewise-Linear Objective Functions [J].

Liu, Qingshan ;

Wang, Jun .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2011, 22 (04) :601-613

[16] Consensus and cooperation in networked multi-agent systems [J].

Olfati-Saber, Reza ;

Fax, J. Alex ;

Murray, Richard M. .

PROCEEDINGS OF THE IEEE, 2007, 95 (01) :215-233

[17] Robust distributed model predictive control [J].

Richards, A. ;

How, J. P. .

INTERNATIONAL JOURNAL OF CONTROL, 2007, 80 (09) :1517-1531

[18] Cooperative distributed model predictive control [J].

Stewart, Brett T. ;

Venkat, Aswin N. ;

Rawlings, James B. ;

Wright, Stephen J. ;

Pannocchia, Gabriele .

SYSTEMS & CONTROL LETTERS, 2010, 59 (08) :460-469

[19] A novel recurrent neural network for solving nonlinear optimization problems with inequality constraints [J].

Xia, Youshen ;

Feng, Gang ;

Wang, Jun .

IEEE TRANSACTIONS ON NEURAL NETWORKS, 2008, 19 (08) :1340-1353

[20] Incorporating state estimation into model predictive control and its application to network traffic control [J].

Yan, J ;

Bitmead, RR .

AUTOMATICA, 2005, 41 (04) :595-604

← 1 2 3 →