Real-time control of systems with unknown and varying time-delays, using neural networks

被引：11

作者：

Ng, GW

Cook, PA

机构：

[1] DSO Natl Labs, Artificial Intelligence Lab, Singapore 118230, Singapore

[2] Univ Manchester, Inst Sci & Technol, Dept Elect Engn & Elect, Control Syst Ctr, Manchester M60 1QD, Lancs, England

来源：

ENGINEERING APPLICATIONS OF ARTIFICIAL INTELLIGENCE | 1998年 / 11卷 / 03期

关键词：

real-time system control; unknown time-delays; varying time-delays; neural networks;

D O I：

10.1016/S0952-1976(97)00080-8

中图分类号：

TP [自动化技术、计算机技术];

学科分类号：

0812 ;

摘要：

This paper presents evidence that a neural network (NN) controller is more robust than PID and self-tuning controllers for a plant with unknown and varying time-delays. Previous works have shown, through simulation, that the neural-network controller is less sensitive to variations in the time-delay. This paper extended the simulation results to a practical case study. The controllers are compared in the real-time control of a heat process. (C) 1998 Elsevier Science Ltd. All rights reserved.

引用

页码：401 / 409

页数：9

共 16 条

[1]

Cybenko G., 1989, Mathematics of Control, Signals, and Systems, V2, P303, DOI 10.1007/BF02551274

[2]

Hunt Kenneth J., 1995, NEURAL NETWORK ENG D

[3] DIRECT NEURAL MODEL-REFERENCE ADAPTIVE-CONTROL [J].

LIGHTBODY, G ;

IRWIN, GW .

IEE PROCEEDINGS-CONTROL THEORY AND APPLICATIONS, 1995, 142 (01) :31-43

[4]

LIGHTBODY G, 1993, ANN R AUT P, V17, P37

[5]

MILLER WT, 1990, NEURAL NETWORKS CONT

[6] Intelligent control using neural networks [J].

Narendra, Kumpati S. ;

Mukhopadhyay, Snehasis .

IEEE Control Systems Magazine, 1992, 12 (02) :11-18

[7]

NG GW, 1995, UNPUB INT J ADAPT CO

[8]

NG GW, 1996, UKACC INT C CONTROL, V1, P188

[9]

NGUYEN D, 1989, P IEEE INT JOINT C N, V2, P11357

[10] NEURAL-NET COMPUTING AND THE INTELLIGENT CONTROL OF SYSTEMS [J].

PAO, YH ;

PHILLIPS, SM ;

SOBAJIC, DJ .

INTERNATIONAL JOURNAL OF CONTROL, 1992, 56 (02) :263-289

← 1 2 →