Application of neural network prediction in process control

Xiaodong Xu; Mengyin Fu; Zhihong Deng; Wenyan Xu

doi:10.1109/ANTHOLOGY.2013.6784757

Application of neural network prediction in process control

Xiaodong Xu, Mengyin Fu, Zhihong Deng, Wenyan Xu

School of Automation

Research output: Contribution to conference › Paper › peer-review

Abstract

In process control, there exists obvious time-delay phenomenon, which brings phase-delay to the system and reduces the stability of the system. There also exists unknown time-varying state which has no mathematical model in the system. PID controllers can't control delay and nonlinear state effectively, although they are applied to the control system design widely. In this paper, neural network characteristic of adaptive learning was used to build accurate model and compensate for time-delay and nonlinearity of the system. Forecast based on the feedback control algorithm was made to solve the problems about time delay and nonlinearity of the water tank level control. Accurate control was achieved.

Original language	English
DOIs	https://doi.org/10.1109/ANTHOLOGY.2013.6784757
Publication status	Published - 2014
Event	2013 IEEE Conference Anthology, ANTHOLOGY 2013 - , China Duration: 1 Jan 2013 → 8 Jan 2013

Conference

Conference	2013 IEEE Conference Anthology, ANTHOLOGY 2013
Country/Territory	China
Period	1/01/13 → 8/01/13

Keywords

Forecast
Neural Network
Process-Control
Time-Delay

Access to Document

10.1109/ANTHOLOGY.2013.6784757

Cite this

Xu, X., Fu, M., Deng, Z., & Xu, W. (2014). Application of neural network prediction in process control. Paper presented at 2013 IEEE Conference Anthology, ANTHOLOGY 2013, China. https://doi.org/10.1109/ANTHOLOGY.2013.6784757

@conference{03f3c0e7f42d4073b210a69be018fe48,

title = "Application of neural network prediction in process control",

abstract = "In process control, there exists obvious time-delay phenomenon, which brings phase-delay to the system and reduces the stability of the system. There also exists unknown time-varying state which has no mathematical model in the system. PID controllers can't control delay and nonlinear state effectively, although they are applied to the control system design widely. In this paper, neural network characteristic of adaptive learning was used to build accurate model and compensate for time-delay and nonlinearity of the system. Forecast based on the feedback control algorithm was made to solve the problems about time delay and nonlinearity of the water tank level control. Accurate control was achieved.",

keywords = "Forecast, Neural Network, Process-Control, Time-Delay",

author = "Xiaodong Xu and Mengyin Fu and Zhihong Deng and Wenyan Xu",

year = "2014",

doi = "10.1109/ANTHOLOGY.2013.6784757",

language = "English",

note = "2013 IEEE Conference Anthology, ANTHOLOGY 2013 ; Conference date: 01-01-2013 Through 08-01-2013",

}

TY - CONF

T1 - Application of neural network prediction in process control

AU - Xu, Xiaodong

AU - Fu, Mengyin

AU - Deng, Zhihong

AU - Xu, Wenyan

PY - 2014

Y1 - 2014

N2 - In process control, there exists obvious time-delay phenomenon, which brings phase-delay to the system and reduces the stability of the system. There also exists unknown time-varying state which has no mathematical model in the system. PID controllers can't control delay and nonlinear state effectively, although they are applied to the control system design widely. In this paper, neural network characteristic of adaptive learning was used to build accurate model and compensate for time-delay and nonlinearity of the system. Forecast based on the feedback control algorithm was made to solve the problems about time delay and nonlinearity of the water tank level control. Accurate control was achieved.

AB - In process control, there exists obvious time-delay phenomenon, which brings phase-delay to the system and reduces the stability of the system. There also exists unknown time-varying state which has no mathematical model in the system. PID controllers can't control delay and nonlinear state effectively, although they are applied to the control system design widely. In this paper, neural network characteristic of adaptive learning was used to build accurate model and compensate for time-delay and nonlinearity of the system. Forecast based on the feedback control algorithm was made to solve the problems about time delay and nonlinearity of the water tank level control. Accurate control was achieved.

KW - Forecast

KW - Neural Network

KW - Process-Control

KW - Time-Delay

UR - http://www.scopus.com/inward/record.url?scp=84899413329&partnerID=8YFLogxK

U2 - 10.1109/ANTHOLOGY.2013.6784757

DO - 10.1109/ANTHOLOGY.2013.6784757

M3 - Paper

AN - SCOPUS:84899413329

T2 - 2013 IEEE Conference Anthology, ANTHOLOGY 2013

Y2 - 1 January 2013 through 8 January 2013

ER -

Application of neural network prediction in process control

Abstract

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this