A novel NN-fuzzy-SPC feedback control system

Author

Wang, LiRen ; Rowlands, Hefin

Author_Institution

Dept. of Eng., Univ. of Wales Coll. Newport, UK

fYear

2001

fDate

15-18 Oct. 2001

Firstpage

417

Abstract

It is a difficult challenge to develop a feedback control system for statistical process control (SPC) because there is no effective method that can be used to accurately calculate the magnitude of the feedback control actions in traditional SPC. Suitable feedback adjustments are normally generated from the experiences of process engineers. In this paper, fuzzy logic and neural network (NN) techniques are used to develop a NN-fuzzy-SPC control system. The fuzzy inference is used to generate the numeric feedback control actions and the neural network optimises the fuzzy membership functions in order to increase the control accuracy. A combined forecaster with EWMA chart and digital filtering is also developed for the NN-fuzzy-SPC system to reduce the control delay. Simulation results show that the NN-fuzzy-SPC system can provide high control accuracy and satisfactorily short control delay.

Keywords

backpropagation; feedback; feedforward neural nets; fuzzy logic; fuzzy neural nets; production control; statistical process control; EWMA chart; backpropagation; digital filtering; feedback; feedforward neural network; fuzzy inference; fuzzy logic; fuzzy membership functions; neural network; statistical process control; Control systems; Delay; Digital filters; Feedback control; Fuzzy control; Fuzzy logic; Fuzzy neural networks; Neural networks; Neurofeedback; Process control;

fLanguage

English

Publisher

ieee

Conference_Titel

Emerging Technologies and Factory Automation, 2001. Proceedings. 2001 8th IEEE International Conference on

Conference_Location

Antibes-Juan les Pins, France

Print_ISBN

0-7803-7241-7

Type

conf

DOI

10.1109/ETFA.2001.996397

Filename

996397