Linguistic information feed-forward-based dynamical fuzzy systems.II. Evaluation

Author

Gao, X.Z. ; Ovaska, S.J.

Author_Institution

Inst. of Intelligent Power Electron., Helsinki Univ. of Technol., Espoo, Finland

fYear

2003

fDate

23-25 June 2003

Firstpage

81

Lastpage

84

Abstract

Part I of this paper introduces the linguistic information feed-forward-based dynamical fuzzy system as well as an adaptive learning algorithm for the feed-forward parameters. In Part II, we are going to evaluate the effectiveness of the proposed LIFFDFS and its learning algorithm using the well-known Box-Jenkins time series (1976). The Box-Jenkins time series are generated from a gas furnace, whose input u(k) is the gas flow rate, and output y(k) is the CO₂ concentration. the LIFFDFS is employed in our simulation to predict y(k), based on the past values of both u(k) and y(k).

Keywords

computational linguistics; dynamic programming; feedforward neural nets; fuzzy control; fuzzy systems; learning (artificial intelligence); time series; Box-Jenkins time series; LIFFDFS; adaptive learning algorithm; carbon dioxide concentration; dynamical system; feed-forward parameter; feed-forward-based system; fuzzy systems; gas flow rate; gas furnace; linguistic information; single-step ahead prediction; system evaluation; Feedforward systems; Fluid flow; Furnaces; Fuzzy sets; Fuzzy systems; History; Power electronics;

fLanguage

English

Publisher

ieee

Conference_Titel

Soft Computing in Industrial Applications, 2003. SMCia/03. Proceedings of the 2003 IEEE International Workshop on

Print_ISBN

0-7803-7855-5

Type

conf

DOI

10.1109/SMCIA.2003.1231348

Filename

1231348