Skip to main content
U.S. Department of Energy
Office of Scientific and Technical Information

Fuzzy logic -- artificial neural networks integration for transient identification

Conference ·
OSTI ID:10102417
;  [1];  [1]
  1. Tennessee Univ., Knoxville, TN (United States). Dept. of Nuclear Engineering

A methodology is presented that integrates pretrained artificial neural networks (ANNs) with rule-based fuzzy logic systems, for the purpose of distinguishing different transients in a Nuclear Power Plant (NPP). In general this approach appears to provide timely, concise and task specific information about the status of a system under consideration. The pretrained neural network typifies different transient scenarios and derives membership functions which independently represent individual transients. The overall system successfully performs transient identification, in a time span faster or at least comparable to that of transient development. In order to examine the proposed methodology simulated accidents are used. The results obtained demonstrate the excellent noise tolerance of ANNs and suggest a new approach for transient identification.

Research Organization:
Tennessee Univ., Knoxville, TN (United States). Dept. of Nuclear Engineering
Sponsoring Organization:
USDOE, Washington, DC (United States)
DOE Contract Number:
FG07-88ER12824
OSTI ID:
10102417
Report Number(s):
CONF-9111215--1; ON: DE93002232
Country of Publication:
United States
Language:
English