Detecting unexperienced events via analysis of error propagation in a neuro-fuzzy signal processing system

R. Kozma, S. Sato, M. Sakuma, M. Kitamura

Research output: Contribution to conferencePaperpeer-review

1 Citation (Scopus)

Abstract

The performance of system identification methods based on multi-layer artificial neural network (ANN) models is analyzed. The state of the system is monitored by multiple ANNs which act as individual neuro agents. An algorithm based on fuzzy logics is applied to combine information from ANNs. In this paper, we elaborate a method which can make a distinction between the occurrence of unexperienced events and any inconsistency in the judgments of agents caused by statistical uncertainties in the actual data. The results are illustrated by analyzing signals of numerical experiments and also actual measurements in a nuclear reactor.

Original languageEnglish
Pages241-246
Number of pages6
Publication statusPublished - 1994
EventProceedings of the Artificial Neural Networks in Engineering Conference (ANNIE'94) - St. Louis, MO, USA
Duration: 1994 Nov 131994 Nov 16

Conference

ConferenceProceedings of the Artificial Neural Networks in Engineering Conference (ANNIE'94)
CitySt. Louis, MO, USA
Period94/11/1394/11/16

ASJC Scopus subject areas

  • Engineering(all)

Fingerprint Dive into the research topics of 'Detecting unexperienced events via analysis of error propagation in a neuro-fuzzy signal processing system'. Together they form a unique fingerprint.

Cite this