Self Organized Feature Maps for Monitoring and Knowledge Aquisition of a Chemical Process

ICANN ’93 ◽  
1993 ◽  
pp. 864-867 ◽  
Author(s):  
Alfred Ultsch
Keyword(s):  
Chemical Process ◽  
Feature Maps ◽  
Self Organized ◽  
Knowledge Aquisition

Topological feature maps with self-organized lateral connections: a population-coded, one-layer model of associative memory

1994 ◽  
Vol 72 (2) ◽  
pp. 103-117 ◽  
Author(s):  
Claus Hillermeier ◽  
Niels Kunstmann ◽  
Bernhard Rabus ◽  
Paul Tavan
Keyword(s):  
Associative Memory ◽  
Topological Feature ◽  
Layer Model ◽  
Feature Maps ◽  
Self Organized

scholarly journals A Hybrid of Self Organized Feature Maps and Parallel Genetic Algorithms for Uncertain Knowledge

2012 ◽  
Vol 60 (6) ◽  
pp. 23-31 ◽  
Author(s):  
Mona Gamal ◽  
Ahmed Abo El-Fatoh ◽  
Shereef Barakat ◽  
Elsayed Radwan
Keyword(s):  
Genetic Algorithms ◽  
Feature Maps ◽  
Parallel Genetic Algorithms ◽  
Self Organized ◽  
Uncertain Knowledge

Phase Transitions in Self-Organized Feature Maps

ICANN ’93 ◽  
1993 ◽  
pp. 597-600
Author(s):  
R. Der ◽  
M. Herrmann
Keyword(s):  
Phase Transitions ◽  
Feature Maps ◽  
Self Organized

Topological feature maps with self-organized lateral connections: a population-coded, one-layer model of associative memory

1994 ◽  
Vol 72 (2) ◽  
pp. 103-117 ◽  
Author(s):  
Claus Hillermeier ◽  
Niels Kunstmann ◽  
Bernhard Rabus ◽  
Paul Tavan
Keyword(s):  
Associative Memory ◽  
Topological Feature ◽  
Layer Model ◽  
Feature Maps ◽  
Self Organized

Self-organized feature maps and information theory

1997 ◽  
Vol 8 (2) ◽  
pp. 215-227 ◽  
Author(s):  
Klaus Holthausen ◽  
Olaf Breidbach
Keyword(s):  
Information Theory ◽  
Feature Maps ◽  
Self Organized

Faithful Representation of Separable Distributions

1997 ◽  
Vol 9 (6) ◽  
pp. 1305-1320 ◽  
Author(s):  
Juan K. Lin ◽  
David G. Grier ◽  
Jack D. Cowan
Keyword(s):  
Geometric Approach ◽  
Data Representation ◽  
Equal Mass ◽  
Distribution Functions ◽  
Faithful Representation ◽  
Feature Maps ◽  
Information Theoretic ◽  
Input Distribution ◽  
Self Organized ◽  
Geometry Problem

A geometric approach to data representation incorporating information theoretic ideas is presented. The task of finding a faithful representation, where the input distribution is evenly partitioned into regions of equal mass, is addressed. For input consisting of mixtures of statistically independent sources, we treat independent component analysis (ICA) as a computational geometry problem. First, we consider the separation of sources with sharply peaked distribution functions, where the ICA problem becomes that of finding high-density directions in the input distribution. Second, we consider the more general problem for arbitrary input distributions, where ICA is transformed into the task of finding an aligned equipartition. By modifying the Kohonen self-organized feature maps, we arrive at neural networks with local interactions that optimize coding while simultaneously performing source separation. The local nature of our approach results in networks with nonlinear ICA capabilities.

Multiple Fault Identification Using Vibration Signal Analysis and Artificial Intelligence Methods

2013 ◽  
Vol 430 ◽  
pp. 63-69 ◽  
Author(s):  
Ninoslav Zuber ◽  
Dragan Cvetkovic ◽  
Rusmir Bajrić
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Power Transmission ◽  
Vibration Signal ◽  
Single Type ◽  
Feature Maps ◽  
Multiple Faults ◽  
Self Organized ◽  
Artificial Neural ◽  
Vibration Signal Analysis

Paper addresses the implementation of feature based artificial neural networks and self-organized feature maps with the vibration analysis for the purpose of automated faults identification in rotating machinery. Unlike most of the research in this field, where a single type of fault has been treated, the research conducted in this paper deals with rotating machines with multiple faults. Combination of different roller elements bearing faults and different gearbox faults is analyzed. Experimental work has been conducted on a specially designed test rig. Frequency and time domain vibration features are used as inputs to fault classifiers. A complete set of proposed vibration features are used as inputs for self-organized feature maps and based on the results they are used as inputs for supervised artificial neural networks. The achieved results show that proposed set of vibration features enables reliable identification of developing bearing and gear faults in geared power transmission systems.

Sign in / Sign up

Export Citation Format

Share Document