Bushing monitoring using MLP and RBF

Sizwe M. Dhlamini; Tshilidzi Marwala

Bushing monitoring using MLP and RBF

Sizwe M. Dhlamini, Tshilidzi Marwala

Faculty of Engineering and the Built Environment

Research output: Contribution to conference › Paper › peer-review

10 Citations (Scopus)

Abstract

This paper examines the use of artificial neural networks (ANN) for monitoring bushings. The first ANN uses multiplayer perceptron (MLP) while the second uses radial basis activation functions (RBF). In this approach, a decision can be taken to remove or leave a bushing in service based on analysis of bushing parameters using RBF and MLP. The results show that the RBF converges to a solution faster than the MLP. Furthermore, the MLP is found to be the best tool of the two for analyzing large amounts of non-parametric non-linear data.

Original language	English
Pages	613-618
Number of pages	6
Publication status	Published - 2004
Event	2004 IEEE AFRICON: 7th AFRICON Conference in Africa: Technology Innovation - Gaborone, Botswana Duration: 15 Sept 2004 → 17 Sept 2004

Conference

Conference	2004 IEEE AFRICON: 7th AFRICON Conference in Africa: Technology Innovation
Country/Territory	Botswana
City	Gaborone
Period	15/09/04 → 17/09/04

Keywords

Bushing
Diagnosis
Dissolved gas analysis
Multi layer perceptron
Radial basis

ASJC Scopus subject areas

Computer Science Applications
Electrical and Electronic Engineering

Cite this

@conference{1ed405c7da5340a19f137150dedcd2a2,

title = "Bushing monitoring using MLP and RBF",

abstract = "This paper examines the use of artificial neural networks (ANN) for monitoring bushings. The first ANN uses multiplayer perceptron (MLP) while the second uses radial basis activation functions (RBF). In this approach, a decision can be taken to remove or leave a bushing in service based on analysis of bushing parameters using RBF and MLP. The results show that the RBF converges to a solution faster than the MLP. Furthermore, the MLP is found to be the best tool of the two for analyzing large amounts of non-parametric non-linear data.",

keywords = "Bushing, Diagnosis, Dissolved gas analysis, Multi layer perceptron, Radial basis",

author = "Dhlamini, {Sizwe M.} and Tshilidzi Marwala",

year = "2004",

language = "English",

pages = "613--618",

note = "2004 IEEE AFRICON: 7th AFRICON Conference in Africa: Technology Innovation ; Conference date: 15-09-2004 Through 17-09-2004",

}

TY - CONF

T1 - Bushing monitoring using MLP and RBF

AU - Dhlamini, Sizwe M.

AU - Marwala, Tshilidzi

PY - 2004

Y1 - 2004

N2 - This paper examines the use of artificial neural networks (ANN) for monitoring bushings. The first ANN uses multiplayer perceptron (MLP) while the second uses radial basis activation functions (RBF). In this approach, a decision can be taken to remove or leave a bushing in service based on analysis of bushing parameters using RBF and MLP. The results show that the RBF converges to a solution faster than the MLP. Furthermore, the MLP is found to be the best tool of the two for analyzing large amounts of non-parametric non-linear data.

AB - This paper examines the use of artificial neural networks (ANN) for monitoring bushings. The first ANN uses multiplayer perceptron (MLP) while the second uses radial basis activation functions (RBF). In this approach, a decision can be taken to remove or leave a bushing in service based on analysis of bushing parameters using RBF and MLP. The results show that the RBF converges to a solution faster than the MLP. Furthermore, the MLP is found to be the best tool of the two for analyzing large amounts of non-parametric non-linear data.

KW - Bushing

KW - Diagnosis

KW - Dissolved gas analysis

KW - Multi layer perceptron

KW - Radial basis

UR - http://www.scopus.com/inward/record.url?scp=20744432894&partnerID=8YFLogxK

M3 - Paper

AN - SCOPUS:20744432894

SP - 613

EP - 618

T2 - 2004 IEEE AFRICON: 7th AFRICON Conference in Africa: Technology Innovation

Y2 - 15 September 2004 through 17 September 2004

ER -

Bushing monitoring using MLP and RBF

Abstract

Conference

Keywords

ASJC Scopus subject areas

Other files and links

Fingerprint

Cite this