Fault diagnosis of a centrifugal pump using MLP-GABP and SVM with CWT

Maamar Ali Saud Al Tobi, Geraint Bevan, Peter Wallace, David Harrison, K. P. Ramachandran

Research output: Contribution to journalArticle

2 Downloads (Pure)

Abstract

This paper presents a comparative study of Multilayer Feedforward Perceptron Neural Network which is trained with Back Propagation (MLP-BP) and also using hybrid training using Genetic Algorithm (GA) (MLP-GABP), and Support Vector Machine (SVM) classifiers to classify the fault conditions of a centrifugal pump. Continuous Wavelet Transform (CWT) with three different wavelet functions (Morlet, db8 and rbio1.5) is used to extract the features. GA is also used to optimize the number of hidden layers and neurons of MLP. From the results obtained, MLP-BP has shown better performance than MLP-GABP and SVM using a lower number of features. SVM has performed better using polynomial kernel function using a smaller number of features and parameters. A centrifugal pump test rig has been specifically designed and built for this work in order to create the desired faults.
Original languageEnglish
Pages (from-to)854-861
Number of pages8
JournalEngineering Science and Technology, an International Journal
Volume22
Issue number3
Early online date22 Jan 2019
DOIs
Publication statusPublished - Jun 2019

Fingerprint

Centrifugal pumps
Wavelet transforms
Failure analysis
Support vector machines
Genetic algorithms
Neural networks
Backpropagation
Neurons
Multilayers
Classifiers
Polynomials

Keywords

  • Genetic Algorithm (GA)
  • Multilayer Feedforward Perceptron (MLP)
  • Support Vector Machine (SVM)
  • Continuous Wavelet Transform (CWT)

Cite this

@article{0fbec0fb20a349bab86e16a66a51ccb5,
title = "Fault diagnosis of a centrifugal pump using MLP-GABP and SVM with CWT",
abstract = "This paper presents a comparative study of Multilayer Feedforward Perceptron Neural Network which is trained with Back Propagation (MLP-BP) and also using hybrid training using Genetic Algorithm (GA) (MLP-GABP), and Support Vector Machine (SVM) classifiers to classify the fault conditions of a centrifugal pump. Continuous Wavelet Transform (CWT) with three different wavelet functions (Morlet, db8 and rbio1.5) is used to extract the features. GA is also used to optimize the number of hidden layers and neurons of MLP. From the results obtained, MLP-BP has shown better performance than MLP-GABP and SVM using a lower number of features. SVM has performed better using polynomial kernel function using a smaller number of features and parameters. A centrifugal pump test rig has been specifically designed and built for this work in order to create the desired faults.",
keywords = "Genetic Algorithm (GA), Multilayer Feedforward Perceptron (MLP), Support Vector Machine (SVM), Continuous Wavelet Transform (CWT)",
author = "{Ali Saud Al Tobi}, Maamar and Geraint Bevan and Peter Wallace and David Harrison and Ramachandran, {K. P.}",
note = "Acceptance from webpage Fwd to 'for validation' in July 2019 by author. OA article",
year = "2019",
month = "6",
doi = "10.1016/j.jestch.2019.01.005",
language = "English",
volume = "22",
pages = "854--861",
journal = "Engineering Science and Technology, an International Journal",
issn = "2215-0986",
publisher = "Elsevier B.V.",
number = "3",

}

Fault diagnosis of a centrifugal pump using MLP-GABP and SVM with CWT. / Ali Saud Al Tobi, Maamar; Bevan, Geraint; Wallace, Peter; Harrison, David; Ramachandran, K. P.

In: Engineering Science and Technology, an International Journal, Vol. 22, No. 3, 06.2019, p. 854-861.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Fault diagnosis of a centrifugal pump using MLP-GABP and SVM with CWT

AU - Ali Saud Al Tobi, Maamar

AU - Bevan, Geraint

AU - Wallace, Peter

AU - Harrison, David

AU - Ramachandran, K. P.

N1 - Acceptance from webpage Fwd to 'for validation' in July 2019 by author. OA article

PY - 2019/6

Y1 - 2019/6

N2 - This paper presents a comparative study of Multilayer Feedforward Perceptron Neural Network which is trained with Back Propagation (MLP-BP) and also using hybrid training using Genetic Algorithm (GA) (MLP-GABP), and Support Vector Machine (SVM) classifiers to classify the fault conditions of a centrifugal pump. Continuous Wavelet Transform (CWT) with three different wavelet functions (Morlet, db8 and rbio1.5) is used to extract the features. GA is also used to optimize the number of hidden layers and neurons of MLP. From the results obtained, MLP-BP has shown better performance than MLP-GABP and SVM using a lower number of features. SVM has performed better using polynomial kernel function using a smaller number of features and parameters. A centrifugal pump test rig has been specifically designed and built for this work in order to create the desired faults.

AB - This paper presents a comparative study of Multilayer Feedforward Perceptron Neural Network which is trained with Back Propagation (MLP-BP) and also using hybrid training using Genetic Algorithm (GA) (MLP-GABP), and Support Vector Machine (SVM) classifiers to classify the fault conditions of a centrifugal pump. Continuous Wavelet Transform (CWT) with three different wavelet functions (Morlet, db8 and rbio1.5) is used to extract the features. GA is also used to optimize the number of hidden layers and neurons of MLP. From the results obtained, MLP-BP has shown better performance than MLP-GABP and SVM using a lower number of features. SVM has performed better using polynomial kernel function using a smaller number of features and parameters. A centrifugal pump test rig has been specifically designed and built for this work in order to create the desired faults.

KW - Genetic Algorithm (GA)

KW - Multilayer Feedforward Perceptron (MLP)

KW - Support Vector Machine (SVM)

KW - Continuous Wavelet Transform (CWT)

U2 - 10.1016/j.jestch.2019.01.005

DO - 10.1016/j.jestch.2019.01.005

M3 - Article

VL - 22

SP - 854

EP - 861

JO - Engineering Science and Technology, an International Journal

JF - Engineering Science and Technology, an International Journal

SN - 2215-0986

IS - 3

ER -