0
Technical Brief

Experimental Time-Domain Vibration-Based Fault Diagnosis of Centrifugal Pumps Using Support Vector Machine

[+] Author and Article Information
Janani Shruti Rapur

Department of Mechanical Engineering,
Indian Institute of Technology Guwahati,
Guwahati 781039, India
e-mail: shruti.rapur@gmail.com

Rajiv Tiwari

Professor
Department of Mechanical Engineering,
Indian Institute of Technology Guwahati,
Guwahati 781039, India
e-mail: rtiwari@iitg.ernet.in

1Corresponding author.

Manuscript received May 26, 2016; final manuscript received November 24, 2016; published online June 13, 2017. Assoc. Editor: Mohammad Pourgol-Mohammad.

ASME J. Risk Uncertainty Part B 3(4), 044501 (Jun 13, 2017) (7 pages) Paper No: RISK-16-1087; doi: 10.1115/1.4035440 History: Received May 26, 2016; Revised November 24, 2016

When the hydraulic flow path is incompatible with the physical contours of the centrifugal pump (CP), flow instabilities occur. A prolonged operation in the flow-instability region may result in severe damages of the CP system. Hence, two of the major causes of flow instabilities such as the suction blockage (with five levels of increasing severity) and impeller defects are studied in the present work. Thereafter, an attempt is made to classify these faults and differentiate the physics behind the flow instabilities caused due to them. The tri-axial CP vibration data in time domain are employed for the fault classification. Multidistinct and multicoexisting fault classifications have been performed with different combinations of these faults using support vector machine (SVM) algorithm with radial basis function (RBF) kernel. Prediction results from the experiments and the developed methodology help to segregate the faults into appropriate class, identify the severity of the suction blockage, and substantiate the practical applicability of this study.

FIGURES IN THIS ARTICLE
<>
Copyright © 2017 by ASME
Your Session has timed out. Please sign back in to continue.

References

Figures

Grahic Jump Location
Fig. 1

(Left) Experimental setup and (Right) mounted accelerometers

Grahic Jump Location
Fig. 2

Pump with faulty impeller

Grahic Jump Location
Fig. 3

Recirculation in the pump—a schematic representation

Grahic Jump Location
Fig. 4

Data processing flowchart

Grahic Jump Location
Fig. 5

Speed versus classification accuracy of different features, case D, B1

Grahic Jump Location
Fig. 6

Blockage level versus average classification accuracy over entire speed range, case D

Grahic Jump Location
Fig. 7

Blockage level versus average classification accuracy over entire speed range, case E

Grahic Jump Location
Fig. 8

Individual classification accuracy for each fault at different speeds for ((i)) B1, (ii) B2, (iii) B3, and (iv) B4; with (μ–σ–S) feature, case F

Grahic Jump Location
Fig. 9

Classification accuracy by testing with one fault at a time for different speeds with (μ–σ–S) feature, case G

Tables

Errata

Discussions

Some tools below are only available to our subscribers or users with an online account.

Related Content

Customize your page view by dragging and repositioning the boxes below.

Related Journal Articles
Articles from Part A: Civil Engineering
Related eBook Content
Topic Collections

Sorry! You do not have access to this content. For assistance or to subscribe, please contact us:

  • TELEPHONE: 1-800-843-2763 (Toll-free in the USA)
  • EMAIL: asmedigitalcollection@asme.org
Sign In