Skip to main navigation Skip to main content
  • E-Submission

JKSPE : Journal of the Korean Society for Precision Engineering

OPEN ACCESS
ABOUT
BROWSE ARTICLES
EDITORIAL POLICIES
FOR CONTRIBUTORS
REGULAR

원심펌프 이상 진단을 위한 데이터 수집 및 머신러닝 기반 분석

A Machine Learning-Based Signal Analytics Framework for Diagnosing the Anomalies of Centrifugal Pumps

Journal of the Korean Society for Precision Engineering 2021;38(4):269-277.
Published online: April 1, 2021

1 충남대학교 기계공학부

2 한국산업기술대학교 경영학부

1 School of Mechanical Engineering, Chungnam National University

2 School of Business Administration, Korea Polytechnic University

#E-mail: sh.park@cnu.ac.kr, TEL: +82-42-821-5649
• Received: January 11, 2021   • Revised: February 11, 2021   • Accepted: February 25, 2021

Copyright © The Korean Society for Precision Engineering

This is an Open-Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0) which permits unrestricted non-commercial use, distribution, and reproduction in any medium, provided the original work is properly cited.

  • 8 Views
  • 0 Download
  • 3 Crossref
  • 4 Scopus
prev next

Citations

Citations to this article as recorded by  Crossref logo
  • A Study on 3D Printing Conditions Prediction Model of Bone Plates Using Machine Learning
    Song Yeon Lee, Yong Jeong Huh
    Journal of the Korean Society for Precision Engineering.2022; 39(4): 291.     CrossRef
  • Deep Learning-Based Analysis for Abnormal Diagnosis of Air Compressors
    Mingyu Kang, Yohwan Hyun, Chibum Lee
    Journal of the Korean Society for Precision Engineering.2022; 39(3): 209.     CrossRef
  • A Cost-Aware DNN-Based FDI Technology for Solenoid Pumps
    Suju Kim, Ugochukwu Ejike Akpudo, Jang-Wook Hur
    Electronics.2021; 10(19): 2323.     CrossRef

Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:

Include:

A Machine Learning-Based Signal Analytics Framework for Diagnosing the Anomalies of Centrifugal Pumps
J. Korean Soc. Precis. Eng.. 2021;38(4):269-277.   Published online April 1, 2021
Download Citation

Download a citation file in RIS format that can be imported by all major citation management software, including EndNote, ProCite, RefWorks, and Reference Manager.

Format:
Include:
A Machine Learning-Based Signal Analytics Framework for Diagnosing the Anomalies of Centrifugal Pumps
J. Korean Soc. Precis. Eng.. 2021;38(4):269-277.   Published online April 1, 2021
Close

Figure

  • 0
  • 1
  • 2
  • 3
  • 4
  • 5
  • 6
  • 7
  • 8
  • 9
  • 10
A Machine Learning-Based Signal Analytics Framework for Diagnosing the Anomalies of Centrifugal Pumps
Image Image Image Image Image Image Image Image Image Image Image
Fig. 1 Framework for fault diagnosis of centrifugal pump
Fig. 2 Schematic diagram for piping system
Fig. 3 Schematic diagram for assembly of mechanical seal
Fig. 4 Experimental mechanical seal for (a) Normal and (b) Wear
Fig. 5 Feature extraction of seven signals on time series
Fig. 6 Scatter plot for (a) f1 (Peak to peak), (b) f2 (Mean), (c) f3 (Standard deviation), (d) f4 (Root mean square), (e) f5 (Crest factor), (f) f6 (Skewness), (g) f7 (Kurtosis) from seven sensors data
Fig. 7 Confusion matrix
Fig. 8 Three-dimensional distribution for f3 (y-Axis acceleration), f3 (Inlet), f3 (Outlet) (Best-case)
Fig. 9 Three-dimensional distribution for f4 (z-Axis acceleration), f4 (Outlet), f4 (y-Axis acceleration) (Best-case)
Fig. 10 Three-dimensional distribution for f6 (z-Axis acceleration), f6 (Current), f6 (Inlet) (Worst-case)
Fig. 11 Three-dimensional distribution for f7 (Outlet), f7 (Current), f7 (x-Axis acceleration) (Worst-case)
A Machine Learning-Based Signal Analytics Framework for Diagnosing the Anomalies of Centrifugal Pumps

Numbers of sensors per type

Sensor types Number of sensors
Acceleration sensor (x, y, z) 3
Pressure transducer (in, out) 2
Current sensor 1
Flow meter 1

Accuracy of seven models for features

f 1 f 2 f 3 f 4 f 5 f 6 f 7
Accuracy
[%]
99 99 100 99 96 92 92
Table 1 Numbers of sensors per type
Table 2 Accuracy of seven models for features