Mechanical vibration fault detection for turbine generator using frequency spectral data and machine learning model: Feasibility study of big data analysis

Long Yi Chang, Yi Nung Chung, Chia Hung Lin, Jian Liung Chen, Chao Lin Kuo, Shi Jaw Chen

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

The frequency spectra of vibration signals can be used to monitor the mechanical conditions of a turbine generator. Frequency-based features are extracted by fast Fourier transformation (FFT). The changes in frequency spectral data and amplitude are used to separate the normal condition from the fault conditions. These features indicate that the characteristic frequencies are 1 × f, 2 × f, 3 × f, and two other frequency bands, < 0.4 × f and > 3 × f, where the frequency f is the rotor frequency. The power spectral data shows the mechanical vibration fault at particular characteristic frequencies. Then, radial-based color relation analysis (CRA) is applied to identify mechanical faults, including normal condition, oil-membrane oscillation, imbalance, and no orderliness. Using practical records, the experimental results will show that the proposed method has a higher accuracy in mechanical vibration fault detection.

Original languageEnglish
Pages (from-to)821-832
Number of pages12
JournalSensors and Materials
Volume30
Issue number4
DOIs
Publication statusPublished - 2018 Jan 1

Fingerprint

fault detection
machine learning
Turbogenerators
turbines
Fault detection
learning
Frequency bands
Learning systems
Oils
generators
Rotors
Color
Membranes
vibration
Big data
fast Fourier transformations
rotors
oils
membranes
color

All Science Journal Classification (ASJC) codes

  • Instrumentation
  • Materials Science(all)

Cite this

@article{987217684de74744aed7d0a29d820033,
title = "Mechanical vibration fault detection for turbine generator using frequency spectral data and machine learning model: Feasibility study of big data analysis",
abstract = "The frequency spectra of vibration signals can be used to monitor the mechanical conditions of a turbine generator. Frequency-based features are extracted by fast Fourier transformation (FFT). The changes in frequency spectral data and amplitude are used to separate the normal condition from the fault conditions. These features indicate that the characteristic frequencies are 1 × f, 2 × f, 3 × f, and two other frequency bands, < 0.4 × f and > 3 × f, where the frequency f is the rotor frequency. The power spectral data shows the mechanical vibration fault at particular characteristic frequencies. Then, radial-based color relation analysis (CRA) is applied to identify mechanical faults, including normal condition, oil-membrane oscillation, imbalance, and no orderliness. Using practical records, the experimental results will show that the proposed method has a higher accuracy in mechanical vibration fault detection.",
author = "Chang, {Long Yi} and Chung, {Yi Nung} and Lin, {Chia Hung} and Chen, {Jian Liung} and Kuo, {Chao Lin} and Chen, {Shi Jaw}",
year = "2018",
month = "1",
day = "1",
doi = "10.18494/SAM.2018.1783",
language = "English",
volume = "30",
pages = "821--832",
journal = "Sensors and Materials",
issn = "0914-4935",
publisher = "M Y U Scientific Publishing Division",
number = "4",

}

Mechanical vibration fault detection for turbine generator using frequency spectral data and machine learning model : Feasibility study of big data analysis. / Chang, Long Yi; Chung, Yi Nung; Lin, Chia Hung; Chen, Jian Liung; Kuo, Chao Lin; Chen, Shi Jaw.

In: Sensors and Materials, Vol. 30, No. 4, 01.01.2018, p. 821-832.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Mechanical vibration fault detection for turbine generator using frequency spectral data and machine learning model

T2 - Feasibility study of big data analysis

AU - Chang, Long Yi

AU - Chung, Yi Nung

AU - Lin, Chia Hung

AU - Chen, Jian Liung

AU - Kuo, Chao Lin

AU - Chen, Shi Jaw

PY - 2018/1/1

Y1 - 2018/1/1

N2 - The frequency spectra of vibration signals can be used to monitor the mechanical conditions of a turbine generator. Frequency-based features are extracted by fast Fourier transformation (FFT). The changes in frequency spectral data and amplitude are used to separate the normal condition from the fault conditions. These features indicate that the characteristic frequencies are 1 × f, 2 × f, 3 × f, and two other frequency bands, < 0.4 × f and > 3 × f, where the frequency f is the rotor frequency. The power spectral data shows the mechanical vibration fault at particular characteristic frequencies. Then, radial-based color relation analysis (CRA) is applied to identify mechanical faults, including normal condition, oil-membrane oscillation, imbalance, and no orderliness. Using practical records, the experimental results will show that the proposed method has a higher accuracy in mechanical vibration fault detection.

AB - The frequency spectra of vibration signals can be used to monitor the mechanical conditions of a turbine generator. Frequency-based features are extracted by fast Fourier transformation (FFT). The changes in frequency spectral data and amplitude are used to separate the normal condition from the fault conditions. These features indicate that the characteristic frequencies are 1 × f, 2 × f, 3 × f, and two other frequency bands, < 0.4 × f and > 3 × f, where the frequency f is the rotor frequency. The power spectral data shows the mechanical vibration fault at particular characteristic frequencies. Then, radial-based color relation analysis (CRA) is applied to identify mechanical faults, including normal condition, oil-membrane oscillation, imbalance, and no orderliness. Using practical records, the experimental results will show that the proposed method has a higher accuracy in mechanical vibration fault detection.

UR - http://www.scopus.com/inward/record.url?scp=85046300241&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85046300241&partnerID=8YFLogxK

U2 - 10.18494/SAM.2018.1783

DO - 10.18494/SAM.2018.1783

M3 - Article

AN - SCOPUS:85046300241

VL - 30

SP - 821

EP - 832

JO - Sensors and Materials

JF - Sensors and Materials

SN - 0914-4935

IS - 4

ER -