Data-Driven Methods for Vibration-Based Monitoring Based on Singular Spectrum Analysis

Irina Trendafilova, David Garcia Cava, Hussein Al-Bugharbee

Research output: Chapter in Book/Report/Conference proceedingChapter (peer-reviewed)peer-review

Abstract / Description of output

This chapter studies the application of data-driven methods and specifically principal component analysis (PCA) and singular spectrum analysis (SSA) for purposes of damage assessment in structures and machinery. In this study, data analysis methods PCA and SSA are applied to the measured vibration signals in order to extract information about the state of the structure/machinery and the presence of a fault in it. Two applications are offered, one for damage assessment on a wind turbine blade and another one for fault diagnosis in rolling element bearings. The results demonstrate strong capabilities of the investigated methodology for both structural damage detection and rolling element fault diagnosis. Eventually, a discussion about the capabilities of the studied methodology and the way forward regarding extending its capabilities and applications is offered.
Original languageEnglish
Title of host publicationVibration-Based Techniques for Damage Detection and Localization in Engineering Structures
Subtitle of host publicationComputational and Experimental Methods in Structures Book 10
EditorsAli S. Nobari, M H Ferri Aliabadi
PublisherWorld Scientific
ISBN (Print)978-1786344960, 1786344963
Publication statusPublished - 4 May 2018
Externally publishedYes

Publication series

NameComputational and Experimental Methods in Structures
PublisherWorld Scientific
ISSN (Print)2044-9283

Keywords / Materials (for Non-textual outputs)

  • VSHM
  • Singular spectrum analysis
  • Principal component analysis (PCA)
  • outlier principle
  • structural and machinery monitoring
  • rolling element fault detection
  • Wind turbine blade


Dive into the research topics of 'Data-Driven Methods for Vibration-Based Monitoring Based on Singular Spectrum Analysis'. Together they form a unique fingerprint.

Cite this