Portal de Periódicos da CAPES

Problem of Detecting Damage Through Natural Frequency Changes

2018; Imperial College Press; Linguagem: Inglês

10.1142/9781786344977_0004

2044-9283

Gilbert-Rainer Gillich, Nuno N. N. Maia, Ion-Cornel Mituleţu,

Ultrasonics and Acoustic Wave Propagation

Computational and Experimental Methods in StructuresVibration-Based Techniques for Damage Detection and Localization in Engineering Structures, pp. 105-139 (2018) No AccessChapter 4: Problem of Detecting Damage Through Natural Frequency ChangesGilbert-Rainer Gillich, Nuno N. N. Maia, and Ion Cornel MituletuGilbert-Rainer GillichDepartment of Mechanical Engineering, University "Eftimie Murgu" of Resita, P-ta Traian Vuia 1–4, 320085 Resita, Romania, Nuno N. N. MaiaDepartment of Mechanical Engineering, University of Lisbon, IDMEC-Instituto Superior Técnico, Av. Rovisco Pais, 1049–001 Lisbon, Portugal, and Ion Cornel MituletuDepartment of Mechanical Engineering, University "Eftimie Murgu" of Resita, P-ta Traian Vuia 1–4, 320085 Resita, Romaniahttps://doi.org/10.1142/9781786344977_0004Cited by:3 PreviousNext AboutSectionsPDF/EPUB ToolsAdd to favoritesDownload CitationsTrack CitationsRecommend to Library ShareShare onFacebookTwitterLinked InRedditEmail Abstract: In damage detection, an important aspect is the accurate frequency evaluation which permits observing the modal parameter changes at the earliest stage. The frequency resolution improvement, achieved by extending the time interval of the analysis, is in the case of standard frequency estimation the key to increasing the accuracy of the study. This is not always possible due to the rapid decay of higher-order modes. Many methods to improve frequency readability are nowadays available in the literature, such as those increasing the spectral lines density or those based on several-point interpolation. We debate herein the limitations of the actual methods and introduce an advanced algorithm that uses for interpolation three peaks from three different spectra, achieved from the acquired signal that is stepwise cropped. Processing the signal in this way allows identifying minor natural frequency changes, thus allowing for the observation of damage occurrence in the earliest stage. The algorithm was successfully tested against generated and real signals. Keywords: Damage detectionNatural frequencyDigital signal processingFrequency resolutionInterpolation-based methodsDense overlapped spectrum FiguresReferencesRelatedDetailsCited By 3Challenges in applying vibration-based damage detection to highway bridge structuresBerhanu Tefera, Adil Zekaria and Abrham Gebre9 February 2023 | Asian Journal of Civil Engineering, Vol. 232A Method for Improving the Accuracy of Natural Frequency Measurement Using In-the-loop ComputingAdam Kotowski13 August 2021 | Measurement Science Review, Vol. 21, No. 4Improving the Capability of Detecting Damages in the Early State by Advanced Frequency EstimationNicoleta Gillich, David Lupu, Codruta Hamat, Gilbert-Rainer Gillich and Dorian Nedelcu11 January 2021 Vibration-Based Techniques for Damage Detection and Localization in Engineering StructuresMetrics History KeywordsDamage detectionNatural frequencyDigital signal processingFrequency resolutionInterpolation-based methodsDense overlapped spectrumPDF download