Enhancing Vibration Analysis with Singular Spectrum Analysis in a System with Passive Damper Attached

Abstract

Vibration analysis is a process of monitoring and searching for anomalies or changes in the vibrational behavior of a system. In this context, the Fourier Transform has been widely employed to study systems whose responses to external disturbances are analyzed through vibrations. However, considering that vibration signals are often represented as time series, the application of Singular Spectrum Analysis (SSA) has been a promising methodology in this type of analysis. In this work, the SSA technique was used to investigate the vibrational responses of a single degree of freedom system with an attached tuned liquid column damper (TLCD). The system responses were captured by an accelerometer and processed by a data acquisition module, followed by analysis using the R programming language. The experimental results demonstrated the efficiency of the damper in attenuating the system vibration; it was also possible to identify the dominant frequencies through the Fourier Transform, specifically its computational version, the Fast Fourier Transform. In this work, the SSA methodology was applied which allowed the identification of distinct signals, including possible signals from the vibrations of aluminum rods coupled to the system, not detected by frequency domain analysis. Thus, the application of the SSA technique enabled the identification and quantification of different signals in the system with an attached passive damper and proved to be an excellent methodology to be added to the classic techniques of vibration analysis.