Operational modal analysis (OMA) is a powerful tool in structural health monitoring (SHM), enabling the estimation of dynamic characteristics (natural frequencies, mode shapes, and damping ratios) of structures based on their vibration response under operational conditions. The performance of OMA techniques depends on specific structural characteristics and identification procedure selected. To this aim, a variety of OMA techniques, including enhanced frequency domain decomposition (EFDD), stochastic subspace identification (SSI-COV and SSI-DATA), PolyMAX, and Ibrahim time domain (ITD), are implemented in a new computationally efficient Python library, Modalyzer, developed and presented here for the first time to unify and streamline the application of OMA techniques to different structural contexts. Flexibility and robustness of Modalyzer are demonstrated by applications to three different case-study structures with different conditions and measurement systems, and results obtained by Modalyzer are compared to those obtained by independent calculations and methods for validation purposes. The results highlight strengths and limitations of each method in different structural scenarios, offering guidance on selecting the most suitable OMA technique for specific SHM applications. Moreover, Modalyzer demonstrates significant time efficiency compared to longer computational times of established tools like ARTeMIS Modal, and incorporates user-friendly features further facilitating its adoption within the SHM community.
Shamsaddinlou et al. (Fri,) studied this question.