Experimental evaluation of various Operational Modal Analysis methods in the presence of simultaneous random and harmonic force excitation

In the present research, performance of harmonic components identification and elimination techniques in practical operational modal data is investigated. Firstly, classical modal testing has been carried out on a cantilever steel beam and modal data are extracted. Four bending modes were available in the frequency band of interest. Afterwards, the same structure underwent operational modal tests with simultaneous random and harmonic excitation forces. In OMA, measurement data obtained from the operational responses are used to estimate the parameters of models that describe the system behavior. Extracted modal parameters from these data compared with baseline parameters. Extraction of modal parameters from operational data was fulfilled through different common OMA techniques. Each set of obtained results was evaluated in comparison with reference results. Accuracy and capability of each method in eliminating spurious modes caused by harmonic input components have been studied. For data analysis, in addition to developed MATLAB codes, commercial software of PULSE has been utilized in parallel as a tool for verification. The results of this survey demonstrate that the accuracy of the Stochastic Subspace Identification method is higher compared to Frequency Domain Decomposition and Enhanced Frequency Domain Decomposition methods. However, when the system has low modal damping, Frequency Domain Decomposition methods provide better estimates.