Amirkabir University of TechnologyAmirkabir Journal of Civil Engineering2588-297X54220220421Modal Data Identification of the Prestressed Concrete Bridge Using Variational Mode DecompositionModal Data Identification of the Prestressed Concrete Bridge Using Variational Mode Decomposition1616437610.22060/ceej.2021.19075.7055FAPayamDindarDepartment of Civil Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran0000-0002-7727-0127MirhamidHosseiniDepartment of Civil Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran0000-0003-4261-2120Mohammad RezaMansooriDepartment of Civil Engineering, Science and Research Branch, Islamic Azad University, Tehran, Iran0000-0003-3895-6326Journal Article20201002<span style="color: black;">In this study, the variational mode decomposition (VMD) algorithm was used to identify the modal characteristics of the structure using the decomposition of the acceleration responses recorded by the sensors. This algorithm has advantages over other signal decomposition methods that is resistant to the noise and sampling frequency. Also, the VMD algorithm extracts the natural frequencies of the structure concurrently. In addition, the damping ratios of the structure were estimated by fitting a linear function to the logarithmic diagram of the modal response in the decaying amplitude and calculating the slope of this line. The efficiency and accuracy of this algorithm were investigated by decomposing the acceleration responses obtained from the sensors installed on a prestressed concrete bridge (PSCB) that is located under the load of passing vehicles. The VMD algorithm was used for signal processing in MATLAB to estimate the natural frequencies, damping ratios and mode shapes of the bridge and ARTeMIS was utilized to verify the results. In addition, the finite element modeling and modal analysis of the bridge were performed in ABAQUS and the natural frequencies and mode shapes of the bridge were obtained. The results showed that the mode shapes estimated by the VMD algorithm were in good agreement with the finite element model and ARTeMIS. Also, the damping ratios estimated by this algorithm were obtained close to the damping value of the prestressed concrete bridge. The difference between the frequencies calculated by the VMD algorithm and ARTeMIS was about 1%, and the difference with the finite element model frequencies was close to 5%.</span><span style="color: black;">In this study, the variational mode decomposition (VMD) algorithm was used to identify the modal characteristics of the structure using the decomposition of the acceleration responses recorded by the sensors. This algorithm has advantages over other signal decomposition methods that is resistant to the noise and sampling frequency. Also, the VMD algorithm extracts the natural frequencies of the structure concurrently. In addition, the damping ratios of the structure were estimated by fitting a linear function to the logarithmic diagram of the modal response in the decaying amplitude and calculating the slope of this line. The efficiency and accuracy of this algorithm were investigated by decomposing the acceleration responses obtained from the sensors installed on a prestressed concrete bridge (PSCB) that is located under the load of passing vehicles. The VMD algorithm was used for signal processing in MATLAB to estimate the natural frequencies, damping ratios and mode shapes of the bridge and ARTeMIS was utilized to verify the results. In addition, the finite element modeling and modal analysis of the bridge were performed in ABAQUS and the natural frequencies and mode shapes of the bridge were obtained. The results showed that the mode shapes estimated by the VMD algorithm were in good agreement with the finite element model and ARTeMIS. Also, the damping ratios estimated by this algorithm were obtained close to the damping value of the prestressed concrete bridge. The difference between the frequencies calculated by the VMD algorithm and ARTeMIS was about 1%, and the difference with the finite element model frequencies was close to 5%.</span>https://ceej.aut.ac.ir/article_4376_3929444d5a39b58defbe280311145dfb.pdf