A Variational Mode Decomposition Approach for Modal Identification of Structures Using Optimal Sensor Placement

Modal identification describes the behaviour of the structure under dynamic loading conditions such as earthquake, wind, blast (impact), wave, traffic, etc. A newly developed variational mode decomposition (VMD) technique identifies the modal parameters of structure based on dynamic responses. The vibration-based VMD techniques mainly rely on a sensor-driven data collection approach for the response of the structure. To develop an effective modal identification system, a method for determining the optimal number and location of sensors is proposed. The vibration response of an optimised sensor network lowers the installation of sensors and operational costs, simplifies modal identification system computing operations, and provides the modal parameter estimation. The Spline interpolation methodology for optimising the number of sensors and their positions in the building is proposed in this study. A MAC (Modal Assurance Criteria) value between the exact and estimated mode shapes was utilized as a criterion for determining the minimum number of sensors. In this study, the capability of the VMD technique to determine a structure’s modal parameter with the help of a limited number of sensors is demonstrated in the multi-storey building and the UCLA Factor building, and the results show that the proposed VMD approach can easily identify the modal parameter of the building with better accuracy.