Review and Application of Optimal Sensor Placement Method on Herøysund Bridge

Abstract

The study attempts to find the best sensor configuration for modal identification on the Herøysund Bridge. The objective is to increase the amount of data obtained from structural testing while reducing the number of sensors needed thereby reducing the cost of the sensor system. Four frequently used optimal sensor placement (OSP) methods, including the modal kinetic energy (MKE) method, the effective independent (EFI) method, minimum modal assurance criterion (MinMAC) method, and the information entropy (IE) method is explored and discussed. The EFI method is selected and applied on a beam model of the bridge, both with and without post-tension. An algorithm is written on MatLAB using as input data the modal analysis results of the bridge obtained from ANSYS. A modified EFI method known as the effective independence driving point residue (EFI-DPR) method is also considered, and its final sensor locations set compared to those obtained from the EFI method. Both methods are validated by the condition number, trace, and determinant of the Fisher information matrix (FIM). Additionally, an interface is established to link ANSYS to MATLAB for the purpose of performing the sensor placement methods. The outcome of the Herøysund Bridge case study shows that the EFI-DPR method is the more effective than the EFI method because it maximizes its performance criterion.