FOOTBRIDGE MODEL UPDATING BASED ON MODAL DATA USING SENSITIVITY METHOD AND PARTICLE SWARM OPTIMIZATION
Marcelo André Toso1; Herbert Martins Gomes1
1 Federal University of Rio Grande do Sul
doi:10.20906/CPS/COB-2015-0024
Resumo
Model updating is an important design task when dealing with numerical models that are confronted to experimental data. In this paper, an investigation is carried out in order to update a footbridge finite element numerical model using the Sensitivity Method and Particle Swarm Optimization. The analyzed footbridge has 34.08 m in length, 2.4 m in width and 2.25 m in height. Modal parameters are obtained by the solution of the corresponding eigenvalue-eigenvector problem and numerical simulations with the updated model are performed in the time domain. The numerical model results are compared to experimental measurements in the actual footbridge and used as metric for the updating. The Young´s modulus, material density of the metallic components and corresponding concentrated masses were elected as the parameters to be updated due to uncertainty in their evaluation. The baseline values for the Young´s modulus parameter and material density were 210.00 GPa and 7850 kg/m³ (structural steel), respectively, which were updated to 184.36 GPa and 7883.45 kg/m³ using the Particle Swarm Optimization. For the Sensitivity Method these values were 200.00 GPa and 7782.34 kg/m³. The corresponding natural frequencies presented errors lower than 2.75% for the first seven natural frequencies. The resulted numerical natural frequencies were fairly close to the experimental ones and confirmed the adequacy of the model to represent the main dynamic features of the footbridge.
Palavras-chave: Model updating; Footbridge; Natural frequencies