Vibration control of the College Bridge using tuned mass dampers

One of the challenging tasks for civil engineers is to mitigate the response of structures that are subjected to dynamic loads in order to prevent possible damages that may cause human and enormous economic losses.To minimize vibration, reduction of the external disturbance to a system is always preferred, but in many cases, this may not be possible. Modification of the system to avoid resonance may entail significant redesign. Furthermore, it would be difficult to be applied to existing structures. Thus, vibration control devices, which can be simply attached to the existing system to reduce the vibration without drastically altering the original system, have been developed. Passive tuned mass damper (TMD), whose concept was presented more than a century ago, is undoubtedly a simple, inexpensive and reliable means to suppress unfavorable vibrations of structures but the very narrow band of suppression frequency, the ineffective reduction of non-stationary vibration, and the sensitivity problem due to detuning are the inherent limitations of the passive TMDs. TMDs are usually tuned to the first natural frequency of the structures. Since TMD parameters are constant during the life cycle of the structure, it is important to adjust them properly to achieve a favorable performance. Optimal values of TMD parameters for structures with non-linear behavior require non-linear dynamic analyses. There are many analytical and empirical relations to identify these parameters obtained by structure simplification and loading. In this paper, genetic algorithm (GA) is employed to find optimum TMD parameters for vibration control of the College Bridge in Tehran. With the length of 372 m, this steel bridge has 14 spans. The bridge is modeled in OpenSees environment. Verification of the finite element modelling is performed by comparing the results of the dynamic analyses under four earthquake records by those of another model created in SAP2000 software. In order to mitigate its vibrations, 11 TMDs are considered to be installed on the bridge. The aim of GA is to minimize the displacement of the tallest pier of the bridge in order to decrease the maximum displacement of the structure subject to earthquake excitations. Based on the analyses conducted for near-field and far-field earthquakes, it was concluded that employing GA considerably reduces convergence rate to achieve optimum TMD parameters. To evaluate the performance of a control system during severe earthquakes, incremental dynamic analyses (IDA) for maximum peak ground accelaration (PGA) of 0.1g to 1.0g was conducted. The longitudinal root mean square and maximum displacement of the tallest pier in uncontrolled and controlled cases are obtained and compared. The results of IDA analyses show that for low PGA values, TMDs by themselves absorb and dissipate a large portion of the input energy because in this case the piers remain elastic. However, for higher values of PGA, piers also dissipate some portion of input energy by entering nonlinear region. The percentage of response reduction for different earthquakes are not the same because each earthquake has its own frequency content. According to the numerical analyses for the mass ratio as 4%, the longitudinal displacement and reduced RMS displacement of the largest pier of bridge by tuned mass damper for El-Centro, Kern-County, Kobe and Northridge earthquakes are 24.9 and 34.3, 43.5 and 38.7, 30.6 and 40.4, 13.6 and 28.1 respectively.