Seismic Performance Of Buckling-Restrained Braces And Substructure Testing Methods

Buckling-restrained braces (BRBs) have good potential of seismic application due to stable hysteretic behavior, simple construction and low cost. The configuration and computation model of BRB are important issues for the sake of enhancing the earthquake resistance of structures with BRBs. Modern structures have become larger and seismic control devices more complex than before. The tests for large scale and complex structures are limited by traditional tests methods. Substructure test methods may resolve these problems by performing tests on key elements of the structure at full or large scale, with the physical test coupled to a numerical model of the surrounding structure. The seismic performance of BRBs and substructure test methods are studied in this dissertation in order to solve the key scientific problems encountered when they are applied in practice.All steel BRBs with different configurations are studied through static cyclic tests. The results from the tests show the stable hysteretic behavior and substantial energy absorption capacity of the braces with properly designed construction. The results of the substructure pseudodynamic tests show that the BRB can significantly reduce the seismic response of the structure by dissipating most of the energy through inelastic deformation. The calculated structural responses, by using bilinear model or Bouc-Wen model, are very close to those from tests. The cumulative plastic ductility is calculated from the results of static cyclic tests and pseudodynamic tests. The result shows good fatigue behavior of BRBs.A key element of substructure tests is the numerical algorithm that is used to perform the stepwise integration of the equations of motion. The stability of classic implicit time integration algorithms for nonlinear structures is studied. The theoretical analyses show that these implicit time integration methods are stable for a rather typical class of nonlinear systems in structural dynamics, i.e. systems with nonlinear restoring forces which can be separated into linear forces and bounded nonlinear forces. Such structures typically include those with softening type hysteresis with the Bouc-Wen model. The energy approach is used to theoretically prove that the average acceleration method is unconditionally stable for structures with typical nonlinear damping, including velocity power type damping with bilinear restoring force model as a special case. Numerical examples verify the correctness of the theoretical analysis.Two energy-conserving time integration methods developed by Simo and Hughes are studied. These methods are unconditionally stable for general nonlinear structures. The theoretical analyses show that the balance equation of Simo method has the only one solution, Hughes method may induce multiple solutions. The results of numerical example exhibit advantages of Simo method over Hughes method and average acceleration method. For Simo method, the procedure is presented to calculate the potential energy increment at inflection point in bilinear model. The results of numerical examples demonstrate the effectiveness of the procedure.Substructure test method with energy-conserving time integration algorithm is developed. The equivalent force control (EFC) method is used to solve implicit difference equation. A proper design of the EFC system depends on the accurate computation of the force-displacement conversion coefficient which can remove the steady-state error of the system. An approach to predict force-displacement conversion coefficient is presented. The oscillation of response of equivalent force is analyzed for MDOF structures. The method to reduce oscillation is presented. The available methods of delay compensation usually amplify the input signals to the actuator, which may leads to system instability. The new compensation scheme presented here does not change the equivalent force command, but predicts the restoring force corresponding to the equivalent force command. The results of numerical examples show that the compensation scheme is effective and the oscillation of equivalent force response is suppressed. The substructure tests are conducted with spring specimens and BRB specimens. These tests show that the substructure test with energy-conserving integration is successful for both linear and nonlinear structures. The substructure pseudodynamic tests validate the computation of force-displacement conversion coefficient. The real-time substructure tests show that the 3rd-order polynomial compensation can effectively compensate the actuator delay.