The seismic behavior of the structures equipped with ATMD is often investigated based on the rigid base assumption without considering soil-structure interaction (SSI) effects. The SSI effects significantly modify the dynamic characteristics of the structures, while these changes may be ignored in the design process of the controllers. The present paper aims to address the issue of the SSI effects on the seismic behavior of the structures and performance of the adopted controllers. For this purpose, a mathematical model is developed for the time domain analysis of tall building equipped with ATMD including SSI effects. Considering the fixed base case and three types of ground states, namely soft, medium and dense soil, the numerical studies are carried out on a 40-story structure subjected to different earthquake excitations. Two well-known controllers, proportional-integral-derivative (PID) and linear-quadratic regulator (LQR) controllers, are employed for tuning control force of ATMD in different conditions of ground state. A particle swarm optimization (PSO) algorithm is used for the optimum design of Tuned mass damper (TMD) parameter and the gain matrices of the controllers in both cases without and with SSI effects. It is found that TMDs are more effective for the higher soil stiffness and their efficiencies are degraded in soft soils. Furthermore, the SSI significantly affects on the optimum design of the PID and LQR controllers. The adopted controllers are significantly able to mitigate the peak top floor displacement of the tall building. In addition that the PID controller is a simple strategy with design variables much less than LQR controller, it performs better than the LQR controller in most earthquakes for different conditions of ground state. The performance of the controllers decreases with increasing soil softness, so that ignoring the SSI effects may result in incorrect and unrealistic results of the seismic behavior of the structures.
Optimum Design of Braced Steel Space Frames including Soil-Structure Interaction via Teaching-Learning-Based Optimization and Harmony Search Algorithms
