Passive dampers can be used to connect two adjacent structures in order to mitigate earthquakes induced pounding
damages. Theoretical and experimental studies have confirmed efficiency and applicability of various connecting
devices, such as viscous damper, MR damper, etc. However, few papers employed optimization methods to find the
optimal mechanical properties of the dampers, and in most papers, dampers are assumed to be uniform. In this study, we
optimized the optimal damping coefficients of viscous dampers considering a general case of non-uniform damping
coefficients. Since the derivatives of objective function to damping coefficients are not known, to optimize damping
coefficients, a heuristic search method, i.e. the genetic algorithm, is employed. Each structure is modeled as a multi
degree of freedom dynamic system consisting of lumped-masses, linear springs and dampers. In order to examine
dynamic behavior of the structures, simulations in frequency domain are carried out. A pseudo-excitation based on
Kanai-Tajimi spectrum is used as ground acceleration. The optimization results show that relaxing the uniform dampers
coefficient assumption generates significant improvement in coupling effectiveness. To investigate efficiency of genetic
algorithm, solution quality and solution time of genetic algorithm are compared with those of Nelder-Mead algorithm.
Elevated civil structure systems, such as communication towers and water tanks, are prone to higher mode vibration and
earthquake induced damages. To mitigate damages, however, the structures are retrofitted with conventional (e.g. steel
casing) and/or emerging techniques (e.g. smart structures). Smart structure entails integration of system behavior, control
design and actuators. In this paper, utility of smart structures is illustrated through an elevated water tank concrete
column. The concrete column is modeled as a continuous system, using the Lagrangian formulation, and linear quadratic
regulator (LQR) is used for the control system, and shape memory alloy (SMA) for actuation. The water tank is excited
with the 1940 El-Centro earthquake record. A sensitivity analysis is performed on the controller error and penalizing
constants, as well as actuator location and angle of the connection. The four control variables that can be analyzed for
the controller are: R<sub>r</sub>, Q<sub>r</sub>, R<sub>e</sub>, and Q<sub>e</sub>, which are the control penalty, error penalty, measurement noise and process noise,
respectively. The connection height on the beam and angle of the actuator is also analyzed for optimal performance.
From the sensitivity analysis, the most efficient controller configuration is identified for further analysis of the structure.
Optimal actuator configuration can be found based on the reduction of displacement versus the amount of energy used. It
has been shown that using the SMA, the seismic demand on the concrete column is reduced using the SMA.