Page 195 - Dynamic Loading and Design of Structures

Page 170

corresponding conventional one). Such a comparison is strictly not valid as it completely
ignores cost/benefit issues relating to future savings due to much lower level of damage in the
isolated structure, which can be substantial (Mayes et al. 1990). On an initial cost basis,
isolation can offer more economical solutions if the design force level is high (e.g. important
structures in high seismicity areas) or if, as a result of using isolation to control damage, the
structure is detailed for less ductility than in the case of conventional buildings; the latter is an
option that is not explicitly recognized by current seismic codes. Finally, the isolation solution
can become attractive when it leads to lower cost of insurance (i.e. lower premiums or no
mandatory insurance against earthquakes in high seismicity areas).

Active control
Whereas in passive control specially provided devices absorb most of the energy input into
the structure, the devices used in active control introduce an energy (or force) source into the
structure. Active control systems have been developed during the last two decades for
reducing the response of buildings (particularly tall ones) to wind and earthquake loading.
In a structure subjected to seismic loading and incorporating an active control system, the
ground motion and/or the structure’s response have to be monitored with appropriate sensors
during the earthquake. Records from the sensors are then fed into a controller (computer) that
activates devices for modifying the structure’s response continuously during its excitation.
These devices are either hydraulic actuators acting against masses in a direction that opposes
that of the earthquake forces or they change the dynamic properties of the structure in order to
reduce its response.
This is an attractive concept, but when applied to massive civil engineering structures such
as tall buildings (instead of mechanical engineering structures) several practical problems
arise, for instance the provision of adequate reaction systems to resist the large control forces
produced by the actuators. Another serious problem is that since active control systems
depend on power supply, it has to be ensured that this supply will not be interrupted during a
strong earthquake (as it often happens), otherwise the whole system will remain idle exactly at
the time that it will be required to function.
Currently used active control systems include active mass drivers, active tendons (wherein
tension in the prestressed tendons is varied during the earthquake excitation in a way to
reduce the structure’s response), active adjustable stiffness systems (joints between the braces
and the structure are either engaged or disengaged by closing or opening a control valve), and
pulse generators (systems of pneumatic actuators and nozzles). Combinations of the above
systems have also been suggested (Soong et al. 1991), offering some advantages.
Despite the attractiveness of the concept and the high quality interdisciplinary research
carried out over the last two decades, the practical application of active