Evolutionary Optimization of Passive Compensators to Improve Earthquake Resistance - Design Optimization of Active and Passive Structural Control Systems - page 252

Bertero 2004, pp.10-11 - 10-19). Dampers may
absorb some of the impact and help to limit the
damage of the building. On the other hand, the
deflection required to make the dampers consume
energy corresponds to large relative displacements
for example between floors, so the local damage
could be essential, limiting the further use of the
edifice.
The same holds for locally weak designs. They
restrict the region of severe destruction to some
predefined parts which are capable of withstanding
the large deformation. But it might be difficult to
repair the deformed sections to re-establish the
performance of the original structure. Neverthe-
less, both ideas help to cut the losses during at
least one earthquake. This can be an essential
advantage compared to the risk of large destruc-
tion without any prevention.
Another approach deals with compensator
systems (Chopra 2000, pp. 470-471, Den Hartog
1956, pp. 87-121), which are also called tuned
mass dampers or absorbers. It is well known that
the Eigen frequencies, the Eigen forms and the
amplitudes at a given excitation of a dynamic
system change if additional masses and springs are
added to the initial system. A qualified selection
of springs and masses may reduce the earthquake
impact on the building. Some compensating
systems work with fluids, being driven through
U-shaped piping systems instead of masses and
springs. Their function and performance is com-
parable to the mass-spring systems.
Two main approaches are dealt with in practice.
Active systems control the displacements of the
compensator masses if sensors indicate ground
motion (Chen and Wu 2001, Reiterer and Ziegler
2005, Teuffel 2004). This can be very efficient but
requires fast control systems, including actuators
which are capable of accelerating large masses in
a very short amount of time. Compensators using

active control have the advantage to respond in a
specific way to the external event.
Passive systems do not need sensors, actors and
energy supply but they require large masses and a
lot of space for their oscillations. Basically they
aim to influence the natural modes of the structure.
So a dimensioning may be done by compensation
the modal contributions as outlined by Den Hartog
(1956, pp. 87-121). Both approaches are used for
high buildings and for other dynamically excited
structures as well.
Passive systems are installed in some of the
most popular high buildings. Examples are the
Taipeh101 tower (Eddy 2005), where a 660 metric
ton mass is hanging close to the roof at a height
of 450m or the Burj al Arab in Dubai (Nawrotzki
and Dalmer 2005), a hotel where 11 compensators
with a mass of 5 metric tons are installed along
the 321m height of the skyscraper.
Especially for bridges there are many ideas
known that reduce the vibrations caused by wind
or by the traffic on the bridge. One of the most
popular examples is the stabilisation of the London
millennium bridge (Nawrotzki and Dalmer 2005),
which had strongly vibrated but is now stable
after a passive compensation system has been
installed. But for even more simple structures,
compensators are used to improve performance
and to avoid damage. Bachmann writes about a
diving tower which showed material damage due
to large oscillations caused by children swinging it
(Bachmann. et al. 1994). A compensator reduced
the maximum accelerations from about 3 m/sec 2
to less than 0.5 m/sec 2 , so the diving tower could
be used again; no further damage was observed.
There is no reason to prefer a special proposal
to improve the capability of buildings to withstand
earthquakes or other dynamic loading. The deci-
sion, which method to use should be based on an
open discussion of historical, technical, economi-
cal and political aspects.