Passive protection of a structure as regards the seismic risk can be achieved with three different design methods:
- the structure is sufficiently strong to withstand the earthquake while remaining in the linear elastic range;
- the structure has a sufficient post-linear deformation capacity (often inaccurately referred to as ductility) to resist the seismic loading by accepting a certain level of damage;
- the seismic excitation is filtered at specific locations using specific devices. As a result, the damage to the structure is much lower for the seismic design level. This filtering can be done by elastic or inelastic deformation. This is the seismic isolation at the base (e. g. elastomer or friction pads at the base of the structure) and its variations (e. g. metal devices plasticizing at strategic locations).
Seismic isolation is currently a rapidly-expanding method. More than 3000 buildings with earthquake-resistant supports exist throughout the world, particularly in areas with high seismicity (Japan, United States of America, Southern Europe). A large number of bridges are also seismically isolated. On the other hand, few applications of the method have been made to date on special risk structures, such as nuclear power plants. As an illustration, Figure 1 shows the Jules Horowitz experimental reactor (Cadarache, France) built on earthquake-resistant supports.
Despite the increasing number of applications of the seismic isolation method, a number of questions remain unanswered. Briefly, these questions concern:
- the dynamic behaviour of the supports, which can be strongly non-linear, possibly coupled with thermal for certain types of supports such as, for example, shape memory devices or elastomeric supports with lead core (Figure 2);
- technological aspects of design, installation and replacement of the supports;
- the amplification of higher modes and their influence on the response of equipment, in particular for industrial installations, such as nuclear power plants, or for buildings of special interest or open to public (e. g. hospitals);
- the sensitivity of these systems, in particular in the event of seismic excitations different from those planned for their design;
- the design and use of supports with an insulating effect in the vertical direction.
In addition to passive isolation, methods combining passive and semi-active devices have emerged in the last two decades, with a number of achievements in the last decade, particularly to optimize the response of structures to the wind. While there is for now few actual applications of this technique to improve the seismic response of structures, the research is however active on this subject. Indeed, the use of mixed insulation could remedy the weaknesses of passive insulation and increase its efficiency. Nevertheless, several technological, theoretical and numerical aspects need to be studied and further developed for the effective application of such techniques. Without insisting on the technological issue, concerning the development of high-capacity and performance semi-active devices, we can mention, among others, the following scientific obstacles:
- identification and modelling of semi-active systems that can be highly non-linear;
- the development of control algorithms for complex systems that are efficient, robust and fast since the aim is to control semi-active devices in real time;
- difficulties in the control procedure related to “real world” noise as well as the limited number of measures (use/development of efficient observers).
Given their increasing development and effectiveness, passive and mixed seismic isolation, as well as other techniques for mitigating the seismic response of structures (e. g. passive or semi-active vibration dampers), are an important research focus for the SEISM Institute.