Final Report Summary - E-FAST (Design study of a European facility for advanced seismic testing)
Executive summary:
High performance experimental facilities are necessary to meet the objectives of earthquake risk mitigation and to make progress in methods for the design and assessment of buildings and infrastructures. Therefore, in order to be positioned within the avant-garde of earthquake research it is important for Europe to build a new high performance experimental facility. For that reason, the European Commission (EC) supported, as a part of the Seventh Framework Programme (FP7), a design study of a new generation seismic testing facility. This is the E-FAST (European Facility for Advanced Seismic Testing) project. Five European partners with a large experience in seismic and dynamic testing were involved in E-FAST: Commissariat à lEnergie Atomique (coordinator, France), the Gheorghe Asachi Technical University of Iasi (Romania), Eucentre (Italy), the University of Kassel (Germany) and the joint research centre (EC).
The first step was the determination of the performance requirements of the new facility. Then a lay-out of was proposed which goes much further than existing shaking table facilities in Europe. It meets the requirements of modularity, flexibility and operational ease with technological choices that minimise the techno-economic risk. Several aspects have been studied related to the preliminary design. Amongst others the studies during the E-FAST project focussed on the following issues:
Technology: design of the hydraulic power supply system, shake tables, reaction mass, modular reaction structure, telepresence room, evaluation of shake table control methods.
- Advances in experimental techniques: carrying out of real time substructure tests with linear and non-linear physical substructures, evaluation of hardware for fast computer networking, development and utilisation of a no-contact vision measurement system.
- Dissemination and access: design of a web portal enabling efficient access and networking, evaluation of the cost of physical access (based on the cost related to the organisation of test campaigns).
- Management and operational issues: overall construction cost estimate, study of operational conditions (types of tests, tests duration, maintenance, necessary staff, safety issues) and operational cost, consideration of alternative configurations with decreased performances and evaluation of the corresponding cost savings, identification of the potential project risks and evaluation of their impact, proposal of a schedule and road map for the detailed design and construction phase, determination of criteria for the optimum construction site of a future facility.
The medium to long term impact of a new advanced seismic testing facility in Europe will be mainly the reduction of the seismic vulnerability. This will promote sustainable economic development of Europe's seismic regions, but also of the entire Europe, through savings on the total financial loss due to future earthquakes. In fact, the enhanced capabilities of an advanced testing facility will lead to:
- a further insight into the earthquake response behaviour of structures, in general;
- the improvement of the numerical simulation tools via their validation with the experimental results;
- the improvement and validation of regulations and recommendations;
- safer design and qualification of industrial structures and equipment, especially those of nuclear and chemical industry;
- the development and validation of new construction methods, materials and devices;
- the support of European companies through the aforementioned validation of new technologies and design concepts;
- demonstration tests for policy makers and public awareness and dissemination purposes;
- a 'natural' excellence centre for advanced knowledge in earthquake engineering through training, transnational access and dissemination.
Project context and objectives:
Seismic events of the recent past have proved that European and the neighbouring countries, especially those comprised in the Mediterranean area, are exposed to a high seismic risk. Surprisingly the number of victims and the overall economic losses are important compared to industrialised country like Japan and United States often faced with higher levels of shaking. This fact can be explained by considering the higher population density, and that the high number of damaged buildings is due to the large presence of monuments or ancient masonry buildings often vulnerable to earthquake loading. It is readily apparent that, in developed countries, although the numbers of victims of major earthquakes is tending to drop, the costs of the consequences are constantly rising. The costs of the consequences, resulting in significant damage and widespread disorganisation in the area, are constantly rising. Considering damage to plants, loss of data and drops in productivity, are extremely costly. A recent example is the social and economic impact of the L'Aquila (Italy) earthquake in April 2009. It is therefore indispensable for Europe to intensify the research and development in the field of earthquake engineering.
In the last decades considerable advances have been achieved in the earthquake engineering field. The research results have contributed to the preparation of the modern design codes, to the identification of several problems in the existing structures and to innovative solutions for the structural assessment. Despite this huge amount of improvements there are still several open problems. For instance, predictive models have frequently been calibrated on the experimental results obtained from scaled structures, several innovative technologies for building constructions are entering the market and require careful evaluations to verify the level of safety, the experimental validation of the behaviour of large infrastructures (bridges or retaining walls) often requires multi-support excitation, a further insight into soil structure interaction requires testing of heavy models. Available data and future results need to be organised in databases, in order to disseminate them, optimise their use, and provide relevant information for risk oriented approaches. These researches require a large amount of analytical and experimental studies. Moreover, a look at the international earthquake engineering landscape reveals that, outside Europe, there are several high performance seismic testing facilities either already operating or under construction. As an example, in 2000-04 Japan, already boasting the most powerful experimental RTD infrastructures in earthquake engineering, spent 350 million to build the largest three-dimensional (3D) shaking table in the world (20mx15m, 1200t payload). Regarding pseudo-dynamic testing, the ELSA laboratory of the joint research centre of EC in Italy with its reaction wall 16 m high and 20 m long is one of the main seismic testing facilities in the world. However, the situation is different for European shaking tables facilities, having considerably lower performances than that of the major shaking tables laboratories in the world. There is also a trend, mainly in the United States and Asia (China, Japan, Korea) towards facilities with an array of shaking tables which increases operating ease and enable multi-support excitations. The objective is to test structures at the largest possible scale in order to avoid scaling effects.
This means that, unavoidably, if the situation does not change, Europe will cumulate a considerable lag in experimental earthquake engineering, with respect to the USA and Asian countries. Europe should not trail these foreign countries, in particular United States and Japan, in experimental research in earthquake engineering and rely on their Research and technology development (RTD) results. It should compete and share results with them as equal, to serve its own needs and promote its own interests. To see the reasons, we should recall first the difference of buildings in Europe from those in Japan or the United States. In Europe new residential construction uses mostly concrete framing, often with masonry infills, especially in the seismic southern countries. Europe is also rich in cultural heritage buildings, mainly of masonry construction. By contrast, in US and Japan heritage buildings are not common and new buildings are mostly of timber. Moreover, concrete and masonry construction in US and Japan is very different from Europe. So, the focus of RTD in these countries does not fully serve Europe's needs. In the other major type of civil engineering works, namely civil infrastructures, the technology and the materials are fully global. There, Europe is the world leader: in niche technologies (post-tensioning, stay cables, marine or off-shore construction, etc.), in overseas construction (it boasts the world top firms: Bouygues, Dragados, Ferrovial, Hochtief, Vinci) and in overseas consultancy and design, with its huge engineering services sector, etc. As most of the overseas activity is in seismic areas (east, south or south-east Asia, central Asia, northern Africa, Middle East and Latin America), European construction firms and engineering services cannot retain their competitive edge in seismic markets and their reputation as leaders in technology, unless the EU as a whole establishes itself as equal to the USA and Japan in earthquake engineering RTD.
In addition, it is worth noting that there is an emergence of advanced experimental techniques, such as real-time substructuring and advanced measurement techniques that are being explored in the most innovative laboratories. This is an important point since the new experimental methods, based on the substructuring technique, have the advantage of reducing the specimen size allowing a better use of the hardware resources.
For all the aforementioned reasons a new platform for dynamic seismic testing in Europe is not just useful but necessary. A new high performance testing facility will enable studying a large variety of structures and systems. In fact, such a facility is an indispensable tool to calibrate and validate new conceptual approaches in modelling and simulations developed for performance based analysis and design of new structures or retrofitting interventions of safe structures in Europe and even worldwide. It will also contribute to increase worldwide the competitiveness of European science and industry.
Therefore, EC granted, as a part of the seventh framework project, the design study project E-FAST. The main objectives of E-FAST are:
- define the needs in experimental research in earthquake engineering in Europe;
- define the features of a new testing facility, complementary to existing research infrastructures in Europe, combining high capacity, flexibility and operational ease;
- make progress in advanced testing methods such as real time sub-structuring techniques and carry out demonstration tests;
- study the technical feasibility of this facility;
- study financial issues related to the construction cost, operating and maintenance cost and access cost.
Project results:
Required general performances of a new European seismic testing facility
The first international E-FAST workshop pointed out some of the fields that need further experimental research. In particular, more experimental evidence is needed in the following topics:
- in plan irregular buildings exhibiting torsion response;
- precast and prestressed concrete elements and systems;
- masonry buildings and infills. In particular more experimental data of buildings with more than one storey are needed;
- infrastructures (e.g. bridges implying multi-support excitation capability);
- retrofitting;
- aseismic devices (e.g. isolation bearings, dampers);
- equipment and components. In particular the motion of the floor the equipment / component is mounted on should be reproduced implying high acceleration and displacement capacity;
- soil-structure interaction Due to the considerable weight of such models only elementary configuration could be tested. In any case a high payload table is required.
One common point to all classes of problems is that, in order to conduct a meaningful risk assessment, the actual available margins of structures have to be estimated. This holds for all structures but it is even more critical for structures of major importance (e.g. power generation facilities, hospitals). To this end, tests with excitation level up to failure should be possible in future. Depending on the tested structure of interest (building, equipment or secondary structure), failure can be defined as loss of operational function, collapse or relevant significant damage or collapse. This implies that the new facility should have the capability to apply high intensity excitations (high acceleration, velocity and displacement) to models which will be representative of the prototype structures. Since the pseudo-dynamic testing facility at the ELSA laboratory of the joint research centre of EC in Italy, with its reaction wall 16 m high and 20 m long, is considered as one of the main seismic testing facilities in the world, it is proposed that the new facility should be, mainly, a new generation shaking table facility with the possibility to apply advanced experimental techniques like real time hybrid testing also. In addition, the new facility should comply as far as possible with the requirements of flexibility, adaptability and operational ease.
Table 1 shows some indicative performance parameters for different classes of tests. The given numbers are reasonable rough estimates as a trade-off between needed performance and cost. Obviously it is not feasible, either for technological or financial reasons, to build a facility so big that everything could be tested therein. The objective is to propose a design that will enable to carry out meaningful tests using conventional and / or more recent techniques and technologies which have already demonstrated their efficacy and reliability. The acceleration values in table 1 may seem to be unrealistically high. However, it is worth noting that a) several records of real earthquakes revealed very high acceleration values (e.g. 0.98 g Northridge earthquake, 1994, 0.85g Kobe, 1995) b) in the case of scale models, if a velocity similitude is considered, the table acceleration should be multiplied by the inverse of the scale ratio i.e. table acceleration will be higher than ground acceleration of the prototype and c) the values in table 1 are conventional acceleration values corresponding to a rigid specimen. Consideration of the dynamic amplification of the specimen results in a higher demand of force capacity which is equivalent to a higher demand of conventional acceleration capacity. In the case of soil-structure interaction tests, the major part of the mass on the table is due to the weight of the soil itself and its container which will have, in general, a weak dynamic amplification. Therefore, in that case the required shaking table acceleration could be smaller. Regarding secondary structures and equipment, because of the amplification of the shaking motion at the floor level, floor accelerations to reproduce on the table may be much higher than ground accelerations.
Velocity values are also in agreement with actual recorded velocities (e.g. 1.4 m / s Northridge earthquake, 1994, 1.5 m / s Kobe, 1995). High displacement values are also necessary for the shaking table motion to be representative of strong, low frequency ground motions or of floor motions of low frequency buildings (in the case of secondary structures or equipment tests).
General layout
A lay-out of the facility is proposed which meets the above performance requirements. The underlying philosophy is resumed to the following points:
- The new facility should be a significant step ahead and go much further than existing shaking table facilities in Europe.
- It should be a combination of components and technology that have already been validated by their operational use in other existing facilities. In fact E-FAST is a design study not a pure Research and development (R&D) project therefore there was no room in E-FAST for innovation and adoption of 'revolutionary' technology and techniques. Moreover, the choice of well validated technological solution is imposed because of:
1. Technical reasons: In fact the new facility should be able to carry out accurately big scale, high capacity demanding tests from the 1st day of its operation. A less or more long period of adaptation and / or changes to achieve this goal (accurate big scale seismic testing) is not acceptable.
2. Safety reasons
3. The requirement to minimise the techno-economic risk for potential investors. Actually potential investors desire to minimise divergence from the foreseen date of operating start and from the foreseen budget at the moment of their commitment.
The design is based on the feedback from:
1. those of the E-FAST partners running big shake tables;
2. some of the most experienced manufactures of shake tables in the world.
The proposed solution is a trade-off between dreams and real world (performance vs. cost). In fact, too high cost (construction, maintenance, handling, specimen transport) would kill any chance for the facility to be constructed. In addition, given that the proposed configuration is composed of up-to-date elements and technology but already existing and operating allows us to make a construction and operational cost estimate with a good accuracy.
The general lay-out of the facility is shown in figure 1 to figure 3. The facility consists mainly of:
1. Two 6 Degrees of freedom (DOF) 6 m x 6 m shaking tables: The payload of each table is of about 100 tonnes. The tables can be positioned in any place within the trench. The gap between them can vary from 0 to 20 m. They will be able to operate independently or be linked and operate as a single table with a payload of 200 tonnes. They will be able to have a synchronous or asynchronous motion.
The two shaking tables allow the following configurations:
- two separate tables operating independently (adjustable distance between the 2 tables from 0 m to 20 m);
- two separate tables but working together, linked by means of a special truss or plateau and supporting a large specimen (distance between 2 tables adjustable from 0 m to 20 m);
- two separate tables, operating simultaneously but with different motions to test multi-supported structures (distance between 2 tables adjustable from 0m to 20m);
- two tables linked together to realise a large table of 6 m x 12 m;
- one or two tables (in this case fixed rigidly to each other) can be mounted on bearings fixed at the bottom of the pit to operate as a single axial table (1 DOF along the length of the pit). This would enable to test even higher models since the height of the specimen could be equal to the height of the hall (distance between the ground level and the bottom of the crane) plus the depth of the pit.
Obviously a higher number of tables, possibly in different trenches, would enable testing more complex configurations. However as already mentioned, the proposed facility is a trade-off between performance-capacity and cost, that is why only two tables are proposed.
Their performances are summarised below:
- maximum horizontal displacement ± 1 m in OX and OY;
maximum vertical displacement ± 0,75 m in OZ;
- maximum horizontal velocity 2 m / s in OX and OY;
- maximum vertical velocity 1.5 m / s in OZ;
- maximum acceleration 2 g in OX and OY;
- maximum acceleration 1.5 g in OZ;
- maximum duration of excitation at full power 30 seconds.
If necessary, the maximum payload of the two 6 DOF tables could be increased with minor modifications of the actuators (slight increase of their length providing higher shock absorption capacity). However, unless the actuators' capacity increases, this will be done at the price of a smaller maximum acceleration at full payload.
During periods of maintenance or repair, the tables can be uncoupled from the horizontal and vertical cylinders and stored temporarily on the strong-floor or on the outdoor area (cf. item 6). Dedicated trusses / frames will maintain in a vertical or horizontal position the cylinders during maintenance (those frames can be the same which will be used for the initial placement of the cylinders in the pit).
2. One 1 DOF (horizontal) shaking table 11 m x 11 m with high payload of about 500 tonnes: It will be mounted on hydrostatic bearings fixed at the upper part of the pit walls. This shaking table will be intended for heavy specimen and in particular for soil-structure interaction tests. The maximum acceleration at full payload will be of about 0.6g - 1g. Actually in a first step for economic reasons we consider that this table will be actuated by the actuators of the two moveable tables. Therefore the achieved acceleration will depend on the number of the actuators which will be utilised. This big shaking table will not be in the pit permanently but it will be mounted on when a mono-axial test of a heavy specimen must be carried out. Obviously, in the case of configurations where the gap between the two 6m x 6m shaking tables is 20 m, the 11 m x 11 m table should be taken out of the pit. To this end the crane was designed to have the capacity to lift up such a heavy structure.
3. A reaction mass which consists, mainly, of a pit that hosts the aforementioned shaking tables, a room (of about 30 m x 20 m x 4.5 m) hosting pumps and accumulators and a big strong floor area: This thick (2 m) strong floor slab gives enhanced adaptability to the facility. In fact several experimental set-ups (small to medium shake tables, dedicated testing machines, reaction structures) can be mounted on it. The necessary power will be supplied to the actuators at any position by means of hoses. The reaction mass weighs about 25 000 tonnes. Vibration nuisance analyses showed that vibration isolation of the reaction mass by means of specific devices (e.g. springs and dampers) is not necessary. Therefore and to avoid the considerable cost increase associated with isolation devices the reaction mass is put directly on the ground.
4. A modular reaction structure, which can be placed anywhere around the pit, allowing for real-time hybrid testing possibly combining shaking tables and actuators attached on the wall.
5. A hydraulic system composed of piping, actuators, pumps and accumulators with a capacity consistent to the performance criteria given in table 1.
6. An outdoor area devoted to the construction of specimens, especially reinforced concrete or masonry models;
7. A high capacity crane bridge with 4 hooks each one having a capacity of 50 t (total capacity 200 t). It spans the whole width of the working area and it can move along the whole length of the hall and the outdoor area. The crane will be able to lift and transport heavy models from the construction area and install them on the tables and vice versa. Its capacity allows also handling of the shake tables, even of the big mono-axial table. The crane bridge includes also a cantilever crane of a capacity of 20 t which will be used for handling low to medium weight items (pipes, actuators, light specimens). The foundations of the crane frame rails are designed to be independent of the foundations of the hall and the offices building.
8. In addition to the experimental hall (overall dimensions LxWxH=47 m x 42 m x 19 m) a 2-storys building (offices, control room, meeting room, teleconference room) for about 40 persons of about 1000 m² per story is foreseen.
It is worth noting, that though not investigated here because it was beyond the scope of this design study, a strong interaction between the aforementioned 'purely experimental' facility and a numerical high computational capability facility, on site or remote, is necessary. Actually, high performance and accuracy numerical simulation is necessary not only for advanced experimental methods, like real-time sub structuring involving complex numerical substructures, but also for conventional tests. Being able to quickly obtain accurate results of predictive analyses before testing and interpretation analyses after testing is of a paramount importance for successful experiments of models with complex behaviour. Predictive analyses are necessary to define the whole testing configuration (model geometry, boundary conditions, properties, input characteristics, measurement technology, sensor locations and calibration etc.). From the experimental point of view, interpretation analyses may be useful for the detection of problems or unexpected response that occurred during the test (e.g. unsatisfactory behaviour of sensors, actual boundary conditions different from that considered). This aspect is of particular importance in the case of series of tests where fast numerical interpretation analysis can be used as a tool of quality control between two successive tests.
Studied aspects
After having determined the general characteristics of the future experimental facility, several aspects have been studied related to its preliminary design. Amongst others the studies during the E-FAST project focussed on the following issues.
Technology
- The preliminary design of the hydraulic system has been accomplished. In particular the number and type of pumps, actuators and accumulators was determined. The cooling and piping systems were also determined.
- Several table geometries were studied which resulted in optimum weight / stiffness ratio. The interaction between table and specimen due to the table deformability was also investigated.
- For the reaction structure two alternative modular adaptive reaction systems are proposed. A modular steel reaction system and a modular reaction wall composed of reinforced concrete blocks assembled by means of pre-stressed rods.
- Taking into account the results of the design of the hydraulic system, the shaking tables and the reaction system, the geometry and dimensions of the reaction mass were determined. Soil-structure interaction analyses were carried out to estimate vibration nuisance in potential neighbouring buildings or facilities.
- Shaking table control methods were studied and comparison between some commercially available controllers was carried out.
- An overview and critical analysis of the most common approaches for telepresence rooms was carried out and a design proposal is made (annex II, section 4.1.3.5).
Advances in experimental techniques
- A considerable effort was devoted to real time substructure testing. Real time hybrid testing with linear or non-linear substructures were designed and carried out at the University of Kassel, EUCENTRE and CEA. A summary of the outcome of these tests are presented in annex I (section 4.1.3.4).
- Hardware for fast computer networking was tested.
- A no contact vision measurement system was developed. The accuracy of measurement sensors (e.g. load cells) necessary for real time substructure testing was also investigated and specific, high accuracy, load cells were designed when needed.
Dissemination and access
- Based on a comprehensive study, the design of a web portal enabling efficient access and networking was proposed taking into account both hardware and software aspects.
- Regarding physical access, based on the experience of the E-FAST partners, procedures related to the organisation of the test campaigns are proposed and the corresponding cost is estimated.
Management and operational issues
- On the basis of information given by manufacturers and experts in projects costs, the construction cost of the facility was estimated.
- The operation conditions (types of tests, tests' duration, maintenance, necessary staff, safety issues) were investigated and the operational cost was estimated.
- Alternative configurations with decreased performances (e.g. only one 6 DOF table instead of two etc.) were considered and the corresponding cost savings were estimated.
- The criteria for the optimum construction site of a future facility were determined.
- The potential risks were identified and their economic and time impact on the project was estimated.
- A schedule and road map for the detailed design and construction phase is proposed.
Annex I: Real time hybrid tests
a) Tests carried out at EUCENTRE
EUCENTRE / Italy conducted a real-time substructure test campaign. The reference system is an existing base isolated structure (figure 5a); such a structure is one of the new buildings, built immediately after the 2009, 6 April L'Aquila earthquake, to host the earthquake victims who were living close to the epicentre.
The structural system is made of a rectangular matrix of supporting columns directly seated on a concrete slab foundation; on the top of them, a Friction pendulum system (FPS) isolation device (figure 4b) supports a thick concrete slab which serves as basement of the building. Such structure is particularly suitable to be investigated by means of a real-time dynamic hybrid testing technique with sub-structuring (RTDHTwS), because the expected non-linearities are likely concentrated in a well identified portion (the base isolation system), which constitutes the physical sub-structure to tested experimentally. The rest of the structure is simulated numerically. Even if a numerical model of the experimental substructure is not strictly required, a fully simulated reference solution is useful to optimise the test setup and estimate the response of the system before doing the test.
Hardware and software setup
Compared to more conventional experimental tests, RTDHTwS, in general, require a more complex hardware and software architecture. In figure 5, the main components of the implemented system are sketched.
The xPC target can be a standard PC or Workstation, running the real-time operative system generated by xPC target toolbox from Mathworks. On xPC target the main system, the RT algorithm and the interface for the external communication as well, run in real-time. Everything is previously implemented on a non-RT windows-based PC (Host PC), with Matlab and Simulink software, then downloaded through Transmission control protocol / Internet protocol (TCP / IP) connection to the xPC. The Host PC works also as a Graphical user interface (GUI) for the xPC during the RT simulation. The Bearing testing system (BTS) of the Eucentre TREES lab is an experimental facility made of a 5-DOF shake table, situated beneath a vertical reaction structure which, combined with five vertical actuators, allows the application of the vertical load to the specimen. The experimental facility is controlled by a specifically designed MTS advanced Proportional-integral-derivative (PID) digital controller.
The communication loop of the whole system is made of numerical parts and a real time algorithm which send a command displacement, after a digital / analogue (D / A) conversion, to the BTS controller, which apply the command displacement and send back the feedback of displacement, acceleration, restoring force, to the real time machine.
Results
Since it is not possible to test physically the whole reference structure (superstructure + bearings) the experimental substructure tests are compared to the analytical results for the whole structure. Figure 6 shows that the substructure technique reproduces successfully the dynamic response of the system.
b) Tests carried out at UNIKA
UNIKA / Germany developed a test setup for real-time substructure testing using a hydraulic shaking table and a non-linear Tuned mass damper (TMD). Series of identification tests, reference tests and substructure tests have been performed. The feasibility of using the method of real-time substructure testing using shaking tables was assessed for E-FAST.
UNIKA modified its existing shaking table and TMD. The test system comprising a bi-directional shaking table and a non-linear TMD has been modified for 2-DOF substructure tests. The shaking table has been supplied with a leaf spring in order to use the shaking table as one DOF while the TMD serves as the second DOF (figures 7-8). New multi-directional load cells were designed and used in this test setup. With the new load cells, the interface between shaking table and specimen is represented in high resolution and the coupling force between experimental and numerical parts was reproduced and measured correctly. The constitutive law of the TMD may be adapted by means of a controllable friction device UHYDE-fbr (US Patent number 5456047).
Before carrying out substructure tests of the non-linear TMD using the shaking table, UNIKA performed identification tests to identify the test setup and carried out a series of reference tests for comparison between the response of the substructure tests and their respective references. UNIKA carried out more than 100 substructure tests to investigate the feasibility of the substructure algorithms and error compensation methods developed by UNIKA and the implications of these algorithms in the case of shaking table force real-time substructure testing. The substructure algorithm developed by Dorka was used successfully. The feasibility of the adaptive error force and adaptive phase lag compensations developed by Nguyen and Dorka was tested in real-time substructure testing using the controllable friction device UHYDE-fbr to produce different non-linear coupling effects depending on the applied pressure:
- no pressure: linear substructure;
- constant pressure: elastic-plastic coupling;
- pressure increasing or decreasing with displacement: bi-linear plastic coupling;
- sudden drop of pressure: simulated sudden partial failure of coupling;
- pressure depends on velocity: simulated viscous damping in coupling.
For example, figure 9 shows the comparison between substructure test variations and their reference test for the linear TMD (pressure in friction device p = 0). It demonstrates the effectiveness of the various compensation strategies but also highlights the need for phase lag compensation for a hydraulic system with low dynamic capacity.
In another test with sudden drop pressure, figure 10 shows that force compensation with P = 0.95 (sub117) can slightly reduce error and the adaptive force compensation (Sub118) does reduce significantly the error in the vicinity of the eigenfrequencies of the system (at 1.9 Hz and 3.1 Hz).
Conclusions
1. The substructure algorithm with sub-step control has been tested and the substructure solutions have shown that the algorithm provided very good accuracy and it is stable even in the particular test system with strong coupling and high nonlinearity.
2. Equilibrium errors may occur at the end of the time step, which can destabilise the test. They can be compensated by PID force compensation with a simple proportional gain or an adaptive compensator. The PID force compensation worked effectively and reduced the unbalanced force and thus improved the accuracy of the substructure response.
3. The adaptive force compensation was tested and it can reduce most error force in substructure tests, especially forces in the frequency range of the system (from 2 to 3 Hz). In some tests, although the error in frequency range of the system was well compensated by the adaptive force compensation, certain errors were in the higher frequency range (about 6 to 10 Hz, higher than the frequency range of the system) were large. However, this did not affect the response of the system very much.
4. Phase lag in hydraulic systems, in particular in typical shaking tables used for tests on civil engineering structures, is considerable and it causes large errors in the substructure tests. It may cause instability in other types of substructure tests where small damping is present in the numerical structure.
5. Adaptive phase lag compensation was used to compensate this error and the results show that the phase lag can be compensated efficiently by the newly developed adaptive phase lag compensation which does not work well for linear specimens only, but as these tests have shown, even in the presence of strong and sudden non-linearity. This is important for large hydraulic shaking tables, which all have large phase lag.
6. In light of these results, the new large tables envisioned for E-FAST may well be capable of real-time substructure testing.
c) Tests carried out at CEA
Model of hydraulic actuator
The first task associated to the experiments carried out at CEA / France was to improve the control of actuators through the development and validation of an accurate hydraulic actuators model. A non-linear model of hydraulic actuators has been developed. The goal is a) the numerical simulation of physical tests and b) a more control of real-time substructure tests by taking into account in the control loop the dynamics of the actuator. The model was validated through its comparison with experimental results for a large range of frequency and various excitation signals. The agreement between the model and the test is very satisfactory in the frequency range 0-30 Hz. Figure 11 shows an example of a time history sample demonstrating the accuracy of the model.
Real time hybrid tests
The second task was to perform two hybrid tests on a 2-DOF linear structure (a 2 storey steel fram and an oscillator (tuned mass damper, TMD) with the following substructure configurations:
- TMD as physical part and the 2 storey frame as numerical part;
- the 2 storey frame as physical part and the TMD as numerical part.
For the test shown in figure 13a (TMD as a physical substructure) a PID displacement control was used. It was observed that stability of the hybrid test was only possible for a narrow range of dynamic control gains. Therefore, offline tuning should be an important step of a hybrid test procedure. Figure 14 shows that the results of the hybrid tests do not fit very well the results of the reference test. Further work is needed to improve the hybrid test.
For the test shown in figure 13b ( 2-storey frame as a physical substructure) a PID force control was used. It was observed that the tuning of the gains was much easier than in the previous case. The results in figure 15 demonstrate an excellent agreement between the reference test of the whole system and the hybrid test.
Annex II: Telepresence room
One of the important issues which ensure the virtual dissemination of any advanced laboratory specialised in research and studies in earthquakes engineering is the issue of Telepresence, nowadays developed in some advanced laboratories, especially, in USA and Japan.
We have proposed in our research studies a 'display wall' where each display or group of displays will have a dedicated purpose. Due to the importance of the experiment itself the idea will be to place the displays related to it in the centre and the displays for the telepresence at margin. Table 2 is shows the basic concept we used in our studies for designing telepresence in E-FAST project.
Figure 16 presents a configuration with keyboards on the table. There are 1U rack specific keyboard and monitor units that are hidden in table and that can be opened as required. All supplementary devices like KVM switches, keyboard cable length extenders or echo cancelation units are placed under the table.
After market analysis and considering the proposed technical solutions by other earthquake engineering laboratories, the following technical data resulted for Telepresence system within the structure presented in figure 17.
For E-FAST the following minimal video structure is required due to the large dimensions of the laboratory itself:
- 4 high quality high speed cameras for the specimen;
- 8 high quality PTZ cameras for the specimen;
- 4 high quality PTZ cameras for videoconference component.
Other type of advanced video related specimen analysis (e.g. using laser based measurement systems or techniques based on pattern recognition).
For the telepresence room a high-definition quality solution must be elected. Due to the complexity of input design requirements the solution must be custom. In any combination this approach will further increase the basic costs. Nowadays, even 3D holography systems are available in the market the simulation software used by the earthquake engineering community is not yet prepared to interface it. The cost of high-definition telepresence system is high if we take into account the network hardware. If a medium quality telepresence solution is selected, the costs will dramatically decrease.
Potential impact:
Impact
The medium to long term impact of a new advanced seismic testing facility in Europe will be mainly the reduction of the seismic vulnerability. This will promote sustainable economic development of Europe's seismic regions, but also of the entire Europe, through savings on the total financial loss due to future earthquakes, to be shared by all EU Member States, rich or poor, seismic or not.
Actually, Europe as a whole, including Turkey, has about the same overall seismicity as in the United States or Japan. Among the about 490 million of European inhabitants, more than 20 million (4 % of the total) live in high seismicity areas and another 44 million (9% of the total) in moderate seismicity ones. If Turkey and the western Balkans are included, 41 million (7 % of the total population of about 580 million) live in high seismicity and 64 million (11 % of the total) in moderate seismicity areas. In the last two decades of the 20th century earthquakes caused about 5 000 casualties in the EU (4 500 of them in Italy) and about 19 000 in Turkey alone. The total 20-year toll should be contrasted to that of about 5 600 in Japan and just 130 in the US. During the 20th century earthquakes inflicted a total monetary loss estimated to USD 58 billion in Europe, USD 200 billion in the whole of Asia (most of it in Japan) and USD 46 billion in the Americas.
Regarding Europe and the neighboring countries, the earthquake disasters of 1980 in Irpinia (Italy), 1999 in Izmit (Turkey), 1999 in Athens (Greece) and 1989 in Spitak (Armenia) are among the most costly ones in history. As an example, only the 17 August 1999 Izmit, Turkey, earthquake caused over 18 373 deaths with injuries to another 48 901 people and destruction of immense proportions. There were reported 93 000 housing units destroyed and another 15 000 small business unites badly damaged. More recently the L'Aquila (Italy) 2009 earthquake had a considerable social and economic impact. The earthquake's life toll climbed up to 305 fatalities and thousands of injuries. There were displaced up to 25 000 people and the number of damaged buildings in the region of L'Aquila raised up to 10 000 buildings. The overall total costs of the L'Aquila earthquake, including financial losses and reconstruction costs could rise up to 16 billion.
As urbanisation increases fast in the very seismic areas of north-western Turkey and south-eastern Romania, the seismic risk in Europe will increase unless corrective and / or predictive measures are implemented. Keeping in mind that the most seismic parts of Europe (from west to east: Portugal, Southern Italy, the Balkans, Greece and Turkey) are also those having the longest road to convergence with the rest of Europe, an earthquake disaster in any of these economically more fragile areas will be a major setback in their course to convergence and may require the EU as a whole to foot part of the bill.
In general, the impact of severe earthquake events is related to:
a) Direct economic losses
In most cases, the direct economic losses are the most significant in terms of total loss from an earthquake disaster, mainly due to the damage of buildings and infrastructure. This may also be the type of earthquake economic impact easiest to measure due to the nature of the losses and to the fact that most of the buildings implicated have some sort of measurement method for the losses incurred (e.g. government statistics, private companies specialised in damages evaluation, insurance companies). These parties may also be combined for most cases.
This type of economic losses includes:
- structural damage;
- non-structural damage;
- damages to building contents and inventory;
- costs due to the shutting down of the building due to maintenance (e.g. relocation costs, lost rental income, lost wages, lost income).
The types of infrastructure involved in this section of economic losses are:
- transportation infrastructure (highways, roadways, airports, ports, light and heavy rail, buses, ferries;
- utilities damage (electric power, water, wastewater, communications, oil, natural gas).
The previously mentioned costs may also be increased due to revenue losses associated with outages, costs associated with providing backup services and possible fines for unavailability towards users.
b) Human impact
Perhaps the most important from a society point of view, the human impact may be major in the event of a catastrophic earthquake, both in terms of direct losses and injuries (or casualties) and in terms of the long-term social and economic impact.
The following situations may imply:
- loss of human lives;
- shelter (long-term and short-term);
- quality of life issues;
- healthcare and long-term mental impact;
- unemployment.
This range of economic issues may also be increased by the loss of the social capital. This term has been used when referring to the ties that an individual or community has with that certain place which was affected by the earthquake.
These ties may be:
- friendships;
- professional relationships;
- an internal sense of stability.
c) Emergency response and recovery costs
In a normal earthquake scenario, these costs may include:
- first-responder costs (personnel costs, including overtime) encompassing search and rescue, fire fighting, emergency medical services; police security at damage sites;
- service costs related to building damage, including post-earthquake building safety inspections (e.g. safety-tagging), emergency shoring and demolition, and debris removal.
In addition, measures including loans from the government, private insurance policies, grants etc. must be activated to contain the disaster.
d) Business interruption and other economic losses
The economic impact for businesses may become the most important part of recovering from an earthquake. The property damages, human casualties and the interruption of business activities may damage the economic environment consistently. For example, direct business interruption can result from building or equipment damage, utility outage, lack of employees (due to injury, displacement, or transportation interruption), or supplier interruption. These problems also lead to more cascading problems, due to the interruption of services, cancellation of previous orders from clients affected by the earthquake and the fact that employees affected by this may work less or less productively. Indirect or secondary losses are those incurred in the days, weeks, or months following a disaster and include losses due to business interruption caused by infrastructure disruption (e.g. electric power, gas, water), reduction of critical services to residents in hazard-prone areas, and psychological trauma. As an example, after the Niigata Cheuetsu-Oki earthquake (Japan, 2007) ,the Riken manufacturing company, one of the largest parts suppliers to major Japanese automaker, including Toyota and Honda, was shut down for two weeks because of non-structural damage to equipment. Toyota alone lost production of more than 120,000 cars in the first weeks after the earthquake. In addition, the Kashiwazaki-Kariwa nuclear power plant, which is the largest in the world, stopped completely its production for two years.
E-FAST will contribute to reduce all the above losses. Earthquake risk mitigation will be achieved through improvements of knowledge in earthquake engineering. In fact, progress in earthquake engineering cannot be made without experimental research on large scale models whose behaviour is similar to that of real structures. This new, advanced, versatile, high capacity experimental facility, which will be unique in Europe and one of the most important seismic testing facilities worldwide, will allow the European earthquake engineering community to make significant progress in Research and development (R&D) and will have considerable impact on the state of the art and the practice of aseismic design and construction methods and technology.
In fact, the advanced experimental capacities of E-FAST will result in:
- A further insight into the earthquake response behaviour of structures, in general.
- The improvement of the numerical simulation tools via their validation with the experimental results: Although there is a tremendous evolution of the analysis tools, there is still a lot of work to be done regarding their capacity to reproduce the behaviour of complex structures at a realistic scale and under realistic dynamic loading.
- The improvement and validation of regulations and recommendations: Actually it is widely admitted and confirmed by the feedback from past earthquakes that the role of design and construction codes is of paramount importance for the aseismic protection of human life and structures. Eurocodes are an important step towards this direction but, as recognised at the first international E-FAST workshop, there are still several aspects which need to be completed and / or validated.
- Safer design and qualification of industrial structures and equipment, especially those of nuclear and chemical industry: To this aspect, it is worth noting that the involvement of Europe in nuclear industry is twofold. Actually, some European countries run a large number of nuclear power plants and at the same time European companies are word leaders in the design and construction of nuclear power plants. The E-FAST facility will be able to carry out tests of structures and equipment for R&D and qualification purposes. In fact, special attention was given in the design so that the new facility will be able to test not only 'classical' civil engineering structures but other kinds of structures and equipment also. This will be possible thanks to the high acceleration capacities of the tables (which enables to take into account floor response amplification), the 3D excitation capability, the multi-support excitation capabilities (e.g. tests of equipment such as pipes, crane bridges), the large displacement capability (which avoids filtering of low frequencies) and the large operational frequency bandwidth.
The development and validation of new construction methods, materials and devices which will improve the protection level of structures against earthquakes and lead to a better performance / cost ratio of aseismic structures.
- The support of European companies through the aforementioned validation of new technologies and design concepts. The R&D and qualification tests related to the activities of European companies will gain them with international prestige thus contributing to their export policy. E-FAST will secure and enhance the competitive edge that European construction firms and engineering services currently hold in overseas seismic markets, by establishing Europe as a world leader in earthquake engineering research.
- Demonstration tests for policy makers and dissemination purposes. The benefit from such demonstration tests is well understood in the US and especially in Japan. The high capacities of E-FAST will enable carrying out tests of realistic models which have considerable effects on public awareness.
- A 'natural' excellence centre for advanced knowledge in earthquake engineering through training, transnational access and dissemination. Within the framework of national and international collaborations E-FAST will be a crossroads for exchanging ideas and training on experimental earthquake engineering and in earthquake engineering in general. It will also promote, through networking and distributed databases, a wider sharing of data and knowledge across the field of earthquake engineering and between academia, research and industry.
List of websites: http://E-FAST.eknowrisk.eu
High performance experimental facilities are necessary to meet the objectives of earthquake risk mitigation and to make progress in methods for the design and assessment of buildings and infrastructures. Therefore, in order to be positioned within the avant-garde of earthquake research it is important for Europe to build a new high performance experimental facility. For that reason, the European Commission (EC) supported, as a part of the Seventh Framework Programme (FP7), a design study of a new generation seismic testing facility. This is the E-FAST (European Facility for Advanced Seismic Testing) project. Five European partners with a large experience in seismic and dynamic testing were involved in E-FAST: Commissariat à lEnergie Atomique (coordinator, France), the Gheorghe Asachi Technical University of Iasi (Romania), Eucentre (Italy), the University of Kassel (Germany) and the joint research centre (EC).
The first step was the determination of the performance requirements of the new facility. Then a lay-out of was proposed which goes much further than existing shaking table facilities in Europe. It meets the requirements of modularity, flexibility and operational ease with technological choices that minimise the techno-economic risk. Several aspects have been studied related to the preliminary design. Amongst others the studies during the E-FAST project focussed on the following issues:
Technology: design of the hydraulic power supply system, shake tables, reaction mass, modular reaction structure, telepresence room, evaluation of shake table control methods.
- Advances in experimental techniques: carrying out of real time substructure tests with linear and non-linear physical substructures, evaluation of hardware for fast computer networking, development and utilisation of a no-contact vision measurement system.
- Dissemination and access: design of a web portal enabling efficient access and networking, evaluation of the cost of physical access (based on the cost related to the organisation of test campaigns).
- Management and operational issues: overall construction cost estimate, study of operational conditions (types of tests, tests duration, maintenance, necessary staff, safety issues) and operational cost, consideration of alternative configurations with decreased performances and evaluation of the corresponding cost savings, identification of the potential project risks and evaluation of their impact, proposal of a schedule and road map for the detailed design and construction phase, determination of criteria for the optimum construction site of a future facility.
The medium to long term impact of a new advanced seismic testing facility in Europe will be mainly the reduction of the seismic vulnerability. This will promote sustainable economic development of Europe's seismic regions, but also of the entire Europe, through savings on the total financial loss due to future earthquakes. In fact, the enhanced capabilities of an advanced testing facility will lead to:
- a further insight into the earthquake response behaviour of structures, in general;
- the improvement of the numerical simulation tools via their validation with the experimental results;
- the improvement and validation of regulations and recommendations;
- safer design and qualification of industrial structures and equipment, especially those of nuclear and chemical industry;
- the development and validation of new construction methods, materials and devices;
- the support of European companies through the aforementioned validation of new technologies and design concepts;
- demonstration tests for policy makers and public awareness and dissemination purposes;
- a 'natural' excellence centre for advanced knowledge in earthquake engineering through training, transnational access and dissemination.
Project context and objectives:
Seismic events of the recent past have proved that European and the neighbouring countries, especially those comprised in the Mediterranean area, are exposed to a high seismic risk. Surprisingly the number of victims and the overall economic losses are important compared to industrialised country like Japan and United States often faced with higher levels of shaking. This fact can be explained by considering the higher population density, and that the high number of damaged buildings is due to the large presence of monuments or ancient masonry buildings often vulnerable to earthquake loading. It is readily apparent that, in developed countries, although the numbers of victims of major earthquakes is tending to drop, the costs of the consequences are constantly rising. The costs of the consequences, resulting in significant damage and widespread disorganisation in the area, are constantly rising. Considering damage to plants, loss of data and drops in productivity, are extremely costly. A recent example is the social and economic impact of the L'Aquila (Italy) earthquake in April 2009. It is therefore indispensable for Europe to intensify the research and development in the field of earthquake engineering.
In the last decades considerable advances have been achieved in the earthquake engineering field. The research results have contributed to the preparation of the modern design codes, to the identification of several problems in the existing structures and to innovative solutions for the structural assessment. Despite this huge amount of improvements there are still several open problems. For instance, predictive models have frequently been calibrated on the experimental results obtained from scaled structures, several innovative technologies for building constructions are entering the market and require careful evaluations to verify the level of safety, the experimental validation of the behaviour of large infrastructures (bridges or retaining walls) often requires multi-support excitation, a further insight into soil structure interaction requires testing of heavy models. Available data and future results need to be organised in databases, in order to disseminate them, optimise their use, and provide relevant information for risk oriented approaches. These researches require a large amount of analytical and experimental studies. Moreover, a look at the international earthquake engineering landscape reveals that, outside Europe, there are several high performance seismic testing facilities either already operating or under construction. As an example, in 2000-04 Japan, already boasting the most powerful experimental RTD infrastructures in earthquake engineering, spent 350 million to build the largest three-dimensional (3D) shaking table in the world (20mx15m, 1200t payload). Regarding pseudo-dynamic testing, the ELSA laboratory of the joint research centre of EC in Italy with its reaction wall 16 m high and 20 m long is one of the main seismic testing facilities in the world. However, the situation is different for European shaking tables facilities, having considerably lower performances than that of the major shaking tables laboratories in the world. There is also a trend, mainly in the United States and Asia (China, Japan, Korea) towards facilities with an array of shaking tables which increases operating ease and enable multi-support excitations. The objective is to test structures at the largest possible scale in order to avoid scaling effects.
This means that, unavoidably, if the situation does not change, Europe will cumulate a considerable lag in experimental earthquake engineering, with respect to the USA and Asian countries. Europe should not trail these foreign countries, in particular United States and Japan, in experimental research in earthquake engineering and rely on their Research and technology development (RTD) results. It should compete and share results with them as equal, to serve its own needs and promote its own interests. To see the reasons, we should recall first the difference of buildings in Europe from those in Japan or the United States. In Europe new residential construction uses mostly concrete framing, often with masonry infills, especially in the seismic southern countries. Europe is also rich in cultural heritage buildings, mainly of masonry construction. By contrast, in US and Japan heritage buildings are not common and new buildings are mostly of timber. Moreover, concrete and masonry construction in US and Japan is very different from Europe. So, the focus of RTD in these countries does not fully serve Europe's needs. In the other major type of civil engineering works, namely civil infrastructures, the technology and the materials are fully global. There, Europe is the world leader: in niche technologies (post-tensioning, stay cables, marine or off-shore construction, etc.), in overseas construction (it boasts the world top firms: Bouygues, Dragados, Ferrovial, Hochtief, Vinci) and in overseas consultancy and design, with its huge engineering services sector, etc. As most of the overseas activity is in seismic areas (east, south or south-east Asia, central Asia, northern Africa, Middle East and Latin America), European construction firms and engineering services cannot retain their competitive edge in seismic markets and their reputation as leaders in technology, unless the EU as a whole establishes itself as equal to the USA and Japan in earthquake engineering RTD.
In addition, it is worth noting that there is an emergence of advanced experimental techniques, such as real-time substructuring and advanced measurement techniques that are being explored in the most innovative laboratories. This is an important point since the new experimental methods, based on the substructuring technique, have the advantage of reducing the specimen size allowing a better use of the hardware resources.
For all the aforementioned reasons a new platform for dynamic seismic testing in Europe is not just useful but necessary. A new high performance testing facility will enable studying a large variety of structures and systems. In fact, such a facility is an indispensable tool to calibrate and validate new conceptual approaches in modelling and simulations developed for performance based analysis and design of new structures or retrofitting interventions of safe structures in Europe and even worldwide. It will also contribute to increase worldwide the competitiveness of European science and industry.
Therefore, EC granted, as a part of the seventh framework project, the design study project E-FAST. The main objectives of E-FAST are:
- define the needs in experimental research in earthquake engineering in Europe;
- define the features of a new testing facility, complementary to existing research infrastructures in Europe, combining high capacity, flexibility and operational ease;
- make progress in advanced testing methods such as real time sub-structuring techniques and carry out demonstration tests;
- study the technical feasibility of this facility;
- study financial issues related to the construction cost, operating and maintenance cost and access cost.
Project results:
Required general performances of a new European seismic testing facility
The first international E-FAST workshop pointed out some of the fields that need further experimental research. In particular, more experimental evidence is needed in the following topics:
- in plan irregular buildings exhibiting torsion response;
- precast and prestressed concrete elements and systems;
- masonry buildings and infills. In particular more experimental data of buildings with more than one storey are needed;
- infrastructures (e.g. bridges implying multi-support excitation capability);
- retrofitting;
- aseismic devices (e.g. isolation bearings, dampers);
- equipment and components. In particular the motion of the floor the equipment / component is mounted on should be reproduced implying high acceleration and displacement capacity;
- soil-structure interaction Due to the considerable weight of such models only elementary configuration could be tested. In any case a high payload table is required.
One common point to all classes of problems is that, in order to conduct a meaningful risk assessment, the actual available margins of structures have to be estimated. This holds for all structures but it is even more critical for structures of major importance (e.g. power generation facilities, hospitals). To this end, tests with excitation level up to failure should be possible in future. Depending on the tested structure of interest (building, equipment or secondary structure), failure can be defined as loss of operational function, collapse or relevant significant damage or collapse. This implies that the new facility should have the capability to apply high intensity excitations (high acceleration, velocity and displacement) to models which will be representative of the prototype structures. Since the pseudo-dynamic testing facility at the ELSA laboratory of the joint research centre of EC in Italy, with its reaction wall 16 m high and 20 m long, is considered as one of the main seismic testing facilities in the world, it is proposed that the new facility should be, mainly, a new generation shaking table facility with the possibility to apply advanced experimental techniques like real time hybrid testing also. In addition, the new facility should comply as far as possible with the requirements of flexibility, adaptability and operational ease.
Table 1 shows some indicative performance parameters for different classes of tests. The given numbers are reasonable rough estimates as a trade-off between needed performance and cost. Obviously it is not feasible, either for technological or financial reasons, to build a facility so big that everything could be tested therein. The objective is to propose a design that will enable to carry out meaningful tests using conventional and / or more recent techniques and technologies which have already demonstrated their efficacy and reliability. The acceleration values in table 1 may seem to be unrealistically high. However, it is worth noting that a) several records of real earthquakes revealed very high acceleration values (e.g. 0.98 g Northridge earthquake, 1994, 0.85g Kobe, 1995) b) in the case of scale models, if a velocity similitude is considered, the table acceleration should be multiplied by the inverse of the scale ratio i.e. table acceleration will be higher than ground acceleration of the prototype and c) the values in table 1 are conventional acceleration values corresponding to a rigid specimen. Consideration of the dynamic amplification of the specimen results in a higher demand of force capacity which is equivalent to a higher demand of conventional acceleration capacity. In the case of soil-structure interaction tests, the major part of the mass on the table is due to the weight of the soil itself and its container which will have, in general, a weak dynamic amplification. Therefore, in that case the required shaking table acceleration could be smaller. Regarding secondary structures and equipment, because of the amplification of the shaking motion at the floor level, floor accelerations to reproduce on the table may be much higher than ground accelerations.
Velocity values are also in agreement with actual recorded velocities (e.g. 1.4 m / s Northridge earthquake, 1994, 1.5 m / s Kobe, 1995). High displacement values are also necessary for the shaking table motion to be representative of strong, low frequency ground motions or of floor motions of low frequency buildings (in the case of secondary structures or equipment tests).
General layout
A lay-out of the facility is proposed which meets the above performance requirements. The underlying philosophy is resumed to the following points:
- The new facility should be a significant step ahead and go much further than existing shaking table facilities in Europe.
- It should be a combination of components and technology that have already been validated by their operational use in other existing facilities. In fact E-FAST is a design study not a pure Research and development (R&D) project therefore there was no room in E-FAST for innovation and adoption of 'revolutionary' technology and techniques. Moreover, the choice of well validated technological solution is imposed because of:
1. Technical reasons: In fact the new facility should be able to carry out accurately big scale, high capacity demanding tests from the 1st day of its operation. A less or more long period of adaptation and / or changes to achieve this goal (accurate big scale seismic testing) is not acceptable.
2. Safety reasons
3. The requirement to minimise the techno-economic risk for potential investors. Actually potential investors desire to minimise divergence from the foreseen date of operating start and from the foreseen budget at the moment of their commitment.
The design is based on the feedback from:
1. those of the E-FAST partners running big shake tables;
2. some of the most experienced manufactures of shake tables in the world.
The proposed solution is a trade-off between dreams and real world (performance vs. cost). In fact, too high cost (construction, maintenance, handling, specimen transport) would kill any chance for the facility to be constructed. In addition, given that the proposed configuration is composed of up-to-date elements and technology but already existing and operating allows us to make a construction and operational cost estimate with a good accuracy.
The general lay-out of the facility is shown in figure 1 to figure 3. The facility consists mainly of:
1. Two 6 Degrees of freedom (DOF) 6 m x 6 m shaking tables: The payload of each table is of about 100 tonnes. The tables can be positioned in any place within the trench. The gap between them can vary from 0 to 20 m. They will be able to operate independently or be linked and operate as a single table with a payload of 200 tonnes. They will be able to have a synchronous or asynchronous motion.
The two shaking tables allow the following configurations:
- two separate tables operating independently (adjustable distance between the 2 tables from 0 m to 20 m);
- two separate tables but working together, linked by means of a special truss or plateau and supporting a large specimen (distance between 2 tables adjustable from 0 m to 20 m);
- two separate tables, operating simultaneously but with different motions to test multi-supported structures (distance between 2 tables adjustable from 0m to 20m);
- two tables linked together to realise a large table of 6 m x 12 m;
- one or two tables (in this case fixed rigidly to each other) can be mounted on bearings fixed at the bottom of the pit to operate as a single axial table (1 DOF along the length of the pit). This would enable to test even higher models since the height of the specimen could be equal to the height of the hall (distance between the ground level and the bottom of the crane) plus the depth of the pit.
Obviously a higher number of tables, possibly in different trenches, would enable testing more complex configurations. However as already mentioned, the proposed facility is a trade-off between performance-capacity and cost, that is why only two tables are proposed.
Their performances are summarised below:
- maximum horizontal displacement ± 1 m in OX and OY;
maximum vertical displacement ± 0,75 m in OZ;
- maximum horizontal velocity 2 m / s in OX and OY;
- maximum vertical velocity 1.5 m / s in OZ;
- maximum acceleration 2 g in OX and OY;
- maximum acceleration 1.5 g in OZ;
- maximum duration of excitation at full power 30 seconds.
If necessary, the maximum payload of the two 6 DOF tables could be increased with minor modifications of the actuators (slight increase of their length providing higher shock absorption capacity). However, unless the actuators' capacity increases, this will be done at the price of a smaller maximum acceleration at full payload.
During periods of maintenance or repair, the tables can be uncoupled from the horizontal and vertical cylinders and stored temporarily on the strong-floor or on the outdoor area (cf. item 6). Dedicated trusses / frames will maintain in a vertical or horizontal position the cylinders during maintenance (those frames can be the same which will be used for the initial placement of the cylinders in the pit).
2. One 1 DOF (horizontal) shaking table 11 m x 11 m with high payload of about 500 tonnes: It will be mounted on hydrostatic bearings fixed at the upper part of the pit walls. This shaking table will be intended for heavy specimen and in particular for soil-structure interaction tests. The maximum acceleration at full payload will be of about 0.6g - 1g. Actually in a first step for economic reasons we consider that this table will be actuated by the actuators of the two moveable tables. Therefore the achieved acceleration will depend on the number of the actuators which will be utilised. This big shaking table will not be in the pit permanently but it will be mounted on when a mono-axial test of a heavy specimen must be carried out. Obviously, in the case of configurations where the gap between the two 6m x 6m shaking tables is 20 m, the 11 m x 11 m table should be taken out of the pit. To this end the crane was designed to have the capacity to lift up such a heavy structure.
3. A reaction mass which consists, mainly, of a pit that hosts the aforementioned shaking tables, a room (of about 30 m x 20 m x 4.5 m) hosting pumps and accumulators and a big strong floor area: This thick (2 m) strong floor slab gives enhanced adaptability to the facility. In fact several experimental set-ups (small to medium shake tables, dedicated testing machines, reaction structures) can be mounted on it. The necessary power will be supplied to the actuators at any position by means of hoses. The reaction mass weighs about 25 000 tonnes. Vibration nuisance analyses showed that vibration isolation of the reaction mass by means of specific devices (e.g. springs and dampers) is not necessary. Therefore and to avoid the considerable cost increase associated with isolation devices the reaction mass is put directly on the ground.
4. A modular reaction structure, which can be placed anywhere around the pit, allowing for real-time hybrid testing possibly combining shaking tables and actuators attached on the wall.
5. A hydraulic system composed of piping, actuators, pumps and accumulators with a capacity consistent to the performance criteria given in table 1.
6. An outdoor area devoted to the construction of specimens, especially reinforced concrete or masonry models;
7. A high capacity crane bridge with 4 hooks each one having a capacity of 50 t (total capacity 200 t). It spans the whole width of the working area and it can move along the whole length of the hall and the outdoor area. The crane will be able to lift and transport heavy models from the construction area and install them on the tables and vice versa. Its capacity allows also handling of the shake tables, even of the big mono-axial table. The crane bridge includes also a cantilever crane of a capacity of 20 t which will be used for handling low to medium weight items (pipes, actuators, light specimens). The foundations of the crane frame rails are designed to be independent of the foundations of the hall and the offices building.
8. In addition to the experimental hall (overall dimensions LxWxH=47 m x 42 m x 19 m) a 2-storys building (offices, control room, meeting room, teleconference room) for about 40 persons of about 1000 m² per story is foreseen.
It is worth noting, that though not investigated here because it was beyond the scope of this design study, a strong interaction between the aforementioned 'purely experimental' facility and a numerical high computational capability facility, on site or remote, is necessary. Actually, high performance and accuracy numerical simulation is necessary not only for advanced experimental methods, like real-time sub structuring involving complex numerical substructures, but also for conventional tests. Being able to quickly obtain accurate results of predictive analyses before testing and interpretation analyses after testing is of a paramount importance for successful experiments of models with complex behaviour. Predictive analyses are necessary to define the whole testing configuration (model geometry, boundary conditions, properties, input characteristics, measurement technology, sensor locations and calibration etc.). From the experimental point of view, interpretation analyses may be useful for the detection of problems or unexpected response that occurred during the test (e.g. unsatisfactory behaviour of sensors, actual boundary conditions different from that considered). This aspect is of particular importance in the case of series of tests where fast numerical interpretation analysis can be used as a tool of quality control between two successive tests.
Studied aspects
After having determined the general characteristics of the future experimental facility, several aspects have been studied related to its preliminary design. Amongst others the studies during the E-FAST project focussed on the following issues.
Technology
- The preliminary design of the hydraulic system has been accomplished. In particular the number and type of pumps, actuators and accumulators was determined. The cooling and piping systems were also determined.
- Several table geometries were studied which resulted in optimum weight / stiffness ratio. The interaction between table and specimen due to the table deformability was also investigated.
- For the reaction structure two alternative modular adaptive reaction systems are proposed. A modular steel reaction system and a modular reaction wall composed of reinforced concrete blocks assembled by means of pre-stressed rods.
- Taking into account the results of the design of the hydraulic system, the shaking tables and the reaction system, the geometry and dimensions of the reaction mass were determined. Soil-structure interaction analyses were carried out to estimate vibration nuisance in potential neighbouring buildings or facilities.
- Shaking table control methods were studied and comparison between some commercially available controllers was carried out.
- An overview and critical analysis of the most common approaches for telepresence rooms was carried out and a design proposal is made (annex II, section 4.1.3.5).
Advances in experimental techniques
- A considerable effort was devoted to real time substructure testing. Real time hybrid testing with linear or non-linear substructures were designed and carried out at the University of Kassel, EUCENTRE and CEA. A summary of the outcome of these tests are presented in annex I (section 4.1.3.4).
- Hardware for fast computer networking was tested.
- A no contact vision measurement system was developed. The accuracy of measurement sensors (e.g. load cells) necessary for real time substructure testing was also investigated and specific, high accuracy, load cells were designed when needed.
Dissemination and access
- Based on a comprehensive study, the design of a web portal enabling efficient access and networking was proposed taking into account both hardware and software aspects.
- Regarding physical access, based on the experience of the E-FAST partners, procedures related to the organisation of the test campaigns are proposed and the corresponding cost is estimated.
Management and operational issues
- On the basis of information given by manufacturers and experts in projects costs, the construction cost of the facility was estimated.
- The operation conditions (types of tests, tests' duration, maintenance, necessary staff, safety issues) were investigated and the operational cost was estimated.
- Alternative configurations with decreased performances (e.g. only one 6 DOF table instead of two etc.) were considered and the corresponding cost savings were estimated.
- The criteria for the optimum construction site of a future facility were determined.
- The potential risks were identified and their economic and time impact on the project was estimated.
- A schedule and road map for the detailed design and construction phase is proposed.
Annex I: Real time hybrid tests
a) Tests carried out at EUCENTRE
EUCENTRE / Italy conducted a real-time substructure test campaign. The reference system is an existing base isolated structure (figure 5a); such a structure is one of the new buildings, built immediately after the 2009, 6 April L'Aquila earthquake, to host the earthquake victims who were living close to the epicentre.
The structural system is made of a rectangular matrix of supporting columns directly seated on a concrete slab foundation; on the top of them, a Friction pendulum system (FPS) isolation device (figure 4b) supports a thick concrete slab which serves as basement of the building. Such structure is particularly suitable to be investigated by means of a real-time dynamic hybrid testing technique with sub-structuring (RTDHTwS), because the expected non-linearities are likely concentrated in a well identified portion (the base isolation system), which constitutes the physical sub-structure to tested experimentally. The rest of the structure is simulated numerically. Even if a numerical model of the experimental substructure is not strictly required, a fully simulated reference solution is useful to optimise the test setup and estimate the response of the system before doing the test.
Hardware and software setup
Compared to more conventional experimental tests, RTDHTwS, in general, require a more complex hardware and software architecture. In figure 5, the main components of the implemented system are sketched.
The xPC target can be a standard PC or Workstation, running the real-time operative system generated by xPC target toolbox from Mathworks. On xPC target the main system, the RT algorithm and the interface for the external communication as well, run in real-time. Everything is previously implemented on a non-RT windows-based PC (Host PC), with Matlab and Simulink software, then downloaded through Transmission control protocol / Internet protocol (TCP / IP) connection to the xPC. The Host PC works also as a Graphical user interface (GUI) for the xPC during the RT simulation. The Bearing testing system (BTS) of the Eucentre TREES lab is an experimental facility made of a 5-DOF shake table, situated beneath a vertical reaction structure which, combined with five vertical actuators, allows the application of the vertical load to the specimen. The experimental facility is controlled by a specifically designed MTS advanced Proportional-integral-derivative (PID) digital controller.
The communication loop of the whole system is made of numerical parts and a real time algorithm which send a command displacement, after a digital / analogue (D / A) conversion, to the BTS controller, which apply the command displacement and send back the feedback of displacement, acceleration, restoring force, to the real time machine.
Results
Since it is not possible to test physically the whole reference structure (superstructure + bearings) the experimental substructure tests are compared to the analytical results for the whole structure. Figure 6 shows that the substructure technique reproduces successfully the dynamic response of the system.
b) Tests carried out at UNIKA
UNIKA / Germany developed a test setup for real-time substructure testing using a hydraulic shaking table and a non-linear Tuned mass damper (TMD). Series of identification tests, reference tests and substructure tests have been performed. The feasibility of using the method of real-time substructure testing using shaking tables was assessed for E-FAST.
UNIKA modified its existing shaking table and TMD. The test system comprising a bi-directional shaking table and a non-linear TMD has been modified for 2-DOF substructure tests. The shaking table has been supplied with a leaf spring in order to use the shaking table as one DOF while the TMD serves as the second DOF (figures 7-8). New multi-directional load cells were designed and used in this test setup. With the new load cells, the interface between shaking table and specimen is represented in high resolution and the coupling force between experimental and numerical parts was reproduced and measured correctly. The constitutive law of the TMD may be adapted by means of a controllable friction device UHYDE-fbr (US Patent number 5456047).
Before carrying out substructure tests of the non-linear TMD using the shaking table, UNIKA performed identification tests to identify the test setup and carried out a series of reference tests for comparison between the response of the substructure tests and their respective references. UNIKA carried out more than 100 substructure tests to investigate the feasibility of the substructure algorithms and error compensation methods developed by UNIKA and the implications of these algorithms in the case of shaking table force real-time substructure testing. The substructure algorithm developed by Dorka was used successfully. The feasibility of the adaptive error force and adaptive phase lag compensations developed by Nguyen and Dorka was tested in real-time substructure testing using the controllable friction device UHYDE-fbr to produce different non-linear coupling effects depending on the applied pressure:
- no pressure: linear substructure;
- constant pressure: elastic-plastic coupling;
- pressure increasing or decreasing with displacement: bi-linear plastic coupling;
- sudden drop of pressure: simulated sudden partial failure of coupling;
- pressure depends on velocity: simulated viscous damping in coupling.
For example, figure 9 shows the comparison between substructure test variations and their reference test for the linear TMD (pressure in friction device p = 0). It demonstrates the effectiveness of the various compensation strategies but also highlights the need for phase lag compensation for a hydraulic system with low dynamic capacity.
In another test with sudden drop pressure, figure 10 shows that force compensation with P = 0.95 (sub117) can slightly reduce error and the adaptive force compensation (Sub118) does reduce significantly the error in the vicinity of the eigenfrequencies of the system (at 1.9 Hz and 3.1 Hz).
Conclusions
1. The substructure algorithm with sub-step control has been tested and the substructure solutions have shown that the algorithm provided very good accuracy and it is stable even in the particular test system with strong coupling and high nonlinearity.
2. Equilibrium errors may occur at the end of the time step, which can destabilise the test. They can be compensated by PID force compensation with a simple proportional gain or an adaptive compensator. The PID force compensation worked effectively and reduced the unbalanced force and thus improved the accuracy of the substructure response.
3. The adaptive force compensation was tested and it can reduce most error force in substructure tests, especially forces in the frequency range of the system (from 2 to 3 Hz). In some tests, although the error in frequency range of the system was well compensated by the adaptive force compensation, certain errors were in the higher frequency range (about 6 to 10 Hz, higher than the frequency range of the system) were large. However, this did not affect the response of the system very much.
4. Phase lag in hydraulic systems, in particular in typical shaking tables used for tests on civil engineering structures, is considerable and it causes large errors in the substructure tests. It may cause instability in other types of substructure tests where small damping is present in the numerical structure.
5. Adaptive phase lag compensation was used to compensate this error and the results show that the phase lag can be compensated efficiently by the newly developed adaptive phase lag compensation which does not work well for linear specimens only, but as these tests have shown, even in the presence of strong and sudden non-linearity. This is important for large hydraulic shaking tables, which all have large phase lag.
6. In light of these results, the new large tables envisioned for E-FAST may well be capable of real-time substructure testing.
c) Tests carried out at CEA
Model of hydraulic actuator
The first task associated to the experiments carried out at CEA / France was to improve the control of actuators through the development and validation of an accurate hydraulic actuators model. A non-linear model of hydraulic actuators has been developed. The goal is a) the numerical simulation of physical tests and b) a more control of real-time substructure tests by taking into account in the control loop the dynamics of the actuator. The model was validated through its comparison with experimental results for a large range of frequency and various excitation signals. The agreement between the model and the test is very satisfactory in the frequency range 0-30 Hz. Figure 11 shows an example of a time history sample demonstrating the accuracy of the model.
Real time hybrid tests
The second task was to perform two hybrid tests on a 2-DOF linear structure (a 2 storey steel fram and an oscillator (tuned mass damper, TMD) with the following substructure configurations:
- TMD as physical part and the 2 storey frame as numerical part;
- the 2 storey frame as physical part and the TMD as numerical part.
For the test shown in figure 13a (TMD as a physical substructure) a PID displacement control was used. It was observed that stability of the hybrid test was only possible for a narrow range of dynamic control gains. Therefore, offline tuning should be an important step of a hybrid test procedure. Figure 14 shows that the results of the hybrid tests do not fit very well the results of the reference test. Further work is needed to improve the hybrid test.
For the test shown in figure 13b ( 2-storey frame as a physical substructure) a PID force control was used. It was observed that the tuning of the gains was much easier than in the previous case. The results in figure 15 demonstrate an excellent agreement between the reference test of the whole system and the hybrid test.
Annex II: Telepresence room
One of the important issues which ensure the virtual dissemination of any advanced laboratory specialised in research and studies in earthquakes engineering is the issue of Telepresence, nowadays developed in some advanced laboratories, especially, in USA and Japan.
We have proposed in our research studies a 'display wall' where each display or group of displays will have a dedicated purpose. Due to the importance of the experiment itself the idea will be to place the displays related to it in the centre and the displays for the telepresence at margin. Table 2 is shows the basic concept we used in our studies for designing telepresence in E-FAST project.
Figure 16 presents a configuration with keyboards on the table. There are 1U rack specific keyboard and monitor units that are hidden in table and that can be opened as required. All supplementary devices like KVM switches, keyboard cable length extenders or echo cancelation units are placed under the table.
After market analysis and considering the proposed technical solutions by other earthquake engineering laboratories, the following technical data resulted for Telepresence system within the structure presented in figure 17.
For E-FAST the following minimal video structure is required due to the large dimensions of the laboratory itself:
- 4 high quality high speed cameras for the specimen;
- 8 high quality PTZ cameras for the specimen;
- 4 high quality PTZ cameras for videoconference component.
Other type of advanced video related specimen analysis (e.g. using laser based measurement systems or techniques based on pattern recognition).
For the telepresence room a high-definition quality solution must be elected. Due to the complexity of input design requirements the solution must be custom. In any combination this approach will further increase the basic costs. Nowadays, even 3D holography systems are available in the market the simulation software used by the earthquake engineering community is not yet prepared to interface it. The cost of high-definition telepresence system is high if we take into account the network hardware. If a medium quality telepresence solution is selected, the costs will dramatically decrease.
Potential impact:
Impact
The medium to long term impact of a new advanced seismic testing facility in Europe will be mainly the reduction of the seismic vulnerability. This will promote sustainable economic development of Europe's seismic regions, but also of the entire Europe, through savings on the total financial loss due to future earthquakes, to be shared by all EU Member States, rich or poor, seismic or not.
Actually, Europe as a whole, including Turkey, has about the same overall seismicity as in the United States or Japan. Among the about 490 million of European inhabitants, more than 20 million (4 % of the total) live in high seismicity areas and another 44 million (9% of the total) in moderate seismicity ones. If Turkey and the western Balkans are included, 41 million (7 % of the total population of about 580 million) live in high seismicity and 64 million (11 % of the total) in moderate seismicity areas. In the last two decades of the 20th century earthquakes caused about 5 000 casualties in the EU (4 500 of them in Italy) and about 19 000 in Turkey alone. The total 20-year toll should be contrasted to that of about 5 600 in Japan and just 130 in the US. During the 20th century earthquakes inflicted a total monetary loss estimated to USD 58 billion in Europe, USD 200 billion in the whole of Asia (most of it in Japan) and USD 46 billion in the Americas.
Regarding Europe and the neighboring countries, the earthquake disasters of 1980 in Irpinia (Italy), 1999 in Izmit (Turkey), 1999 in Athens (Greece) and 1989 in Spitak (Armenia) are among the most costly ones in history. As an example, only the 17 August 1999 Izmit, Turkey, earthquake caused over 18 373 deaths with injuries to another 48 901 people and destruction of immense proportions. There were reported 93 000 housing units destroyed and another 15 000 small business unites badly damaged. More recently the L'Aquila (Italy) 2009 earthquake had a considerable social and economic impact. The earthquake's life toll climbed up to 305 fatalities and thousands of injuries. There were displaced up to 25 000 people and the number of damaged buildings in the region of L'Aquila raised up to 10 000 buildings. The overall total costs of the L'Aquila earthquake, including financial losses and reconstruction costs could rise up to 16 billion.
As urbanisation increases fast in the very seismic areas of north-western Turkey and south-eastern Romania, the seismic risk in Europe will increase unless corrective and / or predictive measures are implemented. Keeping in mind that the most seismic parts of Europe (from west to east: Portugal, Southern Italy, the Balkans, Greece and Turkey) are also those having the longest road to convergence with the rest of Europe, an earthquake disaster in any of these economically more fragile areas will be a major setback in their course to convergence and may require the EU as a whole to foot part of the bill.
In general, the impact of severe earthquake events is related to:
a) Direct economic losses
In most cases, the direct economic losses are the most significant in terms of total loss from an earthquake disaster, mainly due to the damage of buildings and infrastructure. This may also be the type of earthquake economic impact easiest to measure due to the nature of the losses and to the fact that most of the buildings implicated have some sort of measurement method for the losses incurred (e.g. government statistics, private companies specialised in damages evaluation, insurance companies). These parties may also be combined for most cases.
This type of economic losses includes:
- structural damage;
- non-structural damage;
- damages to building contents and inventory;
- costs due to the shutting down of the building due to maintenance (e.g. relocation costs, lost rental income, lost wages, lost income).
The types of infrastructure involved in this section of economic losses are:
- transportation infrastructure (highways, roadways, airports, ports, light and heavy rail, buses, ferries;
- utilities damage (electric power, water, wastewater, communications, oil, natural gas).
The previously mentioned costs may also be increased due to revenue losses associated with outages, costs associated with providing backup services and possible fines for unavailability towards users.
b) Human impact
Perhaps the most important from a society point of view, the human impact may be major in the event of a catastrophic earthquake, both in terms of direct losses and injuries (or casualties) and in terms of the long-term social and economic impact.
The following situations may imply:
- loss of human lives;
- shelter (long-term and short-term);
- quality of life issues;
- healthcare and long-term mental impact;
- unemployment.
This range of economic issues may also be increased by the loss of the social capital. This term has been used when referring to the ties that an individual or community has with that certain place which was affected by the earthquake.
These ties may be:
- friendships;
- professional relationships;
- an internal sense of stability.
c) Emergency response and recovery costs
In a normal earthquake scenario, these costs may include:
- first-responder costs (personnel costs, including overtime) encompassing search and rescue, fire fighting, emergency medical services; police security at damage sites;
- service costs related to building damage, including post-earthquake building safety inspections (e.g. safety-tagging), emergency shoring and demolition, and debris removal.
In addition, measures including loans from the government, private insurance policies, grants etc. must be activated to contain the disaster.
d) Business interruption and other economic losses
The economic impact for businesses may become the most important part of recovering from an earthquake. The property damages, human casualties and the interruption of business activities may damage the economic environment consistently. For example, direct business interruption can result from building or equipment damage, utility outage, lack of employees (due to injury, displacement, or transportation interruption), or supplier interruption. These problems also lead to more cascading problems, due to the interruption of services, cancellation of previous orders from clients affected by the earthquake and the fact that employees affected by this may work less or less productively. Indirect or secondary losses are those incurred in the days, weeks, or months following a disaster and include losses due to business interruption caused by infrastructure disruption (e.g. electric power, gas, water), reduction of critical services to residents in hazard-prone areas, and psychological trauma. As an example, after the Niigata Cheuetsu-Oki earthquake (Japan, 2007) ,the Riken manufacturing company, one of the largest parts suppliers to major Japanese automaker, including Toyota and Honda, was shut down for two weeks because of non-structural damage to equipment. Toyota alone lost production of more than 120,000 cars in the first weeks after the earthquake. In addition, the Kashiwazaki-Kariwa nuclear power plant, which is the largest in the world, stopped completely its production for two years.
E-FAST will contribute to reduce all the above losses. Earthquake risk mitigation will be achieved through improvements of knowledge in earthquake engineering. In fact, progress in earthquake engineering cannot be made without experimental research on large scale models whose behaviour is similar to that of real structures. This new, advanced, versatile, high capacity experimental facility, which will be unique in Europe and one of the most important seismic testing facilities worldwide, will allow the European earthquake engineering community to make significant progress in Research and development (R&D) and will have considerable impact on the state of the art and the practice of aseismic design and construction methods and technology.
In fact, the advanced experimental capacities of E-FAST will result in:
- A further insight into the earthquake response behaviour of structures, in general.
- The improvement of the numerical simulation tools via their validation with the experimental results: Although there is a tremendous evolution of the analysis tools, there is still a lot of work to be done regarding their capacity to reproduce the behaviour of complex structures at a realistic scale and under realistic dynamic loading.
- The improvement and validation of regulations and recommendations: Actually it is widely admitted and confirmed by the feedback from past earthquakes that the role of design and construction codes is of paramount importance for the aseismic protection of human life and structures. Eurocodes are an important step towards this direction but, as recognised at the first international E-FAST workshop, there are still several aspects which need to be completed and / or validated.
- Safer design and qualification of industrial structures and equipment, especially those of nuclear and chemical industry: To this aspect, it is worth noting that the involvement of Europe in nuclear industry is twofold. Actually, some European countries run a large number of nuclear power plants and at the same time European companies are word leaders in the design and construction of nuclear power plants. The E-FAST facility will be able to carry out tests of structures and equipment for R&D and qualification purposes. In fact, special attention was given in the design so that the new facility will be able to test not only 'classical' civil engineering structures but other kinds of structures and equipment also. This will be possible thanks to the high acceleration capacities of the tables (which enables to take into account floor response amplification), the 3D excitation capability, the multi-support excitation capabilities (e.g. tests of equipment such as pipes, crane bridges), the large displacement capability (which avoids filtering of low frequencies) and the large operational frequency bandwidth.
The development and validation of new construction methods, materials and devices which will improve the protection level of structures against earthquakes and lead to a better performance / cost ratio of aseismic structures.
- The support of European companies through the aforementioned validation of new technologies and design concepts. The R&D and qualification tests related to the activities of European companies will gain them with international prestige thus contributing to their export policy. E-FAST will secure and enhance the competitive edge that European construction firms and engineering services currently hold in overseas seismic markets, by establishing Europe as a world leader in earthquake engineering research.
- Demonstration tests for policy makers and dissemination purposes. The benefit from such demonstration tests is well understood in the US and especially in Japan. The high capacities of E-FAST will enable carrying out tests of realistic models which have considerable effects on public awareness.
- A 'natural' excellence centre for advanced knowledge in earthquake engineering through training, transnational access and dissemination. Within the framework of national and international collaborations E-FAST will be a crossroads for exchanging ideas and training on experimental earthquake engineering and in earthquake engineering in general. It will also promote, through networking and distributed databases, a wider sharing of data and knowledge across the field of earthquake engineering and between academia, research and industry.
List of websites: http://E-FAST.eknowrisk.eu
