Fall Seminars 2017 - Professor Gian Michele Calvi
Professor Gian Michele Calvi
Fall Seminars 2017
Sala del Camino, Palazzo del Broletto
1 Engineers understanding of earthquakes demand and structures response 12 September
2 Experiences and trends in seismic design and assessment of bridges 14 September
3 L’Aquila earthquake 2009: reconstruction between temporary and definitive 26 September
4 Concepts and technologies for friction-based isolation of buildings 28 September
5 Seismic assessment and rational renovation of the structural heritage 3 October
6 Energy efficiency and disaster resilience: a common approach 6 October
7 Revisiting seismic demand, structure capacity and design spectra 10 October
8 Education and research in seismic engineering: an inextricable Gordian knot 12 October
All seminars will be presented in English and will be mainly oriented to graduate students, though open to anyone interested.
All seminars will be held in the Sala del Camino at 6 pm and will last approximately 90 minutes.
of earthquakes demand and structures response
The presentation will focus on the engineers understanding of the strength and displacement demands imposed to structures by earthquake motion, and on the structures capacities to withstand these demands, in a historical perspective. Without any claim of completeness or accurate critical assessment of the significance of various seismic events or of the scientific development of knowledge, the essential relevance of the lessons learnt from some seismic events is critically examined in parallel with the development of structural dynamics.
The story moves from a claimed use of some base isolation measure in the temple of Diana at Ephesus in the VI century B.C. (as mentioned by Pliny the Elder in his Naturalis Historia). It continues with the renaissance treaties, where in one case only some emphasis is placed on how to build a safe structure and passing through the first understanding of dynamic equilibrium (by Robert Hooke and Isaac Newton) arrives to the ages of enlightenment and the Lisbon earthquake of 1755.
The breakthrough towards modern seismic analysis is clearly identified with the two earthquakes of San Francisco (1906) and Messina (1908). In particular it is discussed how most of the fundamental principles used for a century had already been stated after the second one.
Spectra, ductility and performance based design are then identified as further milestones derived from earthquake evidence (e.g.: San Fernando, 1971, Loma Prieta, 1989, Kobe, 1995), to conclude with a critical appraisal of some major misunderstanding of structural response, with the merits of displacement based approaches, closing the circle opened with the temple of Diana presenting modern base isolation techniques.
Eventually, the evidence resulted from the most recent earthquakes in New Zealand, Japan etc. is introduced, commenting on what choices should be made to reduce seismic risk and construct safer structures.
Experiences and trends in seismic design and assessment of bridges
The lecture will review and discuss some relevant conceptual considerations for seismic design of different kind of bridges, including isolated and cable-stayed bridges.
The advantages and disadvantages of different structural solutions will be highlighted, reviewing alternatives for deck sections, tower configurations, in both the longitudinal and transverse direction, deck-to-pier connections, and cable arrangements, amongst other issues.
Some simple preliminary sizing procedure will be presented and discussed, with the aim of offering to designers a quick but rational means of identifying reasonable member sizes for cable-stayed bridges that should then be verified through advanced analyses in the detailed design stages of the project.
The subject of repair and strengthening existing bridges will also be touched, more or less extensively, depending on available time.
Some real case study applications will be introduced, comparing preliminary and final design member sizes and showing that intelligent preliminary sizing procedures may be a useful tool for design.
L’Aquila earthquake 2009:
reconstruction between temporary and definitive
The presentation describes the reconstruction intervention after the L’Aquila earthquake of April 6, 2009. Approximately 4,500 apartments were built in 8 months, all in buildings isolated by friction pendulum bearings (more than 7,000 devices were employed). The main aspects related to the logic developed to operate in such a tight time are briefly addressed, with particular reference to structural choices and response of the isolated buildings.
Full scale dynamic testing of more than 400 devices and on site testing of fifteen complete buildings shed light on the behaviour of curved friction sliders and on their combined response with building structures.
Thousands non linear analyses of the buildings, performed using actual experimental data, allowed to estimate the maximum base shear that could be attained in each case, in general lower than 20 % of the weight of the building. Such value was always acceptable considering a low damage performance level, consistent with the full use of the buildings following a 500 years return period ground motion. The small change in the base shear for larger events and the very small interstorey drift demand, together with the displacement capacity of the isolating system, largely in excess of any reasonable demand, make the buildings essentially insensitive to any potential earthquake.
Concepts and technologies for friction-based isolation of buildings
The presentation will start with a critical review of the historical development of base isolation concepts and techniques, discussing the reasons of an undeniable success in practical applications and focusing specifically on various alternatives friction-based technology.
In the last thirty years, devices based on sliding on curved surfaces, characterized by low friction coefficients, have become quite popular. It is well known that such devices are essentially blocked until the acting shear is larger than the vertical force multiplied by the friction coefficient and are then characterized by a stiffness value (K) that depends on borne weight (W) and radius of curvature (r), as: K = W/r. This second stiffness is considered fundamental to contain the residual displacement, but it implies two negative aspects, i.e.: the reduction of the energy dissipated per cycle (which implies a lower equivalent damping and a larger displacement demand) and the increased shear capacity (which implies designing the isolated structure for larger shear demand and more extensive nonstructural damage).
Consequently, some questions arise:
• Is it really fundamental to limit the residual displacement?
• Are alternative ways of limiting this displacement conceivable and applicable?
These and other relevant questions are related to technology advancement and reliability:
• How reliable is the definition of a nominal value for the friction coefficient?
• How relevant is the effect of stick slip? Can it be eliminated?
• How dependable is the friction coefficient at variable velocities and vertical forces?
• How accurately can friction coefficient and radius of curvature of the surfaces be varied?
These subjects will be examined and alternative technological solutions will be proposed, showing that it is theoretically and practically possible to obtain cycles of the kinds shown in the figure below.
Such cycle shapes imply noticeable variation in displacement demand, residual displacement and design shear. These aspects will be emphasized and critically analyzed referring to the results of an extensive numerical investigation.
The driven conclusions will address the problem of developing and applying the most cost-effective solutions, depending on seismicity, use of the building and target performances.
Seismic assessment and rational renovation of the structural heritage
The intelligent application of Newton’s and Hooke’s laws, the development of response and design spectra and of capacity design principles have not been sufficient to protect human life and constructed environment from nature’s whims. Today’s frontiers are related to the impossible comparison of resources and needs, and consequently to the best use of the available resources, in terms of funding, but as well of time, manpower, advanced techniques.
A first difficulty is associated with the problem of defining a common measure of risk, for different kind of structures and infrastructures, including apparently distant effects, such as business interruption, traffic detour, increased pollution, value of the cultural heritage.
A second basic problem relates to our capacity of appropriately evaluating the effect of different, traditional and innovative, strengthening techniques, in the same terms of global economical benefits.
This presentation will discuss these subjects from a critical viewpoint, emphasizing the possible criteria for the mitigation of seismic risk and some of the alternative choices that may be adopted for strengthening, with reference to:
(a) the modification of damage and collapse modes strengthening individual elements or locally increasing the deformation capacity;
(b) the insertion of additional systems resisting to horizontal actions;
(c) the introduction of base isolation, with the objective of capacity-protecting the existing structure;
(d) the reduction of displacement demand by added damping or introducing tuned mass systems.
Examples will refer to major bridges and historical buildings as well as to problems of induced seismicity in historically relevant areas.
Energy efficiency and disaster resilience: a common approach
A proposal for integrated assessment of energy efficiency and earthquake resilience outlines the scope of a multi-disciplinary procedure in which environmental and seismic impact metrics are translated into common financial decision making variables.
In the context of seismic risk assessment, one of the most important outputs for communication with insurance companies, governmental agencies and other decision making entities is the Expected Annual Loss (EALS), which translates the mean value of economic loss (i.e. cost of repair or rebuilding operations) that a building (or group of buildings) will sustain annually over its life-span due to seismic action. The value of EALS should also account for losses related to operational and indirect costs, also referred as downtime (e.g. costs due to business interruption and people temporary re-allocation).
An analogue reasoning is employed when incorporating the energy efficiency analysis in the equation, by defining an energy Expected Annual Loss (EALE) that can directly be compared with its seismic counterpart (EALS). Thus, if one chooses to consider the building value as the common denominator, the EALE can be determined as the ratio between the mean annual cost (referred herein as loss) of consumed energy and total building value, within a compatible base of comparison:
Similarly to what is common practice when evaluating the energy and environmental performance of buildings, discrete classes of both earthquake resilience and energy efficiency are proposed, providing a consistent proxy for building classification - Green and Resilient Indicators (GRI) - as a function of EALE (Green Indicator) and EALS (Resilient Indicator):
This approach allows to directly compare the expected performance of a building in terms of seismic resilience and energy efficiency and the benefit-cost ratios of investment necessary to upgrade a given GRI class.
Revisiting seismic demand, structure capacity and design spectra
For several decades, seismologists and engineers have been struggling to perfect the shape of design spectra, analyzing recorded signals and speculating on probabilities. This research effort produced several improvements, for example suggesting to adopt more than one period to define a spectral shape, or proposing different spectral shapes as a function of the return period of the design ground motion.
However, the basic assumption of adopting essentially three fundamental criteria, i.e.: constant acceleration at low periods, constant displacement at long periods, constant velocity in an intermediate period range, has never been really questioned.
In this contribution, the grounds of a constant velocity assumption is discussed and shown to be disputable and not physically based. Spectral shapes based on different logics are shown to be consistent with the experimental evidence of several hundred recorded ground motions and to lead to significant differences in terms of displacement and acceleration demand.
The main parameters considered to define the seismic input are magnitude and epicenter distance, but the possible influence of other parameters – such as focal depth and fault distance, duration and number of significant cycles, local amplification – are discussed.
Novel forms of ground motion prediction equations and of hazard maps may result from this approach.
The implications on design and assessment will also be addressed, considering case studies derived from recent tectonics events (i.e. the Central Italy sequence of 2016) and from induced seismicity (i.e. the case of the Groningen region in the Netherlands).
Specific points of interest include the generation and adaptation of acceleration and displacement time histories for design, the possibility of including the effects of energy dissipation on the side of capacity rather than on that of demand, the consistent generation of floor spectra for design and assessment of non structural elements.
Education and research in seismic engineering:
an inextricable Gordian knot
At the time of the Imperial Valley earthquake of 18 May 1940 George Housner was a ph. d. student at Caltech. At El Centro the first (and for decades the only) strong motion signal was recorded, thus becoming the paradigm of a strong earthquake epicentral ground motion.
Though the concept of elastic response spectrum – which summarizes the peak response of all possible linear single-degree-of-freedom systems to a particular component of ground motion – was first intuitively conceived in Japan, by Suheiro, in 1926, the most significant advancements in its theoretical development took place at Caltech, with contributions by Von Karman, Biot, Hudson, Popov and others (even Fung, later the father of biomechanics gave some contribution).
How relevant was the interconnection between education and research, to produce one of the fundamental concept in earthquake engineering?
How fundamental was the educational context in the formation of the person who forty nine years later inspired the famous report “Competing against time”, following the Loma Prieta Earthquake?
In the seventies and around, the faculty in earthquake engineering at UC Berkeley was regarded as a sort of “dream team”, never repeated elsewhere in any time. What have been its effects in the progress of knowledge and how can a similar experience be repeated?
What was the link that made possible in a university at the periphery of the world (Canterbury, New Zealand) to have Paulay, Park, Priestley in a row, at an exact spacing of ten years, producing some of the best books still used worldwide in seismic design and the capacity design concept? What is its legacy?
All these questions, and the answers to them, are at the base of the conception and development of the ROSE School in Pavia, Italy, more or less consciously.
The objective of this seminar is to try to extricate and follow each strand of the Gordian knot of research and education in earthquake engineering: an impossible task, though possibly an exciting experience.