CC BY
1016
EARTH SCIENCES / «еЮУУШШУМ-ЛШШаИ» 2021
EARTH SCIENCES
УДК 550.34.01
Семенова Юлия Владимировна к. физ.-мат. наук, старший научный сотрудник, Институт геофизики им. С.И. Субботина НАН Украины, г. Киев, Украина
DOI: 10.24412/2520-2480-2021-390-16-18
МОДЕЛИРОВАНИЕ ВЛИЯНИЯ МЕСТНЫХ ГРУНТОВЫХ УСЛОВИЙ НА КОЛЕБАНИЯ ГРУНТОВ ПРИ ЗЕМЛЕТРЯСЕНИИ НА ТЕРРИТОРИИ КИЕВА
Semenova Yuliia Vladimirovna
PhD, senior researcher, Subbotin Institute of Geophysics of the NAS of Ukraine, Kyiv
MODELING THE INFLUENCE OF LOCAL SOIL CONDITIONS ON THE EARTHQUAKE MOTION
IN THE TERRITORY OF KYIV
Аннотация.
В статье рассмотрены методы моделирования отклика местных грунтовых условий на сейсмические воздействия. Рассматриваются преимущества и недостатки линейного и нелинейного моделирования на примере грунтовой толщи площадки в Киеве. Показана трансформация спектральных характеристик грунтов с учетом их нелинейного деформирования при сейсмическом воздействии и без.
Abstract.
The article discusses methods for modeling the response of local soil conditions to seismic effects. The advantages and disadvantages of linear and nonlinear modeling are considered on the example of the soil .strata of the site in Kyiv. The transformation of the spectral characteristics of soils with and without taking into account their nonlinear deformation under seismic action is shown.
Ключевые слова: сейсмическое микрорайонирование, отклик местных грунтов, линейное и нелинейное моделирование, спектральные характеристики грунтов, сейсмическая опасность.
Keywords: seismic microzoning, local soil response, linear and nonlinear modeling, spectral characteristics of soils, seismic hazard
The task of seismic micro-zoning is to assess the impact of local engineering and geological conditions of the construction site on the intensity, shape and range of oscillations of the soil surface during strong earthquakes. These parameters determine the nature of the destruction of buildings in different soil conditions/
It should be noted that even under favorable seismic conditions of the site, the seismic safety of buildings or structures built on it is not guaranteed, since in some cases resonance effects can occur in soils (a significant increase in oscillations at specific frequencies).
For seismic design, it is necessary to know not only the magnitude of the oscillation intensity and the values of the maximum peak accelerations, but also information how seismic effects are distributed in frequency. It is known that the soil mass under the construction site behaves like a filter: at some frequencies, the soil mass transmits vibrations almost unchanged, while at others it either amplifies or absorbs them. When designing earthquake-resistant buildings and structures, it is important not to allow the maxima of the frequency response of the soil to coincide with the natural frequencies of buildings and structures.
The frequencies corresponding to the resonant amplifications can be determined from the amplitude-frequency characteristics of the soil strata. In this case, it is desirable to consider a wide frequency range from 0.05 to 20 Hz. This range is of the greatest interest in
seismic zoning, since it contains the vibration frequencies of the main types of buildings, structures and their critical structures, as well as the maximums of the vibration spectra during strong earthquakes.
The frequency response of the soil under the construction site can be obtained instrumentally, based on records of vibrations of weak earthquakes and microseisms. Registration of fluctuations in weak earthquakes is not always possible in areas with low seis-micity and gives not very accurate results through industrial noise. With this in mind, seismic observations at a construction site are usually limited to recording microseisms. In the formation of the field of microseis-mic oscillations, as a rule, numerous natural and anthropogenic sources are involved, the contribution of which is difficult to account for. Due to significant errors, the method of recording high-frequency microseisms gives only approximate values of the frequency characteristics of soils. As a rule, this method is used only to assess the increase in seismic intensity in points of the macro-seismic scale.
Calculation methods are an alternative to instrumental methods.
It should be noted that computational methods provide sufficiently accurate results only in the presence of reliable geotechnical and seismological data on the structure of the soil strata of the construction site, as well as on the lithology of geological layers and the physical and mechanical properties of the constituent materials. Based on these data, models of the soil strata
«cqyyomum-jmtmail» #m®m, 2©2h / earth sciences
17
under the construction site are constructed (with arbitrary interfaces or horizontally layered vertically heterogeneous models), which serve as input data when using calculation methods for determining its frequency characteristics.
Obtaining data for building horizontally layered models of the environment is much easier and more realistic. The fields calculated in such models are stable to errors in the initial data. In this regard, the use of just such models is assumed by SBS V.1.1-12: 2014 [2] and Eurocode 8. For calculating the frequency characteristics of horizontally layered models of the geological environment, the Thompson-Haskell matrix method is used [5].
A linear or nonlinear approach can be used to model the response of the soil layer to seismic impacts. Linear - is described by a linear-elastic model of the reaction of the soil layer. It is assumed that at dynamic loads the processes in soils will correspond to the linear stress-strain dependence. But under intense seismic effects in soils, phenomena arise that cannot be described by the linear theory of elasticity. In intense earthquakes, the proportionality between stresses and strains is disturbed, the phenomenon of saturation occurs, this is when the stresses grow more slowly than in smaller values of strains. The value of stresses at which the proportionality of the dependence between stresses and strains disappears is the threshold of elasticity. The threshold of elasticity for different categories of soils is different [1] and is determined, first of all, by the absorbing properties [10]. Therefore, for the analysis of the reaction of the soil to seismic impacts, considerable attention has recently been paid to nonlinear approaches [3, 4, 7].
The mechanisms of linear transformations of seismic waves in the near-surface soil strata, which lead to increased vibrations, are well studied. Unlike nonlinear mechanisms, they are fully taken into account in the practice of seismic microzoning. Since seismic micro-zoning is usually carried out for areas that can be exposed to strong earthquakes, adequate consideration of the nonlinear response of the soil is necessary [8]. Soil
16
response can be considered linear under weak seismic impacts. At high impacts, the contribution of nonline-arity will depend on the magnitude of seismic deformations [7].
The nonlinearity of the response of the near-surface soil strata leads both to a change in the spectral composition of seismic vibrations, sometimes very significantly, and to a change in the amplification of seismic vibrations. At a sufficiently high intensity of oscillations, nonlinear absorption mechanisms begin to operate, which lead to a weakening of oscillations at high frequencies, but do not attenuate low-frequency oscillations. Changes in the spectral composition of oscillations on the surface associated with the nonlinearity of the soil response are manifested in the shift of resonance frequencies to the low-frequency region. The amplification of seismic vibrations on the surface is reduced due to the nonlinearity of the response of the soil compared to the linear response in dry soils (when groundwater is deposited at a depth of 10 m or more). In water-saturated soils (when the groundwater level is at a depth of less than 10 m), such an increase is less noticeable. These conclusions were made both on the basis of the analysis of real records of strong earthquakes and the results of numerical simulation [8].
In equivalent linear modeling, the soil is considered as a linear viscoelastic material, and its nonlinear properties are taken into account by introducing the dependences of the shear modulus and damping ratio with shear strain amplitude. Such dependencies are selected for each layer of the soil model separately according to the data on the lithological composition and depth of occurrence, obtained on the basis of the results of laboratory or field studies. For example, according to the data of [6, 9].
Figure 1 shows the amplitude-frequency characteristic of the soil under the construction site of the stadium of the Central Sports Club of the Armed Forces of Ukraine in Kyiv, obtained by a) linear modeling, c) equivalent linear modeling using the ProShake software package.
-Linear Equivalent 1
near modeling
A
/ A \ / 11
01 0,1 1 10 100
Frequency, Hz
14
12
10
ra
u ^
"E. 6
E
<
4
2 0
0;
Fig. 1 Frequency characteristics of the soil under the construction site of the stadium of the Central Sports Club of the Armed Forces of Ukraine in Kyiv, obtained by a) linear modeling, c) equivalent linear modeling
18
EARTH SCIENCES / «e©yy©®UOUM=J0Umay» #319©)), 2©2fl
Fig. 1 that the frequency response calculated by linear modeling (when the absorption decrement is considered constant for each layer) differs from the frequency response calculated using the equivalent linear modeling (which takes into account that the absorption coefficient and shear modulus depend on the level of deformation).
In linear modeling of the response of the soil to seismic influences, only the amplitude transformation of the frequency components of the output spectrum occurs, while in equivalent linear modeling, as can be seen from Fig. 1, the absolute maxima shift towards lower frequencies. Experience has shown that the use of linear modeling can lead to the appearance of false resonance peaks in the frequency response of the soil mass. The similarity between the results of equivalent linear and nonlinear simulations depends on the degree of soil nonlinearity. Both methods give good results in the real estimation of soil response at small deformations, and at very large deformations, nonlinear modeling gives results closer to those established empirically.
The obvious conclusion is that by choosing the wrong approach to modeling the response of the soil to seismic effects from earthquakes, it is possible to obtain false values of the resonance frequencies of the soil mass, which during an earthquake can lead to the destruction of a building due to resonance effects not taken into account in the design.
Conclusions. The modern paradigm of seismic protection of buildings and structures in seismic regions provides for a transition from the intensive use of increasingly robust buildings and structures designed with a large amount of steel and concrete, to the use of data on the frequency characteristics of the soil at the base of these buildings. This approach is aimed at avoiding the coincidence of the natural periods of seismic vibrations, enhanced by local ground conditions, with their own periods of vibration of buildings and structures. This approach guarantees not only their high seismic resistance but also minimizes the cost of seismic protection measures. Taking into account the fact that with weak seismic effects the results of linear and
equivalent linear modeling are close, and with an increase in the predicted seismic effects in the general seismic zoning indicated on the maps, the equivalent linear modeling gives more accurate results, it is advisable to use it in the practice of seismic design and construction in seismic conditions of the platform part of the territory of Ukraine.
References
1. Alyoshin A.S. Seismic micro zoning of especially responsible objects - M.: LLC "Svetoch Plus", 2010. - 293 p.
2. Building in seismic regions of Ukraine: SBS V.1.1-12: 2014, (2014). Kyiv: Building Ministry of Ukraine, 84 p. (in Ukrainian)
3. Conti, R.; Angelini, M.; Licata, V. Nonlinearity and strength in 1D site response analyses: A simple constitutive approach. Bull. Earthq. Eng. 2020, 18, 4629-4657
4. Groholski, D.R.; Hashash, Y.M.; Kim, B.; Musgrove, M.; Harmon, J.; Stewart, J.P. Simplified model for small-strain nonlinearity and strength in 1D seismic site response analysis. J. Geotech. Geoenviron. Eng. 2016, 142, 04016042
5. Haskell N.A. Asymptotic Approximation for the Normal Modes in Sound Channel Wave Propagation. - J. Appl. Phys., V.22. 1951.- P.157-168.
6. Ishibashi I. and Zhang X. Unified dynamic shear moduli and damping ratios of sand and clay. -Soils and Foundations, V.33, No.1, 1993. - P.182-191
7. Kendzera A.V., Semenova Yu.V. The influence of resonant and nonlinear properties of soils on the seismic hazard of construction sites. - Geophysical Journal - 2016 -№2. - P.3-18 (in Russian)
8. Pavlenko O.V. Seismic waves in the ground layers: a non-linear behavior of soil during strong earthquakes in recent years. - Moscow: "Nauchnyi Mir", 2009. - 260 p.
9. Seed H.B., Idriss I.M. Simplified procedure for evaluating soil liquefaction potential. - Journal of the Soil Mechanics and Foundation Division, ASCE. V.97, № SM9, 1971. - P. 1249-1273
10. Voznesensky E.A. Kushnareva E.S., Funikova V.V. Nature and the laws of stress wave attenuation in soils. - Moscow: Flint Publisher, 2014. -104 p.
