| The seismic standards in the Loess Plateau have been low for a long time.Despite the corresponding design and treatment for collapsibility and liquefaction in practical projects,the seismic subsidence of loess under unsaturated conditions is often neglected.When the earthquake comes,disasters like landslides,liquefaction and magnification effect of ground motion arise a lot of phenomena which can easily cover the phenomena of seismic subsidence,therefore It is difficult to distinguish and extract seismic subsidence data in the field investigation.Then,the study of improvement and treatment of seismic subsidence site foundation needs to be further improved.In this paper,the seismic subsidence of loess with low water content in Northwest China is studied by shaking table test,combining with field investigation,laboratory test and numerical simulation,and the effectiveness of several common ground treatment methods for seismic subsidence disaster is analyzed.The main conclusions of this paper are as follows:(1)Indoor seismic subsidence tests were carried out on soil samples in the study area.The results show that the soil sample has high seismic subsidence coefficient when the water content is low and other physical and mechanical parameters are similar to the regional average level.The earthquake of ? intensity could cause serious damage to the site.Three kinds of anti-seismic subsidence treatment methods were selected,including physical improvement(dynamic compaction),chemical modification(adding cement and fly ash)and physical-chemical improvement(dynamic compaction-cement and dynamic compaction-fly ash).Residual deformation analysis of samples was carried out by means of laboratory tests.The results of relevant tests were compared and analyzed.Seismic subsidence of 20 m thickness is predicted by seismic subsidence coefficient: when the seismic intensity is equal to ?,the seismic subsidence generated by improved methods is less than 4.00 cm except for the method of adding fly ash,which can be considered as eliminating the seismic subsidence;when the seismic intensity is equal to ? degree,the seismic subsidence produced by dynamic compaction-fly ash method and dynamiccompaction-cement-soil method remain below 7.00 cm,while the seismic subsidence effect of physical improvement-dynamic compaction method begin to show its shortcomings;when the seismic intensity is equal to ?,the seismic subsidence of dynamic compaction-cement-soil method is still less than 7.00 cm,in other words,this method has completely eliminated the seismic subsidence of field.(2)By means of electron microscope scanning,the seismic subsidence mechanism of soil samples in this site is explained from the microstructure point of view,and it is more important to reveal the mechanism and improvement effect of the5 physical,chemical and composite Foundation improvement methods applied from the microscopic level to the seismic subsidence,and to explore the influence of the improved method on the microstructure of soil and its intrinsic relationship with the seismic.Results show that:1)Physical improvement is the most effective way to eliminate macropores and porous pore structures,moreover,the particle size distribution and structure are also adjusted.2)Chemical modification affects the strength of loess from the aspects of particle contact and intergranular cementation,moreover,the participation of different chemical reactions can produce unique glass beads or flocculent fine structures,which can greatly reinforce the strength of loess and play a role of filling,cementing or buffering role in strong earthquakes respectively.(3)In order to carry out the model test smoothly,the traditional vibration table test method is partially improved or redesigned.Include: In order to maintain the same overhead structure of the original soil,the indoor production method of large undisturbed loess test block in the field is used instead of layered compaction,and the soil stress application device covering three sides of XYZ is designed and made to restore the soil layer stress in the real field.(4)Shaking table tests of large undisturbed loess blocks and corresponding seismic subsidence improvement methods were carried out in four groups.Large-scale physical simulation is used to reproduce the catastrophic process of seismic subsidence of undisturbed loess under strong earthquakes,which reveals the dynamic characteristics of seismic subsidence loess site under strong earthquake loads,and gives out the stress-strain relationships respectively,then calculates and predicts the seismic subsidence of the original field,finally confirmed the feasibility of seismicsubsidence of undisturbed loess foundation with shaking table.The results corrected the indoor test results which are too large.The value of seismic subsidence varies greatly with the input of different waveforms.At the same time,the data of acceleration,settlement and lateral deformation of the original model and three kinds of seismic subsidence elimination measures are compared,and the seismic subsidence prevention effect of these improved methods is further evaluated.(5)According to the material ratio,physical and mechanical parameters of models obtained from shaking table test and laboratory test,the numerical simulation of MIDAS GTS NX is carried out.The results show that after 10 hours consolidation,the vertical stress and displacement data of the models are expressed as layered distribution.The maximum vertical stress is located at the bottom of the model,and the settlement is increasing from both ends to the middle.The original model has the largest settlement,and improved models are small or negligible.The dynamic calculation results show that the settlement trend of the model is consistent with that of indoor and outdoor tests.With the elevation rising,the settlement increases gradually,and the maximum settlement is located at the top of the model,and there is a phenomenon of non-uniform settlement,and the settlement increases gradually from the four sides to the middle in the surface.When the intensity is?,the maximum vertical settlement of the model is 25.18 cm for the original model,9.94 cm for the fly ash model and less than 1 cm for the cement model.This result is much smaller than that of the laboratory test,which further confirms that the shaking table test can correct the conservative estimation results of seismic subsidence in the laboratory.The magnification coefficients of these models are 2.30 for the original model,1.50 for the fly ash model and 1.00 for the cement model.Moreover,both the original model and the improved model have magnification effect on ground motion,and the numerical simulation results show that the most significant elimination of the magnification effect is the the method of cement on foundation.For a long time,the study of seismic subsidence is mainly based on laboratory tests,and the completion of shaking table tests for large undisturbed loess samples provides a new way of thinking for seismic subsidence research than the previous methods,and then provides a possibility for the verification of new seismic subsidence improvement methods in large-scale physical simulation.In this paper,indoor,model tests and numerical methods are used at the same site.The results obtained are helpful for exploring the theory of seismic subsidence and evaluating the effect of seismic subsidence treatment,achieving the purpose of earthquake disaster prevention and reducing earthquake losses,and providing reference and basis for earthquake subsidence prevention and control in the loess region of Northwest China. |
PDF Full Text Request