Back to search

National Technical University of Athens (NTUA)

Διερεύνηση παραγόντων αναπτύξεως των διαφορικών καθιζήσεων και εκτίμηση των συνεπειών τους

Abstract

dc:description

A major consideration of the design of shallow foundations is the estimation of the expected soil deformations. Foundation design requires sufficient safety factors against soil failure and settlements and differential settlements within allowable limits. According to Eurocode 7, it seems that an average value of the relative rotation or angular distortion β= θ ± ω = Δs/l ± ω =1/500 is generally acceptable for serviceability limit state (SLS), whereas a maximum value of β=1/150, above which superstructure damage is observed, should be considered in order to prevent an ultimate limit state (ULS). The assessment of these values, which appear to be acceptable for many purposes, lies in observations of superstructure damages in various types of structure. It is noted that the design criterion should be closely associated with the selection of the suitable calculation method, ensuring the safety and the economic viability of structures. The design of a shallow foundation can be carried out either by the conventional non-interaction analysis, in which the influence of the structural stiffness on the calculation of settlements is neglected and the superstructure is assumed to rest on unyielding support, or by the realistic approach of soil-structure interaction in which superstructure, foundation and soil are considered as components of a single system for analysis and design of the structure and its foundation. The conventional method of analysis is usually performed in order to simplify mathematical models, but it may result in a design that is either unnecessarily costly or unsafe. Therefore, sometimes it becomes imperative to perform soil – structure interaction analysis. The selection of more realistic models for soil behaviour also seems to be very important. Additionally, soil – structure interaction studies usually do not take into account the non-linearity of the soil behavior, which can sometimes lead to a conservative approach of the design. A rational analysis should include actual support flexibility and recognition of the non-linear and heterogeneous nature of the soil, together with non-linear soil-structure interaction effects. This would result in overall stiffness of the soil-foundation-structure system, realistic to the existing conditions. Furthermore, the significant contribution of the numerical analysis, using finite elements or finite differences under 2D or 3D conditions, on the foundation design undoudtedly, should not be forgotten. The overall scope of the present thesis is to thoroughly investigate the key factors affecting soil-foundation-structure interaction and the developed settlements and differential settlements, considering the general approach of the available acceptance criteria, and to identify possible discrepancies derived when following the traditional requirements for the design of shallow foundations.The accomplishment of the above objectives is based on:•A detailed literature survey covering the main topics in the static design of shallow foundations with particular emphasis on the available methodologies used in current practice for the prediction of settlements; on the various methods which perform soil-structure interaction analysis; and on the allowable deformation limits specified by various researchers as well as recommended according to Eurocode 7.•The execution of 2D finite element analyses for typical cases of independent shallow foundations resting on heterogeneous soils with linear elastic or linear elastic-perfectly plastic behaviour. The aim here is to provide a detailed examination of the effect of soil simulation on the induced settlements, given that there is considerable influence of the soil behaviour on the aforementioned magnitudes.•A broad parametric investigation performed with the aid of 2D numerical analyses using finite elements, for typical framed structures founded on shallow footings, aimed at identifying the general trends in the variation of the predicted differential settlements and critical forces of the superstructure and foundation resting on homogeneous linear elastic or elastoplastic soils. An intention is to correlate the angular distortion between adjacent footings or the variations of critical forces with the relative rigidity of the superstructure. The main parameters examined are the type and rigidity of the foundation, the simulation of the soil behaviour through the effect of the global safety factor, and the modulus of elasticity of the soil.•Comparisons between results obtained by 3D and 2D numerical analyses of soil – structure – interaction, using finite element mesh, for typical framed structures founded on a grid of strip footings, in order to detect and discuss possible differences. The main parameters under examination are the relative stiffness of the structure expressed by the variation of the elastic modulus E of the soil, the simulation of the soil behaviour, and the thickness of the compressible soil layer.•A parametric investigation performed through a number of 2D and 3D numerical analyses, simulating isolated footings or strip spread footings resting on linear elastic or linear elastic-perfectly plastic soils, in order to define correction factors to eliminate the corresponding differences observed between 2D and 3D results. Hence a real foundation problem could be properly simulated under plane strain conditions. In this way the complexity often presented in the use of 3D models can be avoided.•An extensive investigation of the time-dependent soil-structure-interaction effects of multi-storey framed structures which aims to find differences between single-phase and multi-phase simulations and between consolidation process and rapid drainage of soil. The most important parameters under examination are the type and rigidity of the foundation, the stiffness of the superstructure, the compressibility of the soil, and the simulation of the building based on its gradual construction or in only one phase.The main conclusions conducted in the context of the present thesis are summarized and given below.1.An extensive investigation of the soil simulation effect on the developed settlements of typical independent shallow footings resulted in the following:1.1.In cases of two-layered soil models consisting of a softer top layer than the underlying one, elastic solutions taking into account stress distribution in homogeneous soils could be reasonably used to predict settlements beneath flexible footings. However, these approximate solutions seem to underestimate settlements below rigid footings, especially for small thickness of the top soil layer. In cases where the top soil is less compressible, the precise simulation of the two-layered soil seems necessary due to lower predicted settlements.1.2.In order to design shallow foundations resting on heterogeneous soils whose elastic modulus changes linearly with depth, such as on normally consolidated clays or silts or even sands, the simplest Winkler model could be easily implemented. This could not be applied in cases of overconsolidated clays, owing to the constant elastic modulus E.1.3.The study on the development of plastic zones in homogeneous soils shows that there is considerable effect of the safety factor on the settlements, hence the simulation of the soil’s behavior according to the elastic perfectly plastic idealization seems indispensable especially for low values of the safety factor (SF<2,0). This conclusion concerns cases of purely cohesive soils. In cases of purely cohesionless soils, settlements predicted by the elastoplastic Mohr-Coulomb soil model, are significantly higher than those obtained by the linear elastic soil simulation even for high SF values especially below rigid footings, because of extensive plastic zones.2.The results obtained from the systematic examination of factors affecting soil-structure interaction as well as settlements and differential settlements developed in typical framed structures by means of 2D and 3D finite element numerical analyses, provide interesting guidelines for the design of shallow foundations, which are summarized as follows:2.1The available performance-based criteria, associated with the allowable deformation limits, seems to be indicative rather than absolute for every construction project. The well-known criterion recommended for the maximum relative rotation Δs/l≤1/150, aiming at avoiding structural damage, should not be generally adopted. More specifically, rigid or semi-rigid superstructures could potentially suffer serious damage, if the design model is based on analyses in which interaction is not included, even when the induced relative rotations are much lower than the aforementioned limit, because of the developed large magnitudes of their critical forces. In contrast, flexible superstructures might sustain maximum relative rotations of Δs/l=1/150. It is worth noting that conventional non-interaction analyses could be performed only in cases of rigid foundations (approximately 10-3<RF<1) or very flexible superstructures (approximately Rb<0,009). The values of RF and Rb are calculated for the specific cases of the examined frames.2.2An attempt to correlate the predicted rotations as well the relative variation of critical forces with the relative stiffness of typical framed structures built on isolated strip footings or raft foundations, it is found that the level of reliability and validity of the proposed rigidity criteria might be diminished because they practically ignore both the stiffness of the superstructure and of the foundation.2.3Consequently, the methodology used in today’s engineering practice seems to be sufficient in cases of rigid foundations, or very flexible superstructures or “uncompressible” soils, otherwise an accurate i.e. numerical analysis is needed.2.4Correlation functions assessing the critical forces of the superstructure and the foundation are formed with regard to the relative stiffness of the superstructure, in that cases where the corresponding variation is not affected by the number of storeys of the frames based on the soil-structure interaction effect.2.5The appropriate simulation of the soil behavior seems to be very important in most cases employing soil-structure interaction analyses due to the considerable influence of the global safety factor against soil failure on the differential settlements. For purely cohesive soils, in cases of global safety factors SF>2,5, soil-structure interaction analyses could be performed under the simplest model of linearly elastic soil behavior. This is not concluded for purely cohesionless soils, presenting significant corresponding differences even for higher SF values.2.6The investigation of the nonhomogeneity (for soils whose elastic modulus increases linearly with depth under constant rate m) shows the significant effect of the rate increase of the soil modulus E on the soil-structure interaction problem. SSI analysis seems to be essential for the assessment of Δs and critical forces of the frames mainly in cases of semi-rigid and rigid superstructures, especially for low rates of increase.3.Investigating the impacts derived from the evaluation of the corresponding results from calculation analyses considering single-phase as against multi-phase simulation of multi-storey buildings, remarkable discrepancy is observed regarding both differential settlements and critical forces of the structures. The parametric investigation indicated that the above discrepancy is strongly influenced by the different superstructure’s rigidity which increases gradually after each construction phase, as against the total load from the entire frame building applied on the foundation, the stiffness of the foundation elements and the elastic modulus E of the soil.4.Interaction analyses of independent shallow foundations (isolated footings or strip footings) result in proposed expressions of correction factors to decrease the real elastic modulus of the soil, and to adjust the applied loads, as so to be the input parameters under 2D conditions, in order to reduce the divergences. 3D simulation of shallow footings for high values of SF (SF>2,5), in cases of purely cohesive soils, could be performed under the assumption of linear elastic soil.5.Comparing the corresponding differential settlements and critical forces of typical buildings by applying 2D and 3D numerical simulations based on the finite element method, significant divergences are obviously concluded, especially in cases of low elastic modulus E, of small thickness of the compressible soil, and of not large enough aspect ratio. The 3D simulation of the soil’s behavior according to the elastic-perfectly plastic idealization seems indispensable even for high values of the safety factor (SF=2,5).6.A final outcome clearly indicated from the parametric investigation is that SSI plays an important role in achieving precise results; hence, it is deemed more necessary to be taken into account in most calculation analyses rather than to accurately estimate the elastic modulus, especially in cases of soft soils. It is also confirmed that SSI analyses are necessary to be performed in cases of flexible foundations (indicatively for values of the relative stiffness between the foundation and the soil 5x10-5<RF=EbxJ/ExL3<10-2, depending on the total width L of the foundation) and in cases of rigid or semi-rigid superstructures (indicatively for values of the relative stiffness between the superstructure and the soil Rb=ΣEbxJ/Exl3>0,02). SSI analyses are essential in cases of low values of the global safety factor, too. Results show that cohesionless soils seem to present more sensitive response than that of cohesive ones. It is evident that, in cases of very stiff soils, results derived by the conventional design procedure agree well with those of SSI analyses. Special attention should be paid on the spandrel type of settlement curve, which means the development of negative differential settlements, occurring in cases of low SF values. Results indicate that allowable deformation criteria should be more stringent as against the available ones proposed by Eurocode 7 for the concave type of settlement curve. 3D SSI analyses seem indispensable in most cases, mainly for not large enough values of elastic modulus E, for not large enough thickness of the compressible soil and for not large enough aspect ratio.To sum up, a limited number of framed structure types was examined in the context of the present thesis, in order to investigate and define general trends in the variations of specific magnitudes as well as to qualitively assess the effect of significant parameters on the SSI issue. Consequently, the derived correction factors could be used in such framed structures which match those examined herein. This thesis will be a guide to help Engineers comprehend what is important and what is not important so that they can choose the appropriate method of analysis.

Degree

thesis:*
Grantor dc:publisher
National Technical University of Athens (NTUA)
Year dc:date
2022

Author and committee

dc:creator, dc:contributor.*
Authors dc:creator
  • Arapakou, Angeliki
  • Αραπάκου, Αγγελική

Subjects

dc:subject × 7

Rights

Language dc:language
gre

Identifiers

dc:identifier.*
Identifier
10.12681/eadd/51341

Chain of custody

source
Harvested from
Greek National Archive of PhD Theses
Base URL
phdtheses.ekt.gr/eadd_oai/request
Last updated
2026-11-18
Source record
OAI-PMH GetRecord
citation

Arapakou, Angeliki; Αραπάκου, Αγγελική. Διερεύνηση παραγόντων αναπτύξεως των διαφορικών καθιζήσεων και εκτίμηση των συνεπειών τους. National Technical University of Athens (NTUA), 2022. http://hdl.handle.net/10442/hedi/51341