Design and execution of deep excavations in urban areas are topics of interest for geotechnical engineering in Romania, but not only. In addition, given the continuous development of the construction in the urban areas, the remaining unbuilt spaces present more and more difficulties regarding the neighboring sites, as the existing buildings are often located at the property limit and might be sensitive to the settlements and vibrations induced by the execution process, the access conditions and especially by the lithological conditions. Furthermore, together with the advancement of the calculation methods and the experience on similar projects, an optimization of the designed solutions is desired, the monitoring works being relevant in the process.
The back-analysis of the design models based on in situ measurements of the behaviour of deep excavations retaining systems has been previously adopted by the authors of this paper (POPA et al. 2015), (POPA et al. 2018) and by other authors, since it is a powerful approach to increase the knowledge on the behaviour of such geotechnical structures, as well as on the ground models used in the analyses.
The scope of the current paper is to present some particularities of a project for a deep excavation performed in Bucharest, Romania in terms of design and execution. Also, the monitoring works implemented for this will be described and some of the results from this process will be given.
At last, based on the comparison of the initial design models and the results of the measurements obtained from the monitoring process, a back-analysis is performed to check for the most plausible reasons that could justify the differences between the estimated and the actual behaviour of the retaining structure obtained on some area.
PROJECT DESCRIPTION
The analysed project involved the execution of two office buildings with 2B+GF+11F+Technical Floor located in Bucharest, Romania. It was provided that the infrastructure of about 7,650 m2 for the two buildings to be executed in the same stage, including the excavation and its retaining system, and the structural elements of the infrastructure, while the superstructures were provided to be executed in different stages.
- Ground conditions
The geotechnical investigation on the site was performed in three stages and comprised a Preliminary Geotechnical Report based on 6 geotechnical boreholes (20 m deep), followed by a Design Geotechnical Report based on 10 geotechnical boreholes (2 boreholes 50 m deep, 5 boreholes 30 m deep and 3 boreholes 25 m deep) and 3 additional boreholes for pumping tests (10/15 m deep) and a Detailed Geotechnical Report with 3 supplementary boreholes (1 borehole 50 m deep and 2 boreholes 25 m deep each). In addition, to have better understanding of the dynamic parameters of the soil one Downhole test was performed in the 50 m deep borehole following the provisions of norm NP 074 and standard SR EN 1997-2.
The resulted soil stratigraphy is typical to Bucharest area and consisted of a layer of anthropic filling in the surface up to 3.5 m thick, followed by a thick cohesive layer of silty clay (6 to 10 m thick), a layer of sand with gravel with variable thickness from 1 to 4.5 m, a layer of clay up to the depth of approx. 20 m and then alternative layers of cohesive and uncohesive layers.
Following the comprehensive investigation programme, it was assumed that the level of knowledge of soil variation and properties was high, leading to a higher confidence level in the parameters chosen for design, both for the retaining system of the excavation and the foundation system of the new buildings.
- Design considerations
As a result of the different site conditions, respectively those related to the adjacent constructions, the excavation depth, and the lithology, eleven characteristic sections were considered in the design, as shown in fig. 1.
Figure 1: Characteristic sections for the deep excavation design
The retaining system was design based on the provisions of the Romanian norms NP 120 (2014), NP 124 (2014) and the European standard adopted in Romania as SR EN 1997-1 (2004) and consisted of 60 cm diameter piles, which were generally supported by inclined steel struts, and self-supported on one relatively reduced area.
The present paper will provide details on three of these sections, considered most relevant for the purpose – two sections for the wall supported by struts and one section for the self-supported wall (fig. 2).
Figure 2: Characteristic sections: left – wall supported by struts – S3 and S4, right – self-supported wall – S10
- Monitoring works
The monitoring works for the entire investment comprised a complex monitoring that started prior to the beginning of the execution works and continued throughout the entire execution period and, for some of the monitoring elements, on the operation period.
Throughout the execution of the deep excavation, the main monitoring works consisted of:
- horizontal displacement of the retaining wall measured in 9 inclinometer columns installed through the retaining wall, including 5 m below the pile base and on 20 3D geodetic markers installed on the capping beam;
- vertical displacements of the ground measured in 2 extensometer columns, installed in boreholes inside the excavationpit;
- vertical displacements of the neighbouring constructions measured by means of topographical measurements on 35 settlement marks installed on the constructionsand evolution of cracks.
For the present paper, the inclinometer and topographical measurements are used to compare and calibrate the calculation results.
ANALYSIS OF THE RETAINING SYSTEM
For the scope of this study, three different sets of parameters were considered for the layers over the excavation depth (filling and silty clay layers), as described below.
- Mean values considered as “best estimate” (most probable) were used for the main geotechnical parameters for the layers over the excavation depth, described further in the results as “mean”.
- Inferior representative values of the main geotechnical parameters for the layers over the excavation depth, described further in the results as “inf. param.”.
- Superior representative values of the main geotechnical parameters for the layers over the excavation depth, described further in the results as “sup. param.”.
For the rest of the layers encountered below the excavation depth, representative values were considered for the geotechnical parameters for all models.
Based on all the available information, for the analysis in the present paper the following parameters were considered for the soil (Table 1) and the structural elements (Table 2):
Table 1: Lithology and main geotechnical parameters
where: γnat – unit weight at natural moisture content; E50=Eoedref – the triaxial loading stiffness considered equal to the oedometer modulus at the reference pressure; c – cohesion in effective stresses; Φ – internal friction angle in effective stresses; pref – reference pressure; G0ref – small strain shear modulus at the reference pressure, mean – mean values; inf. – inferior characteristic values; sup. – superior characteristic values
Table 2: Properties of the linear elastic elements
where: E – deformation modulus; A – cross section area; I – cross section inertia modulus; ν – Poisson coefficient
The inferior and superior representative values of the geotechnical parameters were obtained either by statistical description when sufficient data was available (characteristic values), or by cautious estimate (nominal values).
Furthermore, a sensitivity analysis was performed for the main geotechnical parameters that would influence the behavior of the retaining wall.The variation of the parameters was performed only for the layers over the excavation depth (filling and silty clay layers), which are dominant for the case.
The analysis of the excavation retaining system was performed using a Finite Element model in plane strain state using Plaxis 2D software, with an elasto-plastic constitutive law for the ground – small strain hardening soil. The soil-structure interface was modelled using Mohr-Coulomb law, associated with the soil strength parameters reduced by Rinter factor considered 0.7. The concrete and steel elements (piles and steel struts) were modelled using linear elastic laws and properties.
RESULTS EVALUATION AND COMPARISON TO THE INCLINOMETER MEASUREMENTS
- Pile wall supported by steel struts
The analysis was performed for two critical stages, the final excavation and the strut dismantling stages. The displacements obtained from inclinometer measurements were situated below the design estimation using mean values for the main geotechnical parameters, for all de execution stages. It can be observed that for the final excavation stage, when the wall was supported by struts, the displacement curve is closer to the more optimistic scenarios, while after execution of the raft and dismantling of struts, the curve gets closer to the one obtained using the mean values of the main geotechnical parameters. The differences obtained can be justified by more favourable ground properties or site conditions than expected, but also by a partial mobilization of the ground pressure. All in all, it can be concluded that the estimates with the mean values of the main geotechnical parameters are in a good match with the measurements for S3 and S4 sections in this project.
Figure 3: Horizontal displacements of the retaining wall – Section 3 (left – final excavation stage, right – strut dismantling after raft execution)
Figure 4: Horizontal displacements of the retaining wall – Section 4 (left – final excavation stage, right – strut dismantling after raft execution)
In addition, a sensitivity analysis was performed for the main parameters of the soil layers encountered on the excavation depth (fig. 5).
Figure 5: Sensitivity analysis performed for the soil layers encountered on the excavation depth
The sensitivity analysis showed that the parameters with the greatest influence in the retaining system behaviour are the small strain shear modulus followed by the shear resistance parameters: cohesion and internal friction angle, while the unit weight variation is the least influential. Also, the parameters for the filling layer are less influencing, especially for S3 section.
- Self-supported pile wall
The calculation for the self-supported retaining was initially performed using the same scenarios as for the retaining wall supported by struts. In contrast to the previous analysed sections, for the self-supported wall the results showed a high difference between the estimates based on calculation and measurements – the measured displacement was about 3 times higher than the displacement estimated using mean parameters for the main geotechnical parameters of the first layers (fig. 6).
Figure 6: Horizontal displacements of the self-supported retaining wall at the final excavation stage – Section 10
Therefore, for this section, some adjustments of the model had to be considered to fit the results of the measurements. At first, given the fact that the retaining wall was near the property limit and no geotechnical information was available from the neighbouring site, a worse scenario for the lithology was considered by increasing the thickness of the filling layer from 1.30 m to 2.5 m (“MEAN &Hfill=2.5 m”). So, the horizontal displacement of the pile wall increased by 26%, but it was still very low compared to the inclinometer measurement.
Afterwards, because on this area some difficulties were encountered during the execution of the piles,despite the works were done under rigorous quality control, that led to high deviations and, thus, to the necessity of breaking part of the concrete cover to accommodate the infrastructure works, a reduction of 20% in the rigidity of the piles was considered (diameter and deformation modulus reduction) (“MEAN PARAM. & Reduced rigidity (20%)”). This assumption led to an increaseby 28% of the horizontal displacement of the pile wall.
The abovementioned two hypothesis combined (“MEAN PARAM. &Hfill=2.5 m &Reduced rigidity (20%)”) led to about 37% difference of the horizontal displacement from the measured one, while considering these with the pessimistic values of the main geotechnical parameters, it led to about 6% overestimation of the displacements compared to the measurements.
The latter scenario led to very close estimate of the horizontal displacement in relation with the measured displacements, thus, concluding that such a combination of unfavourable conditions could lead to these differences between initial design assumptions and actual site conditions.
Figure 7: Sensitivity analysis performed for the soil layers encountered on the excavation depth
The same sensitivity analysis was performed for the soil layers encountered on the excavation depth, as for the case of supported wall. This also showed that the parameters with the greatest influence in the retaining system behaviour are the small strain shear modulus followed by the shear resistance parameters and that the filling layer is less influential than the silty clay layer, although it is more significant than in the case of the previously shown sections S3 and S4 (fig. 7).
CONCLUSIONS
The analyses performed for the case of the retaining wall supported by steel struts showed a good approximation of the calculation estimations relative to the inclinometer measurements for both analysed sections. The displacement obtained from inclinometer measurements were situated below the estimates using mean values for the geotechnical parameters, considered to be the best estimate scenario, for all de execution stages.
In contrast, for the self-supported wall, similar analyses showed a high difference between calculation and measurements – about 3 times higher measured displacements. So, it was found that only a combination of factors, such as soil conditions and properties and retaining wall execution, could lead to such differences.
As it can be seen from the analysis presented above, even though many investigation works were available from the site and the execution was made under strict quality control, in some characteristic sections the registered displacement of the pile wall was higher than estimated by calculation, but overall, in most of the design sections, the registered displacements were smaller than the ones estimated cautiously using representative values or even mean values “most probable”.
As a conclusion, there are many variables to be considered in the design of geotechnical structures and retaining walls in particular. Thus, measurements are of most importance for these works and when they show important differences to the design estimations, diverse hypothesis must be reconsidered to be able to intervene in time, but also to gain valuable knowledge of such cases.
REFERENCES
[1] NP 074 (2014) Technical Romanian Norm regarding the geotechnical documentation for constructions;
[2] NP 120 (2010) Technical Romanian Norm regarding design, construction and monitoring of deep excavation in urban areas;
[3] NP 124 (2010) Technical Romanian Norm regarding the geotechnical design of retaining structures;
[4] PLAXIS 2D (2020) Reference manual;
[5] POPA, H., ENE, A., MARCU, D. (2015): Back-analysis of an anchored retaining structure of a deep excavation. Proceedings of the XVI ECSMGE Geotechnical Engineering for Infrastructure and Development; September 13-17, 2015, Edinburgh, Scotland. 3995-4000;
[6] POPA, H., ENE, A., MIRIȚOIU, R., IONESCU, I., MARCU, D. (2018): Back-analysis of an embedded wall for a deep excavation in Bucharest. Proceedings of the XVI Danube – European conference on geotechnical engineering; June 07-09, 2018, Skopje, Republic of Macedonia. 743-784;
[7] SR EN 1997-1:2004/NB:2016. Eurocode 7: Geotechnical design. Part 1: General Rules. National bulletin;
[8] SR EN 1997-1:2004. Eurocode 7: Geotechnical design. Part 1: General Rules;
[9] SR EN 1997-2:2004. Eurocode 7: Geotechnical design. Part 2: Ground investigation and testing.
Autors:
Ionela CIOCANIU – Popp & Asociații Inginerie Geotehnică,
Alexandra ENE – Popp & Asociații Inginerie Geotehnică; Technical University of Civil Engineering Bucharest,
Horațiu POPA – Technical University of Civil Engineering Bucharest, Romania
[Proceedings of the 17th Danube European Conference on Geotechnical Engineering (17DECGE), June 7-9, 2023, Bucharest, Romania – https://17decge.ro/]
…citeste articolul integral in Revista Constructiilor nr. 205 – august 2023, pag. 62-65
Daca v-a placut articolul de mai sus
abonati-va aici la newsletter-ul Revistei Constructiilor
pentru a primi, prin email, informatii de actualitate din aceeasi categorie!
Lasă un răspuns