The purpose of this paper is to present long-term measurements in a full-scale study on a basal reinforced piled embankment that make it possible to validate calculations used for the design of the geosynthetic reinforcement (GR). These calculations are normally carried out in two steps. To validate steps 1 and 2 together, it is necessary to measure GR strains. To validate calculation steps 1 and 2 separately, arching A needs to be measured, which is the pressure on the pile cap above the GR. An extensive monitoring project was conducted over a period of four years, in a basal reinforced piled embankment on 17 m of soft clay and peat. This study presents the measured GR strains and load distribution including arching, accompanied by measured groundwater levels and deformations. The subsoil support of the geosynthetic reinforcement disappeared quickly, arching developed over the first three months, and an annual cycle in the load distribution became apparent. Arching effects increase during the summer when conditions are relatively dry, resulting in a larger load on the piles and a reduction in the load on the GR. Additionally, the measured changes after an extremely rainy week are presented.

A constitutive model for granular materials which considers grain crushing effects is developed in the framework of hypoplasticity. As grain crushing occurs the behaviour of granular material can usually be significantly affected. Several empirical relations between peak strength, uniformity coefficient and stiffness of sand depending on stress level or amount of grain crushing have been derived in the past. In this paper, such relations are employed to improve a basic hypoplastic constitutive model based on the changes of stress level or grain size distribution. In the proposed modified hypoplastic model only two additional physical parameters, namely uniformity coefficient and mean grain size are incorporated. The validation of the modified model for three different sands under triaxial test response with cell pressures up to 30 MPa is presented and shows a significantly better correspondence with regard to the original basic hypoplastic model.

Three soil-arching evolution patterns in unreinforced piled embankments were observed in a series of two-dimensional (2D) model tests using a multitrapdoor test setup. These include the triangular expanding pattern, the tower-shaped evolution pattern, and the equal settlement pattern. The inclination of the slip surfaces and the height of the vertical slip surfaces that enclose the tower-shaped arches were found to be the critical parameters describing the arching evolution. Three analytical models were proposed to describe the evolution processes of the three arching-evolution patterns and to find the stress distributions of the corresponding processes. Load distribution equations were also derived from these models. Using the empirical relationships between the inclination of the slip surfaces and the tower height and settlement, the stress distribution ratio during the entire evolution process was calculated. The models matched the model tests well.

Distributed temperature sensing (DTS) can be used to monitor the production process of diaphragm walls. DTS is able to differentiate between already present or fresh bentonite suspensions during refreshing of the bentonite slurry in the trench. During concrete casting, DTS is able to differentiate between the bentonite suspension and concrete. As a result, the continuity of the casting process and the arrival of good grade concrete at crucial locations in the trench can be monitored. Tests conducted on laboratory models provided reference information for interpretation of field data. Field experiences have shown the benefits of DTS tests and the predictive value of the reference measurements. Results are compared with crosshole sonic logging measurements at the same location.

Experimental data of laboratory testing of sands from plane strain compression tests are much less available than those from triaxial tests. In order to get the strength parameters of the sands used in 2D piled embankment model tests and the micro parameters used in the DEM simulations of these tests, a series of tests are done using plane strain test apparatus. These tests consist of three groups with different grain sizes: two coarse sands and a medium sand. For these tests, the relative density was set to 85%. A relative density of 50% was applied to assess the influence of the density. The sample size is large, 600mm^∗ 600mm^∗ 300mm (length^∗ width^∗height) in order to eliminate the side friction effects of the container. The results prove that relative density has a significant impact on the experimental results for the same type of sand, while the grain size has negligible impact. The high peak friction angles are attributed to the feldspar component, confining condition of plane strain tests, and high relative density. Tests of sands with different grain sizes and relative density also provide detailed data for the DEM numerical simulations. The high friction properties can be calibrated using the non-circular particle clumps.