Settings

Now showing1 - 7 of 7
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Research Data, Access status: with files ,
    Dataset of Simulated Undrained Cofferdam Models for Parameteric Study
    (2026-07-06)
    Cerek, Kacper
    ;
    Hadjiloo, Elnaz
    ;
    Grabe, Jürgen
    This dataset includes the numerical results of a undrained cofferdam box considering a parameter variation. The simulations have all been conducted using the Finite Element Method with the commercial software PLAXIS 2D. The investigated cofferdams represent a cofferdam box used as an excavation enclosure. The Hardening Soil Model with Small-Strain Stiffness is used as the material model. The results show the wall deformations, vertical stress distributions, and earth pressures on the cofferdam walls for different cofferdam dimensions and various parameter combinations.
    Data Type: Simulation Data
      11  9
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Research Data, Access status: Embargo ,
    Synthetic Dataset of Anchored Backfilled Quay Walls for Surrogate Model Development
    (2026-05-29)
    Cerek, Kacper
    This data publication comprises a synthetic databank obtained through numerical simulation of an anchored, backfilled sheet pile quay wall. The simulations were executed using the finite element method (FEM) within the commercial software package PLAXIS 2D in version 2024.3 driven via a remote Python scripting interface. The databank is designed to facilitate foundations for supervised machine learning architectures, such as neural network-based surrogate modelling, for soil-structure interaction problems without requiring computationally intensive continuum calculations.
    Data Type: Simulation Data
      27  22
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Research Data, Access status: with files ,
    Experimental dataset of triaxial tests with cemented sand
    (2026-03-02)
    Lennart Sekulic
    ;
    Hadjiloo, Elnaz
    ;
    Cerek, Kacper
    ;
    Grabe, Jürgen
    This extensive dataset contains the results of 175 triaxial tests on dense cemented sand samples with measurements of a height of 72 mm and a diameter of 36 mm. The sands used include Hamburg Sand (HH), known for its coarser grain size, and Cuxhaven Sand (CX), which has a finer grain size as also shown in Figure 2. The cement paste consists of Dyckerhoff MIKRODUR R-U Mikrozement and water. The tests were conducted to analyze the mechanical properties of cemented Hamburg sand and cemented Cuxhaven sand under various conditions, specifically focusing on the effects of confining pressure and cementation degree, as shown in Table 1. The sample preparation involved layering and compacting the sand-cement-water mixture in Plexiglas molds, followed by a curing process with initial sealing and subsequent water immersion for 28 days.
    Data Type: Experimental Data
      18  50
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Research Data, Access status: with files ,
    Micro-CT Dataset of Cemented Sand Specimens under Uniaxial Compression with Varying Cementation Degree
    (2026-01-15)
    Hadjiloo, Elnaz
    ;
    Sekulic, Lennart
    ;
    Dao, Duy Anh
    ;
    Cerek, Kacper
    ;
    Grabe, Jürgen
    Computed tomography (CT) is an X-ray–based imaging technique that enables high-resolution analysis of the internal structure of materials. During scanning, X-rays are attenuated while passing through the specimen (depending on density and composition), and a three-dimensional volume is reconstructed from multiple two-dimensional projections acquired at different rotation angles. In this study, CT was used to investigate the microstructure of cemented sand specimens subjected to uniaxial loading, with a focus on identifying the spatial distribution of grains and cement binder as well as the pore structure. The resulting volumetric data support microstructural characterization and serve as input for subsequent numerical (FEM) modelling. The dataset provides the raw, unprocessed CT image stacks as acquired. The CT data consist of reconstructed slice images (i.e., cross-sectional slices already reconstructed from the original projection views). The CT data were acquired at the 3SR Laboratory at Université Grenoble Alpes. Note that the first and last slices are typically not suitable for quantitative evaluation (e.g., due to boundary effects and reconstruction artefacts) and may be excluded during post-processing.
    Data Type: Experimental Data
      75  120
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Research Data, Access status: with files ,
    Dataset of Simulated Cofferdam Models for Parameteric Study
    (2024-12-05)
    Cerek, Kacper
    ;
    Hadjiloo, Elnaz
    ;
    Grabe, Jürgen
    This dataset includes the numerical results of a cofferdam box considering a parameter variation. The simulations have all been conducted using the Finite Element Method with the commercial software PLAXIS 2D. The investigated cofferdams represent a cofferdam box used as an excavation enclosure. The Hardening Soil Model with Small-Strain Stiffness is used as the material model. The results show the wall deformations, vertical stress distributions, and earth pressures on the cofferdam walls for different cofferdam dimensions and various parameter combinations.
    Data Type: Simulation Data
      49  41
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Research Data, Access status: with files ,
    Python Implementation of Bidirectional LSTM for Sequential Data Processing
    (2024-08-02)
    Cerek, Kacper
    ;
    Gupta, Arjun
    ;
    Dao, Duy Anh
    ;
    Hadjiloo, Elnaz
    ;
    Grabe, Jürgen
    This repository contains a Python script for processing and analysing sequential data using a Bidirectional Long Short-Term Memory (LSTM) model. The script is designed to perform various configurations, evaluate model performance, and save results for further analysis. It specifically targets parameter studies conducted on simulated CRS element tests. The script applies different configurations to train the model and evaluate its performance on test data, including varying data percentages, point skips, number of time steps, and batch sizes.
    Data Type: Source Code
      61  36
  • Some of the metrics are blocked by your 
    consent settings
    Item type:Research Data, Access status: with files ,
    Dataset of Simulated CRS Tests for Advanced Soil Parameter Identification
    (2024-04-02)
    Cerek, Kacper
    ;
    Dao, Duy Anh
    ;
    Hadjiloo, Elnaz
    ;
    Grabe, Jürgen
    This extensive dataset contains the results of 30,000 simulated Constant Rate Strain (CRS) tests, comprising three constitutive models with 10,000 simulations each: linear-elastic perfectly-plastic soil model (Mohr-Coulomb), Hardening Soil, and Hypoplasticity with intergranular strain. These simulations are all conducted using the finite element software, Plaxis. Within the simulations, the model parameters were altered in order to generate an extensive data bank with various system responses. The primary purpose of this dataset is to serve as a training foundation for an Artificial Neural Network (ANN). The goal of the ANN is to efficiently and accurately identify suitable parameter sets for the three widely used soil models, based on data sets provided from numerical CRS tests. This identification is critical for enhancing the precision of soil behaviour predictions in numerical structure-soil interaction models.
    Data Type: Simulation Data
      104  1636