10 results
Now showing1 - 10 of 10
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Workshop numerical methods in geotechnics: AI in geotechnics, deformation based design, numerical optimization(Technische Universität Hamburg, 2026-09-17); ; ; ; The Workshop “Numerical Methods in Geotechnics”, held on 17–18 September 2026 in Hamburg, Germany, brings together researchers and practitioners to discuss recent developments in computational geotechnics. The workshop focuses on three main topics: artificial intelligence in geotechnics, deformation-based design, and numerical optimization. The contributions presented in these proceedings cover data-driven and AI-assisted modelling, surrogate models, advanced finite element methods, uncertainty quantification, soil–structure interaction, and optimization approaches in geotechnical engineering. Together, they highlight current advances in numerical methods and their growing importance for more predictive, efficient, and reliable geotechnical analysis and design.Publicationtype: Conference Proceedings125 66 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Development of neural network-based surrogate models in geotechnics: a state-of- the-art reviewPurpose -This state-of-the-art paper provides an overview of surrogate modelling methods suitable for geotechnical engineering, covering the entire cycle of model development. Design/methodology/approach -The paper discusses neural network-based surrogate modelling and its applicability in geotechnical engineering, including its limitations. It addresses techniques for data set generation, sampling, tuning, selecting activation functions and assessing accuracy Findings - Each step during the development of surrogate models offers various options and therefore requires a systematic approach. Besides standardised benchmarking and richer data sets key areas for future research include advanced or hybrid architectures, such as physics-informed long short-term memory networks and transformers to improve generalisability, efficiency and physical consistency.</jats:p> Originality/value -This paper reviews and highlights the most promising methods for surrogate model development in geotechnical engineering, considering the most critical aspects for building high-performing predictive models and lays an outlook for future research.Publicationtype: Journal Article7 3 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Final report of DFG project 448085183: numerical modelling of partially cemented soils in the stagnation zone(2026-05-20); ; ; This final report summarizes the results of DFG project 448085183, “Numerical Modelling of Partially Cemented Soils in the Stagnation Zone”. The project investigated the micromechanical behaviour of partially cemented sands by combining experimental testing, X-ray micro-computed tomography, image processing, and advanced numerical modelling. In the first project phase, the focus was on the linear-elastic response of cemented sand specimens. CT-based microstructures were converted into numerical models and analysed using the Finite Element Method (FEM) and the Finite Cell Method (FCM), including numerical homogenization to determine effective material properties. In the second project phase, the numerical frameworks were extended towards fracture and failure processes, including matrix cracking, grain-scale stress localization, and phase-field modelling. Additional experimental data were obtained by Digital Volume Correlation (DVC) to support the validation of the simulations. The project contributes to a better understanding of load transfer, stiffness development, and failure mechanisms in cemented granular materials and provides automated CT-based simulation workflows for future applications in geotechnical modelling.Publicationtype: Research Report70 47 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Optimising geometries of drag embedment anchors for floating offshore structures using a multi-objective evolutionary algorithmFloating offshore renewable energy systems rely on secure anchoring, e.g. drag embedment anchors (DEAs). DEA design involves balancing multiple objectives and considering various geometric parameters. Although multi-objective optimisation packages are widely available, DEA optimisation is still frequently handled heuristically. This study introduces a workflow that couples established semi-analytical DEA kinematics in MATLAB with multi-objective evolutionary optimisation (gamultiobj, based on NSGA-II) to identify anchor geometries that minimise anchor volume while maximising bearing capacity and penetration depth. Radar charts and the visualisation of the Pareto front across the design space provide interpretable structures that are not directly apparent from the governing equations alone. Three case studies are presented. The first identifies Pareto-optimal fluke and shank lengths and examines the influence of the boundary on the solutions. The second expands the design space to include fluke thickness, junction length, and fluke–shank angle. Clustering of Pareto solutions reveals three consistent geometric groups: (i) generally large anchors with small junctions, (ii) anchors with larger junctions and variable proportions, and (iii) anchors with broad flukes. A high fluke-shank angle is consistently present across all groups and appears beneficial for all objectives. The third case applies a weighted-sum formulation to produce a single design. Overall, this study provides a workflow for optimisation-driven exploration of DEA design, including a classification of Pareto-optimal geometries that offers a compact way to assess trade-offs from concept to sizing.Publicationtype: Journal Article22 10 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Hybrid surrogate models of quay wallsThe use of hybrid surrogate modelling techniques for the life-cycle management of anchored quay walls was investigated. A synthetic dataset was generated using a simplified structural model based on classical earth pressure theory, representing a range of geometrical and hydraulic boundary conditions. Two neural network (NN) architectures were developed and compared: (a) a baseline feedforward neural network (FNN) using static input features and (b) a hybrid model combining bidirectional long short-term memory (BiLSTM) layers with dense layers (BiLSTM–FNN), which incorporates sequential displacement data. Both models were tuned across multiple trials with varying architectures, activation functions and learning rates. The final architectures were deployed in supervised learning to train surrogate models. The BiLSTM–FNN model outperformed the FNN, achieving significantly lower validation loss and superior predictive accuracy, but at a higher computational cost. This modelling approach provides an effective tool for estimating internal structural forces such as maximum bending moments, thereby supporting predictive maintenance and optimised design. The results demonstrate the potential of hybrid NN architectures within digital twin frameworks for port infrastructure, contributing to enhanced resilience and more efficient resource use.Publicationtype: Journal Article12 16 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Optimization of drag embedment anchors applying multi-objective evolutionary algorithm NSGA-IIEstablishing renewables on a floating platform in the deep sea needs secure anchoring to the seabed, commonly achieved with drag embedment anchors (DEAs). The conventional design process relies heavily on empirical testing and is often time and resource-intensive, potentially leading to suboptimal designs. This research aims to overcome these limitations by applying an evolutionary optimization algorithm to existing analytical solutions for DEAs, identifying optimal anchor fluke and shank lengths. By leveraging an optimization strategy, we aim to enhance the design process while diminishing the dependency on exhaustive physical testing and high computational cost. We employ the Non-Dominated Sorting Genetic Algorithm II (NSGA-II) to optimize anchor shapes, with a focus on three key objectives: maximizing embedment depth and bearing capacity, and minimizing anchor volume. The methodology presents a Pareto front, encompassing all optimal solutions based on the formulated objectives, and demonstrates the efficiency of NSGA-II as a tool for optimizing anchor shapes.Publicationtype: Journal Article17 10 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Predicting soil stress–strain behaviour with bidirectional long short-term memory networksPurpose Artificial intelligence, particularly deep learning (DL), has increasingly influenced various scientific fields, including soil mechanics. This paper aims to present a novel DL application of long short-term memory (LSTM) networks for predicting soil behaviour during constant rate of strain (CRS) tests. Design/methodology/approach LSTMs are adept at capturing long-term dependencies in sequential data, making them suitable for predicting the complex, nonlinear stress–strain behaviour of soil. This paper evaluates various LSTM configurations, optimising parameters such as step size, batch size, data sampling rate and training subset size to balance prediction accuracy and computational efficiency. The study uses a comprehensive data set from numerical finite element method simulations conducted with PLAXIS 2D and laboratory CRS tests. Findings The proposed LSTM model, trained on data at lower stress levels, accurately forecasts soil behaviour at higher stress levels. The optimal LSTM setup achieved a median error of 3.59% and 5.10% for numerical data and 3.86% for laboratory data, presenting the setup’s effectiveness. Originality/value This approach reduces the required time to complete extensive laboratory testing, aligning with sustainable industrial practices. The findings suggest that LSTM networks can enhance geotechnical engineering applications by efficiently predicting soil behaviour.Publicationtype: Journal Article8 12 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Investigation of shear band strengthening by using different strengthening criteriaTo prevent a critical failure mechanism, it is possible to simply strengthen the potential shear bands, which is a patented idea. The localization of potential shear bands that are most likely to form shear can be calculated. Before these shear bands can appear, they can be strengthened by injecting a cement suspension or using jet grouting at the locations where the most likely shear bands would occur. As a result, it is possible to increase the bearing capacity of the geotechnical system by focusing solely on the shear bands that are moste likely to form, minimizing the use of materials. Numerical modeling is employed to analyze the increase in bearing capacity. Four criteria for strengthening the shear bands are applied by incorporating them into a hypoplastic soil model, taking into consideration the material transition from soil to cement, as well as intermediate materials that can arise from jet grouting, which are mixtures of soil and cement. Two fundamental geotechnical systems, namely a shallow foundation and a retaining wall, are analyzed using ABAQUS/Standard for finite element analysis. The results demonstrate the potential of numerical modeling to identify the most critical localized areas in the soil that require strengthening, as well as the substantial increase in bearing capacity achieved after reinforcing the identified areas.Publicationtype: Conference Paper55 53 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Towards the saving potentials using a hybrid topology optimization: application of a coupled deterministic and empirical approach solving a geotechnical optimization problemThis article presents the application of the TOSS (topology optimization for stiffness and strength) algorithm, a hybrid optimization strategy for topology optimization, used for the design of a strip foundation as a basic geotechnical structure. One of the challenges of this field is to improve structural performance by using less material than conventional structures. Therefore, the SIMP (solid isotropic material with penalization) is applied to optimize the foundation concerning its deformational behavior. The downstream SKO (soft kill option) then solves for optimality concerning the foundation's stress distribution. Numerical experiments with the TOSS algorithm reveal great potential for material savings. The coupled solution presents the benefit of reducing (Formula presented.) emissions due to material savings of more than 20% in comparison to the individual SIMP and SKO algorithms. A positive by-product of the method is that stress peaks under the foundation are reduced due to the homogenization of stresses as a result of SKO optimization, thus creating the conditions for better durability of the foundation.Publicationtype: Journal Article45 24 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: with files , Workshop Numerische Methoden in der Geotechnik : 12th & 13th of September 2022 Hamburg, Germany : conference proceedings(2022-09-12); ; ; ; Numerical methods have developed into standard tools for the investigation of geotechnical struc tures. With increasing computational power, hybrid and continuum-based approaches supported by sophisticated constitutive models are becoming more and more important. The Workshop “Nu merical Methods in Geotechnics 2022”, hosted by Hamburg University of Technology (TUHH) in cooperation with the working group Numerics (AK Numerik of DGGT) and the Federal Waterways Engineering and Research Institute (Bundesanstalt für Wasserbau, BAW), brings together international scientists and experts in order to present and discuss newest findings with regard to the development of numerical methods in geotechnical engineering. These proceedings have been compiled based on the different topics presented in the framework of the workshop. They are supposed to preserve the findings for their future scientific and practical use and to share them with the geotechnical community.Publicationtype: Conference Proceedings575 4112