4 results
Now showing1 - 4 of 4
- Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Hybrid surrogate modeling of a quay wall: an automatically tuned framework integrating Long Short-Term Memory and Feedforward Neural NetworksThis study presents the development of a hybrid surrogate model combining Long Short-Term Memory (LSTM) and Feedforward Neural Network (FNN) architecture to predict the maximum bending moment in sheet pile walls. The model integrates static input parameters with wall deformation data as sequential input, enabling accurate mapping of complex geotechnical behavior. To identify the most effective architecture, an extensive hyperparameter tuning process was conducted, investigating various activation functions and network configurations using automated search techniques. The optimized model demonstrates excellent predictive performance, achieving a coefficient of determination R² of 0.999. These results highlight the potential of hybrid neural networks as efficient and reliable tools for surrogate modeling in geotechnical applications.Publicationtype: Conference Paper7 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Numerische Untersuchungen zur Definition von KastenfangedämmenNumerical investigations for the definition of box-type cofferdams. Cofferdams in cell and box form serve as temporary structures for enclosing large excavation pits or as permanent structures for bank reinforcements and breakwaters. Box cofferdams consist of two mutually anchored and parallel sheet pile walls, with fill material in between. Box cofferdams are often subjected to unilateral loads and are designed to withstand large horizontal forces such as water pressures, impact loads, or earth pressures. The recommendations of the Working Committee on Bank Reinforcements [1] provide guidelines for the design of cofferdams, primarily based on the studies by [2]. In the stability verification of box cofferdams, according to [1], it is assumed that a compact soil block exists between the opposing cofferdam walls. However, [1] does not explicitly define when two opposing walls are considered a box cofferdam structure. To define the applicability ranges for box cofferdams, extensive numerical investigations in the serviceability state were carried out for this article.Publicationtype: Journal Article13 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Numerical shear band strengthening with contact modelling for the increase of load capacity of geotechnical structuresGeotechnical constructions often fail due to a localized shear failure. Shear zones are formed in the soil as a result of external loads. To increase the load bearing capacity of geotechnical constructions and to prevent a critical failure mechanism, it is possible to only strengthen the potential shear bands by injecting cement in the beginning of the potential shear bands. This local strengthening makes it possible to increase the load bearing capacity of constructions with a minimum use of material. The soil strengthening is numerically modelled for a strip foundation using FEM and implementing different strengthening criteria within a hypoplastic soil model which considers material transition from sand to strengthened soil. The necessity of realistic contact modelling is shown, and the results both with and without contact modelling are shown. The numerical results show a significant increase in the load bearing capacity compared to the results without shear band strengthening.Publicationtype: Conference Paper11 - Some of the metrics are blocked by yourconsent settings
Item type:Publication, Access status: without files , Sustainable dike adaptation measures using finite element method and optimization algorithm NSGA-IIGiven the context of climate change, it is imperative to adapt the majority of existing flood defence systems using sustainable methods. Multi-objective evolutionary algorithms, like NSGA-II, offer a valuable decision support tool for exploring different designs based on predefined objectives. This paper specifically examines the application of the NSGA-II optimization algorithm in geotechnical problems by integrating it with FEM simulations, where slope horizontal runs are considered as variables. Throughout the study, different sets of termination criteria are employed to optimize the reduction of structural exploitation and the expected CO2 footprint.Publicationtype: Conference Paper32