In this article, the authors consider a completely new approach in design, which is related to the use of previously un-adapted technologies known to bridge engineering in geotechnical issues for prestressing of diaphragm wall during deep excavations execution. The bridge technology described here is the prestressing of concrete structures. Hazards related to deep excavations and methods of digging them, such as the ceiling method and top&down method, are presented. Current problems in supporting deep excavation slopes are related to the use of extensive quantities of materials (such as steel struts, ground anchors, or concrete and reinforcement steel). The authors’ method helps to achieve a higher level of sustainability, which is important in a modern approach to geotechnical engineering. The non-linear arrangements of the cables according to the occurrence of the prestressing moments for a given phase are presented. Results related to numerical analysis—showing the correctness of the method and cost optimization results, showing possible savings are presented. The article is a part of the set. In the second (already published) article titled “Modern Methods of Diaphragm Walls Design”, the authors present the concept of the calculation methodology for diaphragm wall design.
Slope stability prediction is a critical task in geotechnical engineering, but machine learning (ML) models require large datasets, which are often costly and time-consuming to obtain. This study proposes a domain-driven teacher–student framework to overcome data limitations for predicting the dry factor of safety (FS dry). The teacher model, XGBoost, was trained on the original dataset to capture nonlinear relationships among key site-specific features (unit weight, cohesion, friction angle) and assign pseudo-labels to synthetic samples generated via domain-driven simulations. Six student models, random forest (RF), decision tree (DT), shallow artificial neural network (SNN), linear regression (LR), support vector regression (SVR), and K-nearest neighbors (KNN), were trained on the augmented dataset to approximate the teacher’s predictions. Models were evaluated using a train–test split and five-fold cross-validation. RF achieved the highest predictive accuracy, with an R2 of up to 0.9663 and low error metrics (MAE = 0.0233, RMSE = 0.0531), outperforming other student models. Integrating domain knowledge and synthetic data improved prediction reliability despite limited experimental datasets. The framework provides a robust and interpretable tool for slope stability assessment, supporting infrastructure safety in regions with sparse geotechnical data. Future work will expand the dataset with additional field and laboratory tests to further improve model performance.
This study deals with the behavior of composite blends constituted of rigid and impervious grains included in saturated clay paste of kaolin, considered as permeable and deformable. Permeability tests performed during standard oedometr tests (before each load step) highlight the key role of the original and actual state of the clay paste, and show the existence of a threshold of sand grain concentration above which a structuring effect influences its permeability. In the light of these experiments some usual homogenization methods (with simplifying assumptions to make the problem manageable) are considered in order to model the mixture permeability. Qualitative and quantitative comparisons with experimental data point out their respective domain of interest and limitations of such approaches
Accurate prediction of soil shear strength is critical for safe and cost-effective geotechnical design. This study investigates the application of four machine learning (ML) models—Random Forest (RF), Decision Tree (DT), Artificial Neural Network (ANN), and Support Vector Regression (SVR)—to predict the shear strength of soils from Bahir Dar city using laboratory-obtained geotechnical data. A total of 298 soil samples and 13 geotechnical parameters were collected from depths of 0.13–35 m, encompassing both disturbed and undisturbed conditions. The dataset was divided into training (80%) and testing (20%) sets, and models were trained with optimized hyperparameters. The RF model achieved the highest accuracy (R2 = 0.9992, RMSE = 0.0983), followed by DT (R2 = 0.9974, RMSE = 0.1812). ANN and SVR showed lower predictive accuracy, with SVR demonstrating the largest maximum errors. Predicted vs. actual plots, kernel density estimates, and absolute error per sample analysis confirmed that tree-based models provide the most reliable predictions, while ANN and SVR exhibited sporadic large deviations. SHAP analysis revealed that Cohesion, Clay content, and Plasticity Index are the most influential factors in predicting shear strength. The results demonstrate that ensemble tree-based ML models offer a robust and accurate tool for geotechnical prediction, capturing complex nonlinear relationships in soil behavior.
One of the most important subjects of geomechanics research is finding mathematical relationships which could correctly describe behavior of the soil under loading. Safety of every engineering structure depends strongly on accuracy and correctness of this description. As laboratory tests show, macroscopic properties of soil are complicated. Therefore, working out appropriate load-settlement relationships is considered to be a very difficult geomechanics tasks to solve. A majority of constitutive models proposed to date concern mineral soils and there is very little research related to modelling organic soil behavior under loading. In case of organic soils, due to their very complicated and composite structure, constitutive models are often formulated empirically based on laboratory tests of particular soils. The authors of this paper propose a 1-D rheological structure which accounts for complex behavior of soil related to the settlement process. The model simulates immediate reversible elastic settlement and plastic soil deformation as well as primary and secondary (creep effect) consolidation. Material parameters of the model were determined by a curve fitting procedure applied for a natural scale settlement test of plate foundation. The test was carried out in soil conditions connected with Eemian geological structure of Warsaw, i.e., Eemian glacial tunnel valley in Warsaw called Żoliborz Glacial Tunnel Valley filled with organic soils being up to 20 metres thick. This area has lately become an object of interest of investors as a site for building construction.