Numerical modelling of water jet induced erosion

This research project that is carried out in the wider research context of experimental and numerical geotechnics deals with the investigation of jet grouting. This ground injection method is frequently used in geotechnical engineering for the production of underpinnings and piles, reducing settlements of new and existing foundations, supporting open and underground excavations, and creating water cut-offs for dams. Using a high-pressure pump, a cutting jet of water or binder suspension is used at a pressure of approx. 500 bar to cut or erode the soil. Water or grout mixture are injected through a nozzle in the grout pipe, which is rotated continuously, and the radially propagating jet from the borehole axis erodes subsurface soil. A large number of factors influences the jet grouting process and its efficiency: on the one hand, soli properties play an important role, on the other hand, factors such as the used jet material, the geometry, distance and arrangement of the nozzles, the pressure of the jet as well as the speed of rotation and pulling rate of the nozzle have a big impact. Even though jet grouting is a ground improvement technique which nowadays contributes to the solution of several geotechnical problems, large uncertainties still exist in the prediction of the diameter of jet columns. In the course of the project, we use a combination of experiments and numerical simulations to investigate on the details of the process. The main challenge is that effects that take place on very small scales affect the overall behaviour of the entire process. It is thus necessary to resolve small details with the simulations while at the same time also the entire, large setup must be captured. For being able to depict the presence of both fluids and solids (bulk materials such as soil), two different, established CFD-DEM (coupling between fluid dynamics method and particle simulation method) methods are combined. Particularly the treatment of the interface between the two methods is challenging and must be developed in the course of this project. The experiments are conducted by the research partner at the Institute of Geotechnical Engineering and Construction Management at the Hamburg University of Technology.

The goal of the DACH Cooperation project with FWF project number I 5165 between Prof. Grabe (Institute of Geotechnical Engineering and Construction Management at Hamburg University of Technology (TUHH)) and Dr. Christoph Kloss (DCS Computing GmbH) was the investigation of the jet grouting process and development of numerical tools to model such a process. A key aspect of the jet grouting process is the presence of a wide range of length and time scales, making the simulations of such processes computationally very expensive. In the first project phase we focused on the development of a novel (static) spatially variable hybrid (unresolved/resolved) model. This model reduces the overall computational cost for a multi-scale processes, while maintaining physical accuracy within specific regions. The model has been implemented in the open source simulation environment CFDEMcoupling, and is generally applicable. To ensure a correct implementation, the stability and accuracy of the model have been validated by theoretical results and several benchmark scenarios. Two more realistic application cases were considered: a vertical and a horizontal water jet in particle beds. For the vertical case experiments had been conducted by the project partners at TUHH before the start of the project, while for the horizontal case experiments were conducted by the project partners at TUHH during this project. Setups for these application cases were generated and according simulations were conducted for varying boundary conditions. The soil was represented as a packing of particles, which enabled the consideration of effects on a very fine-grained level. Although due to the complexity of the jet grouting process a quantitative validation against the experiments is not within reach yet, a qualitative validation against available pressure data shows promising results. The newly developed numerical model will be handed over the project partner, who will apply it to more similar setups. The results and outcomes of the research project have been presented at international conferences and to specialists from both the field of geotechnics and numerical simulation.