Assessment of boulder content by stochastic modelling and inverse analysis




Assessment of boulder content by stochastic modelling and inverse analysis


A key feature of pile drivability studies is the risk of pile loss caused by structural failure of the pile. A frequent cause of pile loss in the Scandinavian context is the existence of boulders in the soil strata. These boulders prevent the pile from penetrating into the soil as well as increase the risk of impairing its structural integrity. A systematic assessment of the risk of pile loss encompasses an estimation of the boulder content based on the site investigation data. The boulder content can be expressed as a point estimate, and the volumetric boulder content is defined as an interval estimate for the whole soil strata. Based on the information from the site investigation, an inverse analysis can be carried out. Herein, a numerical model for the sample generation is described. A computerized site investigation with the Nordic soil-rock drilling method is executed with the Monte Carlo method in which the boulders are randomly generated. An interval estimation of the boulder content is then formulated based on the number of site investigations. The method can subsequently inform if further site investigation is needed to reduce the confidence interval of the estimate. Further, suggestions for practical use in pile drivability studies are explored.



Maedeh Alinejad; Anders Beijer Lundberg


10th European Conference on Numerical Methods in Geotechnical Engineering (NUMGE2023)



5. Probabilistic and inverse analysis



https://doi.org/10.53243/NUMGE2023-73