About the event
Physical modelling can be used to understand complex soil-structure interactions, which might be difficult to model numerically. Modern instrumentation (e.g. LVDTs, strain gauges, load cells) allows accurate information to be gathered on the displacement of structural elements. Additionally, modern digital image correlations (DIC) techniques allow displacements within the soil mass to be determined. Using this information, a variety of techniques can be used to determine strain throughout the system. Typically, simple analytical and/or numerical tools are used to interpret the resulting data. However, more sophisticated methods including automated back calculation (i.e. inverse analysis) of soil properties from test responses might yield important insights. In this presentation, an optimization technique capable of fitting a strain field, to the known boundary conditions, simple material assumptions (e.g. rigid body motion, incompressibility) and displacement data, from both instrumentation and DIC, will be described. Furthermore, an inverse method for determining soil properties using the derived strain fields is described.