Abstract:
Stress and displacement measurements are key tasks for geotechnical modelling. Although broad range of techniques are available; the heterogeneity of the soil and the lack of direct observation of the foundation elements cause accurate measurements hard to obtain. Selection of whether a complicated acquisition system with expensive sensors, versus a simple but large number of sensors using open platforms, becomes a major challenge. To address this problem a range of approaches were implemented; a high-end data acquisition board connected to miniature stress transducers to measure the stress in the soil; sensitive lms and a dedicated software to measure soil-structure contact stress; Arduino based IMU sensors for measuring inclination to calculate corresponding displacements and LVDTs for veri cation of the calculated displacements. A stress transducer, connected to a high-end data acquisition board, enclosed in a silicone paci er that is lled with distilled water, was exposed to a set of di erent pressures under the soil. Although the results from experiments held on the same date were coherent, the repeatability at di erent dates was weak. The prescaled sensitive lms and IMUs were attached on the blocks of the segmental pile, and the pile was placed in a large-scale direct shear test box which was then lled with loose sand. The pile head was xed and was allowed to displace up to 5 cm with a rate of 1 mm per minute. It was observed that the obtained horizontal pressure by using the lms was compatible with calculated value using the relation between the force and the area. Also, a series of three-point bending tests with external LVDTs were performed to verify the de ections from the inclination measured with IMUs. The calculated de ections were within the range of 10% error when compared to those measurements from LVDTs.
