Improved Image-Based Deformation Measurement in the Centrifuge Environment

Abstract

This paper describes a new apparatus and techniques for performing deformation measurements using particle image velocimetry in the centrifuge environment. The new system includes camera, lighting, and control equipment that facilitates image capture at least 30 times faster than that in legacy systems. Methods for optimizing the addition of artificial seeding on the exposed plane of a geotechnical model are also set out. These techniques ensure that the precision of the deformation calculations is optimized even in models with multiple layers of different soils, fully harnessing the method's capabilities. An example application of a flat footing penetrating sand overlying clay is used to illustrate the performance of the equipment and the artificial seeding optimization technique. Deformation fields at the point of peak resistance during punch-through are presented in the form of vector fields, normalized displacement contours, and shear strain contours. It is shown that the advances in equipment and artificial seeding allow both macroscopic and grain-scale deformation features to be identified. These analyses highlight not only the benefits of the new technology, but also the need for carefully optimized experimental procedures to maximize the measurement precision.

Synchronised multi-scale image analysis of soil deformations

“…The multi-scale system hardware is shown in Figure 2. It is a further development on the system described by Stanier and White (2013) and comprises of the following components:…”

“…The two control units are connected using an additional cable so that triggering of the two cameras is synchronous, with the 'slave' camera duplicating the activity of the 'master' camera. The controllers also incorporate an LED control unit (Gardasoft RT220) with a LAN interface allowing a TCP/IP link to control the brightness of LED panels independently (see Stanier & White, 2013).…”

“…The multi-scale PIV/DIC system is controlled by an extension of the software described by Stanier and White (2013). Specifically the software comprises of two packages, written and compiled in the National Instruments LabVIEW environment:…”

“…Prior to pluviation, a fraction of the sand was dyed black and mixed uniformly with undyed sand to enhance the contrast between neighbouring particles. Optimal contrast was achieved by following the procedure recommended by Stanier and White (2013) such that the artificial seeding ratio (ASR) was ~0.5.…”

“…To provide a comprehensive analysis of the impact of the 'micro' view on subset texture quality, two subset texture quality parameters have been calculated: (i) the standard deviation of subset pixel intensities ( ) (Stanier and White, 2013); and (ii) the sum of square of subset intensity gradients (SSSIG) (Pan et al 2008). The former is a simple measure of the contrast within the subset with values of > 15 indicating adequate contrast (Stanier and White, 2013) whilst the latter is a measure of the distribution of the contrast with values of SSSIG > 10 5 indicating adequate texture distribution (Pan et al 2008).…”