Electronic Theses and Dissertations

Date of Award

1-1-2020

Document Type

Thesis

Degree Name

M.S. in Engineering Science

First Advisor

Robert Holt

Second Advisor

Lance Yarbrough

Third Advisor

Craig Hickey

School

University of Mississippi

Relational Format

dissertation/thesis

Abstract

Understanding the relationship between excitation sources, buried target (i.e., buried hazard, land mine, acoustic article) response, and soil properties is fundamental to improve laser-ground-vibration sensing methods. This project investigates the natural soil’s behavior under acoustic stimuli and compares soil behavior with a buried target through geostatistical methods. Vibrational velocity of sand is measured with an LDV in a confined box filled with and without a buried target. Geostatistical calculations were performed on standardized data (e.g., background velocity and with-target velocity) sets to observe spatial variability. The standardized background velocity is mean 0 and variance of 1, while the addition of the target increases the variance to 27X the background. The background variability resembled uncorrelated white noise. The with-target variogram reveals structural features indicative of the target size and location in the measurement grid. Sensitivity studies evaluate the impact of fewer data and uncorrelated, correlated, and trending noise in the off-target soils. In a subdomain of the measurement grid, the structure of the target is preserved in the variogram and correlate with the size of the grid and surrounding encounters with off-target points. Systematically removing velocity points preserved the target presence with slight changes in the variogram structure according to new separation distances. When uncorrelated noise replaced off-target observations, the target is interpretable from the variogram up to a variance of over 400. Alternatively, when a random field with fixed correlation lengths is applied, the target is obscured at higher variances. Trended data added to off-target observations attempts to simulate field parameters. At increasing variances, strong trends in the background obscure the target. Geostatistical characteristics revealed through data sensitivity studies provides a robust indicator of target presence up to applications of high variability. Small-scale variation in sand provides features indicative of target presence. This study suggests that understanding the spatial structure of the acoustic response of natural soils is critical to the development of land mine detection technologies using an LDV. Future studies should focus on collecting experimental data from field sites.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.