The key to the stability of any building is its foundation, but it is difficult to test some building sites in advance – such as those on the moon. New research from North Carolina State University is helping resolve the problem by using computer models that can utilize a small sample of soil to answer fundamental questions about how soil at a building site will interact with foundations.
“If you are going to build a large structure, you have to run a lot of tests on the building site to learn how the soil will behave in relation to the building’s foundation,” says Dr. Matt Evans, assistant professor of civil, construction and environmental engineering at NC State and co-author of a paper describing the research. “How stable is it? How much might the foundation settle over time? Traditionally, that testing process involves a great deal of equipment, time and money.”
But in some situations, that equipment, time and money is not available. For example, it would be tough to transport the relevant equipment to the surface of the moon.
“We initiated this project, with funding from the North Carolina Space Grant, to answer questions that are essential to the construction of buildings on the moon,” Evans says. “It’s cost-prohibitive to do traditional testing on lunar sites, so we developed a technique for applying computer models that can use a tiny sample to tell us about the potential interface between moon soil and anything we might build.”
And the model may also have applications closer to home. The model could potentially be used to assess soil conditions for remote building sites where traditional testing is impractical or unduly expensive. For example, it could be useful for military applications or for siting remote research facilities.
The paper, “Analysis of Pile Behavior in Granular Soils Using DEM,” focuses on how the model can be used when incorporating Earth-specific variables – such as gravity. However, those variables can be modified to account for conditions on the moon, or even on Mars.
The lead author on the paper is NC State graduate student Jeremy Kress. The paper will be presented Oct. 13 at the 35th Annual Conference on Deep Foundations in Hollywood, Calif.
NC State’s Department of Civil, Construction and Environmental Engineering is part of the university’s College of Engineering.
Note to Editors: The presentation abstract follows.
“Analysis of Pile Behavior in Granular Soils Using DEM”
Authors: Jeremy G. Kress, T. Matthew Evans, North Carolina State University
Presented: Oct. 13, 2010, 35th Annual Conference on Deep Foundations, Hollywood, Calif.
Abstract: We propose the discrete element method (DEM) as a complimentary alternative to analytical solutions and FEM models for the analysis of pile response. DEM is a novel approach to the engineering design of deep foundations which can be used to provide insight into the mechanics of soil-soil and soil-structure interaction. It is common to use analytical, empirical, and semi-empirical methods to predict pile behavior, where if granular properties, pile length, or load orientation change, then significant adjustments are made in the design sequence. DEM provides a more uniform approach in granular soils. Based on constitutive force-displacement relationships and laws of motion in rigid and elastic bodies, few changes must be made for a wide variety grain types, pile types, embedment, and load orientation. Preliminary results for simulation of a single pile are qualitatively consistent with results from conventional design procedures and also provide significant information not available from other analyses (e.g., local information about interface failure during shear). Importantly, these analyses are portable to a variety of pile types and installation procedures, which may not be the case for empirical approaches. Given consistent increases in computing power and rapid advances in model speed and sophistication, the DEM should prove to be a robust method for design and the analysis of pile response in granular soils.