ITEMS

modern scholarly publishers in the finest tradition

Journals

	Search

Journal of Porous Media

ISSN for PRINT: 1091-028X

Institutional price:	$713.00
Issues per year:	8

Best Paper Award Selection - Editorial Board Site

2007, Volume10

114 pages

Issue price - $99.00

Identification of the Hydraulic Properties of Heterogeneous Rocks from Laboratory Flow-Pump Experiments

Daniel Lesnic
Department of Applied Mathematics, University of Leeds, Leeds LS2 9JT, UK

Simon D. Harris
Rock Deformation Research, School of Earth Sciences, University of Leeds, Leeds, LS2 9JT, UK

Lionel Elliott
Department of Applied Mathematics, University of Leeds, Leeds LS2 9JT, UK

Radu Mustata
Department of Applied Mathematics, University of Leeds, Leeds, LS2 9JT, UK

M. Ben Clennell
Rock Deformation Research, School of Earth Sciences, University of Leeds, Leeds, LS2 9JT, UK

Derek B. Ingham
Department of Applied Mathematical Studies, The University of Leeds, Leeds, LS2 9JT, UK

ABSTRACT

Transient flow analysis has numerous geophysical and geotechnical applications and it is necessary for the understanding and development of laboratory measurement techniques in rock and soil mechanics, particularly in the determination of their permeabilities. In this paper, we model a laboratory flow-pump permeability test and formulate an inversion technique in order to retrieve homogeneous/spacewise dependent/discontinuous properties of materials. The direct problem is solved using boundary element or finite difference methods, while the recovery of the hydraulic properties of rocks is sought through a genetic algorithm (GA) optimization approach. Well-known constrained optimization routines are often trapped in a local optimum due to a poor initial guess, and further require the calculation of the gradient of a least-squares type functional that is assumed to be differentiable. However, GAs find the global optimum and require neither the calculation of the gradient nor the assumption that the functional be differentiable. Both exact and simulated noisy data are incorporated at optimally selected instants in time through the test, the data measurements used being consistent with the sensitivity analysis that is performed prior to inversion.