Abstract :
A numerically based simulator was developed to assist in the interpretation of complex laboratory experiments examining transport
processes of chemical and biological contaminants subject to nonlinear adsorption and/or source terms. The inversion is performed
with any of three nonlinear regression methods, Marquardt–Levenberg, conjugate gradient, or quasi-Newton. The governing equations
for the problem are solved by the method of finite-differences including any combination of three boundary conditions: 1)
Dirichlet, 2) Neumann, and 3) Cauchy. The dispersive terms in the transport equations were solved using the second-order accurate
in time and space Crank–Nicolson scheme, while the advective terms were handled using a third-order in time and space, total
variation diminishing (TVD) scheme that damps spurious oscillations around sharp concentration fronts. The numerical algorithms
were implemented in the computer code INVERTS, which runs on any standard personal computer. Apart from a comprehensive
set of test problems, INVERTS was also used to model the elution of a nonradioactive tracer, 185Re, in a pressurized unsaturated
flow (PUF) experiment with a simulated waste glass for low-activity waste immobilization. Interpretation of the elution profile was
best described with a nonlinear kinetic model for adsorption.
Keywords :
nonlinear regression , transport , simulation , Inverse model , adsorption