Abstract
|
:
|
Developing a landfill model is essential to understanding how the landfill gases are generated, and migrate. In this Thesis, four different aspects of landfill modeling are studied, in order to design a realistic model for a landfill, and an efficient landfill numerical simulator. After an introductory chapter, Chapter 2 of the Thesis deals with developing a dynamic model of a landfill, developing a numerical simulator for the model, and utilizing them to study the effects of various parameters of the landfill on the pressure and concentration profiles of the various landfill gases (LFG). The effects of the heterogeneity of the landfill, in terms of the spatial distributions of the porosity, totousity, permeability, and the potentials for methane and carbon dioxide generation are studied, along with the effect of the mechanical dispersion. The model is described in detail, along with an efficient iterative technique used to solve the set of nonlinear equations that govern the partial pressures and concentrations of the LFG. Chapter 3 studies the problem of optimizing the model of a landfill, using a limited experimental data. The optimization process attempts to adjust the spatial distributions of the porosity, tortuosity, permeability, and the potentials for methane and carbon dioxide generation. The Genetic Algorithm (GA) is used to optimize the landfill model, since the GA has the ability to locate the global minima of complex functions, and and is amenable to parallel computations for a large number of unknowns. Chapter 4 further develops the model studied in Chapter 2, in order to include modeling of two-phase flow in landfills. Water plays a pivotal role in stabilizing landfills but also poses an environmental problem. If contaminated water escapes out of a landfill, it may contaminate any groundwater source located near the landfill. The results of computer simulations of two-phase flow of gas and liquid are presented in Chapter 4. The last chapter of the Thesis deals with assisting landfill operators in finding the optimal conditions in order to maintain landfills properly. An artificial neural networks (ANN) is used in order to analyze the historical data for a large landfill near Los Angeles, and provide short-term forecasting for various important quantities, such the temperature, and concentration profiles of CH 4 , CO 2 , and O 2 .
|