Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation




Mining Engineering

First Advisor

Dr. Andrzej M. Wala

Second Advisor

Dr. Alexandre Martin


The following dissertation introduces the hazard of methane buildup in the gob zone, a caved region behind a retreating longwall face. This region serves as a reservoir for methane that can bleed into the mine workings. As this methane mixes with air delivered to the longwall panel, explosive concentrations of methane will be reached.

Computational fluid dynamics (CFD) is one of the many approaches to study the gob environment. Several studies in the past have researched this topic and a general approach has been developed that addresses much of the complexity of the problem. The topic of research herein presents an improvement to the method developed by others. This dissertation details a multi-scale approach that includes the entire mine ventilation network in the computational domain. This allows one to describe these transient, difficult to describe boundaries. The gob region was represented in a conventional CFD model using techniques consistent with past efforts. The boundary conditions, however, were cross coupled with a transient network model of the balance of the ventilation airways. This allows the simulation of complex, time dependent boundary conditions for the model of the gob, including the influence of the mine ventilation system (MVS).

The scenario modeled in this dissertation was a property in south western Pennsylvania, working in the Pittsburgh seam. A calibrated ventilation model was available as a result of a ventilation survey and tracer gas study conducted by NIOSH. The permeability distribution within the gob was based upon FLAC3d modeling results drawn from the literature. Using the multi-scale approach, a total of 22 kilometers of entryway were included in the computational domain, in addition to the three dimensional model of the gob.

The steady state solution to the problem, modeling using this multi-scale approach, was validated against the results from the calibrated ventilation model. Close agreement between the two models was observed, with an average percent difference of less than two percent observed at points scattered throughout the MVS. Transient scenarios, including roof falls at key points in the MVS, were modeling to illustrate the impact on the gob environment.

Digital Object Identifier (DOI)