Return to search

SIMPLIFYING TECHNIQUES APPLIED TO COMPUTATIONAL FLUID DYNAMICS MODELING OF METHANE EXPLOSIONS

Traditional methods of studying underground coal mine explosions are limited to observations and data collected during experimental explosions. These experiments are expensive, time-consuming, and require major facilities, such as the Lake Lynn Experimental Mine. The development of computational fluid dynamics (CFD) modeling of explosions can help minimize the need for large-scale testing. This thesis utilized the commercial CFD software, SC/Tetra, to examine three case studies. The first case study modeled the combustion of methane in a scaled shock tube, measuring approximately 1 foot by 1 foot, by 20.5 feet long, with a methane cloud of 2.5 feet in length, at a concentration of 9% methane. The numerical results from the CFD model were in good agreement with experimental data gathered, with all pressure peaks within 0.25 psi of the recorded pressure data. However, the model had an extensive run-time of 16 hours to reach the peak pressures. The second case study modeled the same explosion, but utilized a total pressure boundary condition at the location of the membrane, instead of the combustion of methane. A pressure-time curve was assigned to this boundary, recreating the release of pressure by the explosion. This was made possible with the knowledge of the experimental data. The numerical results from the CFD model were in excellent agreement with experimental data gathered, with all pressure peaks within 0.07 psi of the recorded pressure data. Alternatively, this model had a run-time of 40 minutes. The third case study modeled a methane explosion in a large shock tube, measuring 8 feet by 8 feet, by 40 feet long, with a methane cloud of 4 feet in length, at a concentration of 9% methane. The bursting balloon technique was employed, which did not model the combustion of methane, but instead the equivalent energy release. The numerical results from the CFD model were in good agreement with the experimental data gathered, with all pressure peaks within 0.025 psi of the recorded pressure data. Additionally, the numerical results modeled the negative pressure phenomenon observed in the experimental results, caused by suction or negative pressure created by the blast wave, immediately following the positive wave. This model had a run-time of 20 minutes. The results of this researched provided validation that there are alternative ways to successfully model methane explosion, without having to model the chemical reactions involved in the combustion of methane, providing quicker run-times and in this case, more accurate results.

Identiferoai:union.ndltd.org:uky.edu/oai:uknowledge.uky.edu:mng_etds-1052
Date01 January 2019
CreatorsSteeves, Laura
PublisherUKnowledge
Source SetsUniversity of Kentucky
Detected LanguageEnglish
Typetext
Formatapplication/pdf
SourceTheses and Dissertations--Mining Engineering

Page generated in 0.0024 seconds