Date of Graduation

2015

Document Type

Thesis

Degree Type

MS

College

Statler College of Engineering and Mineral Resources

Department

Industrial and Managements Systems Engineering

Committee Chair

William S Wayne

Committee Co-Chair

Gregory Thompson

Committee Member

John Zondlo

Abstract

Liquefied natural gas is seeing increased usage in the over-the-road trucking sector. Stations which refuel LNG vehicles sometimes emit natural gas into the atmosphere through a pressure relief valve due to over pressurization of the storage tank. Methane, the dominant constitute in natural gas, is considered a more potent greenhouse gas than carbon dioxide. However, natural gas has less of a carbon foot print than other fuels such as gasoline, diesel, and coal when burned giving it a global warming potential (GWP) advantage. Emission of methane into the atmosphere degrades the GWP advantage of burned methane or natural gas. Therein a tipping point exists where switching from conventional fuels to natural gas, if natural gas emissions are high, reduces or eliminates benefits in terms of GWP. This has many concerned about the accurate and comprehensive quantization of emissions from natural gas systems to enable well informed policy decisions for the natural gas sector.;A numerical model was developed to predict the pressurization of a LNG vehicle refueling station for the purpose of estimating vented emissions emanating from the pressure release valve (PRV). Factors which were addressed included the behavior of components of a station, local environmental conditions, and the vehicle fleet characteristics. The model formulation used two thermodynamic approaches to predict pressure rise. The thermodynamics were coupled with heat transfer calculated from station components including the bulk storage tank, transfer piping, and dispensers. Energy transfer was calculated by a combination of specific idealizations of cryogenic components and 1-D resistance networks incorporated into computer algorithms detailed in this thesis. Data were collected from two operating LNG stations, including the associated vehicle fleets, and were used for model creation and validation. Comparison of the pressure rise rate between the collected data and the model results indicated an average absolute model error for the first LNG station of 7.3% and for the second station of 25.1%. A range of emissions was deduced from the data indicating the first station emitted 0.1% to 1.5% of fuel dispensed to vehicles and 0.9% to 5.3% for the second station over the course of the 3 week evaluation periods. The analysis of the stations indicated vehicle transactions were a major factor in the pressure trends of the bulk storage tank. Vehicles refueling at the station caused the pressure rise rate to increase driving PRV emissions. However, at a certain point, specific to the design of each station, removal of LNG by vehicles outpaced pressure rise which caused stations to have zero PRV emissions. Therefore, stations emissions found were indicative of stations tendency to venting but not representative of stations on the whole.

Download

Share

COinS