Date of Graduation

2015

Document Type

Dissertation

Degree Type

PhD

College

Statler College of Engineering and Mineral Resources

Department

Mechanical and Aerospace Engineering

Committee Chair

Scott W Wayne

Committee Co-Chair

Nigel N Clark

Committee Member

Hailin Li

Committee Member

Gregory J Thompson

Committee Member

Avinash Unnikrishnan

Abstract

In this work a thermodynamics and emissions model of liquefied natural gas (LNG) tanks and fueling stations was developed, allowing for the calculation of methane (CH4) emitted from tanks and the prediction of methane emissions in future scenarios. The detailed dynamic thermodynamic model determined the thermodynamic state (pressure, temperature, specific volume) and properties (enthalpy, internal energy, specific heat, etc.) of the liquid and vapor phase methane in the storage tank in order to determine the rate of LNG boil off and venting. The model employed differential forms of the energy balance and mass balance and thermodynamic property relations and data to describe the evolution of liquid and vapor quantity, state and properties with time as a function of fueling station activity. In addition, the temperature inside the storage tank was determined by two approaches: a homogeneous or a stratified distribution, where the homogeneous model assumed a uniform saturated temperature throughout the tank and the stratified model determined a temperature profile in the tank. The model accounted for varying ambient conditions, varying mass flow of LNG into and out of the tank as a result of refueling the tank, fuel dispensing, recirculation to chill dispensing equipment, return of vapor from vehicle tanks for vapor balancing, and release of boil off gas (BOG) to maintain safe tank operating pressure. Furthermore, this work could be adapted to develop a comprehensive model for LNG vehicle fuel tanks.;The model was validated with experimental data acquired by the Center for Alternative Fuels, Engines, and Emissions (CAFEE) of West Virginia University (WVU) from vehicle tanks and at LNG fueling stations in the United States. The complete LNG Fueling Station Model achieved an average error of -0.36 psia/day (1.13%) in the rate of pressure change with respect to time using the stratified approach and an average error of -1.67 psia/day (-10.43%) using the homogeneous approach. Two LNG fueling station tanks of 15,000 gallons and 25,000 gallons capacity were used. For fueling stations both approaches presented cases of over prediction and under prediction. The stratified approach had an error in dP/dt between -8.00 psia/day (-39.03%) and 2.75 psia/day (28.23%). The homogeneous approach had an error in dP/dt between -7.49 psia/day (-36.55%) and 2.15 psia/day (17.38%). Validation of pressure rise in vehicle tanks achieved an average error of -1.01 psia/day (-2.64%) using the stratified approach and 14.56 psia/day (204.25%) using the homogeneous approach. Two LNG vehicle tanks of 120 gallons and 150 gallons capacity were used. For vehicle tanks the homogenous approaches always over predicted and the stratified approach presented cases of over prediction and of under prediction. The stratified approach had an error in dP/dt between -6.31 psia/day (-44.90%) and 4.13 psia/day (21.40%). The homogeneous approach had an error in dP/dt between 2.35 psia/day (12.16%) and 39.48 psia/day (883.96%).

Download

Share

COinS