Evaluation and Comparison of Natural Gas and Supercritical CO2 Correlation Models for Thermodynamic Applications
Research output: Thesis › Master's Thesis › Research
Understanding the thermodynamic characteristics but also the fluid behaviour of natural gas and associated components such as sour gas, nitrogen and CO2 is crucial for the safe and efficient planning and operation of facilities throughout the petroleum industry. The same applies to supercritical carbon dioxide when considering petroleum engineering related applications such as carbon capture and storage, CO2 injection in enhanced oil recovery operations or the use of CO2 as a heat transmission fluid in geothermal energy recovery systems. Fluid properties such as gas density along with viscosity are required parameters for many gas-flow equations including wellbore pressure drop calculations for single and two phase flow, calculations in connection with deliverability testing of gas wells and calculations predicting the gas-liquid flow regime in pipes. Wellbore models and simulators for both natural gas production and geothermal energy recovery systems are based on fluid and heat transfer models. These in turn depend on parameters such as density, viscosity, heat capacity and thermal conductivity, pseudo critical properties and Z-factors. Correlations for determining the mentioned properties find widespread use in industry and research applications as deterministic models are either very complex or not available at all. In this thesis such correlations for natural gas and supercritical CO2 are summarized, analysed and finally evaluated in regard to their applicability. The different correlation models are thereby evaluated and compared against each other. Correlations, which are usually dependent on temperature, pressure, density and specific gravity have been programmed. This allows a detailed assessment and verification in terms of accuracy and applicable range of the different correlations. Furthermore the programmed models are readily available and can be accessed directly or be implemented into further applications. The correlation results are plotted and compared with data from National Institute of Standards and Technology (NIST). The NIST data is used as a reference database for this study and additionally checked against experimental data obtained from literature. Evaluation studies on pseudocritical properties, natural gas viscosity and CO2 viscosity are available from the literature and the main findings of those studies are provided in a summarized form and are a part of this work. Finally, a ranking according to accuracy and applicability of all evaluated correlation models is presented and the deviations of the individual correlation results in relation to the NIST data are listed.