EOS Development: A Case Study

Abstract

A single equation of state (EOS) model has been developed for multiple fields in Saudi Arabia based on a PVT fluid characterization study conducted in 2007. Additional PVT samples and experiments have provided from two of the fields included in the study. The main objective of this thesis is to evaluate the developed PVT model against the new data by simulating the experiments using a PVT package. On the other hand, an in-house lab has conducted all laboratory PVT data for the new samples including compositional measurements and PVT experiments. The thesis aims to analyze these lab reports and understand uncertainties present in lab data. Knowledge about lab methodology in conducting PVT analysis is a key factor for developing a reliable EOS model. Areas of uncertainties are present in all lab experiments and can be identified by knowing how the lab construct their PVT reports. The in-house lab used gas chromatography analysis for all samples compositional measurements. All compositions have been reported in molar basis. Molecular weights used by the lab for mass to mole conversion has not been stated in the reports which might not represent the aromaticity or paraffinicity of the samples. Knowing the exact mass amounts measured by the gas chromatography is essential for the EOS model. For the single EOS model to describe the new samples, C_(7+) characterization must be common. A relationship between molecular weight and specific gravity has been developed using Søreide correlation. Søreide factor has been identified and was equal to the one utilized in the 2007 study. The developed EOS model consists of 35 components where single carbon number description has been utilized. Using PhazeComp, an EOS-based simulator by Zick Technologies, the model has been run to simulate all new samples experiments. Individual sample composition adjustments have been used to enhance the model predictions compared to measured data. During EOS model tuning, weight factors and regression variables are introduced based on accuracy and importance of the data. Reported PVT data such as saturation pressures seemed to be of high quality. However, lab reports are missing key data that can help in quality check some of the laboratory experiments, as an example, the final stage oil compositions and properties of the constant volume depletion tests. The in-house lab should modify reporting templates to provide all required data to have an accurate fluid characterization study.