Development of accurate and reliable correlations for various design parameters in oil and gas processing industries

Abstract

The continuing growth in the importance of oil and gas production and processing overall the globe increase the need for accurate prediction of various parameters and their impact on unit operations, process simulation and design. Because of the particular nature of various parameters, sometimes existing methods encounter difficulties. Currently several models are available to predict various design parameters in the oil and gas processing industries. However, their calculations may require rigorous computer solutions. Therefore, developing the new predictive tools to which are easier than the existing methods, less complicated with fewer computations to minimize the complex and time-consuming calculation steps is an essential need. It is apparent that mathematically compact, simple, and reasonably accurate predictive tools, as proposed in this thesis, would be preferable for computationally intensive simulations.In fact, the development of engineering correlations by a modification to the well-known Vogel-Tammann-Fulcher (VTF) [1921-1926] equation and Arrhenius equation (1889) was the primary motivation of the present thesis, which, nevertheless, yielded predictive tools with accuracy comparable to that of the existing rigorous simulations. Hence, some existing approaches lead to complicated equations for the purposes of engineering importance. This problem has been circumvented conveniently by resorting to simpler approaches, as described in this thesis.The purpose of the proposed Dissertation work is to develop and formulate accurate and reliable predictive tools to serve two purposes. First, being conversion of a set of highly correlated variables to a set of independent variables by using linear transformations. Second one is for variable reductions. When a dependent variable is specified, the method is very efficient for dimensional reduction due to the supervised nature of its methodology. The developed tools in this study can be immense engineering value to predict different process design parameters, including the prediction of hydrate forming conditions of natural gases, hydrate forming pressure of pure alkanes in the presence of inhibitors, water-hydrocarbon systems mutual solubilities, water content of natural gas, density, thermal conductivity and viscosity of aqueous glycol solutions, optimum size of inlet scrubber and contactor in natural gas dehydration systems, estimation of water-adsorption isotherms, estimation of equilibrium water dew point of natural gas in triethylene glycol dehydration systems, true vapour pressure (TVP) of LPG and natural gasoline, hydrocarbon components solubilities in hydrate inhibitors, methanol vaporization loss and solubility in hydrocarbon liquid phase for gas hydrate inhibition, storage pressure of gasoline in uninsulated tanks, emissivity of combustion gases, filling losses from storage containers,bulk modulus and volumetric expansion coefficient of water for leak tightness test of pipelines, silica solubility and carry-over in steam, carbon dioxide equilibrium adsorption isotherms, estimation of packed column size, estimation of thermal insulation thickness, transport properties of carbon dioxide, aqueous solubility of light hydrocarbons, estimation of economic thermal insulation thickness, water content of air at elevated pressures, surface tension of paraffin hydrocarbons, aqueous solubility and density of carbon dioxide, aqueous solubility of light hydrocarbons, thermal conductivity of hydrocarbons, downcomer design velocity and vapour capacity correction factor in fractionators, estimation of convection heat transfer coefficients and efficiencies for finned tubular sections, estimation of heat losses from process piping and equipment surfaces, prediction of absorption/stripping factors, correlating theoretical stages and operating reflux in fractionators, design of radiant and convective sections of direct fired heaters and many other engineering parameters.Following the development of predictive tools, experimental work was undertaken to measure the density and viscosity, of ethylene glycol + water, diethylene glycol + water, and triethylene glycol + water mixtures at temperatures ranging from 290 K to 440 K and concentrations ranging from 20 mol % glycol to 100 mol % glycol. Our data were correlated using a novel Arrhenius-type equation based predictive tool and a thermodynamical method (the generalized corresponding states principle (GCSP)). Both novel Arrhenius-type equation based predictive tool and GCSP method, with two adjustable parameters for each property, offer the potential for judicious extrapolation of density and viscosity data for all glycol + water mixtures.In addition, in this thesis, the PreTOG software package has been developed, which covers a wide range of parameters in oil, gas and chemical processing industries and is using PC-based Windows and Matlab graphical user interfaces and tool boxes. The PreTOG software is also available on an stand-alone CD. Finally the following typical case studies for potential benefits to various processing plants industries will be presented and the results of new proposed model are compared with partial least squares (PLS) and principal component analysis (PCA): •Methanol vaporization loss during gas hydrate inhibition •Methanol loss in condensate liquid phase during gas hydrate inhibition •Estimation of potential savings from reducing unburned combustible losses in coal-fired systems •Recoverable heat from blowdown systems during steam generation •Energy conservation benefits in excess air controlled gas-fired systems •Prediction of salinity of salty crude oil