Fred F. Farshad

Predicting temperature profiles in producing oil wells using artificial neural networks

A novel approach using artificial neural networks (ANNs) for predicting temperature profiles evaluated 27 wells in the Gulf of Mexico. Two artificial neural network models were developed that predict the temperature of the flowing fluid at any depth in flowing oil wells. Back propagation was used in training the networks. The networks were tested using measured temperature profiles from the 27 oil wells. Both neural network models successfully mapped the general temperature‐profile trends of naturally flowing oil wells. The highest accuracy was achieved with a mean absolute relative percentage error of 6.0 per cent. The accuracy of the proposed neural network models to predict the temperature profile is compared to that of existing correlations. Many correlations to predict temperature profiles of the wellbore fluid, for single‐phase or multiphase flow, in producing oil wells have been developed using theoretical principles such as energy, mass and momentum balances coupled with regression analysis. The N...

New developments in surface roughness measurements, characterization, and modeling fluid flow in pipe

Abstract The purpose of this paper is to present petroleum and chemical engineers with a simple means of measuring or estimating the absolute surface roughness, ϵ, and relative roughness, ϵ/D, for internally coated pipes. Our research thrust is to develop new relative roughness charts along with corresponding mathematical equations for internally coated pipes. An integral part of the frictional pressure drop due to fluid flow in pipes involves the determination of absolute surface roughness and relative roughness. An accurate determination of the pressure drop due to fluid flow in oil and gas wells is required for optimizing oil and gas production design calculations. Some of these calculations include developing tubing programs to maximize well deliverability, wellbore flow performance, sizing surface flow lines, and designing artificial lift installations. Previous multiphase fluid flow laboratory and field tests, during the past 50 years, provided correlations for calculating pressure gradients in tubulars [Chem. Eng. Prog. 45(1) (1949) 39; Hetsroni, G., 1982. Handbook of Multiphase Systems, Hemisphere/McGraw Hill, 1174 pp.; Brown, K.E., 1984. The Technology of Artificial Methods, vol. 4, Penn Well Publishing, Tulsa, OK, 447 pp.; Tengesdal, J.O., 1998. Predictions of Flow Patterns, Pressure Drop, and Liquid Holdup in Vertical Upward Two-Phase Flow. MSc Thesis, The University of Tulsa, Tulsa, OK; Farshad, F., Garber, J.D., Polaki, V., 2000. Comprehensive model for predicting corrosion rates in gas wells containing CO2. Soc. Pet. Eng. Prod. Facil. 15(3), 183–190]. All the multiphase fluid flow correlations, used in production operations to calculate these pressure gradients and flow regimes, stem from the general energy balance equation and involves the determination of friction pressure losses. Still today, these friction losses only are being evaluated by practicing engineers using the Darcy–Weisbach equation, which involves the Moody friction factor, f [Szilas, A.P., 1975. Production and Transport of Oil and Gas. Elsevier Sci. Publ., Amsterdam, p. 630; Economides, M.J., Hill, D.A., Ehlig-Economides, C., 1994. Petroleum Production Systems, Prentice Hall, Englewood Cliffs, NJ, 611 pp.]. Moody [Trans. AME 66 (1944) 671] prepared a relative roughness chart for a number of common piping materials. This initial relative roughness correlation was based on experiments on pipes artificially roughened with sand grains. At that time, internally coated pipes were not invented. Hence, Moody did not provide the relative roughness for internally coated pipes. In addition, Moody did not perform a regression analysis of the data to generate functional forms of equations by relating roughness, ϵ/D, as a function of internal pipe diameter, D. Currently, internally coated pipes are being utilized worldwide in oil field production systems. Consequently, new absolute surface roughness and relative roughness values of internally coated pipes are needed to properly model the hydrodynamics [Brown, K.E., 1984. The Technology of Artificial Methods, vol. 4, Penn Well Publishing, Tulsa, OK, 447 pp.]. The new relative roughness plots developed for internally coated pipes are presented and show that the correlation fits between the commercial steel and drawn tubing in the Moody chart.

Coated pipe interior surface roughness as measured by three scanning probe instruments

The objectives of this study were to determine: the type of coating that minimized pipe surface roughness and how the choice of metrological instrument could influence pipe surface roughness data. The internal surface of pipe was coated with either phenolic, modified novalac, epoxy, or nylon material. Roughness of coated pipe was assessed with two linear surface profilers, a Dektak3ST® and a Hommel T1000, and a Dimension 3000® atomic force microscope (AFM). Arithmetic roughness (Ra), root mean square roughness (Rq), and mean peak‐to‐valley height (RZD), were statistically analyzed. The ability of RZD to focus on the extremes of height and depth on the surface made it a significant parameter for detecting features that would affect fluid flow in pipes.

A simulation of crystallinity gradients developed in slowly crystallizing injection molded polymers via parallel splitting

In this study, a three‐dimensional (3D) flow model is used to approximate the crystallinity gradients of slowly crystallizing polymers developed in the injection molding process. A generalized second order parallel splitting formula is constructed to achieve both the accuracy and efficiency of the computation. Calculated values of flow‐wise (flow‐thickness plane) and width‐wise (width‐thickness plane) crystallinity distributions are obtained and compared with experimental results. The structure‐oriented simulation method developed is not only capable of describing moldability parameters, but is also able to predict the characteristics of ultimate properties of the final products.

Recovery of heat and carbon dioxide from compressor station exhaust gas

Gas compressor flue gas has sufficient heat content that heat recovery is practical. Steam and electrical power can be generated in this manner which reduces the utility cost of removing carbon dioxide from the flue gas. A survey of various methods of removing carbon dioxide from flue gas has shown that monoethanolamine (MEA) is the best chemical absorbent at the low pressures available for this process. A system is designed to compress, dehydrate, and transport carbon dioxide to the receiving oil reservoir which could be 10 or 100 miles away from the available source. 8 refs.

A Comprehensive Model for Predicting Internal Corrosion Rates in Flowlines and Pipelines

A computer model has been developed which is capable of predicting the physical and chemical condition inside of a flowline or pipeline. From this information it is then possible to provide a prediction of the internal corrosion rate. This “free” software was funded by the US. Department of Energy and has undergone several revisions and is currently in Phase II. The windows based model consists of 4 modules, which calculate the temperature, pressure, phase equilibrium, flow dynamics as well as the pH, scale tendency and the uninhibited corrosion rate. An expert system has been developed which adjusts the predicted corrosion rate based on fluid parameters. A number of field and hypothetical cases for both the flowlines and pipelines have been presented which show the utility of the model. The model has been compared to PipePhase, a commercial package, and the results have been very satisfactory. INTRODUCTION A computer model has been developed in Visual Basic, which allows the user to physically and chemically describe a production system and to estimate the internal corrosion rate. The model, which was recently presented at NACE is a continuation of work done on earlier oil well and gas well – models that have been developed over the past 20 years by the Corrosion Research Center at the University of Louisiana at Lafayette. Unlike the previous models, this model is “free” to the user since it was funded by the US Department of Energy. During the 4-year study, 3 versions of the program were developed with Phase II being the most recent version released on June 26, 2003. In this paper a brief description of the flowlines and pipeline models are given. Emphasis has been placed on Case Studies of pipelines and flowlines, which describe the range of use of the models. The input data required by the program is information that is readily available to the corrosion engineer. This includes production information from the separator, inlet and outlet temperature and pressure, the various coatings and the dimensions of the flowline or pipeline, as well as the slope of various sections of the pipeline. VARIOUS MODULES OF THE MODEL The computer model consists of five modules as seen in Figure 1. The modules are all dependent on each other for data. The physical description part of the model must be looped until there is convergence on the pressure drop through the individual pipeline or flowline. After the pressure drop converges, the chemical module, pH and scale calculation is run and then the corrosion rate calculation is performed. A brief description of each module follows: 1. Temperature Module: In most instances, the pipeline or flowline is loosing heat to the environment. By dividing the pipeline or flowline into small segments, it is possible to determine the heat lost in each segment and the exiting temperature of that segment. The overall heat transfer coefficient for the pipe must be estimated using coating information and then using the fact that the enthalpy loss by the production fluid equals the heat loss, it is possible to calculate the temperature at the end of the segment. Internal and an external coating data can be input together with information about the surrounding medium. Calculations can be made with the pipe in seawater, mud or in air. The starting temperature must be known, and the program will match the known ending temperature if desired. 2. Phase Equilibrium Module: It is critical in corrosion prediction to know the amount of fluids, gas, oil and water, at various points in the pipeline or flowline. The Peng Robinson Equation of State (EOS) does a very good job in the prediction of dew points of hydrocarbon systems. For phase equilibrium to