A Support Vector Machine Approach for the Prediction of Drilling Fluid Density at High Temperature and High Pressure

Abstract

A support vector machine (SVM) approach was presented for predicting the drilling fluid density at high temperature and high pressure (HTHP). It is a universal model for water-based, oil-based, and synthetic drilling fluids. Available experimental data in the literature were used to develop and test this SVM model. Good agreement between SVM predictions and measured drilling fluid density values confirmed that the developed SVM model had good predictive precision and extrapolative features. The SVM model was also compared with the most popular models such as the artificial neural network (ANN) model, empirical correlations, and analytical models. Results showed that the SVM approach outperformed the competing methods for the prediction of drilling fluid density at HTHP.     

The Prediction of Undersaturated Crude Oil Viscosity: An Artificial Neural Network and Fuzzy Model Approach

Abstract

Viscosity is one of the most important governing parameters of the fluid flow, either in the porous media or in pipelines. So it is important to use an accurate method to calculate the oil viscosity at various operating conditions. In the literature, several empirical correlations have been proposed for predicting undersaturated crude oil viscosity. However these correlations are not able to predict the oil viscosity adequately for a wide range of conditions. An extensive experimental data of undersaturated oil viscosities from different samples of Iranian oil reservoirs was applied to develop an Artificial Neural Network (ANN) model and fuzzy model to predict and calculate the undersaturated oil viscosity. Validity and accuracy of these models has been confirmed by comparing the obtained results of these correlations and with experimental data for Iranian oil samples. It was observed that there is acceptable agreement between the ANN model and fuzzy model results with experimental data.     

The Development of a New Empirical Correlation for Predicting Hydrate Formation Conditions

Abstract

Hydrates are known to occur in a variety of natural-gas handling facilities and processing equipment in oil fields, refineries, and chemical plants when natural gas and water coexist at elevated pressure and reduced temperature. Prevention of hydrate formation costs large amounts of capital and results in large operating expenses. In order to avoid costly losses due to the formation of these hydrates, several methods, which include thermodynamic modeling and empirical correlations, can be employed to predict the conditions for hydrate formation. The authors employed SAS (SAS Institute, Cary, North Carolina) to develop a new correlation for predicting the hydrate-formation temperatures for both pure and mixture of hydrocarbon systems using the gravity method. The method correlates the hydrate-formation pressure against specific gravity, pressure, and water-vapor pressure.     

Development of a Neural Fuzzy System for Advanced Prediction of Bottomhole Circulating Pressure in Underbalanced Drilling Operations

Abstract

Precise prediction of bottomhole circulating pressure (BHCP) is the major concern in the designing stage of an underbalanced drilling (UBD) operation. A successful UBD operation hinges tightly on the accurate control of downhole pressures. In order to predict the pressure gradient, complex two-phase flow correlations are employed and the flow pattern is defined. The problems arising from such methods are the significant errors relative to the nature of a UBD and the need for localization of the parameters or the shift from one correlation to another from case to case. In this study, different types of fuzzy inference systems are developed to construct new models. The resulting average percentage relative error (APE) of 13.83% for a Sugeno model and 13.11% for a Mamdani model indicated that the generated fuzzy models have the capability of precise BHCP prediction with higher accuracies in comparison with the popular mechanistic models.     

MEASUREMENT AMD PREDICTION OF CRITICAL PROPERTIES OF CHINESE PETROLEUM FRACTIONS

ABSTRACT

The critical properties (T_c, P_c, V_c) of 102 virgin oil fractions from five Chinese crude oils were measured by sealed tube method and semi-sealed tube method using gallium as pressure transfer medium. Satisfactory correlations for predicting critical parameter with other easy-measured physical properties were measured. The feasibility of group contribution method for predicting critical parameters of the above mentioned oil fractions were also investigated.     

Predicting the Bubble-Point Pressure and Formation-Volume-Factor of Worldwide Crude Oil Systems

Abstract

This paper presents models for predicting the bubble-point pressure (P _b ) and oil formation-volume-factor at bubble-point (B _ob) for crude oil samples collected from several regions around the world. The regions include major producing oil fields in North and South America, North Sea, South East Asia, Middle East, and Africa. The model was developed using artificial neural networks with 5200 experimentally obtained PVT data sets. This represents the largest data set ever collected to be used in developing P _b and B _ob models. An additional 234 PVT data sets were used to investigate the effectiveness of the neural network models to predict outputs from inputs that were not used during the training process. The network model is able to predict the bubble-point pressure and the oil formation-volume-factor as a function of the solution gas–oil ratio, the gas relative density, the oil specific gravity, and the reservoir temperature. In order to obtain a generalized accurate model, back propagation with momentum for error minimization was used. The accuracy of the models developed in this study was compared in details with several published correlations. This study shows that if artificial neural networks are successfully trained, they can be excellent reliable predictive tools to estimate crude oil properties better than available correlations. The network models can be easily incorporated into any reservoir simulators and/or production optimization software.     

The Prediction of Surge and Swab Pressures and Critical Pipe Running Speed While Tripping in a 12 1/4-in. Hole (Gachsaran Formation) of Maroun Field in Iran

Abstract

Maroun field is a large oil field located in southern Iran. The producing formations in this field are deposited under the Gachsaran formation, which has an extremely high value of pore pressure. The mud density through the Gachsaran formation usually exceeds 20 ppg. In this situation the pipe running velocity is a critical issue and careless moving of the drill string may cause the risk of kick or formation breakdown.

The purpose of this study is to develop a model for prediction of critical pipe running speed. The method of study is the basis of simple fluid flow equations and computer programming is used to simplify primary data analyzing. New correlations are developed using multiple regression method and SPSS software for predicting critical pipe running speed using only simple input data such as well geometry and drilling fluid properties, which can be measured using American Petroleum Institute testing methods.     

CORRECTION OF VERTICAL PERMEABILITY ESTIMATED BY HORIZONTAL-WELL PRESSURE TRANSIENT TEST ANALYSIS

ABSTRACT

Different data are required in order to establish a realistic reservoir model capable of predicting the dynamic field behaviour during the development stage of an oil field. Well testing is considered to be one of the most effective methods for obtaining these reservoir and wellbore data. Numerous analytical models are available and utilised in investigating vertical-well pressure transient tests, on the other hand, horizontal-well pressure transient tests have been considered as more difficult to analyse. In this work, well test data of a horizontal-well are simulated for homogeneous isotropic and anisotropic reservoirs and analysed in order to develop empirical correlations that rectify the vertical reservoir permeability estimated by well testing.     

An accurate model to predict drilling fluid density at wellbore conditions

Knowledge about rheology of drilling fluid at wellbore conditions (High pressure and High temperature) is a need for avoiding drilling fluid losses through the formation. Unfortunately, lack of a universal model for prediction drilling fluid density at the addressed conditions impressed the performance of drilling fluid loss control. So, the main motivation of this paper is to suggest a rigorous predictive model for estimating drilling fluid density (g/cm³) at wellbore conditions. In this regard, a couple of particle swarm optimization (PSO) and artificial neural network (ANN) was utilized to suggest a high-performance model for predicting the drilling fluid density. Moreover, two competitive machine learning models including fuzzy inference system (FIS) model and a hybrid of genetic algorithm (GA) and FIS (called GA-FIS) method were employed. To construct and examine the predictive models the data samples of the open literature were used. Based on the statistical criteria the PSO-ANN model has reasonable performance in comparison with other intelligent methods used in this study. Therefore, the PSO-ANN model can be employed reliably to estimate the drilling fluid density (g/cm³) at HPHT condition.     

Doubling Model for Lower Permeability Formation with Hydraulic Fracture

Abstract

A fluid flow doubling model has been built for low-permeability formation and hydraulic fracture based on non-Darcy pressure propagation and fluid flow velocity difference in both matrix and fracture. A cross-flow rate model between facture and formation is derived from submerged fluid flooded in a rock grid matrix. It will be more convenient to describe the time-varying fracture conductivity than banded heterogeneous grids with higher permeability. A numerical model was derived based on a finite difference scheme for calculation, and a fractured well has been applied for productivity comparison with different models. It has been shown that pressure distribution in the fracture is weakly nonlinear because of cross flow, and a doubling model is more consistent with field production data. There exists a greater productivity difference in the previous production period for constant fracture conductivity and in the later part for banded heterogeneous model result from weakening performance difference between fracture and matrix. This article presents a practical model to evaluate the productivity of a fractured well.     

A new correlation for predicting the minimum miscibility pressure regarding the enhanced oil recovery processes in the petroleum industry

Miscible gas injection is a promising technique to enhance the oil recovery from petroleum reservoirs. Natural gas is usually injected because of consistency between the gas and reservoir fluid composition and accessibility of the injected fluid. In this work a new correlation is presented for accurate prediction of the required minimum miscibility pressure (MMP) for multicontact miscible displacement of reservoir petroleum by hydrocarbon gas injection. Accurate estimation of the MMP is required for optimal design of recovery systems and optimization of the production economics. It is shown that the presented correlation is very accurate and reliable for predicting the MMP over wide ranges of oil and gas compositions. Comparison of the suggested correlation with the most important existing correlations shows that the presented correlation outperforms the other alternatives both in accuracy and generality.     

Comparative Study of Yield-Power Law Drilling Fluids Flowing Through Annulus

Abstract

An exact solution for calculating the frictional pressure losses of yield-power law (YPL) fluids flowing through concentric annulus is proposed. A solution methodology is presented for determining the friction factor for laminar and nonlaminar flow regimes. The performance of the proposed model is compared to widely used models as well as the experimental results of 10 different mud samples obtained from the literature. The results showed that the proposed model could estimate the frictional pressure losses with an error of less than 10% in most cases for both laminar and nonlaminar flow regimes, more accurately than the widely used models available in the literature.     

A Mechanistic Model for Predicting Frictional Pressure Losses for Newtonian Fluids in Concentric Annulus

Abstract

A mathematical model is introduced estimating the frictional pressure losses of Newtonian fluids flowing through a concentric annulus. A computer code is developed for the proposed model. Also, extensive experiments with water have been conducted at Middle East Technical University, Petroleum and Natural Gas Engineering Department Flow Loop and recorded pressure drop within the test section for various flow rates. The performance of the proposed model is compared with computational fluid dynamics (CFD) software simulated annulus flow section and various criteria such as crittendon, hydraulic diameter and slot flow approximation as well as experimental data. The results showed that the proposed model and experimental results are in good agreement for almost all cases when compared with the other criteria and CFD software. Also, the proposed model could estimate the frictional pressure losses for both laminar and turbulent flow regimes within an error range of ±10%.     

Numerical Investigation of the Impacts of Wall Fluid Entry on Fluid Flow Characteristics and Pressure Drop along a Wellbore

Abstract

Single-point or limited-point wall fluid entry in a wellbore has been observed in different oil and gas wells all over the world through downhole video logging. The wall fluid entry is expected to affect both the fluid flow characteristics in a wellbore and the pressure drop along the wellbore. In order to investigate the impacts of the single-point wall fluid entry, a numerical simulation model has been set up and a series of numerical simulations have been conducted by using a commercial computational fluid dynamics (CFD) software package. Flow velocity profiles, streaklines, pressure distribution along the wellbore, and so on have been thoroughly investigated. The simulation results clearly demonstrate the influence of the wall fluid entry on the fluid flow as well as the pressure drop along the wellbore. The impacts on the pressure drop along the wellbore have further been quantified.     

ARTIFICIAL NEURAL NETWORK MODEL FOR ACCURATE PREDICTION OF PRESSURE DROP IN HORIZONTAL AND NEAR-HORIZONTAL-MULTIPHASE FLOW

Accurate prediction of pressure drop for multiphase flow in horizontal and near horizontal pipes is needed for effective design of flow lines and piping networks. The increased application of horizontal wells further signified the need for accurate prediction of pressure drop. Several correlations and mechanistic models have been developed since 1950. In addition to the limitations on the applicability of all existing correlations, they all fails to provide the desired accuracy of pressure drop predictions. The recently developed mechanistic models provided some improvements in pressure drop prediction over the empirical correlations. However, there is still a need to further improve the accuracy of prediction for a more effective and economical design of wells and surface piping networks. This paper presents an Artificial Neural Network (ANN) model for prediction of pressure drop in horizontal and near-horizontal multiphase flow. The model was developed and tested using field data covering a wide range of variables. A total of 225 field data sets were used for training- and 113 sets data for cross-validation of the model. Another 112 sets of data were used to test the prediction accuracy of the model and compare its performance against existing correlations and mechanistic models. The results showed that the present model significantly outperforms all other methods and provides predictions with accuracy that has never been possible. A trend analysis was also conducted and showed that the present model provides the expected effects of the various physical parameters on pressure drop.     

Modelling scale saturation around the wellbore for non-Darcy radial flow system

Scaling is a major profit hurting and aching occurrence induced by extensive use of seawater for oil displacement and pressure maintenance; hence it consequently results in production losses and loss of billions of dollars to the petroleum industry yearly. Over the years several models have been developed for predicting the effect of different variables such as pressure, temperature, ions, and pH on the behaviour of mixture of incompatible waters, scaling tendency, amount of scale precipitates which did not account for the quantity deposited around the wellbore. Few research works have been done to show the effect of reservoir and fluid parameter on the magnitude of scale deposition around the wellbore. The most recent formulations for predicting sulphate scale saturation around the wellbore assuming Darcy flow condition was developed and applied to real life scenario by experts in Colombia field. However, flow through a narrow channels often use in describing permeability loss modelling because pore throat and pore spaces become narrower and tighter during scale particle deposition and blockage around the well bore. Then high fluid velocity experiences when it is moving through narrow and tighter channels hence, results in non-Darcy flow that is always experience at the near wellbore region.This paper presents analytical model for predicting sulphate scale saturation near the wellbore under the non-Darcy flow condition. The results obtained show that the previous models under estimated scale saturation at the near well bore region.     

Numerical Study of Stratified Gas-Liquid Flow

Abstract

In this work the effects of empirical correlations of friction factors (wall and interfacial) on the numerical stability and parameters predictions in stratified gas-liquid pipelines flow, was studied using the two-fluid model, where unequal phase pressure effects were considered. To study the effects of such empirical correlations without taking into account the instabilities due to ill-posed initial-value problems or multiple solutions, the two-fluid model was solved using input data, which satisfies a linear stability criterion where multiple solutions do not occur. In general, we found that the empirical correlations are not important for numerical stability but do, however, affect significantly the parameters predictions.     

Experimental and Numerical Investigation of Proppant Placement in Hydraulic Fractures

Abstract

In hydraulic fracturing treatments, a fracture is initiated by rupturing the formation at high pressure by means of a fracturing fluid. Slurry, composed of propping material carried by the fracturing fluid, is pumped into the induced fracture channel to prevent fracture closure when fluid pressure is released. Productivity improvement is mainly determined by the propped dimension of the fracture, which is controlled by proppant transport and proper proppant placement. Settling and convection (density driven flow) are the controlling mechanisms of proppant placement. In this study, proppant transport and placement efficiency of four non-Newtonian fluids with controlled density differences was experimentally investigated and numerically simulated. Small glass model was used to simulate hydraulic fracture and parameters such as slurry volumetric injection rate, proppant concentration, and polymer type (rheological properties) were investigated.

It has been observed that small glass models easily and inexpensively simulated flow patterns in hydraulic fractures and the flow patterns observed are strikingly similar to those obtained by very large flow models used by previous investigators. Convection was observed to be significant flow mechanism even with small density contrast. As viscous to gravity ratio increases, due to increasing slurry injection rate or decreasing proppant concentration, convection settling decreases and proppant placement efficiency increases. Increasing non-Newtonian flow behavior index (n) by using different types of polymers shows more gravity underrunning and less proppant placement efficiency. Therefore, larger slurry volumes are needed to be injected to prop the entire fracture height. Experiments conducted were simulated and some of the simulated experiments were presented. The simulator quantitatively replicates the experimentally observed.     

THE DEPENDENCE ON CHEMICAL COMPOSITION OF THE CETANE QUALITY OF HYDROTREATED LIGHT CYCLE OIL

ABSTRACT

The dependence of the cetane number of hydrotreated light cycle oil on chemical composition has been investigated. Cetane number was related to hydrogen content and aromaticity. The latter was determined by high performance liquid chromatography and also by carbon-13 n.m.r. spectroscopy. Specific correlations were developed in each case and a regression coefficient of 0.99 and a standard error of 1.25 were obtained for the best case. The usefulness of predicting cetane number from fuel composition is discussed.     

Modeling Oil–Water Countercurrent Flow in Deviated Wells Using Mechanistic and Simplified Approaches

Abstract

Water circulation and oil–water countercurrent flow in a deviated or multilateral well has been investigated in the present paper. Flow patterns for the countercurrent flow have been identified. Two practical criteria for determining the occurrence of countercurrent flow in deviated wells are introduced. Five categories of flow patterns for the countercurrent flow in pipes or wells are identified. Transitions among different flow patterns are proposed. Both a mechanistic model and a simplified model are developed for describing the oil–water countercurrent flow in deviated and multilateral wells. Solution procedure for using the newly-developed models is discussed.