Successful acid stimulation requires a method for diverting an acid across the entire hydrocarbon-producing zone. Because most producing wells are not homogeneous and contain sections of varying permeability, being able to completely acidize the interval is a major problem. This paper describes the use of a new low-viscosity system that uses a relative permeability modifier (RPM) that diverts acid from high-permeability zones to lower-permeability zones and inherently reduces formation permeability to water with little effect on hydrocarbon permeability. This system has been used effectively offshore Mexico with success for more than two years. The cases presented in this paper show the first application in a low-permeability carbonate formation where oil production was increased significantly compared to previous traditional acid treatments using conventional diverters.
The other important feature of this work is that the downhole conditions were high-pressure/high-temperature (HPHT). Details from the jobs using this new RPM acid-diversion system will be presented showing pre- and post-job production results.
Matrix acidizing enhances well productivity by reducing the skin factor. The skin factor can be reduced if near-wellbore damage is removed or if a highly conductive structure is superimposed onto the formation. In either case, the result is a net increase in the productivity index, which can be used either to increase the production rate or to decrease the drawdown pressure differential. Although the benefits of an increased production rate are evident, the benefits of reduced drawdown are often overlooked. Decreased drawdown can help prevent formation collapse in weak formations, reduce water or gas coning, minimize both organic and mineral scaling, and/or shift the phase equilibrium in the near-wellbore zone toward smaller fractions of condensate or solution gas. A reduced drawdown pressure can also help ensure that a greater percentage of the completed interval contributes to production.1
In attempts to achieve uniform placement of acid across all layers, various placement techniques have been used.2 The most reliable method uses mechanical isolation devices (such as straddle packers) that allow injection into individual zones one at a time until the entire interval is treated. However, this technique is often not practical, cost-effective, or feasible. Without a packer, some type of diverting agent must be used.
Typical diverting agents include ball sealers, degradable particulates, viscous fluids, and foams. Although these agents have been used successfully, all have potential disadvantages and none address the problem of increased water production that often follows acid treatments. Therefore, it would be a major advantage to have a material that could inherently decrease the formation permeability to water while also providing diversion.
One method of controlling water production uses dilute polymer solutions to decrease the effective permeability to water more than to oil. These treatments may be referred to as relative permeability modifiers (RPM), disproportionate-permeability modifiers, or simply, bullhead treatments. The latter name is so called because these treatments can be bullheaded into the formation without the need for zonal isolation. RPM systems are thought to perform by adsorption onto the pore walls of the formation flow paths.3-5
This paper presents the development and field implementation of a state-of-the-art bottomhole assembly (BHA) program using the industry's first generic algorithm based on Lubinski's equations. The strengths of the new BHA program are accuracy and computation efficiency, as compared to the conventional finite-element based BHA programs. In addition, the new program integrates static and dynamic models so that users can run both models in the same application. Using the new algorithm, the static model is designed mainly for directional drilling applications, such as optimal BHA design for maximum steerability, bending moment calculations to minimize fatigue failure, and BHA sag corrections to improve survey quality. The dynamic model is based on a hybrid of analytical and finite-element methods to calculate the critical rotary speeds of the BHA. This paper describes the significance of applying these features in a user-friendly application to maximize drilling performance.
Bottomhole assembly (BHA) modeling is always an essential part in directional drilling. A state-of-the-art BHA program enables many critical features, such as (i) designing a BHA to optimize directional performance, (ii) optimizing stabilizer locations to minimize vibration and increase downhole tool reliability, and (iii) improving survey data by correcting the BHA sag. Since the 1950s, several different methods have been developed and applied in the drilling industry to build the BHA models. 1-7
In general, the challenges encountered in developing a computationally efficient, flexible, and accurate BHA model can be summarized as follows:
The most commonly used method in BHA modeling is probably the finite-element method because it is easy to develop and use. However, to the knowledge of the author, many commercial finite-element based BHA programs are still based on the small deformation theory. As a result, they have been shown to lack the accuracy required to model steerable assemblies, such as motor or rotary steerable systems. Finite-element modeling is also cumbersome in handling the collars and wellbore contact. To accurately model steerable systems, the semi-analytical methods are usually required, but semi-analytical methods are inflexible and difficult to program. They are often designed to analyze some specific BHA models and are limited to BHAs with rather simple configurations.
The objectives of developing a state-of-the-art BHA program include the following:
This work presents a validation of the use of Experimental Design (ED) techniques in exploratory evaluations. The results generated by a commercial software, which uses ED techniques combined with the Monte Carlo method, have been compared with 10,000 equivalent actual flow simulations. Several oil and gas real reservoirs were studied. The static and dynamic variables were selected from real prospect studies, as well as their respective range of variation. For each case studied, the commercial software planned a small number of flow simulations to run, applying different ED techniques. Based on the simulation results, a proxy model was generated using the Response Surface Method (RSM). Making use of this proxy model, it was possible to quickly simulate 10,000 experiments and to perform statistics calculations with the Monte Carlo method.
In order to verify the quality of the proxy, 10,000 simulation files were generated, with the exact same combinations of parameters used in the Monte Carlo method. A script was written to run automatically all the files and to register their results. Besides comparing the statistics generated by these two methodologies, it was possible to compare the actual results of each one of the 10,000 simulations.
The results indicate that the quadratic technique is the one with the best cost-benefit ratio and suggest that the proposed methodology generates reliable results in most cases. This reliability decreases when there are complexities in the flow model, such as well and group controls, water and/or gas reinjection, etc. However, results show that the range of possible results to be emulated by the proxy model is the most influential parameter in the quality of the results generated by it.
The quality of the risk analysis performed with these ED techniques is obviously influenced by the ability of the proxy model to emulate the flow simulator response. The efficiency of different ED techniques, in different reservoirs, was analyzed and the reliability and limitations of this evaluation methodology has been accessed.
In the evaluation of exploratory prospects in the petroleum industry, one must deal with a large number of variables, each one of them with a certain degree of uncertainty. Therefore, probabilistic approaches have long been used in order to quantify the impact of those uncertainties in the economic evaluation of these prospects.
Nowadays, exploration studies have already incorporated an uncertainty analysis in the most influential static variables, but they still tend to translate all the dynamic uncertainties into a single variable (usually the recovery factor). From this point on, simple programs or even electronic spreadsheets are used to generate production profiles, based on the estimated accumulated production.
The authors hereby recommend using flow simulators together with experimental design techniques and response surface methods, to have a better understanding of the impact of dynamic uncertainties in the prediction of both the accumulated production and the production profile. The experimental design technique is used to plan a certain number of flow simulations and to try to build a response surface, i.e., a proxy model of the objective-function being studied. With this proxy model, it is possible to run thousands of simulations almost instantaneously and perform a risk analysis using the Monte Carlo method.1,2
The quality of the risk analysis will be greatly influenced by the ability of the proxy model to emulate the flow simulator's response.3 The authors have studied three different cases: a non-associated gas reservoir and two oil reservoirs, and compared the 10,000 results of the proxy model used to make the risk analysis and 10,000 actual flow simulations. By doing that, it is possible to compare the efficiency of different experimental design techniques and to access the reliability and the limitations of the evaluation methodology proposed by the authors.
Nowadays two things are absolutely crucial in petroleum industry: (1) an integrated approach from seismic to reservoir simulation (history matching included) and (2) uncertainty and risk analysis. The first one ensures that the model is coherent with all types of data and the second one drives decision and quantifies risk. Current available hardware and software allow simulating stochastic models with statistical analysis of multiple scenarios. It is important to have a standard workflow that takes into account both aspects and can be applied to all projects. Although geophysicists may not be used to, it is important to include uncertainty analysis on the geophysical model too.
A mature field located in Brazil is used as an example to illustrate the approach suggested above. The field has produced more than 20 million bbl of light oil and now a complementary project development is under study. Despite the amount of oil that has been produced, a significant geological uncertainty still remains. To provide the asset managers with a realistic range of possible outcomes of a project development, a thorough geological stochastic modeling was conducted. Four thousand models were generated considering the following varying parameters: reservoir structure, oil/water contact, porosity, net-to-gross, and initial water saturation. The models were ranked by VOIP and the distribution was sampled (P1, P2, ... P100) to go through numerical flow simulation. Permeability uncertainty was introduced by considering two possible scenarios for each model, giving a total of 200 reservoir models. Dynamic data was compared to simulation results through an objective function and those models which gave results too far away from production history were discarded. The VOIP distribution was recalculated.
From this new distribution, the P10, P50, and P90 realizations were picked to be history matched. After that, 27 models were generated through experimental design to consider the variation in the following parameters: (1) geological model, (2) relative permeability, (3) absolute permeability, (4) well productivity index, and (5) the number of wells to be drilled in the project development. The second and third parameters were kept constant in the vicinity of the producing wells so that all 27 models honored production data. A response surface model was generated by interpolation of the 27 flow simulations to obtain 10.000 outcomes for different parameter values. From these results it was possible to perform uncertainty analysis on the prediction.
In last few years, the petroleum industry has experienced a significant increase in the oil prices mainly due to a higher oil demand. As a result, many projects that were uneconomic in the past have been reviewed and considered economic now. Because it's been so difficult to find new oilfields, improved oil recovery projects such as infill drilling, CO2 injection, WAG injection, etc. have received a lot of attention1. Petrobras uses a probabilistic approach in the decision-making process which applies for all projects2. All available data has to be incorporated in an integrated way, including dynamic data. This poses a real problem since all stochastic models have to be history matched. History Match involves, in general, significant human and computational efforts. Risk Analysis is also computationally demanding. Therefore, a methodology is necessary to make this process feasible. Integrated software, automatic workflow, and the use of proxy models that replace flow simulations are almost always necessary3. Furthermore, the same approach should be applied to all projects since they will compete to the same budget.
Dou, Hong'en (RIPED,PetroChina) | Chang, Yu Wen (Research Inst. Petr. Expl/Dev) | Yu, Jun (Liaohe Oilfield E&D Research Inst.) | Wang, Xiaolin (RIPED,PetroChina) | Chen, Changchun (China U. of Petroleum) | Ma, Yingwei
This paper presents a new mathematic model to calculate heated reservoir area based on the heat balance principle in order to determine the well spacing for thermal recovery. The calculation result of the new model was compared with the classical models of J.W Marx and R. H. Langenheim (Petroleum Transactions, AIME Vol. 216, pp. 312-315, 1959), B. T Willman (JPT, July 1961, pp. 681-696), and Farouq. Ali (1970), using data from Liaohe heavy crude oilfield, China. The results showed that the new model is more accordable with oilfield actual condition than the three classical models. Also, this research reveals a new theory on huff ‘n' puff, which is that the heated radius of the heated region is expanded from the first cycle to the fourth cycle of huff ‘n' puff; in this case, the heated front was expanded with the increase of cycle. However, subsequent cycles (from fifth cycle to tenth cycle of huff ‘n' puff) repeat the heating of the previous heated areas, and the expanding heated area of the next cycle is smaller than the last cycle, and the new heating region should hardly be expanded after the 10th cycle. The authors point out the development results of heavy oil, extra heavy oil and super heavy oil deteriorate due to this reason. In addition, the authors emphasize that the traditional huff ‘n' puff for producing heavy oil, especially for extra-heavy oil and super-heavy oil has to be changed using new technique methods after the fourth cycle. Finally, a suitable well spacing for thermal recovery with huff ‘n' puff was obtained. The new theory was proved by heavy crude/extra heavy oilfield development.
Huff ‘n' puff has been used in heavy oil reservoir development since the 1960's. Heavy oil production techniques have been advanced greatly in Canada and Venezuela. In the early 1980s, many thermal recovery techniques were developed, such as insulation tubing, high temperature packer and measurement instrument of thermal parameters. In Liaohe oilfield, China; the reservoirs are at depth from 800 to 2000 m. Heavy oil development was a great success, with production rate reached 700×104 tons per year. During the 25-year production period (1980 to 2005), 20 % of the oil in place was produced. The mechanism of production was a combination of solution gas expansion and huff ‘n' puff, as the cycle of huff ‘n' puff is more and more, the development result become worse and worse. Currently, huff 'n ‘puff has exceeded 15 cycles in some wells. The adjustment of oil development strategy faces great challenges, especially in planning well spacing for different types of reservoirs in the oilfield to reach the maximum thermal recovery. The heating front of steam injection, swept region of hot water and the determination of the heated radius are the main parameters to be taken account for designing the well spacing of the heavy oil reservoirs during huff ‘n' puff. Well spacing for thermal recovery will not be determined if the heated radius should not be calculated accurately. Therefore, after the three classical models of J. W. Marx-R. H. Langenheim (1959), B. T Willman (1961) and Farouq. Ali (1970) was analyzed [1-3], and the paper presents three generalization calculation equations and a new model for calculating the heated radius of the thermal recovery.
Analysis of Classical Model
Many researches have developed the theory for the estimation of the heated radius of the heated region and design of well pattern by some scholars [4-8], and the three classical models of Marx-Langenheim (1959), Willman (1961) and Farouq Ali (1970) were used and introduced widely. However, the three models were not analyzed systematically, and the conclusions and recognizing of the heated radius of actual oilfield were not presented in past published papers. Also, the heated radius calculation of multi-cycle had not been revolved. Three generalization mathematics models of the heated radius with multi-cycle were given on the basis of the three classical models. Different performance characteristics of heavy oil, extra heavy oil and super heavy oil are analyzed by means of the three generalized mathematical models and actual oil field data.
In the Eastern Venezuela gas assets, mid to long term planning business portfolio considers an increasing gas production potential and recoverable reserves under secondary or improved recovery methods. Most gas reservoirs are originally under or near saturation pressure. Native thermodynamic conditions along with typical production practices favor in-situ liquid condensation, leaving behind considerable non-recoverable liquid hydrocarbons. This liquid condensation is triggered by unfavorable pressure gradients developed under natural production as reservoir static pressure drops below saturation pressure, primarily in the near wellbore region. In addition, as gas expands due to the adiabatic real-gas-expansion Joule-Thompson effect, a reduction in the surrounding temperature is also expected. This effect, in turn, often favors a sharp reduction in the relative permeability to gas when "In Situ?? liquid condensation takes place at interstitial level hence promoting a drastic decay in the gas productivity. Consequently, a considerable amount of liquid hydrocarbon is left behind in the reservoirs.
Based on the nature of the experimental findings, the availability for flue gas in the Eastern Venezuelan region, and the environmental concerns, a breakthrough improved Condensate Recovery (ICR) project is being considered by "Cyclic Supercritical CO2 Injection??.
The cyclic nature of the process combined with the favorable equivalent density for the CO2 at reservoir condition is expected to positively impact thermal diffusion and accelerated mass transfer processes respectively mainly in the near wellbore region. In this project, pilot temperatures are expected to be in the range of 350 to 450 oF.
A combined effort to integrate experimental results into a numerical model was undertaken. A number of laboratory tests were programmed to typify both fluid and rock interactions and also to characterize the thermodynamic profile for the CO2 and In Situ live fluid.
The interaction coefficients of the equation of state (EOS) for the enhanced process were experimentally characterized reproducing the multiple contact events between the in situ liquid hydrocarbon and the foreign fluid (CO2) as a function of temperature.
Expected sweep efficiencies and residual liquid saturation after multiple contacts with CO2 were experimentally determined via core displacement tests using actual core samples. Soaking time extent was also optimize for energy diffusion purposes.
At a second stage, a single well model will be assembled, to numerically reproduce the experimental results, to match primary depletion and to predict enhanced recovery production profile. For field implementation, a pilot area was selected in the Santa Rosa Field belonging to Petroleos de Venezuela assets.
"Anaco Gas?? is compounded by two major areas: "Area Mayor de Anaco??, with mostly volatile and Gas condensate bearing sands, and "Area Mayor de Oficina??, with medium to light hydrocarbon sands. As can be seen in Figure Nº 1, Santa Rosa field is part of the "Area Mayor de Anaco??, and is also one of the main areas of the asset in terms of gas remaining reserves. Typical stratigraphic column is 10.000 feet width, distributed among 150 hydrocarbon sands (Figure Nº 2). These sands are mostly associated to gas condensate production with relatively small API gradients both laterally and vertically. Most of these reservoirs are in advanced state of depletion. As pressure drops within the reservoir, a significant volume of unrecoverable liquid hydrocarbon is gradually left behind, mainly given by phase migration and saturation gradient related damage mechanisms which in turn promote a considerable reduction in productivity to gas mainly in the near wellbore region.
Efforts are being undertaken by PDVSA to increase final recovery of liquid hydrocarbon by Improved Condensate Recovery (ICR) techniques. A pilot ICR project is being developed in the Santa Rosa Field, VEE3 Sands, based on the experimental findings, as well as on the availability of flue gas from both upstream and downstream processes.
Most of heavy and extra heavy crude oil reservoirs in Venezuela are non-consolidated sand deposits. Venezuelan oil industry has a great interest to produce these reserves. Cold Heavy Oil Production with Sand (CHOPS) is an alternative for the primary production of some of these deposits; in particular those where other technologies are not applicable (i.e. sands with thicknesses are less than 10 m). In general, at field level, the application of the CHOPS has been successful around the world. In particular, it has experienced great development in Canada, China and United States. The physical and numerical studies show that change in the production rates, implies cavities formation and/or erosion zones near to the well that increase reservoir permeability. In Venezuela, experience in this technology does not exist. Furthermore, the bibliographical review demonstrates that many doubts still exist on the production mechanisms. It is the reason why this work considers the design, construction and evaluation of a sand production physical model. The main objective is to study the physical processes and to evaluate the influence of different variables like pressure, flow rate, among others, in the sand production mechanisms. The obtained data, will allow the development of models to predict reservoir behavior and its properties evolution in time. This physical model is set up with a cylindrical disc, 50 cm diameter and 20 cm thickness. Cylindrical geometry guarantees radial flow. The model is instrumented with sensors to measure pressures (injection, production and pores fluid) and fluid samples are taken to quantify total sand produced. The assembly system provides facilities to apply vertical stress. Ultrasonic wave velocity measurement is used to locate erosion zones. All data is collected by a Labview card and it is processed digitally in the computer in real time. This work shows results for a preliminary test with water and mineral oil used as fluids and calibrated glass bets as the porous media. The results prove that this system can be used to study sand production mechanism but it cannot be scale up to reservoir conditions. However the results will be used to validate numerical models.
The standard use of the centrifuge is to determine capillary pressure in plugs. What is proposed in this work is to extend its application making it possible to determine the pore-throat size distribution in plugs in addition to the capillary pressure indicated above.
To this end we start with a plug properly cleaned and saturated with the corresponding brine. The plug is then placed in the centrifuge and the collected data consists of the total evacuated brine volume VT(n) as a function of the centrifuge rotation speed n.
A simple model, consisting of capillary tubes running from one end of the plug to the other, is proposed to describe the complex system network of pores and the corresponding interconnecting pore-throats. The corresponding theory, based on treating the capillary pressure as that due to the water-air interphase, is worked out in such a way as to link the capillary size distribution to VT(n) vs. n. The mathematical procedures turn out to be straightforward and the solution is unique.
This original analysis of the centrifuge data was successfully applied to a large number of plugs. As an example the corresponding data and analyses are fully given and described.
In a companion paper these results are compared with those obtained by MICP (Mercury Injection Core Porosimetry), and the agreement found may be considered as excellent.
There is a practical limitation of the method proposed in this work. The capillary pressures that can be reached with the centrifuge are not as high as those that can be reached by MICP. This limitation manifests itself in the fact that pore-throat sizes below 1 mm are poorly detected or not detected at all. However, for pore-throat sizes above that value the agreement is excellent.
The main result obtained in this work is that it is shown that similar information to that produced in a MICP run with two main advantages (1) the centrifuge is a non-destructive experiment, and (2) is a non-contaminating experiment.
Porosity of formation rocks is one of the most relevant petrophysical characteristic parameters. Porosity is made up of pores and pore-throats and their sizes span wide range of values from about 10-2 mm up to 103 mm, and their sizes distribution functions are very important for a proper evaluation and management of an oil field.
There are not many experimental techniques to determine these distribution functions in bulk formation rocks (plugs). They are NMR (Nuclear Magnetic Resonance) and MICP (Mercury Injection (or Intrusion) Core Porosimetry). No one of them provides the information we are looking for. Also, in real formation the pores and the pore-throat network interconnecting them is so complex that it is not always possible to clearly discriminate among pores and pore-throats.
NMR measurements provide an information called the T2-distribution function, which rigorously speaking correspond to the addition of both the pore and the pore-throat sizes distribution functions, and it is not possible to separate them out. One of the most appreciated advantages of NMR is that it is a non-destructive technique.
On the other hand, MICP provides a good description of the pore-throats sizes present in the plug weighed by the fraction of the porosity corresponding to the pores interconnected by that given pore-throat size. Additionally, and unfortunately, MICP experiments are destructive and contaminating.
Thus, neither NMR nor MICP are able to provide the pore and the pore-throat size distributions.
In this work we propose to carry on a new experiment, the WRC (Water Removal by Centrifuge), using the centrifuge whose results resemble those obtained by MICP but with the main difference that is a non-destructive and non-contaminating one.
In the following sections the basic theory is developed, the equipment used (the centrifuge) is described, the results obtained in a standard plug are presented, and the conclusion is that a new method to determine the pore-throat sizes which closely resembles the results obtained by MICP. Limitations and advantages are described.
This paper describes the first application of radial perforation technology with hydraulic jet and coiled tubing in deep wells with complex geometry.
Radial perforation technique utilizing hydraulic energy is being used in different places around the world with good results. Its main application is intended for marginal fields, with low productivity and shallow wells with depths between 1000m and 2000m, normally having simple geometry. Basically, several perforations can be made with this technology, in an existing well (mother well), perpendicular to the well axis and at several productive levels, thus improving the production profile around the main well or mother well.
Repsol YPF Bolivia has adopted this technology to perform a pilot test in fields with good to moderate recovery and in deep wells with complex geometry (directional wells with depths between 3500m and 3900m).
This operation was performed in three wells (Surubi A1, Paloma C7 and Surubi Bloque Bajo 109) in the Mamore Block in Bolivia.
The main challenge for this operation was to adapt equipment and tools used up to that time, to much more severe operational conditions due to greater depths, tortuosity, type of formations to be drilled and mechanical conditions of the wells.
The main subjects covered by this paper are:
The results indicate that this technology may continue being an attractive substitute for other stimulation techniques, such as fractures, acids, side tracking, etc, even with higher costs implied by its application to complex wells and with the improvements that will definitively be applied to current tools and equipment
The permanent quest to increase productivity in low recovery wells has been a very good incentive for the development of new technologies to solve the economic equation in marginal fields.
In some cases, such technologies may have the characteristics needed for other applications, with different requirements, however generating results similar to those for which they were originally designed.
The specific case of the radial perforation system with coiled tubing using hydraulic energy and its application to deep wells is an example.
This technique and the associated equipment were created and designed for application in marginal fields and shallow wells with simple geometry. In the practical operation this represents low degree difficulty and reduced cost operations.
This paper details the application of this system but applying the same to non marginal fields and deep wells with complex geometry, representing operations with a high degree difficulty and high costs. Evidently, under such conditions, these projects represent a major investment risk, particularly when "adapting?? a system to more severe conditions than originally designed for.
This paper describes an implementation of method to optimize the production in intelligent wells varying the wells inflow control valves settings using an optimization algorithm coupled to commercial flow simulators. The optimization is based on direct search methods. The optimization algorithm was coupled with two different commercial flow simulators and has been applied in two real Brazilian offshore fields to quantify the benefits of intelligent wells over a base case with conventional completion. The first field has three horizontal wells, two producers and one water injector, completed in two zones totalizing six inflow control devices. In this case, different scenarios have been analyzed varying the downhole valves type - on-off and multi-position. The results have shown that the intelligent wells scenarios increased the recovery factor and reduced the production and injection of water when compared with the base case (conventional completion). The second field has fifteen wells - nine producers with binary valves and six water injectors with six-position valves - producing and injecting in two or three zone totalizing 39 downhole valves to be optimized. In this case, the results have shown a significant increase of the expected cumulative oil production when compared with the base case.
The intelligent well technology provides the capability to remotely monitor and manage multiple production zones independently, reducing the cost of wells interventions, accelerating the production and reducing the injection and production of water. The ability to control multiple production zones comes from downhole inflow control valves. These devices may be binary (on-off behavior), or multi-position, choking the production zone with a discrete number of positions, or infinity variable position.
The benefits of the intelligent wells technology were shown in practical applications1-6 especially for multiple-zone producing commingled. During the operation with intelligent wells, one possible approach is to react when problems occur, for instance, choke the production zones with high water cut. Yeten et al.7 has called this approach as reactive control strategy. Another approach is to use the intelligent completions in conjunction with a predictive reservoir model. This model may be coupled with optimization algorithms to define production strategies that maximize the value of the field.
Some previous authors have studied the production optimization with intelligent wells. Brouwer et al.8, have presented a methodology that maximized sweep in a water flood study. The strategy was based on choking the segments with highest productivity index and redistributing the production in others segments. Brouwer et al.9, have applied the optimal control theory for production optimization. Yeten et al.7 have used a conjugate gradient optimization method coupled with a reservoir simulator to optimize the production with intelligent wells. They have proposed to divide the simulation into several steps of optimization. Ajayi et al.10 have applied an optimization process based on derivatives calculated as the change of the production rate of undesired reservoir fluid, water or gas, with the correspondent change in the desired fluid, oil or gas. The process corresponds to choke the zone with highest derivatives values in each time step. Naus et al.11 have proposed an optimization strategy with infinitely variable inflow control valves using a sequential linear programming to maximize production at a specific moment in time.
This paper presents an implementation of method to optimize the production in intelligent wells varying the wells inflow control valves settings using an optimization algorithm coupled to commercial flow simulators. The optimization is based on direct search methods. This kind of algorithm has some advantages in this case: the algorithm is based only in objective functions evaluations, this fact allows to consider the flow simulator as an external program like a "black box??; the algorithm permits to model binary and multi-position downhole valves, differently of gradient-based algorithms where is difficult to model problems with a finite number of discrete solutions; and the algorithm takes advantage in a grid computer environment, because the objective function evaluations can easily be done in parallel.