Corrosion Control in the Oil and Gas Industry Using Nodal Analysis and Two-Phase Flow Modeling Techniques.
Characterization of corrosion in the oil and gas industry is becoming of increasing importance for safety reasons as well as for the conservation of the production facilities; therefore preventing down time and damage to the environment. This article presents the methodology used by our company to characterize the corrosion behavior of the whole production facilities, taking into consideration the hydrodynamic and thermodynamic conditions of the produced fluids (flow velocities, flow pattern. liquid holdup, pressure, temperature, etc.) as they flow from the reservoir through the surface installations (flowlines, gas/oil gathering and transmission lines, gas processing plants, artificial lift systems, etc.). The methodology uses Nodal System Analysis (NSA) and Two-Phase modeling techniques to: 1) optimize the entire production system to obtain the most efficient objective flow rate taking into consideration the corrosive/erosive nature of the produced fluid and 2) characterize the corrosive nature of the produced fluid as they flow through the above mentioned installations. The NSA and Two-phase modeling was performed using commercially available simulators and CO2 corrosion rates were determined using well known published correlations. For H2S corrosion, NACE MR0175 criteria is applied. The application of this methodology has allowed for corrosion control strategies, protection and monitoring criteria, inhibitor optimization and to increase the effectiveness of already existing corrosion control systems.
Velocity enhanced corrosion problems in oil and gas production equipment are common. These often occur when produced fluids are accompanied by carbon dioxide (sweet corrosion) and/or hydrogen sulfide (sour corrosion). In recent years the presence of carbon dioxide in the produced fluids is encountered more frequently due to the use of enhanced oil recovery techniques involving CO2 injection into reservoirs, due to the occurrence of sweet/sour gas production from deeper wells, and increased exploitation of heavy crudes with higher H2S and CO2 content. The effect of flow velocities on corrosion are more severe and more frequently encountered in multiphase flows and in many cases corrosion rates depend on the flow patterns. These are enhanced in the presence of sand inducing the phenomenon called erosion/corrosion.
Carbon dioxide dissolves in the presence of a water phase forming a weak acid (carbonic acid) which ionizes, thus reducing the pH and corroding carbon steel pipes. As the carbon steel corrodes, it forms a corrosion product scale (ferrous carbonate, FeCO3) which provides a degree of protection of the steel from further corrosion. The protectiveness of the scale depends on environmental factors and the characteristics of the steel. The environmental factors are those that define the medium in which the steel is corroding such as temperature, CO2 partial pressure, solution chemistry, fluid velocity, single-phase or multi-phase flows, flow pattern, pipe geometry, solution pH, and the ratio CO2/H2S. Steel characteristics include chemical composition and heat treatment, which determines the microstructure of the steel. As the partial pressure of CO2 increases, the pH decreases, thus making the environment more corrosive. Temperature directly influences the solubility of the FeCO3 scale; as it increases, corrosion rate decreases because of the formation of a more tenacious scale. P. 501
A new mathematical model for the miscible displacement in fractured-porous reservoirs is developed. A model is obtained by the upscaling of the traditional miscible displacement equations from the scale, which is lower than the fracture opening, up to the scale, which is much larger than the block size.
The model is based on simultaneous use of different mathematical methods: areal averaging of fluid fluxes, asymptotical averaging in systems with periodical heterogeneity, analytical solutions of the problem of interaction between the flux via fractures and the single block. The model takes into account diffusive, gravitational and advective mechanisms of the mass exchange between blocks and fractures and also hydraulic interaction between fluxes via systems of blocks and fractures.
The formula for the modified fractional flow functions which depends on the geometric tortuosity of the system of fractures are proposed. The tortuosity coefficient can be found from the 1-D laboratory displacement data.
The model developed has been validated by comparison with the number of laboratory studies of the miscible displacement in fractured-porous media.
Miscible gas injection into fractured-porous oil and gas-condensate reservoirs is the effective enhanced hydrocarbon recovery method. Nevertheless the efficiency of this IOR method is strongly dependent on parameters of fractured-porous system and on properties of reservoirs and of injected fluids.
The major oil reserves are located in the block system, so the recovery efficiency during the miscible gas injection is determined by mechanisms of the 'fracture-block' mass exchange and by the resulting displacement from blocks.
The displacement efficiency in fractured-porous media with the large opening and the high permeability of fractures and the low permeability of blocks is poor due to the fast breakthrough of the injected gas through the fractured system and low recovery from blocks. For the fractured-porous media with the less contrast between the fracture and the block conductivity the recovery is high.
Nevertheless, the recovery is determined not by the heterogeneity of the fractured-porous system only. If the fractured system conductivity is significantly higher then the one for the block system, but the blocks are small, the recovery could be still high.
The diffusive, gravitational and advective (convective) mechanisms of the fracture-block mass exchange during the two-phase partly miscible flow have been distinguished. The intensity of these three mechanisms are affected differently by the variation of the displacement velocity and also by the viscosity and the density of the injected fluid. Therefore, the comprehensive mathematical model which takes into account all the above mentioned recovery mechanisms is required for the optimization of the miscible gas injection into fractured-porous reservoirs.
Mathematical description of the flow in fractured-porous media are based on the image of the mass transfer in the double-porosity system (two fluxes via systems of fractures and of blocks interact with each other), by G. Barenblatt, Y. Zheltov and I. Kochina. In some other models the flux in blocks is neglected and blocks are treated as the source-sink terms with the given law of the mass exchange.
Another approach is based on the consideration of the fractured-porous media as a periodically heterogeneous media, and the upscaled model is obtained by the asymptotic averaging in periodical systems. The method allows to derive more complex formulae for the fracture-block mass exchange and to obtain the explicit formulae for effective permeabilities for systems of fractures and of blocks.
Diffusive mechanism of the fracture-block mass exchange corresponds to the linear term proportional to the concentration difference in blocks and fractures.
During matrix acidizing stimulation jobs, the use of more acid does not mean more skin reduction during the treatment. The excessive acid volume serves only to dissolve and weaken the matrix rather than remove the damage. The optimum amount of acid needs to be determined in each matrix stimulation case. This is done by calculating the real time skin value which is used to estimate the optimum volume needed.
The acid volume and type used in matrix acidizing stimulation is usually based on the experience of the operator in the field. Trial-and-error or on-site injectivity measurements are used to optimize the acid volume. In order to limit trial-and-error exercises, a real-time on-site treatment evaluation method has been developed to assist the field engineer with a new real-time matrix volume estimation technique. History matching and curve fitting the early stages of the acidizing job, the method determines when the maximum damage removal is obtained and corresponding acid volume.
This work proposes the use of real-time matrix acidizing data in conjunction with real-time skin effect calculations to estimate the optimum acid volume to be used on-site. The results are compared with three acidizing field cases in the paper, but has been verified for sixteen (16) field cases. The method can be used for optimizing acid volume during matrix stimulation in progress and deciding when to stop pumping the acid.
The concept of skin effect has been used as a magnitude of near wellbore flow impairment. The total skin effect is a multi-component quantity including mechanical skin effects such as mud solid and filtrate invasion during drilling and cementing, formation debris during perforation, and fine migration during production. Actually, the skin effect accounts for any deviation from an ideal undamaged, open-hole, vertical well. "Damage" may be caused by a number of phenomena on which traditional acidizing has no effect. Some of these pseudo-damage phenomena are gravel-pack and partial perforation and penetration.
Matrix acidizing stimulation is a treatment intended to remove near-wellbore damage. The determination whether to and how much to stimulate a well and the type of acid treatment should depend on comprehensive pre, real-time, and/or post analysis with high emphasis on quality data gathering. Matrix acidizing job effectiveness can be assessed through real-time or post treatment evaluation of data collected during the acid treatment.
During the matrix acidizing treatment, fluids are injected into the well, causing a pressure response recorded at the wellhead.
HoSim is a flow network solver which can determine rate and pressure at a given point in a flow system based on terminal and inflow relationships.
An analytic model for fractured horizontal wells has been integrated in conjunction with inflow through perforations and different geometries in the horizontal completion. The fractured well model used is based on the premise that a boundary value problem may be conformally mapped onto a complex plane. The model divides each fracture and the wellbore into a number of flow divisions. Flow into a section is based on the length of each section and the associated pressure distribution. The model is pseudo steady state, but has the capability of being integrated with a gridded reservoir simulator the see the reservoir behavior which may include water or gas coning over time.
This paper gives the theoretical development as well as a North Sea field case simulation and results. The simulator is a new tool that gives the completion/reservoir engineer a way to better optimize single or multiple fractured horizontal well designs for maximum recovery.
Evaluation of multiple completion and stimulation alternatives has become more difficult in recent years due to the many options available on the market today. One of the most challenging scenarios for simulation is the multiple fractured horizontal well. The potential for enhanced recovery coupled with large completion and stimulation expenditures creates a need for better understanding of these complex flow environments. The model presented here attempts to model a system composed of a reservoir drainage block, multiple orthogonal fractures intercepting a horizontal wellbore, and a completion within that horizontal wellbore. This model is constructed using the concept of a flow network, where each component of the system is further subdivided into multiple flow divisions.
Description of the System
A horizontal wellbore is exists within a rectangular drainage block which has dimensions of axbxh. The wellbore azimuth is coincident with the local least principal stress direction within the drainage block. Hydraulic fractures which are introduced in this well will therefore lie in a plane orthogonal to the wellbore. It is assumed that induced fractures are spaced evenly along the length of the wellbore, making it possible to divide the drainage block into subsections, with one fracture lying within each block. These subsections are indexed as b1,b2,...,bn. Each fracture is assumed to be rectangular, having dimensions ci, Lfi, hfi. A diagram of this system is shown in Figure 1. This subdivision of the problem into multiple subproblems assumes that there is no flow between adjacent subdivisions. The size of the end blocks is adjusted to account for the effect of the 'greedy' end fractures, but this assumption requires that each fracture under consideration be given the same conductivity profile when discretized in the network model.
The lengths of the subdivisions of the drainage block can be derived based on the geometry of the system described above by assuming that each equally spaced fracture is geometrically centered in its drainage block, with the exception of the first and last fracture. It can then be stated that the length of the first and last block is given by,
In order to produce available fluids, pump jacks used at surface have been engineered to stroke as slow as 6 (six) strokes per minute (SPM) and as fast as 14 (fourteen) SPM. Depending on available production, pump length (stroke) and pump size were the only two other variables taken into consideration. This approach works well until a time is reached when production has declined to a point where production inflow no longer matches pump capacity. As production declined, the typical method of compensation was to shorten the stroke, downsize the pump and remain at a constant SPM somewhere around 10 (ten) SPM.
As production declines this approach eventually results in partial pump fill even with small bore pumps, short stroke and slowing the unit as much as possible with current sheave limitation.
At this point it is physically impossible to fit a large enough sheave on the gear box or small enough sheave on the electric motor to reduce the speed below approximately 6 (six) SPM.
On some lower volume fluid production (100 bbls. and less) beam pumping wells it is difficult, if not impossible, to maintain constant production without tagging bottom at all times to keep them from "gas locking". The unit pumps quite well for a few days with the pump spacing set just off "tag" then stops pumping and the string has to be lowered 12-14" to make it pump again. If the string is not raised up again the "tag" destroys the clutch at the top of the pump (fig. 14) and pump life is short with all too frequent parted rods and tubing failures.
By reducing SPM to match the fluid inflow capability of the formation a more constant effective weight can be kept on the end of the rod string (on top of the traveling ball) so the rods don't contract and cause pump spacing to vary.
By reducing SPM according to the available fluid production of a well and keeping a pump efficiency in the range of 60%+ (appendix) erratic rod loads could be reduced or eliminated and production maintained without continually "tagging" bottom. Slowing down like this should prevent excess gas being driven into the tubing. This would eliminate the possibility of the tubing flowing above the pump. If the weight of fluid over the plunger were more constant and rod elasticity was better controlled the plunger would continually return to a preset point at the bottom of the stroke. At the slower SPM more time would be allowed for gas separation to happen in the annulus instead of causing interference in the pump. This would keep the unit pumping constantly with no additional adjustments at surface. Prime mover horse power and amperage draw could be reduced as well. At the same time tubing wear would be reduced and pump run time extended (ref. 2). (By shutting the unit down at the time of pump off; this paper shows better pump run times).
The well chosen for application of the theory came on line at 150 barrels per day and 500 mcf gas then declined to 12.7 barrels per day and 64 mcf gas. A 100" stroke length beam pumping unit was on the well with a 1.5" zero slip pump installed (fig. 14). Other well information is recorded in (table 1) Pumping was very erratic as it continually needed to be respaced in order to make production. Dynamometer cards taken before the slow down were very erratic and unpredictable.
The development of accurate digital measurement of instantaneous powerduring a pump stroke has made possible a very quick and detailed analysis ofthe efficiency of the pumping system. The efficiency is then used as thebenchmark for determining whether a complete well performance analysis iswarranted from the standpoint of making best use of personnel and economicresources to increase oil production.
In addition, power measurement provides direct information about liftingcost per barrel of fluid and barrel of oil produced, electrical and mechanicalloading of the prime mover, peak power demand, power factor and minimumrequired ratings. These results give operating personnel information regardingpotential problems and give to management a complete picture of thedistribution of pumping costs.
The power measurements are also converted, by the software, to instantaneoustorque and presented as continuous torque curves for the upstroke anddownstroke. This allows determination of the existing level of counterbalanceand provides the most rapid and accurate method for counterbalance adjustmentto achieve lower torque loading on the gear box and reduced energy utilization.One of the principal advantages of this balancing method is that counterbalanceadjustment can be made without need for an accurate description of the pumpingunit's geometry which is often unknown or inaccurate. The effect ofcounterweight displacement on torque and power is observed immediately byrepeating the power measurement after relocating the counterweights.
This paper presents a series of case studies showing the Application ofpower measurement to a variety of pumping systems and components, includingconventional, Mark II, Rotaflexunits and high efficiency motors.
External Parallelization of Reservoir Simulators Using a Network of Workstations and PVM.
This paper shows a simple way to perform external parallelization of reservoir simulators to help engineers in the process of production history matching. The key idea is to use external parallelization without modification of the existent reservoir simulators sequential codes, providing efficient and simple solutions procedures. An homogeneous network of workstations is used to create a virtual machine by using the software PVM.
This work provides a procedure to make sensitivity analysis of different reservoir parameters and an optimization technique to choose the best value for each parameter. The influence of the number of workstations is studied and a comparison between the wall clock time and number of iterations of sequential and parallel algorithms for different cases is presented. The sensitivity analysis and the optimization algorithm are tested and the advantages of external parallelization are presented. It is shown that it is not necessary to modify or adapt reservoir simulators codes to take advantage of parallelization.
This procedure is a possible solution to the lack of parallel software in reservoir simulation and is the first step to perform efficient semi-automatic history matching by using external parallelization. The process shows a good performance and can also be used in other areas of the petroleum industry such us production optimization, well spacing optimization, etc.
The big time consumption of reservoir simulations is forcing the development of codes that take advantage of parallel machines. However, the development of petroleum engineering software is not following the improvement of parallel computer technology. One reason for this gap may be the complexity of the development of good parallel reservoir simulators.
Published papers in the petroleum engineering literature show that great attention has been given to internal parallelization through development of new software or modification of existing simulators codes. Authors choose different parallelizable parts of the simulator code where they apply parallelization techniques. They propose different ways and algorithms to take advantage of parallel computers.
This procedure has many advantages but the development of such software can be very complex. One alternative to this procedure is to take advantage of parallel computers and network of workstations without modification of reservoir simulators codes (external parallelization).
This paper shows some advantages of applying external parallelization to a commercial simulator and shows a methodology to use an existing network of workstations to save time in the history matching process. This type of procedure has the advantage of simplicity, shortening the time of software development.
In most of the petroleum engineering literature about parallelization of reservoir simulation, most of the attention has been given to the solution of the system of equations resulting from the conservation equations. Some attention has been dedicated to other parts that consume computing time. Sometimes, this is the best procedure but in many others, the resulting codes are complex, their development is slow, and their reusability is complicated.
Ouenes and Weiss, for instance, used PVM to perform an "automatic" history matching using internal parallelization. The same type of work is shown in many other papers in the petroleum engineering literature. This procedure is important but the development of such software can require an excessive amount of work.
Because many companies have a network of workstations that are not fully utilized, especially during the night, we tried to develop a technique that is very easy to use, simple to learn and to program, and that can use all the potential of the existing machines.
The objective of this work is to study the viability to do external parallelization of a commercial simulator that was built to run serially. The simulator is run in each workstation available in the virtual machine created by PVM. This kind of parallelization allows the simulation of different input files at the same time and, therefore, it is easy to do a sensitivity analysis followed by an optimization method to choose the value that makes the best history matching. The efficiency of the process is evaluated comparing the performances of parallel and serial algorithms, studying also the number of workstations that gives the best performance for the parallel process.
This paper summarizes an integrated field, experiments and computer simulation research program conducted to support a review of the reserves and development plan for the Hamaca Area in the Orinoco Belt (OB), Venezuela. The impact of the foamy oil mechanism on the Hamaca heavy-extra heavy Oil reserves and the importance of understanding this behavior is presented in this paper. The study was conducted in reservoirs with the largest production history (within the OB). The experimental results showed in-situ formation of non-aqueous oil foam with high gas retention, improving oil mobility and leading, therefore, to high well productivity. An experimental recovery factor over 10% was obtained under primary conditions so it was possible to increase oil reserves by approximately 30% over the currently accepted volumes. Experimental results provided input parameters to perform preliminary simulation runs which would allow the modification of well spacing schemes, the generation of a high-probability production profile, and the optimization of artificial lift systems, incorporating larger capacity equipment.
When the exploitation of Hamaca heavy oil reservoirs began, it was assumed that the primary production mechanisms were sand compaction, solution gas drive and thermal effects due to steam cyclic stimulation which improves oil mobility. However, an unexpectedly high cold production lead to research of the drive mechanisms to explain the special production performance. Similar behavior has been observed in the Lloydminster area of Canada. Despite considerable speculation, a number of authors have studied foamy oil behavior and some research has recently been done, yet the mechanism leading to this behavior still remains to be successfully explained. Reservoir properties that were proposed to explain the behavior in the OB and in Canada include unusually high sand permeability and/or critical gas saturation. In Canada, high critical gas saturation is now an accepted property of some reservoirs, the so called Foamy Oil reservoirs. However, a monitoring field program of sand compaction and land subsidence did not show any evidence that this mechanism has taken place in the Hamaca Area. None of these proposed properties and mechanisms are consistent with field observations. On the other hand, numerical models showed high uncertainty when trying to match the reservoir production pressure behavior. As a consequence, a research program was designed as part of an integrated reservoir study. It included a production behavior analysis, an experimental program for fluid/porous media characterization through conventional and nonconventional PVT tests, and solution gas drive experiments at research centers of Venezuela (INTEVEP), Canada (PRI-CMG) and USA (LAB).
Perforating and pipe recovery operations performed with coiled tubing provide logistical and technical advantages in certain situations. Conventional tubing conveyed firing head adapters have been redesigned to be conveyed by this method which is activated by coiled tubing pressure. Summary results are presented from 26 different operations performed within Trinidad. Horizontal and highly deviated wells can benefit from this option when electric line methods may be limited due to hole angle or dog leg severity. Depth control is addressed by means of a correlation run or use of profile location mechanisms which can be incorporated in the perforating assembly. A review of pressure deployment systems is provided for thru-tubing underbalance perforating of long sections. This system has added value to the production operator by minimizing space constraints of offshore platforms and limiting additional equipment needed for remote locations.
A hydraulic style detonation mechanism is activated by the combination of applied coiled tubing pressure, and hydrostatic pressure (Fig. 1). As pressure in the coil increases to the shear stud value, a piston is driven into the initiator which detonates the charge. Depending on job conditions, the firing head can be equipped with a circulating port after detonation, or it can remain a closed system which prevents fluid loss to the wellbore and formation. For the closed system, a tubing pressure-activated hydraulic disconnect is used which operates on the differential pressure between the coiled tubing and production tubing. The calculation methods shown in the Appendix should be followed to verity that 20 percent or more shear value exists between the hydraulic firing head pressure and the disconnect pressure. This firing mechanism provides the ability to deploy hollow steel carrier perforating guns, strip charges or jet cutters.
Depth Control. Depth correlation is required for perforating and pipe recovery operations. Several methods are used which include a tubing profile locator that can be run in conjunction with the perforating assembly. Also a memory GR/CCL or drift nozzle can be run to locate a tubing profile at a known depth which requires an additional coiled tubing well entry before perforating. Typically, a stripe is painted on the coil when an additional trip is made and can be used to corroborate the depth. It should be noted that elongation of the coil can occur during successive trips that could be as much as 2 in. per 1000 ft. based on field results.
Wellbore Trajectory Problems.
Deviation. Field experience indicates that 70 degree hole angle is the upper limit for electric line conveyance of perforating guns in wells with a kick off point of 3000 ft. and a 9000 ft target zone. However, lower hole angles with higher departures may decrease the electric line access threshold to as low as 60 degrees due to frictional weight loss. Deviations above 70 degrees will require coiled tubing to reach the target depth for most situations.
Corkscrewed tubing or Wellbore Irregularities. Completion practices which result in corkscrewed tubing can also limit the ability of electric line from reaching the objective. Coiled tubing entry in this environment requires extreme caution to prevent becoming stuck due to friction between the coil and production tubing, which may not be apparent from wireline drift runs.
The study of stratified solid-liquid annular flow is relevant in several areas, e.g., mining and oil industries. A typical application is in the study of the removal of the rock formation cut by the bit during the drilling of a horizontal or highly inclined well. Several theoretical models have been proposed for the description of this phenomenum. In these models, some empirical parameters are required, such as the shear stresses at the interface between the fluids and the cuttings bed and the maximum cuttings-wall friction factor which avoids bed movement. This work shows the results of an experimental work developed on a large scale flow loop aiming at the quantification of these empirical parameters. The tests consisted on the visualization of a sandstone bed erosion by different polymeric solutions flowing through an annular section. Pressure losses and steady state bed heights were recorded for several input parameters, such as fluid flow rate, rheology and density, annular geometry, eccentricity and particle size.
The drilling of horizontal wells is already a common practice worldwide, and considerable effort is being spent now at reducing costs and increasing safety for these operations. One way of pursuing these objectives is through the use of computer simulators, which are of great help on designing and troubleshooting.
One important aspect of drilling high angle wells is the removal of the cuttings that tend to deposit on the lower part of the horizontal annular section. The description of the complex solid-liquid flow taking place during this removal, in which non-Newtonian drilling fluid is used, requires information about the interactions among the different materials involved, namely, fluid, rock cuttings, rock formation and pipe. These properties are necessary inputs for mechanistic models, developed to characterize the phenomena. It very often occurs that, due to a lack of experimental data, researchers use information from the mining industry, i.e., data obtained from pipe flow of water and mineral grains. This practice, naturally, is questionable, because those flows do differ considerably from the flow of drilled cuttings and non-Newtonian fluids in an eccentric annulus.
In order to overcome this lack of information, an extensive experimental work has been performed in relation to cuttings transport. In this paper, the data have been used to quantify the friction at the interface between the stream and the cuttings bed, as well as the static force between cuttings bed and bore hole wall, which prevents the bed from moving. The procedure used is an adaptation of the one proposed by Torres in the context of gas-liquid stratified flows.
Experiments have been conducted on a pilot scale flow loop built at the PETROBRAS research center (Martins et al.). The test section consists of a 12m long 0.127 m diameter acrylic pipe where 0.0508m and 0.0635 m PVC internal pipes are introduced to represent two typical annular geometries of horizontal wells. A solids injection system is coupled at the entrance of the test section. Fig. 1 shows a schematic view of the flow loop.
The tests consisted of the following steps:
1) Build up of a constant height solids bed along the test section;
2) Interruption of the solids injection and increase of the fluid flow rate: the bed is eroded until a new equilibrium height is obtained;
3) Measurement and recording of the pressure drop and bed perimeter.
Experimental data have been obtained for different annular geometries, and different values of solid/fluid properties and operational variables. Hence, variables such as eccentricity, particle diameter, rheology, fluid density and flow rate have been changed. The mass flow rate of the polymeric solution has been measured by a Coriolis type equipment, while pressure drop has been determined by a differential pressure transducer. The sandstone bed perimeter has been recorded by direct length measurements.
Test Matrix. A total of 55 different tests were performed using solutions of Xhantam Gum, at different concentrations, to represent the drilling fluids and particles of sandstone of different diameters.