Artificial neuron network (ANN) models are designed to emulate human information processing capabilities such as knowledge processing, speech, prediction and control. The ability of ANN systems to spontaneously learn from examples, reason over inexact and fuzzy data, and provide adequate responses to new information not previously seen, has generated increasing acceptance for this technology in the engineering field and resulted in numerous applications. A preliminary investigation into the use of this novel technology is presented towards predicting formation damage by quantifying wettability and two-phase relative permeability of oil reservoirs.
An artificial neuron network model based on the Back Propagation technique is trained with a number of variables from experimentally established relative permeability (relperm) curves. The reservoir core input data covers an extensive range of porosities and permeabilities from different lithologies having diverse wettabilities. The trained model is then tested with only a couple of easily obtainable input variables such as the Swc and Sor and predictions are made on the wettability and relperm curves. A change or shift in the relperm curves is associated with changes in wettability, and perhaps to formation damage in the drilling process.
The wettabilities of the rock-fluid system are predicted to within 90% of the experimentally determined values. The relperm curves, particularly the end-points are predicted to within 85% of the measured results. The accuracy of the predictions are significantly enhanced with model training using more precise reservoir data and better defined formation lithologies. Neural networks have immense potential in predicting relperm curves and thereby assessing formation damage in reservoirs.
Formation damage is usually associated with the decrease in permeability of hydrocarbon reservoirs. The resulting permeability changes directly influence the relative permeabilities of the hydrocarbons in-place. The important reservoir parameter, wettability, may also be altered, depending on the type and cause of formation damage. Drilling and completion fluids, and additives, are known to alter wettability around the wellbore, due to improper fluid and/or additives selection. Accurate prediction of the wettability and the relative permeabilities at any instant will therefore provide a valuable tool for formation assessment and the ensuring changes due to possible formation damage.
The evaluation of a present acid injection profile used in acid flowbackanalysis is an important tool for optimizing acid treatments, as it affects theacidizing process by improved results of hydrocarbon production rather thansubjecting the well to yet another optimized acid treatment in the future. Inthe standard method of acid flowback analysis, the engineers interpret theplots of the return fluid ionic content versus return volumes at certainincrements i.e. every 10-20 bbl. In our work. we have found that this methodmay pose more questions than it answers.
In order to simplify the acid flowback analysis method and provide a moredefinitive interpretation. we introduced a simple derivative technique in theanalysis of the dissolved formation or the scale dissolved ionic content. Inthis technique the ratio of incremental ionic concentration per incrementalproduced volume (the derivative) was plotted against the cumulative producedvolume. The significance of using this technique is that the derivative ofionic concentration with respect to the produced volume is very sensitive tothe minute changes in the ion concentration per unit volume or cumulative acidflowback volume.
We applied our method to two wells, which we shall call them wells # 1 and#2. In the case study of well # 1, first we found that the acid stimulationtreatment dissolved significant amounts of material containing Potassium,Sodium. Magnesium, Iron, Aluminum, Silicon, Barium. and Calcium. But thetreatment did not remove the Zinc completely. The source of Zinc could havebeen either the formation of zinc hydroxide from heavy brines or the 60% Zinccontent of some pipe dopes. Incomplete dissolution of material containing Zincresulted in an inefficient treatment. Consequently, the treatment was able torestore only the original production rate but failed to result in any increasein incremental production. Secondly. from the disordered dissolution profilesof Magnesium, Calcium and Barium, we found that these three elements existed inthe formation as individual mineral rather than scale.
Nonionic surfactants are commonly used during well stimulation for several reasons. They reduce interfacial tension between the acid and oil phases, thus improving acid/rock interaction. They are also used to form a stable foam which improves the sweep efficiency during acidizing. However, these surfactants should be employed at temperatures below their cloud point (defined as the temperature at which the surfactant solution becomes cloudy). This temperature signifies the onset of the surfactant salting out, which will reduce the efficiency of the stimulation and may damage the formation.
An experimental study was conducted to assess the effect of various acids and stimulation additives on the cloud point of nonionic surfactants. The influences of acids (inorganic and organic), mutual solvents, friction reducers, hydrogen sulfide scavengers, sequestering agents, short chain alcohols, simple salts, scale inhibitors, anionic surfactants on the cloud point of several nonionic surfactants were examined over a wide range of parameters.
The results indicated that the cloud point monotonically increased with the acid concentration. However, the rate of increase depended on the acid type and the number of ethylene oxide groups of the surfactant. Salts depressed the cloud point of nonionic surfactants at all hydrochloric acid concentrations examined. Alcohols, methanol and isopropanol enhanced the cloud point of nonionic surfactants. The effect of mutual solvents was found to be a function of the number of ethylene oxide groups of the surfactant, acid and mutual solvent concentrations. Anionic surfactants depressed the cloud point of nonionic surfactants at all sodium chloride concentrations examined. Clay stabilizers (cationic polymers) and hydrogen sulfide scavengers depressed the cloud point whereas scale inhibitor and phosphonic acid did not affect the cloud point significantly.
It is extremely important to measure the cloud point of nonionic surfactants before performing a stimulation job. It is also recommended to use the acid formulation and mixing waters that will be used in the field.
Well stimulation is a process aimed at the removal of near-wellbore impairment due to deposition of particulate solids during drilling, workover or production operations.1,2 In this process an acid or mixture of acids is injected into the well to dissolve and remove this damage. Apart from removing deposited material, the acid can react with the rock matrix and enlarge pore sizes.3 This will allow insoluble fines to be flushed out when oil production is resumed (producing well) or during backflow (injection well). Enlarging pore sizes will improve the permeability in the wellbore area, hence the productivity or injectivity of the well will improve.
Scale formation during waterfloods can damage reservoirs far beyond the wellbore region. A comprehensive analysis with geochemical modeling can improve waterflood design in the selection and/ or mixing of source waters and thus, mitigate formation damage arising from injecting incompatible fluids. This method can also predict the types of scales and their severity at various production stages. This will help optimize treatment schedules and thus reduce operation costs.
In the case study of Zone 4, Prudhoe Bay Unit (North Slope, Alaska), the geochemical model was validated with the laboratory analyses of Zone 4 produced water samples. Then it was used to evaluate the impact of mixing formation brine and seawater on rock-fluid interactions and scale formation. The prediction is consistent with the observation of calcite formation in early production at Prudhoe Bay. The model also indicates the precipitation of iron carbonate and iron sulfide scales as the waterflood matures.
This paper presents the results of a study of sludge formation from thereaction of "real-world" hydrochloric acid with several crude oils.Dissolved ferric iron has previously been shown to be a major factor in thegenesis of sludges in acidizing treatments. This work compares the sludgeabatement capabilities of a novel iron reducing agent with those of severaliron control agents in common use in the oilfield. Simple tests based onestablished API procedures are used. Results of field treatments using the newreducing agent are included.
The reaction of common oilfield acids in their pure state with components ofsome crude oils to produce insoluble solids known as sludge has long beenrecognized as a significant obstacle to successful acid stimulation in manygeographic areas. Chemical technology has been available for many years toaddress the problems of less-than-optimal production stimulation caused by thissludge formation. The American Petroleum Institute recommends an "acidsludge test," distinct from emulsion testing, to evaluate chemicalanti-sludging agents. These materials are often based on dodecylbenzenesulfonic acid (DDBSA) or its derivatives; they frequently give excellentresults in the API test, even with difficult-to-treat oils such as thenotorious light, asphaltic oils from Alberta.
Another chemical problem in acidizing, first recognized in the 1930s, comesfrom dissolved iron (particularly when in the ferric, or Fe+3, state) in theacids reaching the formations to be stimulated. The true nature and severity ofthe problems caused by dissolved iron, however, was not widely appreciateduntil the mid-1980s when some significant papers were published dealing withtwo distinct facets of the issue.
Coulter and Gougler, in 1984, reported a field investigation whichdocumented very high levels of iron in hydrochloric acid "as delivered"to the zone to be acidized in a typical acid treatment. They showed that themajor source of this iron is the tubular goods in the well. The chief concernexpressed in their paper, as well as in the patent literature of around thattime, was in preventing precipitation of inorganic ferric iron compounds, suchas ferric hydroxide. This precipitation is well known to occur as acid spendsto a pH of around 2, and was widely held to be the major mechanism of formationdamage attributable to iron. Many systems to suppress iron solids depositionwere then, and are still, in use. Most depend on chelation or reduction offerric iron as the acid spends toward and past this pH value.
The Mittelplate oil field is located in the Tidelands National Park in theGerman North Sea. Its location in an ecologically sensitive area was achallenge during well completion planning. The horizontal well A 7a penetratesthe Dogger-delta sandstone in the Jurassic trough on the flank of a salt dome.The well's productivity is a very important factor in the economics of fielddevelopment. The specific reservoir conditions were decisive for well designand completion fluid selection. To retain all options for future productionphases, a concept using a cemented liner and subsequent perforation wasselected. Formation damage control was supported by short workover time due tonew completion techniques. To avoid formation damage during the initial stagesof perforation and completion, a non-damaging completion fluid was evaluated inlaboratory investigations. The objective was to verify the rheologicalproperties, carrying capacity, and time stability. Minimizing fluid loss duringcompletion guarantees optimum well productivity. A xanthan-based carrier fluidwith oil-soluble resins serving as bridging agents was selected. Intensivequality control of the chemicals ensured the use of correct batches in thefield, contributing to the positive results of the project.
The formation damage caused by the injection of water containing suspended particles, which are stable and are not adsorbed spontaneously onto pore surfaces under Brownian motion, has recently been analyzed at a pore scale level. Formation damage is the result of four more or less overlapping successive steps: (1) deposition on a grain surface, (2) formation of mono- or multiparticle bridges with subsequent accumulation upstream from the bridges, (3) internal cake formation as soon as the nonpercolation threshold has been reached near the core entrance, and (4) external cake formation. The surface deposition is not uniform over the grain surface and varies from the upstream stagnation point to the near pore throat zone according to a function depending on flow rate and surface forces. The bridging of pore throats is strongly dependent on the effective pore throat-to-particle size ratio, and the pore-throat size is often reduced by previous surface deposition.
A new model has been developed to predict formation damage while taking into account these different steps. The dominant mechanism in each step is governed by parameters that have a clear physical meaning. However, due to the complexity of natural systems, these parameters cannot be quantitatively predicted from theoretical considerations but can easily be determined by specifically designed lab experiments. The model predicts the retention by deposition, by bridging and by subsequent accumulation upstream from bridges, the concentration in flowing particles and the local permeability reduction as a function of the distance from the inlet, as well as the overall permeability reduction, and the beginning of external cake formation.
This new model appears to be an effective tool for analyzing the consistency of a set of laboratory data and for selecting the values of the parameters that must be introduced in a near-well bore field simulator for the proper prediction of formation damage in a given application.
The early models proposed in the oil literature aimed to fit the decrease in relative permeability observed when a particle suspension is injected into a permeable core. Thus these models are empirical or at best purely phenomenological in the sense that they simulate observed phenomena by using equations without any clear physical meaning. They are very attractive however for petroleum engineers since they are very simple and easily introduced in conventional field simulators without increasing computing time too much. A good review of these empirical models can be found in Ref. 1.
This paper describes laboratory experiments demonstrating that the use of polymer linkage specific enzymes is effective in breaking the starch polymer constituent of sized salt fluid loss control materials (FLCMs). It is further demonstrated that the linkage specific enzymes are efficient in reducing the near-wellbore damage induced by the starch polymer that makes up roughly 75% of the FLCM's polymer content. As a result of the laboratory work presented here, it is shown that return permeabilities are in the 80-98% range without the use of acid systems.
Sized (or graded) salt systems are effective, relatively non-damaging FLCMs when properly applied. However, when improperly used, these same FLCMs may be the source of significant completion damage interpreted from pressure buildup analysis as high skin values or reduced perforation efficiency. Burnett1,2 demonstrated that fluid loss control materials are relatively impossible to remove when no leakoff occurs in the injection direction; this typically occurs when the FLCMs are doing the job they were designed to do. McLeod and Crawford3 pointed out that a contributing factor in poorly performing gravel packed completions was the inefficient removal of fluid loss control materials, since insufficient leakoff prevented the perforation tunnels from being fully packed with gravel. FLCMs must be removed prior to gravel packing to improve leakoff and gravel pack efficiency4. Improper cleanup of FLCMs, especially in gravel packed wells, may cause permanent damage to completion productivity5.
In long or horizontal (or both) completion intervals, sufficient drawdown in the production direction may not exist to remove the filter cake of a drill-in fluid system. Acid sensitivity concerns in openhole sections may prohibit the use of adequate volumes of acid to degrade the filter cake. Facility upsets that sometimes occur when large volumes of acid are produced back to surface as a result of treating long sections may compound the problem of using an acid system for FLCM cleanup.
The laboratory data presented here was designed to replicate downhole conditions and demonstrates that the use of polymer linkage specific enzymes does not require leakoff through the filter cake of a sized salt FLCM for efficient removal. The enzymes will degrade the starch polymer in the FLCM from the point of contact in the wellbore out to the formation matrix. A subsequent overflush with an undersaturated brine will effectively dissolve the remaining salt particulates. An alternative to the use of acid systems for cleanup of fluid loss control agents is presented.
Graded salt systems are widely used by the industry as fluid loss control materials during completion and workover operations, and during the drill-in phase of horizontal and high angle wells. The sized salt systems are comprised of undissolved salt particles of a size distribution specific to the pore throat diameter of the interval being completed or drilled. The salt particulates are transported in a saturated brine and suspended by a crosslinked hydroxypropylated starch derivative and xanthan gum polymer system. The starch polymer constituent comprises approximately 75% of the polymer content, and the xanthan gum polymer the remaining 25%.
A filter cake is deposited on the formation face or in perforation tunnels by dynamic fluid loss during pumping or by the differential between hydrostatic pressure in the wellbore and formation pore pressure. Polymeric damage in the form of unbroken gel residue or dynamically deposited filter cake can significantly reduce well productivity due to impeded flow capacity in the near-wellbore region. The damage may be characterized by unbroken gel residue having limited mobility or by insoluble polymer fragments.
In completion and workover operations, the loss of completion brines due to a positive hydrostatic differential pressure is problematic from a formation damage and economics standpoint. Formation damage is incurred when solid particulates are placed in perforation tunnels and matrix pore throats during the loss of completion brines6. Water sensitive formations and formations with a history of fines movement may also be negatively influenced by excessive fluid loss. From an economics point of view, each barrel of completion fluid lost to the formation is money spent, a particularly important issue when some of the higher density brines cost US$100/bbl or more.
Laboratory studies and field case histories are presented that demonstrate enhanced well productivity when selection and formulation of solids-free, high density completion fluids are optimized prior to field use. Results of return permeability studies and compatibility tests between completion fluids and formation fluids are presented. These results illustrate the tendency for high density completion fluids to form stable emulsions with certain crude oils. Studies supporting the treatment of these fluids with non-emulsifiers, surface tension reducing surfactants or mutual solvents to reduce formation damage are discussed. Preliminary data describing unique chemistry of calcium chloride, calcium bromide and zinc bromide solutions are provided. These properties influence recovery of high density fluids from oil or gas formations.
The chloride and bromide salts of calcium and zinc are highly soluble in water. In combination, solutions of up to 19.2 pounds per gallon (ppg) density can be obtained. These solids-free solutions, commonly referred to as high density (HD) brines, are widely utilized in completion and work-over operations. The ability to achieve hydrostatic well control without solids has made HD brines the work horse of the completion industry. As pure solutions, HD fluids are inherently stable and relatively compatible with both well operations and petroleum bearing formations. Nonetheless, HD brines are most economically utilized when completion and work-over operations are engineered to minimize fluid losses to the productive interval. Reduced fluid costs and enhanced well productivity are the result.
One objective of this paper was to provide experimental data not previously published and that may impact the interaction of HD fluids with formation matrix. Load water recovery, fluid-fluid compatibility and shale/clay stability are areas of particular interest. Data presented in this paper include: (l) activity of water for various HD solutions, (2) viscosity of brines as a function of composition and temperature, (3) surface tension of typical HD brines, and (4) capillary suction time (CST) after exposure to clay minerals. Core flow studies using field and Berea cores were conducted to evaluate recovery of reference fluid (formation water, oil or gas) after saturation with HD completion fluid. In all core flow studies, fluid permeability was calculated. Several tests included ion analysis of the core effluents by inductively coupled plasma emission spectroscopy (ICP).
This paper describes the methodology and results of a series of testsconducted on HydroxyEthyl Cellulose (HEC) gel to determine its effects onreturn permeability. HEC gel is widely used in the oilfield as a carrier fluidin gravel packing operations and is generally considered to cause minimalformation damage. However, results of this study indicate that even when HECgel is efficiently removed from the formation, interaction between the fluidand the formation can result in reduced return permeability. Therefore,particular care must be given to quality control and mixing procedures tominimize potential formation damage.
HEC gel break-back procedures used by oilfield laboratories for returnpermeability measurement are based on displacement studies related to reservoirengineering. In these tests, flow through a Berea core is maintained atconstant pressure and later, after introduction of the test fluid, the changein permeability is calculated as a percent of the initial permeability. Inthese cases, any resulting loss of permeability was assumed to be caused by HECgel trapped in the formation.
To insure complete HEC degradation over time and temperature with andwithout the presence of formation minerals, seven different breaker systemswere evaluated to select the most efficient system. This study concludes thatdamage caused by HEC residue is minimal compared to the damage caused by theinteraction of the fluid and the rock minerals present in various formations.The results of our investigation also indicate that the performance of breakersis influenced by the presence of formation minerals.
This paper will present experimental data of HEC gel break-back in thepresence of formation minerals as well as the testing of chemical additivesdesigned to minimize damage and deliver improved return permeability.
Studies indicate that HEC workover, packer, completion and gravel packfluids cause minimum formation damage and yield high productivity. Viscositybreakers (e.g. enzymes, oxidizers, acids) are particularly useful duringworkovers and gravel pack operations. A fast breaker, consistent with a verylow level of insoluble residues, is essential in preventing wasted rigtime.