This paper highlights an unconventional approach of using DPDP (Dual Porosity Dual permeability) simulation technique for modelling hydraulic fractures in a full field simulation model during the forecast analysis performed on a tight carbonate reservoir in Kuwait. This was a part of integrated study in which ‘multi-stage hydraulic fracturing’ was recommended as the most optimum stimulation technique in order to enhance the productivity of all the proposed horizontal producers. Importance of DPDP model increases multi-fold when contrast between fracture and matrix permeability is in the order of 10 times or more. In the studied case, as average matrix permeability of the reservoir is in the range of 2-3 mD, this contrast is magnified to the order of 1000-10000 times (considering fracture permeability is in Darcies) which further complements the use of DPDP model.
Three different approaches were tried to model the impact of multi-stage hydraulic fracturing in the full field simulation model; 1) ‘Enhance Well PI’ for all the stimulated wells, 2) ‘Enhance Matrix Permeability’ in the vicinity of all the stimulated wells, hereby referred as SPSP (Single Porosity Single Permeability) approach, and 3) build ‘DPDP Model’ by using upscaled fracture porosity and fracture permeability without changing the matrix properties. First two approaches are very common in the industry but most of the times are not able to capture the real impact of hydraulic fracturing on flow behaviour (bi-linear flow), whereas DPDP model is designed to capture the flow through dual medium. In both SPSP and DPDP approaches permeability anisotropy (increased permeability in the direction perpendicular to horizontal section of the well) in the fractured zone was very well captured and was needed to honour the hydraulic fractures direction. Fracture permeability was calculated using the Poiseuille's law; few sensitivity cases were run to address the associated uncertainty.
Field cumulative oil production and recovery factor were analysed for ‘Enhanced Well PI’ case, SPSP cases and DPDP cases. Field oil cumulative production in DPDP cases is 6% more than SPSP cases and around 10% more than ‘Enhanced Well PI’ case. The hypothesis for the higher recovery in DPDP case with respect to other two cases is that bi-linear flow (fractures are getting filled with the matrix fluid and then feeding to well) is better represented in the DPDP model. Impact in this case is more significant due to the big contrast between matrix and fracture permeability.
Low capacity with high conductivity signature of hydraulic fracture is difficult to model in the SPSP or just by enhancing the well PI. Study clearly demonstrated the benefits of DPDP model for modelling hydraulic fractures over the conventional methods.
Accurate estimation of the Original Gas in Place (OGIP) early in the reservoir life is fundamental as field development plans and ultimate recovery strongly depend on it. This is particularly relevant when the conventional material balance is not suitable due to the lack of pertinent shut-in pressure measurements. This paper presents a case history of a tight gas field in which we use flowing material balance technique and type curves for decline curve analysis to calculate OGIP by using only flowing pressure and rate data.
The method uses fundamental pseudo-steady state theory, which determines that a plot of the rate-normalized pressure drop vs the pseudo-time produces a straight line with the slope of such line yielding the OGIP. The use of the pseudo time concept calls for the estimate of gas properties at the prevailing reservoir pressure which in turn is a function of the OGIP and the cumulative production. We propose an iterative scheme based on the Newton-Raphson method to compute the OGIP using the flowing material balance technique coupled with the conventional P/Z material balance.
We illustrate the application of the method with the aid of synthetic examples as well as field cases obtained from low permeability gas reservoirs where no shut-in pressures are available. Results from the technique adequately compare with type-curve matching analysis. Furthermore, we demonstrate the problem can be transformed into an equivalent-liquid system and being analyzed with standard PTA techniques using the constant rate liquid solution.
In absence of shut-in pressure information, the PSS analysis offers an attractive alternative to the conventional material balance method. Besides, the method only requires minimal phase behavior data in the case of gases rendering its application practical and convenient. Also, we describe how to transform the constant pressure problem into a constant rate one in order to apply standard PTA techniques. Additionally, this work demonstrates the importance of having automated wells with permanent gauges by enhancing the value of the information provided by them in the framework of an adequate and judicious reservoir management.
Altemeemi, Bashayer (KOC) | Gonzalez, Fabio (BP) | Al-Nasheet, Anwar (KOC) | Gonzalez, Doris (BP) | Al-Shammari, Asrar (KOC) | Sinha, Satyendra (KOC) | Muhammad, Yaser (Schlumberger) | Datta, Kalyan (KOC) | Al-Mahmeed, Fatma (KOC)
Sound development plans are based on complex 3-D 3-Phase multimillion grid reservoir simulation models. These models are used to run different scenarios where probability distributions are included to understand the impact of uncertainties and mitigate main risks that could raise during the life of the field. With today's available dominant supercomputers, reservoir engineers have the tendency to undervalue the power of classical reservoir engineering. However, in a fully connected reservoir tank that honors the basis of the material balance equation, material balance technique has been long recognized as a powerful tool for interpreting and predicting reservoir performance by estimating initial hydrocarbon in place and ultimate hydrocarbon recovery under various depletion scenarios. In brief, under the right conditions, material balance technique is a suitable tool for field development planning. The power of material balance to predict long term performance is undisputable, especially in the case of a prevailing uncertainty. This is the case of the Magwa-Marrat field, where the development plan has historically been driven by the potential risk of asphaltene deposition in the reservoir.
The objective of this paper is to show a step by step process to integrate data to build a reliable model using material balance and how this model is utilized to progress a field development plan capable of managing uncertainty and provide the tools to mitigate risk.
Pressure data is obtained from repeat formation tester (RFT), static data from shut-in pressures and reservoir superposition pressures from pressure transient analysis. The average reservoir fluids properties are retrieved from a compositional equation of state based on circa 20 PVT studies.
The material balance model was successfully completed, and the resulting stock tank oil initially in place (STOOP) was compared to volumetric calculations. Solution gas, rock compaction and aquifer influx were determined as drive mechanisms. The Campbell Plot, diagnostic tool, was proven to be prevailing defining early energy to determine STOOIP and the aquifer properties were calculated by matching the distal energy
The material balance model was then used to run different development strategies. This methodology captured the impact of depleting the reservoir down to Asphaltene Onset Pressure (AOP) as well as below AOP. The model was also used to define the requirements for water injection rates and startup of a water flooding project for pressure support. Additionally, the material balance work was implemented to support reservoir management and to maximize recovery factor.
This paper presents an innovative approach of integrating asphaltene behavior from laboratory tests and fluid studies, combined with material balance to screen development scenarios for an efficient depletion plan including water injection to manage asphaltene risks and optimize ultimate recovery. Finally, a fully ground-breaking strategy, not reported earlier to the knowledge of the authors, has been established to manage the perceived main risk in the Magwa-Marrat reservoir.
Al-Obaidli, Asmaa (KOC) | Al-Nasheet, Anwar (KOC) | Snasiri, Fatemah (KOC) | Al-Shammari, Obaid (KOC) | Al-Shammari, Asrar (KOC) | Sinha, Satyendra (KOC) | Amjad, Yaser Muhammad (Schlumberger) | Gonzalez, Doris (BP) | Gonzalez, Fabio (BP)
The Magwa-Marrat field started production early 1984 with an initial reservoir pressure of 9,600 psia Thirtysix (36) producer wells have been drilled until now. By 1999, when the field had accumulated 92 MMSTB of produced oil and the reservoir pressure had declined to 8000 psia, the field was shut-in until late 2003 due to concerns on asphaltene deposition in the reservoir that could cause irreversible damage and total recovery losses. The field was restarted in 2003 an it has been in production since then. By April 2018 the field had produced 220 MMSTBO, with the average reservoir pressure declined to 6,400 psia. As crude oil has been produced and the energy of the reservoir has depleted, the equilibrium of its fluid components has been disturbed and asphaltenes have precipitated out of the liquid phase and deposited in the production tubing. There is a concern that the reservoir will encounter asphaltene problems as the reservoir pressure drops further. The objective of this manuscript is to present the process to understand the reservoir fluids behavior as it relates to asphaltenes issues and develop a work frame to recognize and mitigate the risk of plugging the reservoir rock due to asphaltenes deposition with the end purpose of maximizing recovery while producing at the maximum field potential Data acquired during more than 30 years have been integrated and analyzed including 22 AOP measurements using gravimetric and solid detection system techniques, 17 PVT lab reports, 1 core-flooding study and 1 permeability/wettability study. Despite the wide range of AOP measured in different labs, it was possible to determine that the AOP for the Magwa-Marrat fluid is 5,600 500 psia and the saturation pressure is 3,200 200 psia. Results of this fluids review study indicates that it might be possible to deplete the reservoir pressure below the AOP while producing at high rates.
One of the North Kuwait Carbonate fields which starts its production in 1957 has very low recovery factor after 60 years of production although the field was under water flooding since 1997. A workflow was developed to first understand the reason behind the low recovery and second to propose the best way to improve it.
The workflow starts with first building a material balance model to understand the main reservoir driving mechanisms. Second, a fine-scale history matched simulation model was used to understand the main reasons of the current low recovery. A Produce High and Inject Low (PHIL) concept was proposed with locating all the injectors at the deepest zone and the producers at the shallow zones. Finally, the proposed PHIL concept with inverted 5-spot horizontal wells was examined compared to the inverted 9-spot vertical wells and to the peripheral PHIL concept using the simulation model to examine the best approach to maximize the recovery.
Different outcomes from the above-mentioned workflow can be summarized as follows; first, it was found that the main driving mechanism is water injection which represents 70% of the reservoir recovery factor. Hence the importance of creating an artificial aquifer along the whole area of the field to provide the required pressure support which calls for the implementation of the PHIL concept with inverted 5-spot pattern background as the best development concept for the field. Second, the thorough data review used on building the fine-scale model shows that the current recovery is dominated by single zone which represents only 15 % of the in-place and on top of this, it was found that all the developed wells are located only on 30% of the field leaving 70% of the field undeveloped. These are the main reasons behind the low recovery. Finally, the developed PHIL concept with inverted 5-spot background shows that the recovery can be increased by five times with less number of new wells and less water injection volume required compared to the 9-spot vertical wells and the peripheral PHIL concepts. This five-folds increase in recovery encourages the asset to do a pilot to implement the proposed development strategy.
Unlike the commonly used inverted 5-spot vertical wells, this work proposes a novel approach of inverted 5-spot horizontal wells with directing the horizontal injectors at the deepest zones and the horizontal producers at the shallow zones. Hence creating an artificial bottom aquifer with minimizing the water production and maximizing the water injection distribution along the whole area of the reservoir.
Sanyal, Arunava (KOC) | Kumar, Sanjeev (KOC) | Al Awadh, Ahmed (KOC) | Al Samhan, Sarah (KOC) | Al Azmi, Jassim (KOC) | Sikdar, Koushik (Schlumberger) | Sultan, Gamal Ali (Schlumberger) | Das, Sourav (Schlumberger)
Zubair Formation is one of the key producers in North Kuwait; however, the reservoir complexity and hydrocarbon movement along with pressure depletion always poses challenge for determining the perforation and completion strategy to optimize the production. Zubair Formation is broadly divided into three parts e.g. upper, middle and lower, the upper and lower units are of utmost importance for the current study. An integrated approach was adopted utilizing the high-resolution borehole image outputs; which has not only helped in identifying thin bedded reservoir zones but also facilitated the understanding of detail reservoir geology and sand dispersion.
Integrated formation evaluation and workover design is always the key to sustain the production and it becomes even more important when the reservoir is highly heterogeneous in nature and coupled with declining pressure trend. Therefore, an innovative methodology was necessary to address the uncertainties. High resolution borehole images were utilized to determine the sand count, which can detect even the thinnest reservoir layer in the formation. Heterogeneity analysis was also performed to understand the relative sorting of the different reservoir units; sorting has a direct relation with reservoir permeability and thus reservoir productivity. High resolution sedimentary analysis was performed to understand the detailed sedimentology using the borehole image derived dip data; cross bedding types were identified which provides fair idea about depositional energy condition along with depositional environment. All the high-resolution inputs were integrated with openhole logs and volumetric results, which led to a clear deterministic picture of the reservoir, based on which crucial decision was taken.
This integrated approach was adopted in three deviated well sections in Zubair formation, which has facilitated in improving the well performance. Detail sedimentary analysis and cross bedding typing in multiwell helps in fine tuning the sand dispersion in the reservoir model; which in turn found to be helpful for deciding future well locations.
The shortage and high cost of CO2 and/or Hydrocarbon gases, in some areas, makes chemical EOR a practical option for tertiary oil recovery. Alkaline, Surfactant and Polymer, ASP, formulations continue to evolve to withstand challenges in relation to reservoir heterogeneity, complex mineralogy, high temperature and high formation water salinity of carbonate reservoirs. Such advanced ASP formulations have been considered to evaluate the performance of tertiary oil recovery process in a Kuwaiti carbonate reservoir. Successful performance has been seen in the lab through the evaluation of ASP coreflood experiments using composite carbonate cores. This paper presents the results of these coreflooding experiments and the steps followed to build representative ASP flooding simulation models as well as the workflow to calibrate these models to the observed experimental data. Moreover, the paper highlights the challenges associated with ASP coreflooding process and its modeling in the difficult environment of carbonate reservoirs. The paper also presents the techniques followed to overcome some of these challenges.
The modeling of two corefloods are presented in this paper, the first is for high-pressure live oil ASP coreflood, and the second is for low-pressure, surrogate oil ASP coreflood. The carbonate composite cores were first flooded with seawater down to residual oil saturation, Sorw. The ASP coreflood started with pre-flushing phase using softened seawater, followed by an ASP slug, and ended by injecting a number of pore volumes of polymer solution for mobility control. The representative ASP flooding simulation models of this paper captured the vital mechanisms involved in the ASP chemical EOR process, such as:
Micro-emulsion phase behavior, surfactant solubility ratios and resulting IFT changes Saponification process by the reaction of naphthenic acids of the crude oil with the injected alkali Adsorption of surfactant and polymer on the carbonate rocks as a function of pH and time The geochemistry of aqueous and oleic phase reactions Updating reservoir capillary number, resulting from the changes to IFT and wettability Effect of changes in capillary number is reflected by different sets of interpolation Kr curves The rheological behavior of polymer solutions Optimum salinity and salinity gradient effect.
Micro-emulsion phase behavior, surfactant solubility ratios and resulting IFT changes
Saponification process by the reaction of naphthenic acids of the crude oil with the injected alkali
Adsorption of surfactant and polymer on the carbonate rocks as a function of pH and time
The geochemistry of aqueous and oleic phase reactions
Updating reservoir capillary number, resulting from the changes to IFT and wettability
Effect of changes in capillary number is reflected by different sets of interpolation Kr curves
The rheological behavior of polymer solutions
Optimum salinity and salinity gradient effect.
Assisted history matching software was employed in the calibration of the two corefloods following a stepwise procedure by first matching the water flood results, then matching the surfactant production values, and finally matching the remainder of the ASP flood results. This paper discusses the parameters that needed to be tuned in order to attain a match of both waterflood and ASP flood results. The matched results included the oil recovery, flow pressure differential, and the concentration of chemical effluents traced during the experiments. The profile of ASP oil recovery in these carbonate composite cores is more gradual, and is different from those observed in sandstone corefloods.
Ortegon, Luis Rodrigo Diaz Teran (Schlumberger) | Al-Shammari, Nouf Abdullah (KOC) | Al-Qattan, Abrar (KOC) | Al-Samhan, Amina (KOC) | Al-Enizi, Nawaf Khalaf (KOC) | Duvivier, Giles (BP) | Brown, Richard C. (BP) | Perez, Godofredo (BP) | Ritchie, Sarah (BP)
Fluid characterization and mapping in the Greater Burgan field was performed using an extensive database of PVT analysis reports. This enabled an enhanced understanding of the distribution of fluids, in which a lateral compositional gradient was discovered.
Summary information of 381 samples that had been acquired from 1938 to 2017 was gathered and plotted against both true vertical depth subsea (TVDSS) and spatial directions (northing and easting). Correlations of average fluid characteristics of all these samples against TVDSS and the direction in 15° north-east was determined. The API gravity, oil viscosity and solution gas to oil ratio were parameterised as a function of horizontal direction and depth. The saturation pressure was modelled using a correlation, being a function of the other variables modelled in space. Over 30 correlations published in the literature were ranked, and the best-fitting correlation was selected to predict the saturation pressure distribution spatially. All these properties were mapped by formation.
This is the first study in which the existence of this lateral gradient is fully described. This work is being used as a reservoir management tool to predict zones close to or below the saturation pressure and reduce the production offtake from those zones and to develop an appropriate sampling plan. This work has also help manage access to the heavy oil zones.
Flow zonation and permeability estimation is a common problem in reservoir characterization; usually, integration of openhole log data with conventional and special core analysis solves the latter. We present a Bayesian based method for identifying hydraulic flow units in uncored wells using the theory of Hydraulic Flow Units (HFU) and subsequently compute permeability using wireline log data.
First, we use the F-test and the Akaike's criteria coupled with a nonlinear optimization scheme based on the probability plot to determine the optimal number of HFU present in the core dataset with the regression match giving the pertinent statistical parameters of each flow unit. Second, we cluster core data into its respective HFU by using the Bayes' rule. Finally, we apply an inversion algorithm based on Bayesian inference to predict permeability using only wireline data.
We illustrate the application of the procedure with a carbonate reservoir having extensive core data. The results showed the Bayesian-based clustering and inversion technique delivered permeability estimates in agreement with core data as well as with results obtained from pressure transient analysis.
Among the applications of the workflow presented are better productivity index assessments, enhanced petrophysical evaluations, and improved reservoir simulation models. Coupling of Nonlinear optimization with Bayesian inference proves a robust way for performing data clustering providing unbiased estimations
This paper describes seismic attributes approaches and rock typing in channelized reservoirs of North Kuwait. Seismic facies with a range of frequencies from 10 - 50 Hz, along with four other attributes: semblance, RMS frequency, instantaneous phase, relative acoustic impedance and RMS amplitude of shallow upper Cretaceous channelized system are calculated and the channel infill and overbank deposits are represented in maps. Five classes are used and found to be sufficient in the unsupervised classification method. The seismic facies classification was matched against the above-mentioned four attributes and found to correspond to them. The major channel components are illustrated with the
Steam injections in shallow heavy oil targets come with a risk of breaching the thin cap shale sealing layer and not fully understanding the thickness and continuity of shale barriers within the reservoirs. This paper presents an investigation to mitigate those risks through different seismic attributes and rock typing for shallow Tertiary and Cretaceous reservoirs in North Kuwait which extended to a deeper heavy oil targets in neighboring fields. This work that studies all the key risk elements in such heavy oil reservoirs mitigates the drilling and steam injection risks of heavy oil field development. Those who utilize seismic data to map heterogeneities must realize that the changes we observe in our seismic events can be due to one of the following items: depositional environments, sweet spots, stratigraphic features, lithology, petrophysical properties such as porosity or fluid or clay content, geohazards, etc.
From a previous study presented this year at Geo 2018 for Tertiary targets, we concluded that Simultaneous geostatistical inversion (SGI) added much value in terms of delineating the shale barriers and enhancing resolution, in addition to estimating effective porosity for well releases.