Integrated Data Analysis: Key to Understanding the Reservoir

The primary objective of this paper is to introduce an innovative, integrated workflow designed to quantify the expected initial productivity of oil and gas wells. This is achieved through the meticulous analysis of openhole logs and wireline formation testing data. The workflow is particularly beneficial for reservoir engineers and key decision-makers, offering them critical dynamic insights that are instrumental in shaping the future development strategies for different zones within a reservoir, thereby facilitating effective wellbore and reservoir management practices. This workflow is a multi-disciplinary endeavour, incorporating data from geological, petrophysical, and reservoir engineering domains. The first pivotal step involves the creations of a discretized geological model, constructed using a plugin within an industry standard simulation software. Basic logs such as triple-combo and resistivity are employed after LQC to make the foundational structure of the static reservoir model; these logs assist in identification of significant lithological boundaries and distribution of reservoir properties across grid cells. Advanced datasets like Nuclear Magnetic Resonance (NMR), routine core analysis, formation pressure and mobility data from the wireline Formation Tester (FT) tool(s), etc. are utilized to capture finer-scale heterogeneities with superior resolution. A comprehensive depth-wise characterization of all zones is then formulated, to ascertain the representative permeabilities for each zone. Fluid models from PVT data of the field are assigned to different zones according to expected fluid types from FT fluids data and offset well data analysis. Once all data is imported into the simulation software, a Neural Network analysis is employed within the platform to segment the formations into distinct hydraulic flow units, using calibrated permeability, effective porosity, and saturation data. To further refine the model, Special Core Analysis (SCAL) data is integrated and assigned to the identified flow units, ensuring an accurate representation of reservoir behaviour. The implementation of this fit-for-purpose workflow results in the construction of a robust 3D simulation model. The model is instrumental in making estimates concerning the initial zonal productivities. These simulation results were further corroborated through an independent multilayer network model developed using similar input parameters. This integrated suite of outputs not only offers invaluable insights that can aid choosing of optimal completion strategies, such as commingling or single selective completion, but also provides a platform for conducting sensitivity analyses. These analyses assess the potential impact of up to 10 uncertain production parameters, including water contacts and permeability anisotropy. In conclusion, this research serves as a benchmark in innovation: it successfully develops a tailored workflow capable of predicting the dynamic attributes of different zones in a reservoir using only a single well's static data. This is the first time such an integrated end-to-end workflow has been created, relying on a borehole centric approach that helps customers take critical decisions from the minimum amount of data that is available.

A brown field is an oil or gas asset that has matured to a production plateau or even progressed to a stage of declining production, where significant oil or gas reserves have been produced. Acquisition of brown fields all over the world was on the increase prior to the current volatility of crude oil price in the international market. On acquisition of brown fields, the new investors are always eager to revive the field, explore for more reserves and increase production to maximize profit by seeking to extend the economic producing life of the field using cost-effective, low risk, adaptable technologies. Brown fields' re-entry are usually associated with enormous challenges and opportunities that requires proper solutions to unlock such opportunities to maximize production, break even and maximize profit. This paper focuses on re-entry/re-opening challenges, opportunities, solutions and learnings of XYZ field in Niger delta. Areas to be discussed include; wells selection, estimated vs actual start-up rates, well reactivation, wellheads and facility refurbishment.

Identifying oil-saturated versus water-saturated sands in shallow, unconsolidated, viscous-oil-bearing terrigenous-clastic reservoirs of Kuwait field is challenging. Field appraisal was based upon seismic, core and wireline-log data from 19 wells. Static and dynamic models incorporating all subsurface data were built to estimate oil-in-place and forecast production. Estimating and modeling fluid saturations in reservoir zones was accomplished by integrating core, dielectric-resistivity, Nuclear Magnetic Resonance (NMR) and Wireline Formation Tester (WFT) data. Wells were drilled along a northwest/southeast-trend, thus geologic and reservoir-property variability in east and west parts of the field are uncertain. Stratigraphy and lithologic properties in these Miocene-age fluvial to shallow-marine strata impart a complex 3–D fluid distribution in the field. Repeated shoreline progradations and retrogradations deposited a stratigraphic succession defined by five facies-associations (i.e., shoal, tidal flat, tidal channel, lagoon, sabkha). Five lithofacies (i.e., shale, shaly sandstone, sandstone, carbonate-cemented sandstone, evaporite) were identified from core, elemental spectroscopy logs, and X-ray diffraction (XRD) data. Facies associations and lithofacies models were built using a combination of multiple-point statistics and sequential-indicator simulation. Lithofacies distribution in the static model was constrained by the facies-association distribution; reservoir-property distribution (e.g., porosity, permeability) was conditioned by lithofacies. Discrete reservoir zones were defined to separate oil-saturated versus water-saturated sands. The volume and position of oil-bearing sands are controlled by the defined zones and permeability distribution. The oil-filling process in these viscous oil-bearing reservoirs is typically controlled by the pore throat distribution with the migrating oil taking the path of least resistance. Due to the presence of stratigraphic-flow baffles, fluid contacts vary from sand-to-sand vertically and laterally. Log data, core descriptions, ultraviolet photographs, WFT and pressure volume temperature (PVT) data guided the interpretation of lowest known oil and highest known water levels, thus reducing fluid saturation uncertainty in the field.

Traditionally, the Gulf of Thailand (GoT) has been known for high temperature, small borehole size, variable CO2, and highly compartmentalized reservoirs. In particular, it is a very challenging environment for Wireline Formation Testers (WFT). Owing to cost constraints, since the start of exploration and development campaigns in this area, usage of newer technologies has been highly selective. However, this has been changed significantly in the past few years where the right WFT technology has been applied to the right environment. This paper is the first to present a work process to derive clean fluid sampling in the very challenging environment of the GoT. Time per station used to collect downhole fluid samples using WFT has been a major concern for a costly offshore operation. In addition, borehole stability is also another factor limiting WFT time. Given the time constraint, collected fluid samples usually have high drilling mud filtrate contaminations, in the ranges of 25 to 85wt%, and have not been suitable for further laboratory analysis and field development purpose. Balancing between the time used for fluid sampling and the quality of the collected fluid samples is not a simple task to manage. This paper shares a successful story of downhole fluid sampling in exploration wells for one of the operators in the GoT. Instead of using a conventional probe with one Downhole Fluid Analyzer (DFA), the concentric shaped probe with two synchronized pump-out modules and two DFAs are used in this case. However, this is not as simple as other fields already presented in the literature because an unconsolidated sand character introduced complications into this sampling technique. Several attempts have been developed to make focused sampling work in these challenging environments. At the end, "Less than 5% OBM contaminated samples in oil reservoirs are successfully achieved in a timely manner." The average time used per each sampling station is approximately 30 minutes. This paper will also discuss the advantages and disadvantages for each technique applied and the final results. In addition, more improvements have been suggested to enhance this focused sample technology in more complex fluids, such as gas condensate reservoirs to make sure that less contaminated fluid sample can be collected in the limited time per station.

Globally the challenge to meet the increasing energy demand is on with the application of game-changing technologies to maximize recovery from proven reserves in mature assets. Production studies have shown that some Niger delta fields have heterogeneous reservoirs with low to fair recovery factor derived in most cases by software simulations without adequate field acquired reservoir parameters before embarking on the secondary development plan for such reservoirs. Failures recorded in most secondary recovery strategies for heterogeneous reservoirs are accountable for lack of in-depth studies of the reservoir characteristics. There is a direct relationship between reservoir recovery factor and the petrophysics of the reservoir. A sand body in the field can exhibit variable petrophysical changes at different positions in the field. Therefore optimum secondary recovery plan for such reservoirs are designed with the combination of field acquired (not simulated) petrophysical data (porosity and permeability), the environment of deposition, special core analysis and formation evaluation studies. This study emanates from the recently developed project in a field in Niger delta.

Produced water often contains chemicals, oil and sometimes naturally occurring radioactive materials which could be harmful to the environment. Its management is aimed at reducing the toxicity of discharged volumes and mitigates the effects in the environment. This study evaluated the management method being carried out, the physio-chemical properties and trace metals constituents of produced water from a well in the Niger-Delta to determine the extent of compliance with standards of discharge. The results obtained showed that the pH values, Zinc, copper and cadmium were within acceptable regulatory limits for near-shore and inland discharge; the concentration of trace metals like iron, lead, chromium were not within acceptable limits for near-shore and inland discharge. The constituent of the produced water of most importance was the concentration of the dispersed oil in the water and the management method which is being used reduces the oil concentration to a minimum of 250mg/l before disposal which does not comply with the regulatory standard of 10mg/l for inland disposal. A better management method was proposed which reduces the oil concentration to 50-75mg/l based on the relationship between oil droplet size and oil concentration and the economics of operation. Also, due to the oil and water production rates, it was proposed that some of the produced water be used for waterflooding to obtain additional recoverable reserves thereby optimizing the water management approach.

Derived elastic attributes has been used to discriminate rock and fluid properties in EK Field using well logs data. These derived rock attributes were analyzed in cross-plot space for target reservoirs. The log analysis for delineated reservoir B20 shows an average volume of shale (7.5%), total porosity (33.9%) and water saturation (29.3%). Cross-plots of elastic rock attributes (Vp/Vs, Lambda-Rho (λρ), Mu-Rho (μρ), Poisson ratio and acoustic impedance) were used as fluid and lithology indicators and in reservoir characterization. The cross plots results shows distinct separation of hydrocarbon sand, brine sand and shale. Low Poisson’s ratio (0.2-0.26), Lambda-Rho (7 GPa*g/cc -10 GPa*g/cc), Vp/Vs (1.6-1.8), low acoustic impedance and high Mu-Rho values indicate hydrocarbon sands. The intermediate values of Poisson’s ratio (0.2-0.26), Lambda-Rho (17 GPa*g/cc - 21GPa*g/cc) , Vp/Vs ratio (2.05-2.3), relatively high acoustic impedance and Mu-rho indicated brine sand while high Poisson’s ratio (0.35-0.41), Lambda-rho (24 GPa*g/cc -27 GPa*g/cc), Vp/Vs ratio (2.3-2.5), high acoustic impedance and low Mu-Rho indicated shale. The cross-plot models all show similar result of hydrocarbon sand characterized by high porosity, low water saturation and volume of shale. The well based elastic attribute analyses established useful relationships between elastic derived seismic attributes and reservoir properties in delineating lithology and reservoir fluid for better understanding of reservoirs in the Niger Delta field.

Accurate knowledge of reservoir fluid properties, especially reservoir saturation pressures and reservoir type, is paramount for the estimation of reservoir volumetrics, well design and placement, well and reservoir performance management, field development planning and ultimately economic evaluation of the reservoir. However, most of the existing methods such as PVT fluid sampling, compositional grading, and empirical models have proven to be either ineffective or expensive and sometimes, leads to ungeneralizable results. This paper discusses the application of machine learning (ML) techniques to develop a robust model for prediction of reservoir fluid properties such as saturation pressures in a Niger Delta Field and the subsequent classification of the reservoirs as under-saturated or saturated. Reservoir data including the PVT data, compositions (C1-C7+), temperature & pressure data, fluids contacts of known reservoir type were considered initially in order to train a model using a multi-features regression ML algorithms for the determination of the saturation pressure and a Two-class boosted decision ML algorithms to determine the determine the type of reservoir (saturated or undersaturated). Of the 34 parameters considered, it was found that the reservoir fluid composition has significant impact in determining the accuracy of the Pb-hypothesis. The key performance indicators such as MAE, RMSE, and RSE are within 0.02-0.05 and Co-efficient of determination of about 95% for Pb determination. When compared with the Standing, Glaso, Petrosky-Farshad etc. correlation, the AAE was significantly less than both cases. AAE for the Standing and Glaso correlations were respectively18.5% and 25.1% while that of the ML model was 2.3% using both the data from training set and test set. For the classification algorithms in determining type of reservoir, the model performed within the 73-100% accuracy, precision & recall. The Area under the curve (AUC) of the Receiver Operator Characteristic (ROC) chart of approximately 97% indicated the robustness of the model. The results showed that the use of a properly trained and accurately validated ML model can deliver better predictions of reservoir fluid properties and subsequent reservoir type when compared to conventional methods.

Latin Hypercube Sampling (LHS), unlike other simulation methods such as Monte Carlo Simulation (MCS), has not enjoyed wide applications in determination of reserve despite its capability to realize same accuracy as with the latter even with much fewer simulation trials. The parameters used in these distributions are extracted using software ‘digitizer’ from different plot of values derived from measurements of reservoir and fluid properties. This work dwells on analysis of two depleted gas fields for which the ultimate recovery is known. The distribution of porosity, water saturation, reservoir thickness and reserve size are input to crystalbal software to generate cumulative distribution function (CDF) and Probability density function (PDF) of original gas in place (OGIP) for each of the fields. The accuracy of Latin Hypercube Sampling is analysed by comparing the actual ultimate recovery and the OGIP predicted by LHS. The calculated recovery factors of 68.7% and 93.3%, for water driven and depletion driven gas reservoirs respectively, compare favourably with those of known gas reservoirs. Therefore, LHS is a good estimator of gas reserve. It is recommended that similar study should be carried out on condensate and oil reservoir, particularly on Niger Delta fields.

This article, written by JPT Technology Editor Chris Carpenter, contains highlights of paper SPE 181526, “New Wireline, In-Situ, Downhole Fluid Compositional Analyses To Enhance Reservoir Characterization and Management,” by Gabor Hursan and S. Mark Ma, Saudi Aramco, and Wael Soleiman, Sami Eyuboglu, Neeraj Sethi, and Nacer Guergueb, Halliburton, prepared for the 2016 SPE Annual Technical Conference and Exhibition, Dubai, 26–28 September. The paper has not been peer reviewed. This study focuses on recent experience in Saudi Arabia with crude-oil compositional analyses during pumpout with a wireline formation tester (WFT). It summarizes experience with the in-situ measurement of methane, ethane, propane, saturates, aromatics, and gas/oil ratio (GOR) on the basis of multivariate optical computing (MOC) conducted at more than 200 pumpout stations in a total of 37 wells drilled with a variety of inclinations, bit sizes, and drilling fluids in several oil and gas fields. Introduction In reservoir-fluid characterization performed in the laboratory conventionally, samples of representative formation fluids are analyzed to determine bulk fluid properties, fluid-phase behavior, and chemical properties. Exploration and evaluation wells are often drilled exclusively for fluid-analysis purposes for which the only way to analyze or capture formation fluids is a downhole pumpout WFT (PWFT). Capturing high-quality reservoir samples is one of the most important objectives in any PWFT job. The keys to ensure fluid cleanup during pumpout are (1) a set of downhole sensors measuring the pumped fluids and (2) the unambiguous contrast between the drilling-mud filtrate and reservoir fluids as measured by at least one of these sensors. Avoiding such fluid ambiguity is a challenge in many situations. Optical sensors respond with high sensitivity to chemical compositions of fluids. Implemented downhole, these tools can provide a powerful means of differentiating between the oil-based-mud (OBM) filtrate and reservoir oils. The most valuable benefit of these measurements is the prompt availability of in-situ fluid-composition data without the additional steps of acquiring, transporting, and analyzing physical fluid samples in the laboratory. This process requires durable, robust, and accurate optical sensors that operate reliably and consistently in the hostile downhole environment. One recently developed optical-sensor system is based on an optical device known as an integrated computational element (ICE) that performs the mathematical operation of MOC. For each fluid component, an ICE “core” is engineered such that only one particular fluid component or property is accentuated in the detector response and everything else is muted. This detector response is then used to infer the abundance of the fluid component of interest. These optical elements are typically very broadband and may have a response that extends from 400 to 5000 nm. The high bandwidth of these optical elements combined with their intrinsic high signal/noise ratio enables laboratory-grade optical analysis downhole.

An integrated analysis of fluid property data of deep tight gas-condensate stacked reservoirs was coupled with regional petroleum system analysis across North-Central Oman to investigate variations observed in PVT data. The study, focused on the lower Haima reservoirs of the Haima Supergroup, aims to identify the broad links between the petroleum system and the PVT properties in order to delineate areas of varying properties within the basin and develop a common Equation of State (EoS) fluid model for modelling purposes. The approach is based on investigating fluid property variation observed from PVT lab reports across multiple fields and reservoirs. Data trends were plotted against a range of key subsurface parameters for multiple fluid samples. A comparison of the fluid property data from PVT with the maturity and temperature history of the basin was carried out to understand property variation and link it with the regional geology. The process links PVT trends to Geochemistry by investigating HC maturity and mixing of Hydrocarbon sources. It also dives into the reservoirs’ temperature and maturity history within the basin. The study focused on ensuring that the developed approach can help in understanding fluid property variation within the subsurface by integrating PVT Lab data with the charge history. While it was anticipated that reservoir stratigraphy and depth would have a direct relationship with fluid density and composition, the findings revealed that the complexity of charge history, hydrocarbon mixing, and temperature history significantly influence the observed variability. However, coming to this understanding can aid in identifying areas with potential for development, such as delineating uncertainty range of the fluid properties & CGR across the basins. Despite these insights, the common EoS model was found to be reasonably effective across the area of interest. This workflow has provided good insights into the fluid history of the basin by linking PVT data trends and variation with observations from charge history understanding. It has also assisted in understanding the limitations of PVT modelling across complex reservoir fluid systems.

This study uses well logs and seismic data sets to map out and evaluate the structural distribution of siliciclastic reservoirs in the field for identification of possible hydrocarbon bearing zones, continuity, and for the creation of appropriate geological models. This is aimed at using the analysis therefore to mitigate the challenges of reservoir facies witnessed during the development and drilling of hydrocarbons from reservoirs found in the field. In order to delineate the hydrocarbon reservoir facies using the structural distribution, the integration of seismic and petrophysics methods were adopted. The result of the petrophysical analysis shows that the identified reservoirs are composed of the intercalation of sand and shale lithofacies. The estimated porosity and permeability values of more than 12% and1000mD respectively indicates that the reservoir are said to be siliciclastic. From the seismic, three different faults-which corresponds to the three correlated reservoirs were mapped. The structural pattern of the investigation indicates that the hydrocarbons are being trapped by growing faults, which a common trapping mechanism is observed in the Niger Delta field. The vertical orientation, position, and connectivity of the reservoirs are shown by the mapped faults. Hydrocarbon saturation increased in the northeastern direction of covered seismic and this was substantiated by the findings of the time surface contour maps. This is as a result of the increase in the hydrocarbon saturation towards the Northeast direction. Therefore, the prospective hydrocarbon region in this study is enclosed by the fault; which is a rollover anticline. This fault growth aligns with the potential hydrocarbon accumulation area.

3D reservoir modelling of stacked reservoirs is often difficult, due to interplay of various uncertainties related to heterogeneity of the sand units modelled together within the stack. In the past, reservoir simulation had been used to model stacked reservoirs for commingled wells with varying predictive capacity. Typically, dynamic models use conventional approach in assessing uncertainties, involving discrete sensitivity of uncertain parameters and often lack robust subsurface uncertainty management. Additionally, handling dynamic cross flow issues by modelling of control valves in such cases presents production allocation challenges. Huge man-hours are needed for several run times, without achieving relatively good history-match and predictive capacity. As a result, only base-case history-matched model is developed; thus robust assessment of impact of sub-surface uncertainties on the predictions is usually inadequate. This paper details a multidisciplinary approach adopted in integrated dynamic modelling of stacked reservoirs in a Niger Delta field using experimental design methodology. The phased development was aimed at full-scale reservoir development of multi-zone and commingled wells. Instead of single average for each sand unit, facies-dependent kvkh values were required to adequately capture reservoir heterogeneity for the complex depositional environment. Field-wide history-matches for 23 reservoir blocks and over 70 individual conduit matches were done to calibrate the simulation models. Model robustness was conducted through blind-testing and calibration with carbon-oxygen and openhole log data from newly drilled wells. From the model predictions, recoveries for existing conduits from simulation models were benchmarked against Decline Curve Analyses and the results compared closely. Response Surface Models were used in selecting best history-matched realizations. 1P, 2P and 3P models were selected from the ultimate recovery probability distribution curves of these history-matched realizations. The modelling of interval control valves settings within the stacked models and use of smart well routines assisted evaluation of recoveries from planned commingled wells.

Horizontal wells and hydraulic fracturing are the key technologies that allow commercial production from tight oil and gas reservoirs. However, rigorous analysis of production data from these reservoirs requires incorporation of the impacts of stress-dependent permeability and multi-phase flow. Changes in the stress state of the system during production may reduce the absolute permeability. Furthermore, gas phase formation and flow in presence of supersaturated oil phase affects fluid dynamics in tight oil reservoirs. This study provides a rigorous methodology for incorporation of the effects of non-static permeability and multi-phase flow in rate transient analysis (RTA) of tight oil and gas reservoirs producing at variable rate/flowing pressures during transient linear flow period. Analytical solutions for the approximate linearized form of the flow equation have been widely used as the basis for RTA tools for conventional reservoirs during transient flow period. However, they lead to considerable error when applied to tight oil and gas reservoirs. In particular, during the transient linear flow period, the slope of the square -root-of-time plot obtained from numerical solution differs from the slope calculated by analytical methods. Efforts have been made by some researchers to obtain a correction factor from the numerical solution of the flow equation to correct the slope of the square-root-of-time plot for single phase flow of gas during transient linear flow period. In this study, an iterative method is used for evaluation of the slope correction factor in the presence of multi-phase flow and non-static permeability for constant-pressure production during transient linear flow period. Further, the correction factor is used for analysis of production data from tight oil and gas reservoirs producing at variable rate/flowing pressures. The correction factor is used in the analysis of different sets of synthetic production data for tight oil and gas reservoirs. The results show that the correction factor can reduce/eliminate the considerable errors associated with the conventional analytical methods in initial permeability estimation. For multi-phase flow cases, the producing gas -oil ratio (GOR) is used to estimate the oil saturation-pressure relationship in the reservoir, which is required for calculation pseudo-pressure and the correction factor. The method developed in this study alleviates the need for using numerical simulation models to generate empirical correlations for the correction factor for the square -root-of-time plot. The easy-to-implement iterative procedure of this method only requires the pressure dependencies of the constituent elements of the hydraulic diffusivity. Therefore, this method is applicable for analysis of production profiles for variety of reservoirs with nonlinear flow equations.

This paper discusses the application of selective slug sampling using the wireline formation tester – RCX* (Reservoir Characterization eXplorer), in a clastic heavy oil reservoir and its significance to fluid characterization. The low mobility ratio between the reservoir fluid and water-based mud filtrate results in a long cleanup time to obtain samples with an acceptable level of contamination. Tool sticking and wellbore stability issues add to the challenges in open-hole fluid characterization and testing. Increasing the flow rate to reduce cleanup time does not help mostly due to formation of an emulsion between the viscous oil and the mud filtrate. The emulsion is hard to disperse, leading to a contaminated sample. A higher flow rate results in sanding and loss of seal. In selective slug sampling, phase segregation is achieved by lowering and optimizing the flow rate, reducing the pressure drawdown that curtails emulsion formation and movement of grains thus increasing the sealing efficiency. The segregated oil slugs can be selectively collected in sample tanks multiple times downhole. Even a water cut of 60 to 70% is sufficient to collect high-quality samples, which is a significant improvement in sampling time compared to conventional fluid sampling, where 10% or less water cut is required. The technique was applied in an unconsolidated clastic heavy oil reservoir in Vietnam. Four 14- to 22-°API single-phase oil samples were collected in two hours each. Laboratory results indicate less than 10% contamination; water cut being 50 to 75% at the time of the sampling. It was the first time that successful heavy oil samples were collected from these layers. Fluid sampling is critical to identifying compositional variation of reservoir fluids along the depth, resulting in appropriate sampling strategy and eventually correct fluid characterization. Sampling in unconsolidated heavy oil reservoirs is improved using the selective slug sampling technique, making it possible to secure fast, yet high-quality samples, aiding fluid property characterization.