Dissertations / Theses on the topic 'Petroleum and reservoir engineering'

Ce travail propose des solutions innovantes en vue de l'exploitation des modèles d'ingénierie hétérogènes. Il prend pour exemple le domaine de la prospection pétrolière. Les stratégies de prospection sont élaborées à partir de représentations tridimensionnelles du sous-sol appelées modèles géologiques. Ceux-ci reposent sur un grand nombre de données hétérogènes générées au fur et à mesure de la conduite de l'exploration par des activités telles que la prospection séismique, les forages, l'interprétation des logs de puits. A fin d'optimisation, les utilisateurs finaux souhaitent, pouvoir retrouver et réutiliser à tout moment les données et les interprétations attachés aux différents modèles successivement générés. Les approches d' intégration des connaissances susceptibles d'être mises en œuvre pour résoudre ce défi, doivent être dissociées aussi bien des sources et des formats de données que des outils logiciels en constante évolution. Pour cela, nous proposons d'utiliser l'annotation sémantique, technique courante du Web sémantique permettant d'associer la connaissance à des ressources au moyen d' "étiquettes sémantiques". La sémantique ainsi explicitée est définie par un certain nombre d' ontologies de domaine, qui, selon la définition classique, correspondent à autant "de spécifications formelles de la conceptualisation" des domaines considérés. En vue d'intégrer les modèles d'ingénierie considérés, nous proposons une architecture, qui permet de relier des concepts appartenant respectivement à des ontologies locales et à une ontologie globale. Les utilisateurs peuvent ainsi avoir une vision globale, intégrée et partagée de chacun des domaines impliqués dans chaîne de modélisation géologique. Un prototype a été développé qui concerne la première étape de la chaîne de modélisation (interprétation séismique). Les expérimentations réalisées prouvent que, grâce à l'approche proposée, les experts peuvent, en utilisant le vocabulaire de leur domaine d'expertise, formuler des questions et obtenir des réponses appropriées
This work intends to propose innovative solutions for the exploitation of heterogeneous models in engineering domains. It pays a special attention to a case study related to one specific engineering domain: petroleum exploration. Experts deal with many petroleum exploration issues by building and exploiting three-dimensional representations of underground (called earth models). These models rest on a large amount of heterogeneous data generated every day by several different exploration activities such as seismic surveys, well drilling, well log interpretation and many others. Considering this, end-users wish to be able to retrieve and re-use at any moment information related to data and interpretations in the various fields of expertise considered along the earth modeling chain. Integration approaches for engineering domains needs to be dissociated from data sources, formats and software tools that are constantly evolving. Our solution is based on semantic annotation, a current Web Semantic technique for adding knowledge to resources by means of semantic tags. The "semantics" attached by means of some annotation is defined by ontologies, corresponding to "formal specifications of some domain conceptualization". In order to complete engineering model exploitation, it is necessary to provide model integration. Correspondence between models in the ontology level is made possible thanks to semantic annotation. An architecture, which maps concepts from local ontologies to some global ontology, then ensures that users can have an integrated and shared global view of each specific domain involved in the engineering process. A prototype was implemented considering the seismic interpretation activity, which corresponds to the first step of the earth modeling workflow. The performed experiments show that, thanks to our solution, experts can formulate queries and retrieve relevant answers using their knowledge-level vocabulary

The objective of this study is threefold: 1) Build a dual-porosity, geological reservoir model of Niobrara formation in the Wishbone Section of the DJ Basin. 2) Use the geologic static model to construct a compositional model to assess performance of Well 1N in the Wishbone Section. 3) Compare the modeling results of this study with the result from an eleven-well modeling study (Ning, 2017) of the same formation which included the same well. The geologic model is based on discrete fracture network (DFN) model (Grechishnikova 2017) from an outcrop study of Niobrara formation.

This study is part of a broader program sponsored by Anadarko and conducted by the Reservoir Characterization Project (RCP) at Colorado School of Mines. The study area is the Wishbone Section (one square mile area), which has eleven horizontal producing wells with initial production dating back to September 2013. The project also includes a nine-component time-lapse seismic. The Wishbone section is a low-permeability faulted reservoir containing liquid-rich light hydrocarbons in the Niobrara chalk and Codell sandstone.

The geologic framework was built by Grechishnikova (2017) using seismic, microseismic, petrophysical suite, core and outcrop. I used Grechishnikova’s geologic framework and available petrophysical and core data to construct a 3D reservoir model. The 3D geologic model was used in the hydraulic fracture modeling software, GOHFER, to create a hydraulic fracture interpretation for the reservoir simulator and compared to the interpretation built by Alfataierge (2017). The reservoir numerical simulator incorporated PVT from a well within the section to create the compositional dual-porosity model in CMG with seven lumped components instead of the thirty-two individual components. History matching was completed for the numerical simulation, and rate transient analysis between field and actual production are compared; the results were similar. The history matching parameters are further compared to the input parameters, and Ning’s (2017) history matching parameters.

The study evaluated how fracture porosity and rock compaction impacts production. The fracture porosity is a major contributor to well production and the gas oil ratio. The fracture porosity is a major sink for gathering the matrix flow contribution. The compaction numerical simulations show oil production increases with compaction because of the increased compaction drive. As rock compaction increases, permeability and porosity decreases. How the numerical model software, CMG, builds the hydraulic fracture, artificially increases the original oil-in-place and decreases the recovery factor. Furthermore, grid structure impacts run-time and accuracy to the model. Finally, outcrop adds value to the subsurface model with careful qualitative sedimentology and structural extrapolations to the subsurface by providing understanding between the wellbore and seismic data scales.

The Spraberry Formation is a Leonardian age submarine fan deposit restricted to the Midland Basin. The formation consists of very fine-grained sandstone, medium to coarse grain size siltstones, organic shales and carbonate mudstones. These rocks show variability in sedimentary structures and bedding types varied from thinly laminated to convolute laminations. Bioturbations were present in some samples and soft sediment deformation, such as water escape features, sediment loading and flame structures.

The Spraberry Formation is a naturally fractured reservoir with low porosity and low matrix permeability. Porosity measured varied from 2% in rocks with poor reservoir quality such as the argillaceous siltstone and mudstone while good reservoir rocks had an average porosity of 9%. Seven lithofacies were identified based on sedimentary structures, grain size and rock fabrics. Petrographic analysis showed four porosity types: (1) intragraular porosity; (2) dissolution porosity; (3) fracture porosity and (4) intergranular porosity. Fractured porosity was only observed in the argillaceous siltstone lithofacies.

The prominent diagenetic influences on the Spraberry Formation are: quartz cementation, quartz overgrowth, illtization of smectite, feldspar dissolution, clay precipitation, carbonate cementation, formation of framboidal pyrite and fracture formation. These diagenetic features were observed using scanning electron microscope (SEM) and in thin sections. Generally, petrophysical properties, such as porosity and permeability, vary gradually from reservoir rocks to non-reservoir rock. Observed trends where: 1) increasing organic and argillaceous content with decreasing porosity and 2) increasing carbonate sediments and calcite cements with decreasing porosity. Mineralogical analysis from FTIR showed an abundance of quartz and calcite, while illite is the prominent clay mineral observed in all samples.

This thesis describes the advantages of investigating gas shales reservoir description on a nanoscale by using petrographic analysis and core plug petrophysics to characterize the Barnett, Marcellus and Mancos shale plays. The results from this analysis now indicate their effects on the reservoir quality. Helium porosity measurements at confining pressure were carried out on core plugs from this shale plays. SEM (Scanning Electron Microscopy) imaging was done on freshly fractured gold-coated surfaces to indicate pore structure and grain sizes. Electron Dispersive X-ray Spectroscopy was done on freshly fractured carbon-coated surfaces to tell the mineralogy. Extra-thin sections were made to view pore spaces, natural fractures and grain distribution.

The results of this study show that confining pressure helium porosity values to be 9.6%, 5.3% and 1.7% in decreasing order for the samples from the Barnett, Mancos and Marcellus shale respectively. EDS X-ray spectroscopy indicates that the Barnett and Mancos have a high concentration of quartz (silica-content); while the Mancos and Marcellus contain calcite. Thin section analysis reveals obvious fractures in the Barnett, while Mancos and Marcellus have micro-fractures.

Based on porosity, petrographic analysis and mineralogy measurements on the all the samples, the Barnett shale seem to exhibit the best reservoir quality.

This Master Thesis work aims to find a model system combining the positive effects of surfactant and polymer flooding to enhance oil recovery. This report presents the results of 12 core floors performed to enhance recovery of waterflood residual oil. The recovery is enhanced by a visous surfactant flood consistent of one polymer to increase the viscosity, one surfactant for interfacial tension reduction, and one di-alcohol to function as co-surfactant and for salinity control.

The chemical treatment that gave the best result, gave an additional oil production normalized on OOIP of 20%, improving the oil recovery from 45 to 66% mostly by the means of mobility control. Pure viscosity floods gave an additional recovery of 12 to 13% of OOIP.

Novel technology is used to investigate environmental friendly enhanced oil recovery. A biopolymer made out of microfibrils from wooden material was for the first time ever to my knowledge, attempted used in a core flood to enhance oil recovery.

A viscous surfactant tertiary recovery process may help improve oil recoveries from many marginal oil fields or those that face shut-down due to uneconomic operating costs, but still contain significant amounts of oil.

The Pine Prairie Field is situated on a salt dome in northern Evangeline Parish, located in south-central Louisiana. Pine Prairie contains the only known Cook Mountain Formation hydrocarbon reservoir in Louisiana. Operators have targeted and produced hydrocarbons from the Cook Mountain reservoir in eight wells at the Pine Prairie Field. The source and origin of the Cook Mountain’s reservoir properties are unknown. The objective of this study is to determine the origin of the Cook Mountain Formation’s reservoir properties by identifying the processes associated with the formation of a Cook Mountain Reservoir. There are two carbonate outcrops at the surface expression of the Pine Prairie Dome. Samples were taken and thin sections made to determine the relationship, if any, to the Cook Mountain Formation. Thin section analysis of the carbonate outcrop was used to gain a better understanding of the depositional setting present at Pine Prairie Field. Well log, seismic, and production data were integrated to determine that, in all instances, commercial Cook Mountain production is associated with fault zones. The passage of acidic, diagenetic fluids through Cook Mountain fault zones generated areas of vuggy porosity proximal to Cook Mountain faulting. Further, fluctuations in short-term pressure gradients associated with salt diapirism resulted in the vertical migration of hydrocarbons via fault zones. In the Pine Prairie Field, fault seal breakdown occurs in Sparta and Wilcox Reservoirs, subsequently charging the Cook Mountain fault zone. Early hydrocarbon charge from the underlying Wilcox and Sparta Reservoirs prevented additional diagenesis, preserving secondary porosity in areas of Cook Mountain faulting.

Water coning is a complex phenomenon that depends on a large number of variables which include among others: production rate, perforation interval, mobility ratio, capillary pressure, etc. Its production can greatly affect the productivity of a well and the reservoir at large. In fractured reservoirs, the phenomenon is more complex owing to the high permeability of the fractures in the porous media. With this complexity in mind, water coning behaviour in fractured reservoir was studied by simulating a reservoir supported by a strong aquifer using ECLIPSE-100 Black-Oil Simulator. The water cut (WCT), oil production rate (OPR) and water saturation (BWSAT) at the producing interval (Block 1, 1, 7) were used to evaluate the coning phenomenon in a fractured reservoir. In the course of the study, sensitivity analyses on the modelled reservoir’s anisotropy ratio (kv/kh), production rate (q), storativity capacity (ω), fracture width (b) and fracture permeability (kf) were conducted to evaluate their effect on coning behaviour in fractured reservoir. The results obtained depict that while the anisotropy ratio is very significant in water cut and water saturation at the perforating interval it has no adverse effect on oil production rate. It was however, observed that the water cut and oil production rate decreased as the production rate (q) increased. Furthermore, the water cut, oil production rate and water saturation (BWSAT) from the fractured reservoir is sensitive to the storativity capacity (ω) depending on the fracture porosity (φf). Conversely, the fracture’s width (b) and permeability (kf) have no significant effect on the coning behaviour of the modelled fracture reservoir. However, anisotropy ratio (kv/kh), production rate as well as storativity capacity (ω) are significant parameters in evaluating coning phenomenon in fractured reservoirs.

Gas coning is the tendency of the gas to drive oil downward in an inverse cone due to the downward movement of gas into the perforations of a producing well thereby reducing oil production and the overall recovery efficiency of the oil reservoir. This work addresses gas coning issues in a naturally fractured reservoir via a numerical simulation approach on a single-well radial cross-section using the ECLIPSE 100 reservoir simulator. Matrix and fracture properties are modelled. Critical rate, breakthrough time and GOR after breakthrough is determined which is used to investigate the effect of matrix and fracture properties on gas coning effective reservoir parameters such as oil flow rate, matrix and fracture porosity, vertical and horizontal matrix and fracture permeability, matrix block size, etc. Results show that reservoir parameters that affect coning include oil flow rate, matrix and fracture porosity, matrix and vertical permeability, anisotropy ratio, perforated interval thickness, density difference and mobility ratio. While matrix block size and fracture spacing have no significant effect on gas coning.

Several types of transient well testing in Norne field are presented in this thesis. One production well from each segment in Norne field was participated in different type of test. The well test data of all cases were generated from reservoir simulation. It allows flexibility in modifying reservoir model condition to understand different behavior of pressure response. The tests were first started by producing the well at a constant rate for 10 days, and then shutting-in the well for at least 24 hours. The importance of reservoir model grid refinement, determination of reservoir communication across the fault, and the complexity of horizontal well test analysis are the three main discussions in this thesis work.Series of buildup tests at well D-1H in C-Segment were performed to recognize the significance level of Local Grid Refinement (LGR) near the wellbore. There are two sensitivities performed in the reservoir model, extension of LGR area and increase of LGR factor. Based on pressure responses, wider area of LGR affected permeability estimation, while increase of LGR factor impacted the storage capacity calculation. In the next discussions, LGR near the wellbore becomes a standard procedure in generating well test data.The next type of transient well testing performed in Norne field is interference test. This test was executed at well E-3H as an observation well in E-Segment; while well E-1H and E-2H acted as interfering wells in D- and E-Segments respectively. According to pressure and production trends, it can be ensured both interfering wells are located in different segments. A reservoir communication across segments was identified through pressure drop analysis at well E-3H; hence presence of a major fault between segments is not fully sealed.Transient well testing in horizontal well gives a special and more complex analysis compare to vertical well analysis. A buildup test was examined at horizontal well E-4AH in G-Segment to determine vertical and horizontal permeability. Two flow regimes existed during the test, early-time radial flow and intermediate-time linear flow. They were discovered from pressure versus time plot and pressure derivative analysis. Interpretation results from both flow regimes show a very low kv/kh ratio in the segment around the well.All data tests were interpreted manually using practical equations after doing comprehensive literature studies. The data were also evaluated quantitatively using F.A.S.T WelltestTM – engineering software of pressure transient analysis from Fekete reservoir engineering software and services. Reservoir properties obtained from pressure transient analysis have similar results with the original data on the reservoir model. To simplify the study, production rate which was used in build-up and interference tests are only from oil production basis. In addition, no injections in Norne field were included during the tests to have the same comparison in all analysis. As the future work, any other types of tests are strongly recommended, both in single-well and multiple-well testing, also in vertical and horizontal wells.

The topic of this report is "Mobility of Residual Gas in Ormen Lange" and it has been prepared as a part of the course TPG4915 Petroleum Engineering - Reservoir Engineering, Master Thesis at the Norwegian University of Science and Technology (NTNU). The work has been performed on Ormen Lange, a natural gas field on the Norwegian continental shelf, operated by A/S Norske Shell.Substantial volumes of residual gas are present in the Ormen Lange field as a result of the hydrodynamic aquifer flow9. Total residual gas volume in 2007 is 80±30 Bcm gas depending on the residual gas saturation, Sgr. Depending on scenario 15±6 Bcm of the residual gas is recovered in 2040, and the recovery factor of residual gas is 15±5%. In general, with increasing residual gas saturation, residual gas volume increases and the recovery factor of the residual gas zones decreases.The main challenge and purpose of this project has been to understand the physics of residual gas during depletion, and to assess the potential recovery from the residual gas in different parts of the reservoir. A literature study aiming to identify the main driving parameters with respect to residual gas mobility has been conducted and used as a basis for subsequent simulation work.The need to understand the charge history of the field became important in 2008 when an appraisal well in the northern part of the field encountered only residual gas saturations in the crest of the structure and in the middle of the Direct Hydrocarbon Indicator (DHI). Core analyses, well logs and geological interpretation obtained during drilling and gas production in Ormen Lange so far allows the assessment of residual gas, and gives indications of where it resides in the reservoir. Residual gas saturations (Sgr) in the range of 0.21 to 0.41 have been observed in water-flood core measurements. Recovery of residual gas depends on the final reservoir pressure. The recovery from the residual gas zone in the south, where the reservoir is well depleted, is good compared to the north where the pressure depletion is limited. Applying the base case residual gas saturation value of 0.3 and assuming no mobility threshold above residual gas saturation (critical gas saturation is equal to residual gas saturation), the total recovery in 2040 of residual gas is 19%, see Figure 1. Future development plans will increase total residual gas recovery, as the pressure will be further depleted.In the South where an acting aquifer is present, recovery from residual gas highly depends on the strength of the aquifer. Earlier breakthrough of water with a potential stronger aquifer results in earlier shut-in of the wells, hence higher abandonment pressure and lower recovery of residual gas in the south. The critical gas saturation is one of the parameters that has been extensively investigated in this project. The critical gas saturation is the saturation at which a continuous gas flow can be first observed, coinciding with a non-zero gas relative-permeability. Most literature indicates that residual gas requires approximately 5% increase of gas saturation units in order to reach critical gas saturation. The implementation of the latter mobility threshold for residual gas in Ormen Lange reduces the total residual gas recovery by 2%. The flow rate of the remobilized gas depends on how fast the gas relative permeability increases during secondary drainage. Hence changing the slope of the gas relative permeability curve, Ng, and the endpoint value, krg, also affects the ultimate recovery of residual gas.

En god reservoarmodell, som både representerer de statiske parameterne og strømningsegenskaper, er avgjørende for å optimalisere produksjonen fra hvilket som helst reservoar. Mye tid og krefter blir brukt til å beskrive reservoaret så godt som mulig og store økonomiske beslutninger hviler på prognosene fra denne modellen. Prognoser utført ved bruk av flere realisasjoner basert på samme modell blir stadig mer populære for å fange usikkerhet. Historietilpasningsmetoder som Ensemble Kalman Filter er godt egnet for dette.EnKF ble foreslått av Evensen i 1994 som en data-assimilasjon metode innen oceanografi, og har blitt utviklet og testet flere ganger innen petroleumsindustrien siden da. Filteret bruker et ensemble av vektorer for å beskrive reservoarparameterne og hver av disse vektorene beskriver en realisasjon av reservoaret. Kovariansen mellom disse vektorene brukes til å representere både spredning og reservoarets respons til parameterverdier.I denne oppgaven har EnKF blitt brukt til historietilpasning av PUNQ S3, en syntetisk reservoarmodell, for å se effekten av tilgjengelig informasjon. Dette ble gjort gjennom to ulike simulering, hvor den første ble gjennomført med grenseverdier for å begrense de statiske parameterne. I det andre tilfellet ble utviklingen av en historietilpasning presentert gjennom ulike tidsskritt. Modellene ble evaluert på bakgrunn av sin prognose for fremtidig produksjon, samt sine avvik i parameterverdi sammenliknet med de sanne parameterne. Selv om prognosene ble forbedret for samtlige modeller, ble det ikke observert noen forbedring i reservoarparameterne. En utvikling mot en falsk løsning ble observert. Denne løsningen hadde feil parameterverdier, men gav en prognose for fremtidig produksjon som var veldig lik sannheten. Den geologiske kunnskapen ble ikke anvendt i oppdateringen, noe som førte til at de oppdaterte modellene var lengre unna sannheten enn det opprinnelige utgangspunktet.

World proved oil reserve gradually decreases due to the increase production but decrease new field discovery. The focus on enhance oil recovery from the existing fields has become more interesting in the recent years. Since waterflooding has been used in practices in secondary recovery phase for long time ago, the low salinity waterflooding is possible to apply as tertiary recovery phase. Another effective enhance oil recovery method is chemical flooding especially, nowadays, when the price of chemical is not a big issue compared to oil price. Both low salinity and chemical flooding method have been trialed and success in laboratory studies and some field tests. Moreover the salinity sensitivity on chemical flooding has been studied and both positive and negative results were proposed. Because new technology has been developing day by day in order to get higher oil recovery, the new technology as the combination of low salinity waterflooding and chemical flooding has been studied in this report. In this thesis, the literature of low salinity water flooding, alkaline flooding, surfactant flooding, polymer flooding and alkaline-surfactant-polymer flooding (ASP) have been reviewed. The mechanisms of each method that affect to oil recovery and salinity sensitivity on each chemical flooding method have been summarized. All of those studies showed the benefit of chemical to the low salinity water flooding. the result of literature reviews has turned to the numerical simulation part.The simulation has been carried out on a 3 dimensional synthetic model by using Eclipse 100 as the simulator. The model is heterogeneous with patterns variation in permeability and porosity. The effect of low salinity in water flooding, alkaline flooding, surfactant flooding, polymer flooding and ASP flooding have been observed in many aspects.The main role of low salinity effect in water flooding is wettability changing from oil-wet to water-wet. The low salinity water in the first water flooding phase give the positive effect but not much different compared to overall recovery. The low salinity in chemical solution influences an additional oil recovery in all combinations. Mainly, low salinity increases polymer solution viscosity that can improve sweep efficiency of polymer flooding. In alkaline flooding and surfactant flooding, the salinity is need to be optimized to optimum salinity condition corresponding to optimum alkaline concentration and surfactant concentration, where creates the lowest IFT. The range of secondary flooding for alkaline and surfactant flooding is when they reach the optimum concentration. In case of polymer, the viscous polymer solution can impact longer as the polymer injection range. In term of low salinity in tertiary water flooding, it influences better oil recovery than high salinity water flooding. Therefore, it can be concluded that low salinity water flooding gives a positive effect to overall result when combined with chemical flooding. The recommendations are also available for further study.

Bean Bazar gas field was discovered by Pakistan Shell Oil Company (PSOC) in 1960 and initial production started since 1999. The field has two wells-BB1 and BB2 and two sand groups- Upper Gas sand (UGS) and Lower Gas Sand (LGS). This is one of the condensate rich fields in Bangladesh. The field is produced by water drive. A huge amount of water is produced from the two sands. The proven gas reserve of this field was estimated approximately 230.80 Bcf. The total gas produced till December, 2011 was 75.65 Bcf. That is one-third gas had already been produced. The remaining gas is required to recover from the wells by predicting the present well and reservoir performance for a certain time based on the current production data. That is why, this task was liked by me when the authority proposed me.In this thesis work, a simulation model was constructed based on the latest production data. Vertical Flow performance (VFP) for BB1 and BB2, Change of transmissibility, Change of angle of aquifer etc. improved the recovery. Most of the geological data was taken from the "Simulation Study of Beani Bazar Field" by RPS Energy, U.K.2009. The simulation model was then run to forecast the future field performance to find out an optimal development plan for the field and to determine the reserve estimation.Simulation results showed that the ultimate recovery is very high in drilling wells but it involves a lot of cost. But there is no way out. The water must be controlled. The final recommendation for future work on Beani Bazar simulation model is that the water rise should be controlled by drilling a new well in the present reservoir a few km away from the existing wells. The quick gas production can bring huge water which should be handled by re-installing the plant infra-structure.

A great percentage of oil is observed to be left in the reservoir after the traditional primary and secondary recovery methods. This oil is described as immobile oil. Alkaline-Surfactants are chemicals used to reduce the interfacial tension between the involved fluids, while polymer is used in making the immobile oil mobile.Norne C-segment is in the decline stage and is facing considerable challenges regardingvolume of oil bye-passed due to water flooding. There is need for developing cost efficient enhanced oil recovery (EOR) methods that would be suitable for Norne fluid and rock properties and therefore improve sweep efficiency significantly. Based on literature and screening criteria, alkaline-surfactant-polymer can be used as an enhancing agent to produce extra oil and reduce water-cut significantly in the C-segment.The objective of this work is to evaluate the possibilities of using alkaline, surfactant and/or polymer to increase the oil recovery factor and prolong the production decline stage of Norne field. An initial study was conducted using heterogeneous synthetic models (with Norne Csegment fluids and rock properties) to assess the suitability of alkaline/surfactant/polymer (ASP) flooding. All the chemical cases simulated gave substantial incremental oil production and water-cut reduction. However, history matched Norne C-segment reservoir model was used to simulate alkalinesurfactant-polymer flooding using Eclipse 100. Appropriate chemical quantity for injection was ascertained by simulating several cases with different concentration, injection length and time of injection. Different sensitivity analyses were made and simulations revealed that the most effective method was not the most profitable. Having established most profitable method which was injecting ASP slug with a concentration of 7Kg/m3, 2Kg/m3 and 0.3Kg/m3 into C-3H (injector) for 4-years in a cyclic manner, an incremental recovery factor of 2.61% was recorded and Net Present Value (NPV) was calculated to be 1660 x103MNOK

SummaryThe displacement of oil from reservoir rock pore spaces is a function of many interacting variables, amongst which the reservoir wetting state has been shown to be one of the important affected by the rock lithology, oil chemistry and brine salinity. A finding from previous research says that the injection brine into oil saturated core plug increased oil recovery. Based on this the objective of this master thesis is to investigate the effect of brine concentration on flow properties in two types of carbonate rocks for enhanced oil recovery (EOR) through imbibition and water flooding processes.The methodology used to evaluate the effect of brine concentration (BC) and chemical composition (CC) for oil recovery consisted on two stages. The first stage covers the literature review regarding the effect of brine concentration and chemical composition, including carbonates (chalk and limestone) characteristics. The second stage is related to the laboratory experiment which was performed using n-Decane oil, six (6) brines with different concentrations and chemical composition and the six (6) core plugs where four (4) “chalks” from Ekofisk (Norway) and the other two (2) “limestones” from Iranian field. The experiment was carried out in the laboratory of Institute of Petroleum and Technology (IPT), the materials, chemicals products, apparatus and equipments, methodology and procedures were provided by the IPT laboratory.To carry out the laboratory experiments, initially the two cores from Iranian were cleaned before being used. Different properties of brines, cores and oil were measured using different methods and procedures; and results were computed. Next, each core was saturated with one type of brine and after that flooded by n-Decane oil for establishment of initial water saturation and determination of volume of oil produced by drainage process at room temperature conditions at one bar. After that, all cores were aging about 15 days at room temperature condition. Finally, each core was flooded using brine by imbibition process at room temperature conditions.Results achieved were computed and discussed based on the literature review and compared with “A salinity (AS) Ekofisk core reference case” and similar studies. From this study was observed that the matrix block has a high porosity. The average porosity was about 40.24% of the volumes of large pores. The average absolute permeability was about 3.73 mD which is low because the microporous dominate the pore network. The average brine density (ρ) was about1.026 g/cm3 and pH was about 7.25. The initial water saturation varies between 14.58 to 28.50% and residual oil saturation among 22.49 to 62%. The sleeve pressure in the cylinder was kept from 15 to 28 bar. During waterfloodig was observed that the breakthrough pressure drop and time to increases when the oil recovery increase.The highest original oil in place (OOIP) was achieved in the low salinity (LS) core which was about 68.46% and the lowest was recorded in the C salinity (CS) core which was around 26.71%. The reason of the high and the low recovery is related with the effect of brine concentration and chemical composition of Sodium, Calcium, Magnesium and Sulphate, added in the solution. The main driving mechanism for low salinity waterflooding is believed to be multi component ionic exchange made possible by the expansion of electrical double layer. The permeability and porosity of the cores can be pointed as other factor. In general, it was showen that there is an increase in oil recovery as the salinity decreases.

I performed manual History Matching on the E-segment of the Norne field based on available production & pressure data. I used the obtained final match for future prediction. I performed Production Optimization by experimenting with increased and decreased water injection rates.

This research is a simulation study to improve total oil production using ASP flooding method based on simulation model of Norne field C-segment. The black oil model was used for simulations. Remaining oil in the reservoir can be divided into two classes, firstly residual oil to the water flood and secondly oil bypassed by the water flood. Residual oil mainly contains capillary trapped oil. Water flooding only is not able to produce capillary trapped oil so that there is a need for additional technique and force to produce as much as residual oil. One way of recovering this capillary trapped oil is by adding chemicals such as surfactant and alkaline to the injected water. Surfactants are considered for enhanced oil recovery by reduction of oil–water interfacial tension (IFT). The crucial role of alkali in an alkaline surfactant process is to reduce adsorption of surfactant during displacement through the formation. Also alkali is beneficial for reduction of oil-water IFT by in situ generation of soap, which is an anionic surfactant. Generally alkali is injected with surfactant together. On the other hand, polymer is very effective addition by increasing water viscosity which controls water mobility thus improving the sweep efficiency.In the first place, ASP flooding was simulated and studied for one dimensional, two dimensional and three dimensional synthetic models. All these models were built based on C-segment rock properties and reservoir parameters. Based on test runs, well C-3H was selected and used as a main injector in order to execute chemical injection schemes in the C-segment. Five studies such as polymer flooding, surfactant flooding, surfactant-polymer flooding, alkaline-surfactant and alkaline-surfactant-polymer flooding were considered in the injection process and important results from simulator were analyzed and interpreted. Sensitivity analyses were done especially focusing on chemical solution concentration, injection rate and duration of injection time. The polymer flooding project in this study has shown a better outcome compared to water flooding project. Economically best ASP solution flooding case is the flooding with concentration of alkaline at1.5kg/m3, surfactant at 15kg/m3 and polymer at 0.35 kg/m3 injecting for 5 years. AS flooding case for 4 years with alkali concentration at 0.5kg/m3 and surfactant concentration at 25 kg/m3 gave highest NPV value. It was found that surfactant flooding has a promising effect and it is more profitable than polymer flooding for the C segment in terms of NPV. Economic sensitivity analysis (Spider diagram) for low case, base case and high case at different oil prices, chemicals prices, and discount rate were also presented. It was found that change in oil price has significant effect on NPV compared to other parameters while polymer price has the least effect on NPV for high and low cases.

This study presents a detailed analysis of producing gas-oil ratio performance characteristics from conventional reservoir to unconventional reservoir. Numerical simulations of various reservoir fluid systems are included for comparison. In a wide sense of the word, the term of unconventional reservoir is including tight gas sand, coal bed methane, gas hydrate deposits, heavy oil gas shale and etc. In this study we specify the unconventional reservoir to only mean the low and ultra low permeability reservoir, which is including tight or shale reservoir. As an emerging research topic in the E&P industry, shale reservoir’s long-term well performance characteristics are generally not well understood (Anderson et al. 2010). Research methods and techniques for conventional reservoir are usually directly used in this unconventional reservoir analysis. These methods, however, have proven to be too pessimistic (Anderson et al., 2010). Fit-for-purpose approaches or solutions should be introduced in this new topic. Recently, hydraulic fracturing treatment is commonly used in the low matrix permeability reservoir to attain an economic production rate. The difference of well production performance between conventional reservoir and unconventional reservoir is not well known. In this study, we are trying to give a quantitative analysis in order to answer this question.In this study, a “generic” reservoir from field data with constant reserves and size were assumed. This reservoir model is homogeneous and of constant porosity, permeability and initial water saturation. In order to compare the production performance, fluid systems are varied from volatile oil to near critical oil, to gas condensate and to wet gas. The permeability of the reservoir model is also designed from high (conventional reservoir) to ultra low (unconventional), which ranges from 101 to 10-5 mD. Influence from fracture is especially considered because fractures in the low permeability reservoir provide a high conductivity that connects the reservoir matrix to the horizontal well. Fractures in the model are designed with identical geometrical characteristics (length, thickness) and of inner homogeneous properties (porosity, permeability).A black-oil model is used for each reservoir, and its PVT properties are generated with a 31 components EOS model using Whitson-Torp procedure (Whitson et al., 1983). Reservoir fluid systems equilibrium calculation in the black-oil model is done using the initial gas-oil ratio. We have compared the well’s production performance for each fluid system.Based on the industry experience, two standards are used in reservoir simulation control: gas production rate and cumulative revenue. The gas production rate with 10 ×106 ft3/day in the first 10 days or the cumulative revenue equal to 5 ×105 USD from the first 10 days is set as the standard for the commercial well rate. All of these simulations are run under the control of these two types which have just been mentioned. A case of liquid rich gas reservoir is analyzed systematically, to compare its production performance when reservoir permeability is changed from high to low. We are interested in how much oil or gas condensate can be extracted from the “reservoir” if same initial fluids in the reservoir but of a different permeability. This study is useful and practical, particularly for the industry in the era of “high” oil price and “low” gas price in North America.The simulation results show that we can extract more liquid from the reservoir if the matrix permeability is higher, particularly for the reservoir with initially large oil contents (volatile oil reservoir, near critical reservoir and gas condensate reservoir). Fracturing treatment in unconventional reservoir is required to attain an economic production rate. We also realize that for the required number of fractures and reservoir’s matrix permeability, there exists linear correlation in log-log plot in the low-permeability reservoir. In this study, the unique optimization software Pipe-It and reservoir simulator SENSOR are used. Optimal simulation results of permeability combination are obtained by the module Optimizer in Pipe-It.

In this study, based on the work of Vogel, we generated the Inflow Performance Relationship (IPR) curves and its dimensionless form at any stage of depletion using black-oil simulator results. The IPR was generated for horizontal well with gas and water coning problems, producing from thin oil reservoir sandwiched between gas cap and aquifer. Two empirical IPR equations adopted from SPE paper by Whitson was also presented here. The first empirical relationship was developed based on simulated data for each reservoir pressure (stage of depletion) while the second relationship was developed based on all generated data.A fully implicit black-oil Cartesian model with total grid number of 1480 and 150 ft total thickness was used as reservoir model. The horizontal well extends through the full length of reservoir in y-direction with only one grid number along the horizontal section which makes the model a 2D problem. Sensor reservoir simulator and Pipe-It software were utilized to generate the IPR data.This work also includes a sensitivity study to understand the effect of several parameters to gas and water coning behavior, well placement optimization, coning collapse study, and the effect of coning to maximum well production rate. In coning collapse study, a relationship between flowing bottom-hole pressure and reservoir pressure when the cone collapse is provided in graphical form. This could be useful in field application where chocking the well to lower flowing bottom-hole pressure has become one alternative to reduce coning problems.

AbstractInside of the Norne E-segment remains a considerable amount of residual oil even after applying the primary and secondary oil recovery methods (water injection). Recently, several methods have been studied based on simulations to decrease the residual oil trapped by capillary forces and consequently improve the oil recoverability. Additionally, Norne E-segment is severely affected by stratigraphic barriers and faults of nature not sealing, semi sealing and completely sealing. Water Alternating Gas (WAG) and Simultaneously Water Alternating Gas (SWAG) injection techniques are presented as potential candidates to increase oil productivity in the Norne E-Segment by decreasing the gas mobility and capillary forces guarantying effective microscopic displacement due to gas flooding and macroscopic sweep created by water injection.In the first part of this study, based on simulations (Eclipse 100, Black oil simulator), sensitivity analyses of WAG cycles and WAG ratio were investigated combining with low injection rate and high injection rate. However, three WAG cycle were suggested (3 months, 6 months and 1years injection cycles) and different values of WAG ratio were studied based on low and high injection rates of water and gas. According to the results, WAG cycle doesn’t affect the fluids rates productions when low injection rate is used, but a slightly effect is noticed when high injection rate is applied, thus a slightly optimal WAG ratio was found to be 1:3 when high WAG ratio is used.As a sequence, examination of three different injection patters scenarios were simulated to optimize the oil recoverability using both techniques WAG and SWAG, namely: injection studies using the injection wells already existed; injection studies using the injection wells already existed by doing a new completion within Ile and Tofte formations; injection studies placing a new injection well plus new completion of the injection well. As a result, the last scenario using SWAG technique presented oil recovery around 73%, whose was approximately 5% higher than oil recoverability when WAG injection technique (68%), when high injection rate is applied.

The world needs energy – and over the short and medium term it is clear that much of our global energy consumption will come from fossil sources such as oil, gas and coal. With the current growing demand for oil led by major energy consuming countries such as China and India, securing new oil resources is a critical challenge for the oil industry. Each year, new production is needed to compensate the natural decline of existing wells, and the additional production required to satisfy the yearly demand for hydrocarbon energy that will represent approximately 9% of the worldwide total production. For this growth to be sustainable, a strong focus will have to be placed on finding new discoveries and/or optimizing oil production from current resources. The cost associated with the first option is significant. Therefore, reservoir management teams all over the world will have to cater for this demand mainly by maximizing hydrocarbon recovery factors through Enhanced Oil Recovery (EOR) processes. EOR consists of methods aimed at increasing ultimate oil recovery by injecting appropriate agents not normally present in the reservoir, such as chemicals, solvents, oxidizers and heat carriers in order to induce new mechanisms for displacing oil. Chemical flooding is one of the most promising and broadly applied EOR processes which have enjoyed significant research and pilot testing during the 1980s with a significant revival in recent years. However, its commercial implementation has been facing several technical, operational and economic challenges. Chemical flooding is further subdivided into polymer flooding, surfactant flooding, alkaline flooding, miscellar flooding, alkaline-surfactant-polymer (ASP) flooding. ASP flooding is a form of chemical enhanced oil recovery (EOR) that can allow operators to extend reservoir pool life and extract incremental reserves currently inaccessible by conventional EOR techniques such as waterflooding. Three chemical inject in the ASP process which is synergistic. In the ASP process, Surfactants are chemicals that used to reduce the interfacial tension between the involved fluids, making the immobile oil mobile. Alkali reduces adsorption of the surfactant on the rock surfaces and reacts with acids in the oil to create natural surfactant. Polymer improves the sweep efficiency. By simulating ASP flooding for several cases, with different chemical concentrations, injection length, time of injection, current well optimization and new well placement, this report suggests a number of good alternatives. Simulations showed that the most effective method was not the most profitable. From the simulation results and economic analysis, ASP flooding can be a good alternative for the Norne E-segment. But the margins are not significant, so fixed costs (such as equipment rental) will be of crucial importance.

In petroleum fields, the essence of well placement is to develop and maintain petroleum reservoirs in order to achieve maximum production for economic benefit. Maximum production can be achieved with more oil wells, but few optimal numbers of wells in good location reduces economic costs and increase recovery. The best location for the placement of oil, gas or water wells depends on reservoir and fluid properties, well and surface equipment specifications, as well as economic parameters [1].The objective of the study is to determine the net present value from few well placements in the Norne C-segment reservoir by either obtaining the same or more oil production/recovery compared to the base case wells. New well placement in a reservoir simulation model uses an industrial standard ECLIPSE reservoir simulator. Manually simulation approach is used to locate high oil saturation grids for the new well placement. From the base case simulation result, a total number of thirteen wells were discovered, nine producers and four injectors. The production and injection wells were classified with a suffix according to the production templates B, D, K and injection template C respectively.The base case wells removed and new well placed from exhaustive simulation runs for two different scenario cases. A total number of ten wells, six producers and four injectors were placed in each scenario. In order to obtain maximum oil recovery, the producers are placed horizontally while injectors remain the same as those from the base case. The new well placements in the scenario cases are identified with the suffix “P-H” for producers and “I-H” for injectors. Simulation results, the total oil produced for wells in each field case from the start year 1997 to December 2015, (end of production) can be seen in Table 10, 11 and 12 in chapter 6. The cumulative oil produced from each field case is the same as the total oil produced from all the wells in each case. The cumulative field oil and gas production from the start of production, November 1997 to December 2015 is 41.3 million Sm3 oil and 260 million Sm3 of gas for base case, 42.8 million Sm3 oil and 269 million Sm3 of gas for scenario 1 case, 43.2 million Sm3 oil and 272 million Sm3 of gas for scenario 2 case. The recovery factor for base case is 28%, scenario 1 & 2 are 29.0% and 29.3%. Each field case uses drive mechanisms, gas injection and water injection to support oil production and maintain pressure in the each field case. The total gas and water injected in the base case field were 9.6 billion Sm3 and 78.8 million Sm3 respectively. In scenario 1, a total of 8.6 billion Sm3 of gas and 81.6 million Sm3 of water was injected and in Scenario 2, 8.6 billion Sm3 of gas and 81.3 million Sm3 of water was injected. The Net present values for the three cases were calculated taking into account the economic costs such as well cost, cost of gas and water injection. Sensitization was done on the oil price ($25, $35 and $45). The NPV results from Table 19 prove that all case projects are acceptable, but scenario 2 is the most economical as it has the highest NPV of $4,026 million based on $35-medium oil price that was considered.

Orientador: Denis José Schiozer
Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica e Instituto de Geociências
Made available in DSpace on 2018-08-20T19:01:11Z (GMT). No. of bitstreams: 1 Cotrim_HenriqueAraujo_M.pdf: 2298003 bytes, checksum: 2551e58ec65d13716630850ca27cd695 (MD5) Previous issue date: 2012
Resumo: No contexto da engenharia de reservatórios de petróleo, é bem conhecida a importância do trabalho de seleção de estratégias de produção. No caso específico em que há restrição de tratamento de fluidos na superfície, a alocação de vazões de produção e injeção dos poços é mais um item a ser otimizado. Quando o problema envolve diversas unidades segregadas de reservatórios, cada qual caracterizado por seu próprio modelo de fluxo, as simulações não devem ser consideradas isoladamente, pois elas estão efetivamente acopladas pelos limites das vazões totais de produção e/ou injeção. Neste cenário de múltiplos reservatórios, o presente trabalho propõe o estudo de alocação das vazões de produção utilizando abordagens tradicionais e também aplicando a Modelagem Integrada de Produção (MIP) simplificada considerando uma estratégia de produção fixa. A MIP simplificada consiste no uso de um aplicativo que gerencia o acoplamento explícito entre simulações de reservatórios e sistemas de produção, sendo este último elemento representado de forma simplificada. Neste trabalho, foi selecionado o limite de escoamento de gás como restrição operacional de superfície em dois casos de estudo e aplicados os procedimentos tradicionais de simulação segregada com rateio manual das vazões e união dos modelos para simulação de uma malha única, além da MIP simplificada. Em cada abordagem, foram aplicados três métodos de alocação de vazões disponíveis no simulador comercial adotado e implementados em rotina específica (RAV). Os resultados obtidos mostram que o uso de diferentes algoritmos de alocação de vazões pode levar a alterações na previsão de VPL superiores a 12% e indicam que a MIP simplificada é válida e pode ser considerada a melhor alternativa em cenários específicos, quando a otimização manual do rateio e a união das malhas não se apresentarem como soluções simples
Abstract: The selection of production strategies for petroleum field development is an important task. The specific case in which there are constrained surface facilities, the allocation of production and injection well rates is one more item to be optimized. When the problem involves several segregated reservoir units, each of which characterized by its own geological model, the simulations should not be considered as isolated because they are effectively coupled by the limits of the total flow of production and injection. In this scenario of multiple reservoirs, this work proposes the study of production rates allocation using traditional approaches and also applying the simplified Integrated Production Modeling (IPM), considering a fixed position for the wells. The simplified IPM consists of using an application that manages an explicit coupling between reservoir simulators and production systems, the latter element represented in a simplified form. In this work, the total gas rate was selected as the surface constraint in two case studies. The traditional procedures of segregated simulation with manual optimization and amalgamating the simulation models into a single model are used, besides the simplified IPM. In each approach, three methods are applied for well rates allocation, available in the adopted commercial simulator and implemented in the specific routine (RAV). The obtained results show that the use of different flow allocation algorithms can lead to changes in estimated NPV above 12% and indicate that the simplified IPM is valid and it can be considered the best alternative in some specific scenarios
Mestrado
Reservatórios e Gestão
Mestre em Ciências e Engenharia de Petróleo

Orientador: Denis José Schiozer
Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica e Instituto de Geociências
Made available in DSpace on 2018-08-16T18:22:07Z (GMT). No. of bitstreams: 1 Machado_AndreFrancisco_M.pdf: 5201739 bytes, checksum: 085087757151932cbfb475a71ef60dab (MD5) Previous issue date: 2010
Resumo: A técnica de ajuste de histórico consiste em modificar o modelo de simulação de modo que este reproduza o histórico de produções e pressões observadas. No entanto, a aplicação dessa técnica nos primeiros anos de produção de um campo fica bastante limitada, principalmente, devido à reduzida quantidade de informações disponíveis. Atualmente, a ferramenta que tem ajudado a minimizar o problema da escassez de informações é a sísmica 4D. Alguns trabalhos recentes têm contribuído para o desenvolvimento de metodologias que integrem informações sísmicas ao processo de ajuste de maneira quantitativa, resultando em modelos ajustados mais confiáveis. O objetivo principal deste trabalho foi continuar com os estudos realizados por Risso (2007) e Ida (2009) no sentido de avaliar a utilização dos mapas provenientes da sísmica 4D no processo de ajuste de histórico de produção. O foco desta pesquisa consiste na aplicação dos mapas de saturação e pressão de forma simultânea no processo de ajuste durante a fase de parametrização do problema para melhor diagnosticar e determinar as heterogeneidades do reservatório. Foi também avaliada a influência da informação do mapa de pressão na fase de otimização como um parâmetro de ajuste no sentido de melhorar a precisão da função-objetivo. Foram estudados dois modelos durante a pesquisa, um para a validação da metodologia e outro para a aplicação da mesma. A validação da metodologia proposta foi realizada num modelo bidimensional five-spot com duas barreiras geológicas e um canal de alta permeabilidade e a aplicação foi realizada num modelo modificado do Campo de Namorado. Nos dois casos, foram constatados ganhos de qualidade no ajuste de histórico proporcionado pela incorporação quantitativa dos mapas de saturação e de pressão
Abstract: The history matching technique consists in modify the simulation model so that it reproduces the historical production and pressure observed. However, applying this technique in the early years of production of a field is rather limited, mainly due limited amount of information available. Currently, the tool that has helped minimize the problem of scarcity of information is seismic 4D. Some works, such as Risso (2007) and Ida (2009), have contributed to the development of methodologies that integrate seismic data to the history matching process in a quantitative way, promoting adjusted models more reliable. The focus of this research was application of pressure map in the history matching process during the parameterization of the problem in order to supplement the information of saturation map to better determine the heterogeneity of the reservoir. In the background was evaluated the influence of pressure map information during optimization as a parameter in order to improve the accuracy of the objective function. The validation of methodology proposed was done in a simple synthetic model and the application was done in the Namorado Field, modified with one geological fault and one channel with high permeability
Mestrado
Reservatórios e Gestão
Mestre em Ciências e Engenharia de Petróleo

Orientadores: Denis Jose Schiozer, Celio Maschio
Dissertação (mestrado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecanica, Instituto de Geociencias
Made available in DSpace on 2018-08-14T07:27:37Z (GMT). No. of bitstreams: 1 Ida_Mauro_M.pdf: 14939670 bytes, checksum: 3bac33f4177c36ec35c53933f345273b (MD5) Previous issue date: 2009
Resumo: Boas decisões no gerenciamento de um campo de petróleo em produção dependem fortemente da confiabilidade da previsão de produção que demanda um modelo de escoamento que reproduza com boa precisão o histórico de produção. Devido à complexidade do processo de ajuste de histórico de produção, vários modelos podem resultar em ajustes aceitáveis, porém ainda com incertezas na previsão de produção, principalmente pelo fato de existirem algumas heterogeneidades de grande impacto não observadas na fase de caracterização do modelo. Para reduzir as incertezas, surgiu a tecnologia de sísmica 4D que a partir de análises qualitativas permite identificar frentes de saturação de água, resultando em grandes avanços na caracterização do reservatório e, conseqüentemente, no ajuste de histórico. Apesar de muito útil, em alguns casos podem levar a interpretações equivocadas, sendo necessário um procedimento adicional para o uso da informação dessa tecnologia. O objetivo principal deste trabalho é propor uma metodologia que utiliza simultaneamente os dados de produção e quantitativamente a impedância acústica da sísmica 4D para identificar as heterogeneidades do reservatório visando melhorar a qualidade do ajuste de histórico. A metodologia proposta consiste de duas etapas principais: parametrização e otimização com algoritmo genético. A validação da metodologia proposta foi realizada num modelo bidimensional five-spot com duas falhas geológicas e um canal de alta permeabilidade e a aplicação foi realizada num modelo modificado do Campo de Namorado com as seguintes características: total de 44 poços, uma falha geológica e um canal de alta permeabilidade. Em todos os modelos, foram constatados ganhos de qualidade no ajuste de histórico proporcionado pela incorporação quantitativa da sísmica 4D.
Abstract: Good decision making related to oilfield management depends on reliability of production forecast which demands calibrated reservoir simulation models. Due to the complexity of the production history matching process, there are many models with reasonable match but, many times, with different forecasts, mainly due to important heterogeneities that are not observed during the reservoir characterization phase. In order to reduce this uncertainty, a new technology named 4D seismic became available which uses quantitative analysis to identify water saturation front resulting in huge advance in reservoir characterization and consequently in history matching. Although this technology is very useful, in some cases, it can result in mistaken interpretation and it needs additional work to use this technology. The main objective of this work is to propose a methodology to use production data and quantitatively acoustic impedance from 4D seismic to identify reservoir heterogeneities to increase the quality of the history matching. The methodology proposed is divided in two main steps: parametrization and optimization with genetic algorithm. The validation of methodology proposed was done in a simple synthetic model and the application was done in the Namorado Field, modified with one geological fault and one channel with high permeability.
Mestrado
Reservatórios e Gestão
Mestre em Ciências e Engenharia de Petróleo

We consider the structural and physical properties of complex model morphologies and microstructures obtained by Xray-CT imaging. The Minkowski functionals, a family of statistical measures based on the Euler-Poincaré characteristic of n-dimensional space, are shown to be sensitive measures of the morphology of disordered structures. Analytic results for the Boolean model are given and used to devise a reconstruction scheme, which allows one to accurately reconstruct a complex Boolean structure given at any phase fraction for all other phase fractions. The percolation thresholds of either phase are obtained with good accuracy. From the reconstructions one can subsequently predict property curves for the material across all phase fractions from a single 3D image. We illustrate this for transport and mechanical properties of complex Boolean systems and for experimental sandstone samples. By extending the Minkowski functionals to parallel surfaces using operations from mathematical morphology, a powerful discrimination of structure is obtained. Further the sensitivity of the Minkowski functionals under experimental conditions is analysed. Accurate calculations of conductive and elastic properties directly from tomographic images are achieved by estimating and minimising several sources of numerical error. Simulations of electrical conductivity and linear elastic properties on microtomographic images of Fontainebleau sandstone are in excellent agreement with experimental measurements over a wide range of porosity. The results show the feasibility of combining digitised images with transport and elasticity calculations to accurately predict physical properties of individual material morphologies. We show that measurements of properties based on microtomographic images are more accurate than those based on conventional theories for disordered materials. We study the elastic behaviour of model clean and cemented sandstones. Results are in excellent agreement with available experimental data, and are compared to conventional theoretical and empirical laws. A new predictive empirical method is given for predicting the elastic moduli of sandstone morphologies. The method gives an excellent match to numerical and experimental data.

Production of natural gas from unconventional gas-hydrate reservoirs faces kinds of challenges and uncertainties. One of the main and most common problems in gas-hydrates drilling is the dissociated gas from gas hydrates with decrease of pressure, increase of temperature, or combination of them. A reliable method that can be applied to predict the temperature profile of fluid during circulating in the drilling pipe and the annulus is needed. An analytical model was developed in this study for predicting temperature profiles in drilling gas-hydrate deposits. A case study is provided and indicates a good consistency between model-implications and field observations. According to the sensitivity analyses, the temperature profile of fluid in the drill pipe can be affected by the thickness of drill pipe, density and heat capacity of drill mud, pumping rate of drill mud, geo-thermal gradient, and the surface geo-temperature. The bottom hole temperature is dominated by the temperature and flow rate of the injected drilling fluid, thermal conductivity of cement, heat capacity and density of drill mud, geo-thermal gradient and geothermal temperature at surface, thickness of drill pipe, and cement sheath. Higher geothermal gradient and surface geothermal temperature can lead to a higher temperature profile of fluid in the annulus. The Joule-Thomason cooling effect below the drill bit nozzles will rapidly diminish in a short interval above the bottom hole by the heating effect of geo-thermal gradient. The rate of penetration of drill bit has negligible effect on the fluid temperature profile due to the low percentage of heat flow contributed by the drill cuttings.

The average pore size in producing unconventional, liquids-rich reservoirs is estimated to be less than 100 nm. At this nano-pore scale, capillary and surface disjoining force interactions, such as van der Waals, structural, and adsorption, affect the phase behavior that is not considered to be significantly, different than in conventional reservoirs. In this dissertation, a comprehensive discussion of the thermodynamics required to model phase behavior of unconventional, liquids-rich reservoirs is presented. Three oil compositions from different unconventional reservoirs are used to generate results.

The impact of confinement manifests itself in the form of reduction of the liquid pressure at which the first gas bubble forms when compared to the bulk fluid measurements in PVT cells. It is shown that the suppression of the bubble-point pressure impacts the saturated portion of the liquid formation volume factor and extends the undersaturated portion of the curve. The equilibrium gas composition is different for each supersaturation level and the gas is composed of lighter components as the supersaturation, i.e., the bubble-point suppression, increases. The minimum radius of the pore that is required to form a specified size bubble is also investigated and the range of pore sizes required under different assumptions is reported.

The impact of this phase behavior deviation on the flow of confined fluids is investigated using a black-oil simulator, COZSim, which evaluates gas and oil fluid properties at corresponding phase pressures. The simulator was independently developed in a DOE project with the capability to incorporate the findings of this research. The results of the analysis show that there is a difference in gas production and gas saturation distribution in the reservoir with and without the confinement impact on the PVT properties. The produced GOR is lower when the confinement is considered due to the bubble-point suppression. These results indicate that the use of bulk fluid measurements in modeling and predicting the performances of nano-porous unconventional reservoirs may result in significant underestimation of the reservoir potential.

Orientador: Denis José Schiozer
Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica e Instituto de Geonciências
Made available in DSpace on 2018-08-25T02:58:58Z (GMT). No. of bitstreams: 1 Barreto_CarlosEduardoAndradeGomes_D.pdf: 32862139 bytes, checksum: a675c53341d5c762ef1540a698f013c6 (MD5) Previous issue date: 2014
Resumo: O posicionamento de válvulas de controle ao longo de poços de petróleo é baseado na viabilidade econômica de sua instalação para melhorar o desempenho da produção considerando o controle do escoamento do reservatório para o poço. Para tomar a decisão de instalar essas válvulas, frequentemente, a simulação de reservatórios é utilizada para representar a operação do sistema produtivo e auxiliar a estimativa dos benefícios econômicos e a avaliação de risco de fazer os investimentos requeridos. Entretanto, para casos específicos, o elevado número de variáveis envolvidos na solução deste problema e a relação entre os recursos computacionais disponíveis e o tempo necessário para avaliação fazem com que métodos tradicionais de otimização e seleção demandem muito esforço computacional sendo até inviáveis para algumas aplicações. Este trabalho propõe um método de otimização que utiliza dados técnicos e econômicos gerados na previsão do comportamento do reservatório para auxiliar na otimização e seleção de válvulas de controle em poços produtores. O principal objetivo do método é reduzir o número de variáveis do problema e o espaço de busca das variáveis delimitando regiões com maiores potenciais para serem analisadas. Selecionam-se quais regiões dos poços tem maior potencial de avaliação a partir do uso de indicadores de produção e econômicos. Com esta análise, a otimização é realizada por partes, privilegiando inicialmente as análises nas regiões com maior potencial e seguindo na direção das regiões com menor potencial até não ser mais encontrada a viabilidade de aplicação de válvulas. Cada região é otimizada de forma isolada. Isto evita a análise de regiões com menor potencial mesmo antes de avaliações em regiões potencialmente viáveis. Para avaliar a eficiência do método, os resultados encontrados são comparados com resultados gerados por um algoritmo evolucionário. A metodologia é aplicada em casos simples para avaliar o desempenho do método e melhor comparar com as repostas do algoritmo evolucionário e é aplicada em casos mais complexos para avaliar o desempenho do método em condições mais próximos aos reais. Os resultados para o caso simples mostraram que o método proposto atingiu soluções similares aos encontrados pelo algoritmo evolucionário, mas com um número de simulações reduzido. Já para o caso mais complexo, os resultados apresentados pelo método proposto foram superiores e o algoritmo evolucionário apresentou dificuldade em solucionar o problema com o número máximo de simulações permitidas. O uso de dados técnicos e econômicos auxiliou de forma relevante na evolução do método de otimização, uma vez que reduziram o número de variáveis e o espaço de busca. Em cada caso, as os indicadores das alternativas testadas se mantiveram próximas aos da solução otimizada desde o início da simulação e não houve uma dispersão significativa, comparando com as soluções testadas pelo algoritmo evolucionário. Além disso, as avaliações mostram que o valor presente líquido de um projeto pode aumentar significativamente com a inclusão de válvulas de controle. Observa-se também uma influência significativa do cenário econômico no resultado das otimizações. Portanto, o método proposto pode ser utilizado em problemas reais com relevantes restrições computacionais e de tempo ao invés do uso de métodos tradicionais de otimização. Conclui-se que o método de otimização proposto atingiu os objetivos de ser eficiente na busca da solução ótima com um número reduzido de simulações de reservatórios. O uso de indicadores para guiar o método de otimização reduz o espaço de busca e evita a avaliação de alternativas com baixo potencial de aplicação. A otimização por partes é um instrumento que pode ser utilizado na otimização de válvulas de controle e auxilia na melhoria do desempenho do processo como um todo e, se bem utilizado, não interfere negativamente na obtenção de soluções próximas à ótima. Além disso, o uso de válvulas de controle se mostra viável em casos específicos, principalmente em casos em que existe uma restrição na vazão da plataforma. As válvulas de controle também são eficientes no aumento da produção total de óleo e na redução da vazão de água nos anos logo após a chegada de água, apesar de a produção total de água possa ser maior
Abstract: The selection of number and position of inflow control valves in petroleum wells is complex and involves a high number of analyses to determine their viability of installation. Reservoir simulation can be used to represent valve operation and forecast production to evaluate the benefits in long terms. However, in specific cases, the high number of variables involved in this type of problem associated with the usual time-consuming of a single reservoir simulation made traditional optimization methods inefficient to solve the problem within an adequate time frame. This work proposes a dynamic optimization process that uses economic and technical indicators to speed up the process. The main ideia is to improve the exploration efficiency of the search space by selecting more potential well regions to start the process and to use an optimization workflow that splits the whole optimization in some suboptimization processes to evaluate each region at time. To assess the effectiveness of the proposed method, its results are checked by comparing them with those of an evolutionary algorithm using a simple example. The methodology is also applied in a more complex example with different geological scenarios. The results show that the proposed method achieves good results when compared with the evolutionary algorithm results. The design optimization in the complex example shows that the dynamic process is able to significantly increase the Net Present Value of the field with an acceptable number of simulation runs. It is also shown that the use of economic and technical indicators can be applied to reduce the number of variables of the problem, to define suitable constraints for each variable and to help the initial guess of seeds for the optimization method. It is concluded that the proposed methodology can be efficiently used to optimize inflow control valve design in cases in which computational resources and available time are limited. The use of indicators to better explore the search space is a viable way to selected more potential alternatives to be analyzed. The division of the whole optimization process in smaller pieces is an alternative to reduce the number of variables and to make faster evaluation, whitout loosing precision. In addition, the application of valves can improve total oil production, to retard water breakthrough and to reduce the water flow rate in the begginig of the water production, even with an increase in the total water production
Doutorado
Reservatórios e Gestão
Doutor em Ciências e Engenharia de Petróleo

The dynamic nature of coalbed methane reservoir permeability makes the continuous modeling of the flow process difficult. Knowledge of conventional reservoir modeling is of little value because the gas storage and flow mechanisms in coal are remarkably different. Field observations suggest that the gas productivity does not decline with time as expected. An increasing permeability is a possible explanation. This laboratory study is aimed at measuring the volumetric changes in coal matrix resulting from gas desorption, its impact on coal porosity and permeability, and an estimate of the stage when the effect is significant. Results using cylindrical samples of coal suggest that coal matrix shrinks with desorption of gas. The shrinkage is linearly proportional to the quantity of gas desorbed. Using the measured changes in matrix volume, variations in total and cleat porosity were estimated and found to increase significantly with desorption. These changes should, therefore, be used as input when simulating long-term gas production.

Orientadores: Denis José Schiozer, Célio Maschio
Tese (doutorado) - Universidade Estadual de Campinas, Faculdade de Engenharia Mecânica e Instituto de Geociências
Made available in DSpace on 2018-08-23T20:38:54Z (GMT). No. of bitstreams: 1 Davolio_Alessandra_D.pdf: 20299340 bytes, checksum: 39bd0e5f79603d95c1cb4c74b2616103 (MD5) Previous issue date: 2013
Resumo: O foco deste trabalho é usar dados de engenharia de reservatórios gerados através da simulação de fluxos para melhorar a interpretação quantitativa da sísmica 4D. A ideia é usar os conhecimentos de engenharia para minimizar possíveis informações incorretas geradas pela sísmica 4D, antes de usá-la para atualizar modelos de simulação de reservatórios em um procedimento de ajuste de histórico. Neste trabalho a integração entre estes dois conjuntos de dados é feita no domínio da pressão e saturação. Dessa forma, a primeira parte do trabalho apresenta uma metodologia para estimar variações de pressão e saturação a partir da sísmica 4D através de um procedimento de inversão petro-elástica. Este procedimento pode ser visto como uma ferramenta para aplicar a metodologia de integração da segunda parte, que é uma das principais contribuições deste trabalho. A metodologia de integração usa múltiplos modelos de simulação para restringir a estimativa de pressão e saturação a partir da sísmica 4D. Como resultado, observaram-se mapas menos ruidosos que permitem, portanto, uma melhor interpretação das variações ocorridas no reservatório. Seguindo a sequencia tradicional de trabalho, o mapa de saturação estimado foi então usado como dado de entrada em um processo de ajuste de histórico apresentado na terceira parte da tese. Foi mostrado que o procedimento de ajuste gera melhores resultados quando o dado de entrada, neste caso mapa de saturação, respeita o balanço de massa, o que não acontecia para os dados considerados. Assim, uma metodologia para calibrar o volume de água injetada associado ao mapa de saturação gerado pela sísmica 4D é apresentada na última parte do trabalho, que é outra contribuição importante a se destacar. Para melhor controlar os resultados de todas as metodologias aqui apresentadas foram usados dados sintéticos em todo o trabalho. Embora resultados satisfatórios tenham sido observados para este conjunto de dados, é importante destacar que as principais contribuições deste trabalho não são apenas os resultados observados, mas as metodologias propostas, que apresentam uma perspectiva inovadora para integração entre dados de sísmica 4D e engenharia de reservatórios
Abstract: The focus of this work is to use reservoir engineering data from numerical flow simulation to improve the quantitative interpretation of 4D seismic signals. The idea is to use engineering knowledge to minimize possible incorrect information provided by 4D seismic before using it to update reservoir simulation models in a history matching procedure. In this work the integration between the two dataset is done in the pressure and saturation domain. So, the first part of the work presents a methodology to estimate pressure and saturation changes from 4D seismic through a petro-elastic inversion procedure. This procedure can be seen as a tool to apply the integration methodology of the second part which is one of the main contributions of this work. The integration methodology uses multiple simulation models to constrain the estimation of pressure and saturation from 4D seismic. As a result, less noisy maps were obtained, allowing a better interpretation of the reservoir changes. Following the traditional sequence, the estimated saturation map is then used as input to the history matching process presented in the third part of the work. It was shown that the history matching procedure provides better results if the input data, in this case the saturation map, respects the expected mass balance, which was not the case for the dataset considered. Thus, a methodology to calibrate the volume of injected water associated to the saturation map provided by 4D seismic is presented in the last part of the work, which is another important contribution to be highlighted. To better control the results of all the methodologies here presented, a synthetic dataset was used in the entire work. Although satisfactory results were observed for this dataset, it is important to highlight that the main contributions of this work are not only the results, but the methodologies proposed, that present an innovative perspective for 4D seismic and reservoir engineering data integration
Doutorado
Reservatórios e Gestão
Doutora em Ciências e Engenharia de Petróleo