Neural networks (NNs) are widely used to investigate the relationship among variables in complex multivariate problems. In cases of limited data, the network behavior strongly depends on factors such as the choice of network activation function and network initial weights. In this study, I investigated the use of neural networks for multivariate analysis in the case of limited data. The analysis shows that special attention should be paid when building and using NNs in cases of limited data. The linear activation function at the output nodes outperforms the sigmoidal and Gaussian functions. I found that combining network predictions gives less biased predictions and allows for the assessment of the prediction variability. The NN results, along with conventional statistical analysis, were used to examine the effects of folding, bed thickness, structural position, and lithology on the fracture properties distributions in the Lisburne Formation, folded and exposed in the northeastern Brooks Range of Alaska. Fracture data from five folds, representing different degrees of folding, were analyzed. In addition, I modeled the fracture system using the discrete fracture network approach and investigated the effects of fracture properties on the flow conductance of the system. For the Lisburne data, two major fracture sets striking north/south and east/west were studied. Results of the NNs analysis suggest that fracture spacing in both sets is similar and weakly affected by folding and that stratigraphic position and lithology have a strong effect on fracture spacing. Folding, however, has a significant effect on fracture length. In open folds, fracture lengths in both sets have similar averages and variances. As the folds tighten, both the east/west and north/south fracture lengths increase by a factor of 2 or 3 and become more variable. In tight folds, fracture length in the north/south direction is significantly larger than in the east/west direction. The difference in length between the two fracture sets creates a strong anisotropy in the reservoir. Given the same fracture density in both sets, the set with the greater length plays an important role for fluid flow, not only for flow along its principal direction but also in the orthogonal direction.

Integrated interpretation of three-dimensional seismic and well-logging data reveals a prominent “pop-up” structure associated with the VLE 400 fault on the regional unconformity between the Eocene and Miocene in the VLE 196 field, Maracaibo basin, Venezuela. The VLE 400 fault family, an eastern splay of the left-lateral Icotea fault in the basin, played an important role in hydrocarbon migration and accumulation in the field. Hydrocarbons accumulated to the east of the fault but not to the west. The “pop-up” structure on the Eocene unconformity has a four -way dip closure, straddling the fault and extending to the west of the fault. Structures of the Misoa Formation, which is the main reservoir developed below the unconformity in the basin, differ from the structure of the unconformity. The structure of the Misoa Formation shows a tilted uplift of the eastern block of the fault dipping toward the east caused by thrust tectonic movements. Thrust movement and following strike-slip movements provided additional accommodation space to the west of the fault and generated expanded thickness of Eocene sediments compared to the area east of the fault. The thickness of the Misoa Formation east and west of the fault shows no significant changes. Expanded sediments overlie the Misoa Formation in the western block in lateral contact with Misoa sediments eastern block act as a lateral seal. Ductile movement of the Guasare Formation shale contributed to the lateral sealing of the fault against the reservoir rocks in the eastern block. Mobilization of the Guasare Formation modified the structure of overlying formations including the anticline of the Eocene unconformity. The growth strata provide useful information of reactivation of e xisting faults, especially subtle movements which are not recognized by conventional seismic interpretation. Growth strata isochrons shows subtle reactivation of the VLE 400 fault family during Miocene time.

The purpose of this study was to describe depositional and diagenetic characteristics of the (Jurassic) Smackover formation and subsequently identify and rank the quality of flow units within Grayson field, Columbia County, Arkansas. The field has production from the Smackover, a reservoir which consists mainly of highly altered peloidal grainstones. This was a four part study including a lithological analysis of ten cores, a petrographical study of 97 thin sections, a petrophysical study of reservoir properties from core analyses and borehole logs, and predictive mapping of quality ranked flow units across the field. Examination of the cores and thin sections revealed H1a as the main pore type in Grayson field, which was a hybrid of both depositional and diagenetic processes with dominantly interparticle pores. The lowest ranked reservoir quality corresponded to intraparticle and intercrystalline pore types, which occurred mainly in the wackestone/mudstone and packstone/wackestone facies. The highest ranked reservoir quality corresponded to the H1a pore type which occurred mainly in the grainestone/packstone facies 1 and 2. The reservoir quality maps identified the spatial distribution of the facies within the field, which could be used to locate zones for possible in-fill drilling. These results should aid in the economical development of Grayson field and other similar fields.

Thesis (M.S.)--West Virginia University, 2001. / Title from document title page. Document formatted into pages; contains xi, 149 p. : ill. (some col.). Includes abstract. Includes bibliographical references (p. 118-124).

Sixteen percent of New Zealand’s power comes from geothermal sources which are primarily located within the Taupo Volcanic Zone (TVZ). The TVZ hosts twenty three geothermal fields, seven of which are currently utilised for power generation. Ngatamariki Geothermal Field is the latest geothermal power generation site in New Zealand, located approximately 15 km north of Taupo. This was the location of interest in this project, with testing performed on a range of materials to ascertain the physical properties and microstructure of reservoir rocks. The effect of burial diagenesis on the physical properties was also investigated. Samples of reservoir rocks were taken from the Tahorakuri Formation and Ngatamariki Intrusive Complex from a range of wells and depths (1354-3284 mbgl). The samples were divided into four broad lithologies: volcaniclastic lithic tuff, primary tuff, welded ignimbrite and tonalite. From the supplied samples twenty one small cylinders (~40-50mm x 20-25mm) were prepared and subjected to the following analyses: dual weight porosity, triple weight porosity, dry density, ultrasonic velocity (saturated and dry) and permeability (over a range of confining pressures). Thin sections impregnated with an epoxy fluorescent dye were created from offcuts of each cylinder and were analysed using polarised light microscopy and quantitative fluorescent light microstructural microscopy. The variety of physical testing allowed characterisation of the physical properties of reservoir rocks within the Ngatamariki Geothermal Field. Special attention was given to the petrological and mineralogical fabrics and their relation to porosity and matrix permeability. It was found that the pore structures (microfractures or vesicles) had a large influence on the physical properties. Microfractured samples were associated with low porosity and permeability, while the vesicular samples were associated with high porosity and permeability. The microfractured samples showed progressively lower permeability with increased confining pressure whereas samples with a vesicular microstructure showed little response to increased confining pressure. An overall trend of decreasing porosity and permeability with increasing density and sonic velocity was observed with depth, however large fluctuations with depth indicate this trend may be uncertain. The large variations correlate with changes in lithology suggest that the lithology is the primary control of the physical properties with burial diagenesis being a subsidiary factor. This project has established a relationship between the microstructure and permeability, with vesicular samples showing high permeability and little response increased confining pressure. The effects of burial diagenesis on the physical properties are subsidiary to the observed variations in lithology. The implications of these results suggest deep drilling in the Tahorakuri Formation may reveal unexploited porosity and permeability at depth.

The hydrothermal scheduling problem aims to determine an operation strategy that produces generation targets for each power plant at each stage of the planning horizon. This strategy aims to minimize the expected value of the operation cost over the planning horizon, composed of fuel costs to operate thermal plants plus penalties for failure in load supply. The system state at each stage is highly dependent on the water inflow at each hydropower generator reservoir. This work focuses on developing a probabilistic model for the inflows that is suitable for a multistage stochastic algorithm that solves the hydrothermal scheduling problem. The probabilistic model that governs the inflows is based on a dynamic linear model. Due to the cyclical behavior of the inflows, the model incorporates seasonal and regression components. We also incorporate climate variables such as precipitation, El Ni\~no, and other ocean indexes, as predictive variables when relevant. The model is tested for the power generation system in Brazil with about 140 hydro plants, which are responsible for more than 80\% of the electricity generation in the country. At first, these plants are gathered by basin and classified into 15 groups. Each group has a different probabilistic model that describes its seasonality and specific characteristics. The inflow forecast derived with the probabilistic model at each stage of the planning horizon is a continuous distribution, instead of a single point forecast. We describe an algorithm to form a finite scenario tree by sampling from the inflow forecasting distribution with interstage dependency, that is, the inflow realization at a specific stage depends on the inflow realization of previous stages. / text

East Texas field, a giant U.S. oil-field, produced 5.42 billion stock-tank barrels from discovery in 1930 through mid-2007. The lower part of the siliciclastic Upper Cretaceous Woodbine Group is reservoir rock, and almost all production comes from the upper unit, the operator-termed Main sand. The field could produce 70 million stock-tank barrels (MMSTB) using current strategies, whereas 410 MMSTB of remaining reserves from the Stringer zone (lower unit), along with bypassed pay in both units and unswept oil, is possible. These favorable statistics have increased interest in reservoir characterization of the Woodbine, especially the Stringer zone. This study delineates sandstone geometry and interprets reservoir facies and heterogeneity of the Stringer zone and Main sand in northeast East Texas field. Additional objectives are to define key chronostratigraphic surfaces, such as flooding surfaces and unconformities, and to establish a realistic depositional model for the reservoir succession. To achieve these objectives, well log analysis, core description, and net-sandstone mapping of the Stringer zone and Main sand were conducted. According to sequence-stratigraphic and depositional-system analysis, the Woodbine Group is divided into two genetically unrelated units: (1) the highstand deltaic Stringer zone and (2) the lowstand incised-valley-fill Main sand. Principal reservoir units are Stringer 1 and Stringer 2 sands within the Stringer zone and the Main sand. Stringer 2, best developed in the southwest study area, is the most promising reservoir unit for new production. Well deepening and water-flooding in this more continuous and thicker sand are proposed to increase production in East Texas field. / text

Chemical flooding is one of the most difficult enhanced oil recovery methods and was considered a high-risk process in the past. Some reasons are low and uncertain oil price, high chemical prices, lack of confidence in performance of the chemical flooding process, long project life, and reservoir and process uncertainties. However, with significant improvement in simulation and optimization tools and high oil price, chemical flooding is feasible in terms of economical and carefully implemented design. Optimization of chemical floods requires complex integration of reservoir, chemical, economics properties and also drilling and production strategies. Many of these variables are uncertain parameters and many simulations are required to capture the effect of the uncertain and decision variables. These simulations could become very expensive and may not be feasible to consider all of the required simulation models. The goal of this research is the development of a methodology for optimization and design of chemical flooding of candidate oil reservoirs. We performed a comprehensive sensitivity study of reservoir and fluid properties that have significant influence on the oil production during the chemical flooding by performing a series of reservoir simulation runs. For performing the reservoir simulation runs, this study used the UT_IRSP platform and the multiphase, multicomponent, chemical flooding simulator called UTCHEM. During the study, UT_IRSP and UTCHEM have been modified by adding new modules, functions and variables. For example, a deviated well module was implemented in UTCHEM to study deviated wells. Deviated well module allows the users to introduce deviated wells in reservoir and import the well locations similar to Eclipse or CMG simulators. A time-dependent well schedule module was implemented in the UT_IRSP framework. This enhancement allows the well placement optimization studies to find the best time to add new wells, and change the status of the well for example from a producer to an injector in order to have an optimum development plan. An advanced post processing module was added to UT_IRSP in order to design, screen, and optimize complex cases for chemical enhanced oil recovery processes such as investigating the well patterns, well spacing, and type of the well (horizontal vs. vertical wells). An experimental design and response surface methodology with integrated economic model were utilized in this study to obtain the optimum design under uncertainties and have an optimal combination of the decision variables. This methodology is based on applying multi-regression analysis and ANOVA (analysis of variance) between the objective function (i.e. dependent variable, which is net present value (NPV) in chemical flooding) and other uncertain and process variables (independent variables). The economic analysis model used the discounted cash flow method to calculate net present value at the economic life of process, internal rate of return, and growth rate of return for each simulation case. Also the optimizer, OptQuest, is launched with a goal of maximizing the mean NPV. The range and the risk associated with the optimum design was studied using Monte Carlo simulation of objective function of the response variable and other independent variables. This methodology was applied for complex chemical flood cases such as well placement, change of status of wells as a function of time or well pattern and well spacing to investigate the best well scenario from recovery and economics point of view. / text

Fractures make both opportunities and problems for exploration and production from hydrocarbon reservoirs. It is always difficult to predict how to optimally produce a fractured reservoir due to the complexity and heterogeneity of fluid flow paths. The following behavior is seen in fractured reservoirs: early water breakthroughs, reduced tertiary recovery efficiency due to channeling of injected gas or fluids, dynamic calculations of recoverable hydrocarbons that are much less than static mass balance and dramatic production changes due to changes in reservoir pressure as fractures close down as conduits. These problems often lead to reduced ultimate recoveries or higher production costs. Polymer gels, in particular in-situ gels that can be placed deep into the reservoir, have been widely used for improved conformance control. In this dissertation, we aim to block the high-permeability zones, fractures in particular, with the microgels to increase the sweep efficiency by diverting the waterflood water to the low permeability zones that still contain unswept oil. vii Polyacrylic acid microgels can swell a thousand fold as the pH of the surrounding solution changes, with an accompanying large increase in viscosity. This pH trigger is simpler than chemical cross-linking and thus offers operational advantages. The ability of pH-sensitive polymers to block high permeability fractures is studied by performing several coreflood and batch experiments. The effect of different rock and salt minerals, polymer concentration, polymer salinity, and temperature on polymer performance is studied in this dissertation. Polymer microgels show excellent consistency in the presence of various salt minerals and in contact with different rock minerals. The placement of microgels into the fractures lowered the overall core permeability in all cases. In addition, polymer microgels were stable after being in reservoir for a month with conditions at 58°C. Consequently, using pH-triggered polymers for conformance control and reducing the permeability of high permeability areas in fractured reservoirs merit further investigations. These polymers are inexpensive and are easy to prepare. The polymer concentration, salinity and shut-in time could be set according to the desired PRF value, injectivity, propagation distance and reservoir mineralogy. / text

Surfactant-polymer (SP) and alkali-surfactant-polymer (ASP) flooding has gained little traction among different tertiary recovery strategies such as thermal and miscible gas flooding; however, many mature onshore reservoirs could be potential candidates. More than four decades of research has detailed technical challenges and successes through laboratory experimentation, chemical flood simulation, and some pilot projects, which have provided technical screening procedures to efficiently filter unfeasible projects. Therefore, technical understanding seems sufficient to advance projects through early development stages; however, a project value identification and realization process ultimately dictates project implementation in the oil and gas industry, with technical feasibility merely supporting overall valuation and project feasibility. A quick early screening methodology integrating important project valuation criteria can efficiently assess large numbers of projects. The relatively few studies detailing chemical flooding valuation from just an economic standpoint reflects the need for an integrated process-oriented framework for quick early screening valuation of chemical flooding opportunities. This study develops an integrated process-oriented framework for early screening and valuation, with an overall objective to quickly filter unfeasible projects based on valuation criteria, rather than technical feasibility alone. A reservoir-to-market model was developed, integrating information from laboratory experiments (phase behavior, core flood), field analogues (well performance and layout), facilities, rigs, costs, scheduling, and economics. Recently published ASP flood data of the central Xing2 area in Daqing, China was used for model inputs. A reservoir-to-market benchmark model for a typical mature onshore field was successfully built and tested, and could value projects using standard economic metrics (net present value, internal rate of return, value investment ratio, unit technical cost, and payback period). Model simplification was achieved through global sensitivity analysis. Using a mean-reversion oil price model, the oil price accounted for 98% of the total sensitivity. . Model efficiency was achieved through discretization of input parameter uncertainties, which sped the screening process. Decision-making between model alternatives given information and different states of nature was performed through decision-tree techniques based on overall project valuation. Overall, this study was novel and provided benefit as a robust, integrated process-oriented framework for chemical EOR project screening, valuation, and decision-making. / text