Hydraulic fracturing operations affect reservoir flow dynamics and increase production in unconventional tight reservoirs. The control of fracture growth and geometry presents challenges in formations in which the boundary lithologies are not highly stressed in comparison to the pay zone, thus failing to prevent the upward migration of fractures. Several factors influence the growth and geometry of fractures, including reservoir, wellbore, and fluid/proppant parameters. Successful results require a thorough knowledge of reservoir parameters, including stress distribution and the appropriate use of corresponding wellbore components and fluid/proppant. The success of a hydraulic fracturing treatment is highly correlated with control of the created fracture geometry.This paper discusses a study in which a numerical fracture model is used to design the fractures in a tight oil reservoir. Fracture treatment designs include the selection of fracturing fluids, additives, proppant materials, injection rate, pumping schedule, and fracture dimensions. Using the fracture model, a statistically representative synthetic set of data is generated for each parameter to build data-driven models.The performance of the data-driven models is validated by comparing the results to a numerical model, and considering the significance of parameters, including size, number, location, phasing angle of perforations, fluid and proppant type, rock strength, porosity, and permeability on the fracture design optimization using various fracture models. Data-driven predictive models are generated by using neural networks (NN) and support vector machine (SVM) algorithms. Optimum values of model parameters are also investigated.The SVM and NN models are used to optimize the fracturing treatments per well, and are evaluated based on accuracy and computational complexity. Based on the performances of the models, model parameters are adjusted to obtain fit-for-purpose well-based hydraulic fracturing models.

Economic production from tight-oil formations requires the same hydraulic fracturing techniques used during production of shale gas, and the same horizontal well technology is often used. Tight-oil formations are heterogeneous and greatly vary over relatively short distances. Thus, even in a single horizontal drilled hole, the amount recovered can vary, just as recovery within a field or even between adjacent wells can vary. This can make evaluation and decisions regarding profitability difficult. In relatively thinner pay zones, horizontal or slanted wells can be successful in terms of productivity as long as robust reservoir management is applied with well designs that consider both geology and operational control variables. Thus, it is crucial to understand every control and uncertainty parameter to help maximize efficiency and recovery within such systems.Robust commercial optimization and uncertainty software is coupled with a full-physics commercial simulator that models this phenomenon to investigate the significance of major parameters on the performance of horizontal wells in tight-oil formations.Slanted wells provide more flexibility and access to pays than vertical wells drilled from the same surface location, which is of more significance in tight formations because of less communication between zones. The results of the study not only confirm this, but also show the increased value provided using slanted wells compared to vertical wells in tight formations. The study also illustrates the significance of each optimization and uncertainty variable in terms of the success of recovery from slanted horizontal wells in tight formations.The results and sensitivities are compared and discussed considering a comprehensive literature review of recycling gas-condensate reservoirs using different process optimization methods. The significance of all major parameters is outlined using tornado charts to serve as a practical example for optimization of similar future applications. Vertical, Slanted and Horizontal WellsThe technology for drilling horizontal wells to increase reservoir drainage areas has been used for many years. As directional drilling technology has advanced, the cost of drilling directional wells has been reduced, and the accuracy of drilling has been significantly improved.

Asset managers constantly seek to determine how wells are performing to assess performance of long-term strategy and to achieve expected results. For this purpose, periodic production tests are performed to measure individual flow contribution to total measured platform production in addition to other measurements, including BSW, gas-oil ratio, pressure, and temperature. Daily well production is estimated by back-allocating production measured at fiscal meters to individual wells based on the well’s production potential validated during tests. This paper presents an alternative system for measuring individual well-oil production based on a neural network and online correlation logic using data from sensors, well tests, and simulations. This system permits a closer right-time monitoring of the wells by enabling readings to be taken more frequently and by minimizing the intrinsic estimation errors that normally arise when doing back-allocation of well production based on performance of other wells. This paper describes a methodology for data selection, sensor validation analysis, modeling, online implementation, and quality control of the results. The main benefit of this implementation has been to quickly identify production deviations above or below well potential and to identify and adjust the variables that affect these deviations. The combination of high volumes of measured data that automation technology enables and historical values of testing data made it possible to implement this smart solution where data is constantly transformed into information. This information allows the engineers to analyze and associate results and transform them into events of knowledge. This methodology can be applied to any asset where time and operational constraints do not permit the testing of wells on a daily basis or where it is too expensive to justify the installation of multiphase meters and where a high level of automation is available.

The goal of a field development optimization process, or workflow, is to investigate various options and determine a course of action that will deliver the largest expected value from an asset. The analysis is often complicated by uncertainty in important inputs. Ideally, operators desire workflows and tools that integrate reservoir engineering and optimization principles in a fast-solving model that can be used to explore the full range of the uncertain inputs. This need is acute in the screening and concept selection stage where the primary objective is to determine the sensitivity of competing concepts to the sources of uncertainty. In these early stages, model results can be used to determine whether additional information should be collected, and to narrow down the number of competing options.The objective of this research is the development of a workflow and tool that integrates reservoir response surfaces within a project optimization model that contains real options. The incorporation of real options is critical because a static view of capital investment and facility constraints causes a systematic undervaluation and can introduce error to development decisions. The new workflow and integrated reservoir-economic optimization tool developed in this research leverage methods and engineering work products that are already known to industry, for example experimental design (ED) and response surface methods (RSM).A demonstration is provided for a gas flood project using a stylized reservoir. Specifically, we investigate the selection of initial well configurations and injection capacities while simultaneously accounting for the (real) option to update these decisions after production information is acquired in the early periods of production. The workflow is used to optimize the development of a gas flood. As a second step, the workflow is used to solve a value of information problem. BibliographyLake, Larry W. 1989. Enhanced Oil Recovery, Englewood Cliffs, New Jersey: Prentice Hall.

Water alternating gas (WAG) injection has been widely used for the last 50 years throughout the world. The typical improved oil recovery (IOR) potential for WAG injection compared with water injection is 5 to 10%. It was originally intended to improve sweep efficiency during gas flooding, with intermittent slugs of water and gas designed to follow the same route through the reservoir. Mechanisms in WAG injection include microscopic effects, particularly in cases where three-phase flow and hysteresis are important for the IOR effect. Injection of gas usually aids an ongoing waterflood, and finding technical and commercial methods to reduce gas costs would be useful. Water injection alone tends to sweep the lower parts of a reservoir, while gas injected alone sweeps more of the upper parts of a reservoir because of gravitational forces. Gas represents a large fraction of the total cost, making WAG injection an expensive method. Thus, optimizing WAG injection is not only crucial in terms of recovery but also economics, especially where gas is expensive and/or limited. In this study, the significance of key components in a WAG injection process on SPE's 5th Comparative Solution Project (CSP) is presented that models the WAG process through a pseudo-miscible formulation by means of coupling a full-physics reservoir simulator with commercial optimization and uncertainty software. The results are analyzed and presented in a comparative manner by means of tornado charts showing the significance of each decision and uncertainty variable.