Luigi Saputelli

State of the Art of Artificial Intelligence and Predictive Analytics in the E&P Industry: A Technology Survey

Artificial intelligence (AI) has been used for more than two decades as a development tool for solutions in several areas of the EP (b) approximately 50% of respondents declared they were somehow engaged in applying workflow automation, automatic process control, rule-based case reasoning, data mining, proxy models, and virtual environments; (c) production is the area most impacted by the applications of AI technologies; (d) the perceived level of available literature and public knowledge of AI technologies is generally low; and (e) although availability of information is generally low, it is not perceived equally among different roles. This work aims to be a guide for personnel responsible for production and asset management on how AI-based applications can add more value and improve their decision making. The results of the survey offer a guideline on which tools to consider for each particular oil and gas challenge. It also illustrates how AI techniques will play an important role in future developments of IT solutions in the E&P industry. Introduction While there is hardly a rigorous definition of the term artificial intelligence (AI) that is unequivocally accepted, the tools of AI and its intended uses have been well studied for decades and many applications have appeared. Loosely speaking, AI is the capability of machines (usually in the form of computer hardware and software) to mimic or exceed human intelligence in everyday engineering and scientific tasks associated with perceiving, reasoning, and acting. Since human intelligence is multifaceted, so is AI, comprising goals that range from knowledge representation and reasoning, to learning, to visual perception and language understanding (Winston 1992). AI techniques have been present in the E&P industry for many years. A quick literature search reveals application of AI in SPE scientific and engineering papers as early as in the 1970s. There are numerous references about the applications of neural networks, fuzzy logic, genetic algorithms, expert systems, and other artificial techniques in the resolution of problems in diverse areas, such as reservoir simulation, production optimization, process control, and fault detection and diagnosis, among many others. AI is an area of great interest in the E&P industry, mainly in applications related to production control and optimization, proxy model simulation, and virtual sensing. The most popular techniques are artificial neural networks, fuzzy logic, and genetic algorithms, with interesting developments in hybrid and nontraditional techniques. There has been recent increase in such AI-based commercial applications for production management. While the full impact of such applications is still being realized, there are already solutions in the market with a positive impact in the E&P industry.

Design Optimization of Horizontal Wells With Multiple Hydraulic Fractures in the Bakken Shale

Hydraulic fracturing is currently the completion method of choice in most tight reservoirs; however, the ultimate performance of fractured wells is severely affected by the interfering effects inside the fracture and interfractures. Previous simulation studies investigated the effects of well spacing and fracture length on well productivity in low-permeability oil and gas reservoirs. It was shown that the most important parameters for determining the optimum fracture length are the formation permeability and the stimulated reservoir volume (SRV). Although a number of studies have examined the performance of horizontal fractured wells and the fracture geometry effect, fracture spacing and intersecting angles in vertical and horizontal wells should be further investigated. This study presents the results of a tight oil reservoir analogy. Reservoir parameters considered include local rock stresses, rock compressibility, absolute and relative permeability, and porosity. The well-completion parameters included fracture length, height, width, conductivity, number and spacing between fractures, fracture intersecting angle, and cased- vs. openhole completion. Fracture modeling considered rigorous description of the hydraulic fracture properties and finite difference reservoir modeling. Economically attractive reserves recovery was modeled through multiple fracture placements in a 10,000-ft horizontal well. Numerical simulation showed that oil recovery increased between 8 to 15%, while net present value (NPV) increased 8 to 24%, as the number of fractures increased. Based on the critical assumptions in the study (permeability, natural fracture distribution, and stress orientation), an optimum number of fractures was identified. Openhole completions provided better performance in most cases, and recovery was greater for a higher number of contributing perpendicular vs. longitudinal fractures. The results of the study hopefully can be used to improve the understanding of the role of fracture geometry, spacing, and open/cased-hole completion strategy to enhance an operator’s optimum completion design.

Waterflood Recovery Optimization Using Intelligent Wells and Decision Analysis

The many critical decisions in a waterflood recovery process range from well number, architecture, and completion configuration to scheduling and facilities capacity planning. Project success is also affected by subsurface uncertainties, such as reservoir heterogeneity and compartmentalization, as well as surface events including equipment uptime and availability. Managing the complexities of a waterflood recovery is traditionally a sequential and intermittent process in terms of data acquisition, modeling, and workflows. However, traditional project planning, execution, and monitoring may help an operator reach production targets and budgets, but these activities do not necessarily align technical and business goals to create a comprehensive development plan that increases the probability of optimal business success. This paper describes a process to maximize long-term economic return by optimizing key decisions, such as well number, placement, and use of intelligent wells and operating schedules, as well as evaluating surface capacity parameters, such as export pipeline diameter and pump inlet pressure. An integrated asset model (IAM) simultaneously simulates the reservoir flow, wells, and facilities, and calculates the economics associated with each potential decision scenario. An automated optimization workflow is presented as an efficient procedure to handle a large number of decisions. A well-known industry waterflood case study is discussed in the paper. A development plan that includes processes for maximizing oil recovery and economic return by implementing intelligent-well operational procedures is presented. The results from this work for the case examples presented show that the number and location of the wells are more significant factors in the success of the field development plan than the selection of export pipeline diameter or pump inlet pressure. The optimum selection of well number, location, and producer to injector ratio could lead to an increase of 136%, when compared to the optimized primary recovery case. In the secondary recovery example, a proactive operating scheduling of intelligent well valves improved the waterflood project value by 130% with respect to the case with no intelligent completion; this was achieved by the efficient reduction in produced (~9%) and injected (~8%) water, and the increase of produced oil (~3%). Applications of this workflow and integration by professionals from multiple surface and subsurface disciplines provide a means of minimizing the risk of investment and setting optimal surface parameters that maximize production while reducing capital and operating costs. In this way, the return on investment of the project is optimized. The decision framework presented may be applied in any type of hydrocarbon field, from early to mature development, and may be extended to other important decision variables, including the selection of the reservoir recovery process and market strategy.

Real-time Production Optimization in the Okume Complex Field, Offshore Equatorial Guinea

This paper addresses real time production optimisation (RTPO) by implementing a digital framework using several available commercial applications (IPM and IFM) with respect to the Okume complex field in offshore Equatorial Guinea. There are two primary challenges; gaslift is continuously allocated in frequently changing field conditions while minimising production losses as well as understanding and maximising the field production plateau. The operators on the field are required to maintain and increase mature asset production by optimising the artificial lift system and managing the well operating constraints which requires a deep understanding of the well performance. A full picture of the well performance comes from the well test validation and updates as well as accurate well rate estimation and volume allocation. These steps take a large amount of time out of the engineers’ day and away from the technical work to carry out the necessary repetitive tasks. There is therefore a need for tools or methodologies which can continuously provide optimisation settings for the asset. RTPO also assists operators in better understanding the gap between the potential and actual production and helps to identify the potential changes in the process operations.

Digital Oilfield Technologies Enhance Production in ESP Wells

Two common challenges to sustain production in North Kuwait (NK) include (1) management of artificial lift equipment run life and (2) reduction of production decline through pressure maintenance by water injection. Standard electric submersible pump (ESP) management guidelines promote practices to protect ESP run life while sacrificing well production rates. However, using operating rates that are not always safe can permit optimum reservoir recovery. Better management of well production targets is required to maintain safe ESP operation ranges while pushing the limits to increase production. A methodology has been developed to continuously optimize the well production potential without surpassing its safe operating envelope, keeping pump intake pressure (PIP) above the bubble point, ensuring optimum delivery at the surface, and minimizing pump downtime. This workflow was developed to assist the engineers in expediting data analysis and interpretation. This paper describes the developed production optimization methodology, which enables short- and long-term production enhancement while honoring reservoir, ESP, and surface constraints. The process involves historical and real-time data collection and production test interpretation. From this analysis, the best estimations of short- and long-term well production potentials are identified, and the correspondent action plans are determined. The developed workflow and the result of such an iterative production optimization process are illustrated. To date, as a result of this effort, an average of 8% in production gains has been certified in a number of wells in Sabriyah, NK.