Gustavo Carvajal

Engineering Workflow for Probabilistic Assisted History Matching and Production Forecating: Application to a Middle East Carbonate Reservoir

Traditional reconciliation of geomodels with production data is one of the most laborious tasks in reservoir engineering. The uncertainty associated with the great majority of model variables only adds to the overall complexity. This paper introduces an engineering workflow for probabilistic assisted history matching that captures inherent model uncertainty and allows for better quantification of production forecasts. The workflow is applied to history matching of the pilot area in a major, structurally complex Middle East (ME) carbonate reservoir. The simulation model combines 49 wells in five waterflood patterns to match 50 years of oil production and 12 years of water injection and to predict eight years of production. Initially, the reservoir model was calibrated to match oil production by modifying permeability and/or porosity at well locations and by fine-tuning rock-type properties and water saturation. The second level history match implemented two-stage Markov chain Monte Carlo (McMC) stochastic optimization to minimize the misfit in water cut on a well-by-well basis. While relative to evolutionary algorithms or the ensemble Kalman filter (EnKF), the McMC methods provide a statistically rigorous alternative for sampling posterior distribution; when deployed in direct simulation, they impose a high computational cost. The approach presented here accelerates the process by parameterizing the permeability using discrete cosine transform (DCT), constraining the proxy model using streamline-based sensitivities and utilizing parallel and cluster computing. While probabilistic assisted history matching (AHM) successfully reduced the misfit for most producing wells, the computational convergence was sensitive to the level of preserved geological detail. The optimal number of representative history-matched models was identified to capture the uncertainty in reservoir spatial connectivity using rigorous optimization and dynamic model ranking based on forecasted oil recovery factors (ORFs). The reduced set of models minimized the computational load for forecast-based analysis, while retaining the knowledge of the uncertainty in the recovery factor. The comprehensive probabilistic AHM workflow was implemented at the operator’s North Kuwait Integrated Digital Oilfield (KwIDF) collaboration center. It delivers an optimized reservoir model for waterflood management and automatically updates the model quarterly with geological, production, and completion information. This allows engineers to improve the reservoir characterization and identify the areas that require more data capture. Introduction As part of a comprehensive strategy to transform the Kuwait Oil Company (KOC) through the application of digital oilfield (DOF) concepts, KOC initiated an assessment of the major Sabriyah-Mauddud (SaMa) reservoir for conversion to an integrated digital oilfield (iDOF) master platform, with the goal of increasing effectiveness through automating work processes and shortening observation-to-action cycle time. The group of nine first-generation production engineering workflows focuses on production and operational activities and was launched at KwIDF in 2012. The workflows are introduced in Al-Abbasi et al. (2013) and described in greater detail in Al-Jasmi et al. (2013) and references therein. With the vision to drive future KOC operations to the next level of excellence and to realize a large return on the investment in iDOF, the operator’s senior management endorsed the development of a family of advanced integrated asset management (IAM) workflows, referred to as “smartflows,” to optimize and integrate the subsurface models with well models and network surface systems in various time horizons.

Next Generation of Workflows for Multilevel Assisted History Matching and Production Forecasting: Concept, Implementation and Visualization

Traditional reconciliation of geomodels with production data is one of the most laborious tasks in reservoir engineering. The uncertainty associated with the great majority of model variables only adds to the overall complexity. This paper describes the conceptualization, implementation, and visualization characteristics of the multilevel assisted history matching (AHM) technique that captures inherent model uncertainty and allows for better quantification of production forecasts. The workflow is applied to history matching of the pilot area in a major, structurally complex Middle East (ME) carbonate reservoir. The simulation model combines 49 wells in five waterflood patterns to match 50 years of oil production and 12 years of water injection and to predict eight years of production. Initially, the reservoir model was calibrated to match oil production by modifying permeability and/or porosity at well locations and by fine-tuning rock-type properties and water saturation. The second level history match implemented two-stage Markov chain Monte Carlo (McMC) stochastic optimization to minimize the misfit in water cut on a well-by-well basis. The inversion process is dramatically accelerated by the efficient parameterization of permeability, constraining the proxy model using streamline-based sensitivities and using parallel and cluster computing. The optimal number of representative history-matched models was identified to capture the uncertainty in reservoir spatial connectivity using rigorous optimization and dynamic model ranking based on forecasted oil recovery factors (ORFs). The reduced set of models minimized the computational load for forecast-based analysis, while retaining the knowledge of the uncertainty in the recovery factor. The comprehensive probabilistic AHM workflow was implemented at the operator’s North Kuwait Integrated Digital Oilfield (KwIDF) collaboration center. It delivers an optimized reservoir model for waterflood management and automatically updates the model quarterly with geological, production, and completion information. This allows engineers to improve the reservoir characterization and identify the areas that require more data capture. Introduction With the vision to transform the Kuwait Oil Company (KOC) through the application of integrated digital oilfield (iDOF) concepts and drive the future KOC operations to the next level of excellence, the operator’s senior management endorsed the development of a family of advanced integrated asset management (IAM) workflows, referred to as “smart flows,” to optimize and integrate the subsurface models of the major Sabriyah-Mauddud (SaMa) reservoir with well models and network surface systems in various time horizons. The objective is to increase the effectiveness through automating work processes and shortening observation-to-action cycle time. The group of nine first-generation production engineering workflows focuses on production and operational activities and was launched at KwIDF in 2012. The workflows are introduced in Al-Abbasi et al. (2013) and described in greater detail in Al-Jasmi et al. (2013) and references therein. The second generation of smart flows combines subsurface waterflooding optimization (SWFO) (Khan et al. 2013), integrated production optimization (IPO), and simulation model update and ranking (SMUR). The preceding publication, Maucec et al. (2013), briefly discusses the outstanding challenges of the model reconciliation and history matching and reviews the recent approaches the oil industry is taking to quantify the uncertainty and increase the accuracy of reservoir models. Moreover, in Maucec et al. (2013), the engineering concepts of SMUR smart flow are described in detail, combining the processes of building the high-resolution geocellular model and the associated reservoir simulation model, leading into a history-matching case study of the SaMa field. The design of the SMUR smart flow is leveraged with the technology for

Introduction to Digital Oil and Gas Field Systems

World energy demand will grow from about 550 QBTU in 2012 to 850 QBTU in 2040 according to 2016 projections from the International Energy Agency (IEA). Although renewable energy sources will grow by a large percentage, petroleum-based liquids (oil) and natural gas will continue to be the largest contributors to energy utilization by the worlds population, representing about 55% of the total in 2040. As any current oil and gas production naturally declines, the continued growth of petroleum fuels will be made possible only by forward leaps in technology in finding, drilling, and producing those resources more efficiently and economically. One of the great stories in oil and gas production is the industrys implementation of new digital technologies that increase production for less unit cost. This “revolution” of the “digital oil field” is the subject of this book.

Data Filtering and Conditioning

This chapter presents the major features of such a system, which includes (1) data processing, (2) basic error detection, conditioning, and alerting, (3) well and equipment status detection, (4) advanced validation, and (5) workflow-based conditioning. This chapter is a condensed tutorial on how to validate and condition data appropriately for digital oil field (DOF) systems. The process flow of the chapter is summarized in Fig 3.1, which has the main steps for a DOF data validation and conditioning system. You can also reference a myriad of specialty material on signal processing (e.g., Vetterli et al., 2014 ), which is not covered here.

Smart Wells and Techniques for Reservoir Monitoring

This chapter introduces concepts associated with smart well technology and its application to maximize the oil recovery factor and improve the financial indicator for the oil company. In 1997, the first successful completion incorporating permanently installed, downhole pressure and temperature measurements integrated with remotely controlled, high fidelity flow control valves was installed in a well in the Norwegian (in 1997) sector of the North Sea. Konopczynski and Ajayi (2008) stated that this event marked the genesis of the intelligent well era. The use of intelligent well technology has “crossed the technology adoption chasm” in many regions over the past decade as oil and gas producers have increasingly incorporated this technology in field developments to capture the benefit of enhanced reservoir management that intelligent well technology delivers. Nowadays, technical challenges remain without a direct answer particularly with control strategies to operate the valves during the water or gas breakthrough. In this chapter, the readers are guided with technical aspects to optimize oil production with smart wells.

Workflow Automation and Intelligent Control

The previous two chapters discussed how we must manage, condition, and analyze all the data gathered in digital oil field (DOF) operations. Next, we look at how the data can be used to automate workflows and provide intelligent control of equipment used in oilfield operations and DOF systems. This chapter introduces concepts associated with process control and then explains how process control is used in the context of automated DOF workflows. The subsequent sections describe these engineering components of automated DOF workflows: virtual multiphase metering, smart production surveillance, well-test validation, and well diagnostics. As a product of the automated workflows, advisories and tracking actions are generated for engineers.

The Future Digital Oil Field

The oil and gas (OG Murray and Neill, 2017 ). However, the changes coming over the next few years have significant potential for even bigger step changes. Gartner (2016) published an extensive report on the opportunities for digitalization of upstream O&G. The report covered the projected operator impact for drilling and wells, field development, and production and operations. In the report, Exxon Mobils chief computational scientist is quoted as saying, the “oil patchs digital transformation will be comparable to horizontal drillings tech revolution” ( Endress, 2017 ) (which, of course, was the game-changing technology that enabled the shale gas revolution, among other major industry achievements). Furthermore, the report states that “…he predicts the digitalization of oilfield equipment and operations will continue for the foreseeable future, due to future competitive advantages and untold economic value” ( Endress, 2017 ). This final chapter highlights a few of the exciting technologies that are in development or are being envisioned for the digital oil field (DOF) of the future.

Transitioning to Effective DOF Enabled by Collaboration and Management of Change

Previous chapters presented a suite of technologies related to the digital oil field (DOF) and pointed to examples of industry investment in sensor, communication, and automation technologies. But technology is only part of the requirements for effective DOF. The chapter cites that the rate of return on DOF investments in technology is often limited to less than 25%; 75% of expected value can only be achieved through the implementation of DOF with respect to work processes, competency, and role transition and how people work and collaborate using technology. This chapter introduced collaboration and work processes as critical components of DOF. This chapter presents details on challenges in delivering high value through high-performing teams in DOF, and then discusses the value chain of components and characteristics for success in change management and collaboration. These components are: (1) the physical space, the collaboration work environment (CWE); (2) team composition and roles; (3) interskill and team collaboration through change management and work processes; and (4) competency development and sustainability.

Instrumentation and Measurement

One primary reason for the recent proliferation of digital oil field (DOF) is the proliferation of DOF infrastructure. As instrumentation, well-control systems and supervisory control and data acquisition (SCADA) have advanced and become less expensive they have been more broadly applied throughout the oilfield. As they become more broadly applied, then more data and control is available to use in surveillance, automation, and optimization activities that are the hallmark of DOF. This chapter presents details for the prevailing trends for well instrumentation, wellhead control, and SCADA systems. It first presents some general trends in each category. The asset control network discussed here includes field instrumentation, field control devices, telemetry, SCADA systems, and data historians. Engineers and operators engaged in companies implementing DOF have become very familiar with the growth of these systems discussed below in oil field operations and conversant with data-driven production tools. With concomitant growth of Big Data and mobile systems, field operators and engineers are able to connect to the field sensor and control systems from multiple systems, for example, networked computers, tablets, phones, etc. from any location in order to make near real-time decisions.

Integrated Asset Management and Optimization Workflows

Digital oil field (DOF) systems conventionally have focused on wells, production, and operations. However, DOF is expanding its footprint into field decisions and management. Thus, the integration of the production systems’ models with the reservoir models is growing in order to optimize production and recovery. DOF systems now deploy three-dimensional (3D) coupled subsurface and surface models that, when calibrated (i.e., history-matched), provide short- and long-term forecasts of asset production and performance. The main objective of this chapter is to provide an overview of the modern integrated asset modeling (IAMod) practices and outline techniques and workflows for optimization and decision-driven forecasts of DOF systems that is integrated asset management (IAM), including their uncertainty. This chapter introduces the engineering principles and technology concepts of IAMod and optimization and the use of the models for IAM. The IAMod developments were introduced in the EP Liao and Stein, 2002 ), combining subsurface models with surface production facilities and networks. They are now becoming a standard means of modeling entire oil and gas assets with the objective to optimize existing facilities and to plan enhancements to production (wells and facilities).