Schedule | SPE Workshop: Drilling Systems Automation, A Systems Perspective | 29-30 October 2019 | San Antonio, TX
Monday, October 28
Data driven modeling is becoming a key differentiation to unlock higher recoveries from existing fields as well as identify new opportunities. The availability of data and democratization of these advanced algorithms is changing the landscape of subsurface workflows – helping create as well as improve existing ones. We are in an exciting phase in the industry where access as well as ease of using these advanced tools is transforming decision making in organizations.
In this course, we will start by introducing advanced analytical tools and techniques - machine learning and data mining algorithms used to identify of trends and patterns in any given dataset and predict future trends. We will showcase how each of these tools and techniques have been successfully applied to subsurface data - formation evaluation data, well testing data, reservoir data as well as data from surface facilities. We shall also present case studies of how integration of this seemingly disparate data can be done through new workflows that help identify opportunities to increase recovery. Finally, we will draw important distinctions between the more traditionally used forward models (physics-based approach such as reservoir simulation) and these statistics-based models. Using a case study that demonstrates integration of these two approaches, we shall conclude by a drawing out a framework for integration of these tools in your existing workflows.
Tuesday, October 29
In an interconnected world, where topside drilling equipment and downhole sensors and actuators work in concert, the automation of the drilling process should be one-step closer to reality. It is important to understand that sensor technology is evolving and that sensors for automation are different in terms of quality, reliability, and capability from sensors for manual drilling. A sensor in automation is not simply a transducer – it can be a smart digital sensor with an embedded server; it can be a virtual sensor derived from a physics-based model; it can be a redundant component in a sensor network. This session explores facets of sensors in drilling automation systems: in fusing data from different sensors, in coupled sensor-actuator systems, in transforming conventional sensors, to function in automation capable systems.
|Leveraging Sensor Data Fusion in Drilling Automation||Pradeep Ashok, University of Texas at Austin|
|Sensors and Actuators in Automatic Trajectory Drilling Systems||Christian Hansen, Baker Hughes, a GE company|
|Making the Dumb Sensor Smart||Soren Oydna, MHWirth|
The topic of digital twins has gained prominence in business literature recently, and the term "digital twin" is currently at the top of the Gartner hype cycle. However, is there something fundamentally new under the hood, or is this simply an accelerated evolution of software modeling? Low cost sensors, high fidelity physics models, machine learning, and high performance computing can now be integrated at ever-larger scales and run in real time. This reduces the need to make performance and functionality trade-offs, providing engineers with powerful diagnostic and prognostic tools. However, this comes with increasing complexity and a multi-domain systems integration challenge. How is the industry responding to this opportunity? What is hype, what can realistically be done with digital twins, and what is the outlook?
|Extracting Value from Data Using an Industrial Data Platform to Provide a Foundational Digital Twin||Per Arild Andresen, Cognite|
|Digital Twin Enables Testing, Training and Validation of Drilling Operations in a Virtual Environment - Allows for Measuring the Unmeasurable||Karl Erik Thoresen, National Oilwell Varco|
Open Algorithm Examples and Value Added
Data Analytics and AI are now a major theme in most of the business optimization discussions these days. However, are they mainly limited to abstract vision statements rather than practical applications? In this session, experts will discuss some practical examples of data analytics and AI that resulted in demonstrated business value, and discuss the very real hurdles that need to be overcome to capitalize on data analytics and AI.
|Building an Applied Drilling Data Analytics Platform - Why do we Fail More Than we Succeed?||Nathan Zenero, Teradata|
|Leveraging Real-Time Data and Intelligent Analytics to Improve Drilling Performance||Reza Asgharzadeh & Ali Karimi, Occidental Oil & Gas|
|Enhancing Real-Time Drilling Analytics System with Digital Transformation Strategy and Machine Learning Methodology||Kate Ruddy, Anadarko Petroleum|
Advancements in real-time streaming analytics are being leveraged to improve the drilling industry. These tools are being used to drive efficiency, consistency, and accuracy as well as enable better remote support. This session explores how streaming analytics are being used inside and outside of our industry with the goal to identify future opportunities. Key questions that will be discussed include: how to enhance analytics when the available sensor information is not enough, what real-time decisions can be enabled through analytics, and what emerging technologies will create additional value.
|Automatic Prediction and Mitigatation of Dysfunctions of Drilling Data Using Streaming Analytics||Philippe Herve, Sparkcognition|
|Distributed Analytics Leverages Sensor Location to Optimize Remote Decision Making||Julien Marck, Halliburton|
|Improving Drilling Parameter Selection Through Data Mining and Machine Learning||Russell Whitney, Corva|
National Oilwell Varco
Performance Drilling Technology
Wednesday, October 30
Drilling is a complex collection of sensors, technologies, processes, and human interactions that collectively deliver a completed oil/gas well. Advanced industries have adopted a systems engineering approach that defines the architecture of the interrelated systems and designs the life cycle of these systems of interest. Systems architecture maps the relationship of the systems of interest while recognized standards are applied to map the hierarchy of data from sensor to enterprise. Experts will share applications, what is possible, and discuss what is next in the application of systems engineering in the complex systems of systems drilling environment.
|Right Sizing Systems Engineering in Drilling – An Aerospace Perspective||Douglas Zimpfer, Shell|
|Systems Engineering and Systems Modeling for Effective Well Construction Systems||Jason Baker, Transocean|
|Where is My System? – A Matter of Perspective||Hans-Uwe Brackel, Baker Hughes, a GE company|
The need to calculate either large sets of data or high-speed data is becoming more critical on rig sites for advanced automation and data analytics. Unfortunately rig sites are not located in the best urban areas for high speed data communication. Most of the time, satellite bandwidth is the primary means of data communication. Experts will discuss the benefits and risks associated with deploying both cloud-systems and edge-systems at rig sites.
|The Operator's View: Whatever Works Best on My Rigs||John Willis, Occidental Oil & Gas|
|Platform as a Business, Providing a Digital Solution Suitable for the Oil Industry||Tony Pink, National Oilwell Varco|
|Cloud, Edge, or Both?||Sean Halloran, Ensign Energy Services|
Introduction to Systems Engineering Verification and Validation in Highly Automated Industries
High technology industries (including aerospace and commercial aviation) that rely upon data accuracy in displays and control systems for performance and safety apply systems engineering verification and validation processes. Paul will describe the application of this well established process in other industries and highlight opportunities for application to drilling systems for sound designs, implementation, and maintenance leading to reductions in cost and improvements in quality, avoiding unintended disasters.
Drilling systems automation is neither an all (autonomous) nor a nothing (manual) proposition. Automated industries, e.g. aviation, have taken and applied the work of T Sheridan & R Parasuraman through the four cognitive functions of acquire, analyze, decide and act, developing a matrix showing levels of automation taxonomy (LOAT) at each functional stage. This matrix, adopted by the Drilling Systems Automation Roadmap industry initiative, enables a transition mapping from current state to future state, and clarifies where humans should be included in an automated world. Experts will present examples and the outlook for human participation in new automated drilling loops.
|The Evolution of the Land Driller's Role: From a Doer to an Automated System Supervisor||Ariel Torre, Precision Drilling|
|Experiences Transitioning Human Roles While Advancing Drilling Automation Applications||Matt Isbell, Hess Corporation|
|Managing Change for Drilling Automation||David Rowatt, Schlumberger Land Rigs|
The advent of disruptive technologies and capabilities in our daily lives and in business by companies such as UBER, SpaceX, and Tesla allow us to witness a change in the way we deliver data and things never-before-seen in our human history.
|It's Time for Digital to Scale||Tommy Inglesby, Accenture|
|Satellite Monitoring of the Permian||Deven Desai, Planet|
|NASA is Sending a Drone to Mars in 2020||Stayne Hoff, Aerovironment|
|Unmanned Cargo Delivery||Steven Athanas, Swanson Group Aviation|
We will close the workshop with an opportunity to reflect on what we have heard and learned. We will also discuss what topics were not covered and which may be suitable for future workshops.