The Industry of the Future: What Does It Look Like?
You have access to this full article to experience the outstanding content available to SPE members and JPT subscribers.
As new technologies, environmental concerns, and consumer priorities disrupt the status quo of energy operations, industry has begun to rethink how it must proceed in order to remain as vital to global development in the future as it is today. But what does rethinking entail? Does the industry need to transition much to adapt to a new reality, or will success in the future be a simple matter of changing perceptions? How much will industry need to continue leveraging innovative technologies to survive in the future? Those questions have been at the forefront of several discussions held in recent months.
In September, Baker Botts and the Center for Energy Studies at Rice University’s Baker Institute for Public Policy hosted its Global Energy Transitions Summit. The summit described a vision of an energy industry in flux. Panels of industry executives examined how economics, policy, and technology will drive change across energy markets.
At one of the panels from the summit, Deborah Byers, managing partner of Ernst and Young’s (EY) Houston office, described the energy company of the future as energy “reimagined” in three dimensions: future demand, the future of operations, and the future of the workforce. US shale production has spearheaded a global abundance of natural gas. Byers said that any ramp up of demand of natural gas will come from the low cost of supply.
However, as developing countries have industrialized, she said industry has seen strong advancement in efficiencies, whether it’s lighter materials for products or greater conservation of natural resources. A greater focus on efficiency, which may manifest in less fuel-dependent technologies and stricter climate policies, may lead to a decoupling of energy intensity that affects how industry proceeds in the future.
“Natural gas is so abundant and so cheap, and it can be landed in Asia at very competitive prices,” Byers said. “It’s about advances in material technology, material sciences, and then distributive power is a big item that I think could decrease demand. Those are some of the things that we think about and then wonder, how could we look at our organizations and overlay everything?”
Steven Martin, vice president and chief digital officer at GE Energy Connections, said renewables will add complexity to the global energy market, particularly as governments enact policies to reduce their carbon footprints. He said GE has seen an increase in orders from clients for solar- and wind-powered technologies in 2016, and while the percentage of total demand is still fairly small, he expected the percentage of new energies entering the market to go up by a noticeable amount in the future.
“We can think this is a good thing, we can think this is a bad thing, but it doesn’t negate the reality that it is happening whether it’s an economically viable option or not,” Martin said.
Byers emphasized the importance of perceptions on energy companies moving forward. An EY study published earlier this year revealed that 71% of people aged 16 to 18 (the bulk of Generation Z) believe that renewable fuels such as solar and wind are the fuels of their generation, while 56% of them believe that oil and gas are the fuels of their parent’s generation. In addition, 67% believe that, because their generation will be around longer, they will have a greater stake in what the oil and gas industry does.
Even though EY’s data showed that across generations support for oil and gas industry is a net positive, Byers said this perception gap translates to a similar divide on the appeal of oil and gas jobs. While this is not an immediate concern for the industry, she said this gap must still be addressed, as the teenagers of today will make up the workforce and policymakers of the future.
“They don’t believe it’s an industry of the future. If you believe that solar and wind are going to be the energy of the future, then why go into an industry from your parent’s or your grandparent’s generation? This is something that has to be changed from a perception standpoint,” Byers said.
Artificial Intelligence: The Brittleness Bottleneck
The oil price downturn has spawned a maxim of “lower for longer” for the industry, hinting at an economic environment that, even after a future recovery, may not be as prosperous as times past. Efficiency in operations is more critical than ever, and an energy company looking to succeed in the future will likely make use of increasingly innovative technology platforms. In particular, artificial intelligence (AI) solutions may help industry process significant amounts of data and help in the decision-making process.
At the SPE Annual Technical Conference and Exhibition, Cycorp founder and CEO Doug Lenat discussed the ways in which AI technology has incorporated slow-thought logic and reasoning processes to overcome problems with information retrieval.
Expert information retrieval software systems can behave in a manner similar to the human brain, in which it resorts to a number of fallback options when faced with situations that require the categorization of data points.
But, Lenat said, these systems often lack a model of what they do and do not know, and they have little understanding of the context in which rules or facts are stated when they try to find patterns in previously observed data. He called this problem the “brittleness bottleneck,” where software programs that traditionally represent documents as groups of words are restricted to learning individual word occurrences in a limited training set.
ArtificiaI Intelligence Applications
Lenat is the original author of Cyc, a platform that assembles a comprehensive ontology and a formally modeled common sense knowledge repository, with the goal of enabling AI applications to perform human-like reasoning. He said that AI platforms that incorporate logic- and causality-based reasoning in addition to pattern recognition will help foster significant developments in the industry.
“It’s this symmetry between the statistical sorts of reasoning, the data analytic reasoning, and the Cyc-like causal reasoning which I think are really going to be the foundation for AI programs going on in the future, the nonbrittle AI that we desperately need when important decisions are being made for large amounts of money for safety decisions and so on,” Lenat said.
The Cyc technology has been used to help interpret data that could indicate impending problems at oil pumping facilities, but Lenat said other industries have used it in ways that nay be adaptable to oil and gas. A financial institution used Cyc to model its information technology facilities and capabilities to help with a migration from stove-piped applications to more unified support for the full information technology life cycle.
The Cleveland Clinic also partnered with Cycorp to develop an interface that builds query fragments through natural language-driven interactions with a clinical investigator.
Machine Learning Implementation
While there has been a lot of advancement in nonbrittle AI systems, full implementation in oil and gas operations is still some time away from being a reality. Lenat said that machine-learning technology, in which algorithms process billions of data points in real time and facilitate the quick identification of trends and patterns that would otherwise be difficult to detect, has gathered much more steam in the industry.
Some principles of machine learning have already been utilized by industry for much of the past decade. For instance, the DrillEdge computer system uses case-based reasoning methods to help analyze real-time drilling data by making links to previously observed data. Developed by Verdande Technology in the mid- to late-2000s, the company received private funding from Statoil Energy Ventures. Hess employed the system for its operations in the Bakken to identify a form of drilling dysfunction induced by an improperly configured auto driller, which the company said could lead to reduced drilling efficiency, motor failure, and other bottomhole assembly failures costing approximately $150,000 per incident.
Deep-learning systems have also begun establishing a foothold in the industry, through the adoption of graphics processing unit (GPU) systems such as MapD and Nvidia’s multi-GPU technology. MapD is a GPU-accelerated in-memory database and visualization platform that can be used as a preprocessing mechanism that feeds data directly from an open source GPU database into a machine-learning pipeline. Multi-GPU technology uses two or more GPUs to calculate and visualize seismic data to help reduce model processing cycle times. Nvidia’s technology has been used in interpretation and modeling software such as GeoTeric, TerraSpark’s InsightEarth, and Paradigm’s VoxelGeo.
While he acknowledged the value of machine-learning and deep-learning AI systems, Lenat expressed concern with a wholesale transition to fully autonomous deep-learning systems in operations, saying that humans will still be needed to provide a logical counter.
“It’s very dangerous if there are systems that are going to be increasingly autonomous with increasingly less chance for humans to counteract, discuss, or unbalance its decisions,” Lenat said. “That’s something that really worries me, and that’s one of the reasons I believe this kind of synergy is important, so that you can get some back-and-forth argumentation about whether this is really the right decision to make.”
Byers expressed similar concerns at the Rice summit. With any new technology, she said it was critical for companies to have people operating them who understand the purpose that technology serves and can make quick, smart business decisions.
“You can have remote sensing technologies, field operations, but if you don’t have people who can run that or understand what it means to the business, or you don’t have a business process that can incorporate that technology, you won’t see any efficiency gains from that. So one of the challenges you have is that, without the right workforce installed, you’re going to suboptimize the use of something, whether it’s digital, data analytics, or robotics,” Byers said.
Industry Standards: Unmanned Systems
Innovation has always been a hallmark of successful oil and gas companies and should continue to be so in the future. As technological development progresses at an increasingly faster pace, industry standards are constantly adapting to address new realities. One area where standards have tried to keep up with the pace of innovation is in subsea operations, where companies have blurred the lines between traditional remotely operated vehicle (ROV) and autonomous underwater vehicle (UAV) systems.
Speaking at a meeting of the SPE Unmanned Systems Technical Section, Charles White examined the changes being made to an API recommended practice (RP 17H) on ROV interfaces for subsea production systems, and how API seeks to keep pace with and further encourage constant innovation in this space. Released in 2008, the first edition RP established standards for the design and operation of ROV-operated system interfaces. A second edition of the RP was released in 2013.
“We took a different approach where we were saying that if you’re putting [an ROV system] out there subsea, you need to know the dimensions extremely well, but you can have a lot of leeway on what’s actually interfacing with that,” said White, vice president of subsea facilities engineering for Doris Engineering. “We did that to help standardize, but also allow some creativity and innovation to prevent things from getting stagnant.”
White outlined two main points behind the updated AUV content in RP 17H written as an acknowledgement of how the roles of AUVs and ROVs have changed in the industry. The first point came from a document Deepstar, an operator-funded research and development consortium, wrote last year titled “AUV Interfaces for Subsea Production Environments.” The document was similar in many ways to earlier editions of RP 17H, but it addressed specifics of subsea docking and recharging protocols, communication, and data manipulation from AUVs and ROVs, information that had not been addressed previously in great detail.
The second point API sought to address was the increasingly interchangeable nature of different types of autonomous vehicles. Traditionally, an ROV was a tethered remote vehicle that needed to be brought to the surface for occasional recharging, while AUVs were untethered vehicles that could operate more freely in a subsea environment. However, since the Swimmer project introduced the concept of ROV deployment in a deepwater field without the need for long umbilicals 16 years ago, hybrid ROV/AUV technologies are commonplace.
In 2015, Aquabotix released the Hybrid ARV, the first ROV/AUV hybrid for shallow-water tasks. The Modus Seabed, which was made commercially available in early 2017, can operate fully autonomously or as a tethered ROV, and features a thruster pattern that enables it to hover and operate with 6° of freedom, which allows inspection and intervention capabilities different from other hybrids. Developed with Saab, the system will be used for survey and inspection, supporting pre-engineering, construction support, and life-of-field condition monitoring.
“Before, with ROVs, you had to fly them into place to manipulate valves and inject fluids and things like that. Today, with the control systems we have, you can just be sitting there and just say you want to inject that thing into that receptacle, and it can account for all of the movements or the currents or whatever else subsea, and it automatically injects it without people flying,” he said.
White said this blurring of lines is a byproduct of necessary innovation within the industry, and in creating the third edition of RP 17H, API wanted to adapt its standard in a way that would not hinder such innovation.
“I think the industry needs that innovation, and we seem to be using intelligence where we can,” White said. “We like to have telepresence. Even if it’s an AUV, you still can have a camera on it and maybe you can inspect something using the cameras. So, again, we tried to address that by not trying to talk about what ROVs do and what AUVs do. It’s a subsea vehicle. It doesn’t matter as much. What matters now are the interfaces.”
White said the third edition of RP 17H is due for release by the end of 2017 or early 2018.
For Further Reading
SPE 141598 Case-Based Reasoning: Predicting Real-Time Drilling Problems and Improving Drilling Performance by H.Z. Raja, Halliburton; F. Sormo and M.L. Vinther, Verdande Technology.
Chadard, Y. and Copros, T. 2002. Swimmer: Final Sea Demonstration of This Innovative Hybrid AUV/ROV System. International Symposium on Underwater Technology, Tokyo, 19 April 2002.
Ernst and Young, 2017. “How Can Oil and Gas Fuel Tomorrow as Well as Today?” (Accessed 31 October 2017).
The Industry of the Future: What Does It Look Like?
Stephen Whitfield, Senior Staff Writer
01 December 2017
Accidental Discovery: Bitumen Pellets for Heavy Oil Transport
Researchers at the University of Calgary have developed a solid pellet that can transport bitumen and heavy oil by railcar instead of pipelines.
BP and Shell Agree: A New Energy Future Is Coming
The world still needs oil and gas, but it is also making room for renewable energy which will change how upstream companies do business.
This Vision of Onshore Seismic May Look Strange at First
Faced with big potential discoveries under terrain that makes good seismic imaging impossible, Total is rethinking how to gather the data it needs, with an idea that could change the face of seismic exploration.
28 November 2017
01 December 2017
01 December 2017
28 November 2017