For oil and gas companies to remain in existence in the second half of the 21st century, they must find ways to dramatically reduce, if not totally eliminate, their output of carbon dioxide and other greenhouse gases. Artificial intelligence technology could provide one tool to help the energy industry accomplish that staggeringly difficult goal.
According to an August 2020 report by the National Oceanic and Atmospheric Administration on the impacts of atmospheric carbon dioxide on climate change, CO2 levels — mostly caused by the burning of fossil fuels — rose to about 410 parts per million in 2019, the highest level seen on earth in more than 3 million years.
Oil and gas companies, especially the big international ones, are under increasing pressure to reduce their carbon footprint in accordance with the Paris Agreement’s goal to limit global warming to well below 2 degrees, preferably to 1.5 degrees Celsius, compared to pre-industrial levels. Most major companies have set carbon-reduction targets, such as BP and Royal Dutch Shell, which both have vowed to achieve net zero carbon emissions by 2050.
ExxonMobil has focused on more modest short-term climate goals, such as reducing greenhouse gas intensity of its upstream operations by 15% to 20% compared with 2016 levels by 2025.
AI technology, which has the ability to harness large volumes of data from divergent sources to come up with solutions to problems, has the potential to not only increase global productivity but also lower overall emissions of carbon and other potent greenhouse gases, according to a recent report published by Microsoft in association with PWC.
“Using AI for environmental applications has the potential to boost global GDP by 3.1 – 4.4% while also reducing global greenhouse gas emissions by around 1.5 – 4.0% by 2030,” the report states.
“These AI technologies can help the industry optimize energy management by using digital twins to better monitor and distribute energy resources and provide predictive forecasting,” said Darryl Willis, corporate vice president of Energy at Microsoft. A digital twin is a digital representation of a physical piece of equipment or an entire system.
“They can also be used to create visualize simulations, improve decision-making, reduce operational costs, and manage and extend the life cycle of physical assets,” he said.
Multiple uses of AI
Companies in the exploration-and-production segment of the industry are using AI in multiple ways to lower their carbon footprint: from performing predictive monitoring of carbon emissions from a particular oilfield; to conducting analysis of the oil-producing potential of a given field, thus reducing the number of wells that need to be drilled; to optimizing the storage of CO₂, which can be used in the production of hard-to-get-at oil. Such enhanced oil recovery results in storing the CO₂ deep underground, rather than releasing it into the atmosphere.
“Just bringing that efficiency to the table, from exploration to bringing the first well to market, that gives not only monetary dividends, but also reduces the CO₂ footprint for every barrel of oil,” said Mike Krause, senior manager of AI software developer Beyond Limits.
Another way in which AI is helping producers such as Shell lower their carbon footprint is in conducting predictive maintenance of pieces of equipment or entire systems, which allows the companies to anticipate and address potential equipment failures before they occur.
“If we can be more proactive and predictive in terms of when things might go wrong, we have fewer likely incidents, more controlled deployment of spare parts, less travel for people to come to the site, less hot-shotting of spare parts. All of those things have a CO₂ impact,” said Dan Jeavons, Shell’s general manager of data science.
Shell is also implementing AI technology across its new lines of businesses, which has potentially major implications for the company’s overall carbon footprint. For example, Shell is using AI at scale in its Quest carbon capture and storage facility in Alberta, Canada, which began operations in 2015. By May 2019 Quest had captured and stored more than four million metric tons of CO2, roughly equal to the emissions from about one million cars, deep underground.
The company also deploys AI to optimize the operations of its wind farms, which supply carbon-free energy in locations around the world, Jeavons said.
Monitoring emissions across vast distances
Ron Beck, energy industry director at Boston-based AI technology company AspenTech, said in the future AI will be critical in helping oil and gas companies take the first step in lowering their carbon footprint, by accurately measuring their greenhouse gas emissions across their operation.
“Public companies are starting to set executives’ pay based on their progress. So, you see companies reporting in their sustainability reports, ‘This is our carbon footprint. This is our flaring,’” he said.
Big oil and gas companies such as ExxonMobil are using AI to sort through large volumes of data collected from sensors deployed across the wide swathes of territory that comprise their areas of operations.
The company is employing the AI technology to reduce its emissions of methane, a potent greenhouse gas, across the Permian Basin in the southwestern U.S. Collaborating with Microsoft, Exxon is using “Internet-of-things” technologies to monitor and optimize a vast number of its widely dispersed Permian field assets.
Working from anywhere and using data collected from its extensive network of sensors and stored in the cloud, Exxon’s engineers, scientists and analysts can strive to decrease emissions, lower costs and increase production from the field, according to the company’s website.
