Using data to digitalize the oil and gas industry

B. Fuentes, Thermo Fisher Scientific

The oil and gas industry has been historically slow to adopt new technologies and truly take advantage of the digital tools that are widely available and extensively used in other sectors. This is partially due to the expansive operations that often span many thousands of kilometers, and the seemingly great expense it would take to update a network. However, a mass overhaul is not always needed, and some devices already in place—such as flow computers, which are ubiquitous in the industry—could potentially be employed to generate and access even more data.

Indeed, data could help to transform the industry to better monitor product flows and create algorithms for predictive maintenance, leading to preemptive leak detection and heightened safety. This will be particularly important at a time when the industry has set clear goals to improve environmental credentials, with a particular focus on better carbon management and the reduction of greenhouse gas (GHG) emissions, two areas that typically rely on careful monitoring. This article focuses on some innovative solutions that promise to truly bring the oil and gas industry into the digital age and help to achieve more sustainable operations.

Making the data flow. The sheer scale of the oil and gas industry makes effectively overseeing entire operations a challenging proposition. Companies are inundated with terabytes of data every day, generated by a multitude of sensors, devices and machines spread across vast geographical locations. However, despite significant investments in physical infrastructure, many companies are not making the most of real-time operating data to improve their functional and business capabilities. Missed faults and inadequate maintenance can result in unplanned downtime costing millions of dollars. In addition, products lost and unaccounted for—due to leakage or inaccurate pipeline control—can cause delivery delays or environmental issues, with further financial implications.

Other industries have been quick to adopt digital tools to help prevent these types of failings, but uptake has been much slower in the oil and gas world despite them offering opportunities to eliminate variables, maximize flow throughout, increase profitability and improve sustainability.

Unleashing the power of the flow computer. The abundance of data currently squandered could easily be harnessed with the infrastructure already in place, offering a way to simplify the digitalization process. For example, the flow computer is a stalwart of the industry, whose ongoing evolution is increasing both its impact and versatility. It was originally used to calculate product volumes going through sites but, owing to the ever-increasing quantity of operational data being generated, these instruments are now able to capture and share more precious information.

The addition of access and control measures has also repositioned the flow computer to the center of operations, giving it the ability to regulate pressures and alter valves. Not only is there now more control, but the quality of the data produced has also improved, including information on ‘health status’ that can affect the system’s capacity. Furthermore, remote access to this data enables technicians to either adjust the system to compensate or assess what the issue might be before traveling to the site, improving efficiency and maximizing the speed at which problems are resolved. This proactive approach can identify and prevent some forms of pipe failure before they occur. Harnessing data from existing equipment is just the beginning, and there are many more opportunities available to drive digitalization even further.

Computing on the edge. Oil and gas pipelines can traverse thousands of kilometers, making access to data a challenge. Attempts to centralize data analysis through cloud computing or hybrid storage are often hampered by the ability to transmit it—particularly by satellite—owing to the significant volumes being produced and the speed of the connections. However, edge computing offers a way to eliminate geographical limitations by processing and acting on data in the field through small, onsite data centers or, increasingly, in devices themselves. Managing data where it is generated provides real-time insights into operation and is not inhibited by distance, the different environments found across operations or the intense temperatures of LNG, making it a promising tool for now and the future.

Improving operations with artificial intelligence. Artificial intelligence (AI) is no longer just a buzzword and is becoming increasingly accepted in many industries, from manufacturing and healthcare to education and transport. AI in the oil and gas industry has been predominantly adopted in the upstream sector, where it has been used to analyze reservoir data to identify reserves and inform drilling decisions. However, it has not yet made inroads in the midstream and downstream sectors, despite the clear opportunities. Using AI would allow routine tasks to shift from humans to digital systems, freeing up the capacity to pursue new, transformative business models. Putting AI at the digital core—by using information collected from pipeline sensors and Internet of Things (IoT) hardware—could help to optimize systems and processes, manage suppliers, respond to real-time product demand, mobilize the workforce effectively and enhance the customer experience.

The power of predictive maintenance. Integrating intelligent digital platforms improves the ability to reduce costs and improve safety through condition monitoring and predictive maintenance. Unusual activity can usually be detected instantly—via flow computers, valves and sensors—and the data is analyzed immediately to identify the cause of any issues. Appropriate actions can subsequently be taken to resolve any issues before significant damage occurs. Furthermore, analyzing historical data with up-to-date information (e.g., external air and internal gas temperatures) can enable technicians to anticipate valve malfunctions at a particular location.

In many cases, implementing these solutions at the field level will require upgrades to both the communication infrastructure (e.g., IoT devices, Bluetooth and ethernet, among others) and the flow computers to ensure they have the processing capacity to manage the huge swathes of data generated. However, these costs should be offset against the benefits of predictive analytics, which provide technicians with an opportunity to take corrective action where and when it is needed before it becomes a critical issue that could impact the wider network. Transitioning from reactive to proactive maintenance has the potential to unlock previously unachievable levels of safety and reliability, as well as greatly increase operational uptime.

The future of data integration. The oil and gas industry currently has ambitious targets for improving its environmental profile. Companies are looking at strategies to reduce GHG emissions and the carbon intensity of their operations. Digitalization offers many essential tools to better monitor operations, leading to better visibility on emissions and allowing leaks to be detected and resolved more efficiently—or even prevented in the first place. Gaining deep insights into carbon use and emissions throughout the entire value chain will be critical to meeting sustainability targets. More broadly, harnessing the vast amounts of data produced offers oil and gas companies the opportunity to improve efficiency, reduce downtime, manage costs and remain competitive. GP

BIO

Benjamin Fuentes has been with Thermo Fisher Scientific for 14 yr of his 24 yr in the oil and gas industry, working in various positions across manufacturing operations, marketing and account management. He has also been a participant in various API petroleum measurement committees, as well as an instructor and speaker at the ISHM and ASGMT measurement schools. 

Related Articles

Comments

{{ error }}
{{ comment.comment.Name }} • {{ comment.timeAgo }}
{{ comment.comment.Text }}