How AI is transforming the oil and gas industry

Vast and remote oil and gas facilities face a variety of risks, including sabotage, theft, vandalism, system failure, fire, violations of environmental regulations, and other issues that can disrupt and even paralyze your business. Artificial intelligence, making video cameras and other sensors smarter than ever before, can help operators mitigate these risks and improve efficiency through the automation of remote monitoring and control processes. 

A recent EY survey reported that 92% of oil and gas companies were currently investing in some form of AI, or at least planned to in the next couple of years. (Source: Neoteric)

To value the potential of AI for the oil and gas sector, it is critical to understand that AI is not just an algorithm, a tool, a piece of software, or a process. Instead, it must be considered as a system of technologies, data, and capabilities that, taken as a whole, is designed to complement or even replace specific human competencies. These capabilities can be applied to a wide range of areas of oil and gas production. 

We will take a closer look at 3 of them: monitoring, automation, and analytics, which are closely aligned in practice.

Monitoring: ensure reliable and safe operation

AI makes the monitoring of operations smarter through cognitive capabilities, empowering IoT devices to support or even take over the tasks of humans in many production areas. AI can detect patterns and signals in sensory data, such as video footage from surveillance cameras, that are beyond normal human perception.

Example: AI-enabled video cameras help monitor flaring

Source: Noctuai

Wherever smart sensing is needed in oil and gas production or transportation, and humans and conventional technologies are stretched to their limits, AI can make a valuable contribution to ensuring reliable operations. Here are some other examples of AI applications for monitoring the operation and safety of facilities and pipelines:

• AI-enabled surveillance cameras detect and report theft and vandalism
• Monitoring liquid levels in tanks using AI-driven thermal imaging
• Automatically detect accidents involving humans and call for help
• Leverage smart video technology to ensure clean flaring
• Detect violations of work safety regulations, monitor PPE compliance, etc.
• Detecting leaks in tanks and pipelines using AI video-enabled drones

Example: Pipeline inspection using drones equipped with thermal imaging

Source: Workswell Infrared

Automation: increase operating efficiency in remote sites

The examples above show what sensors such as surveillance cameras equipped with AI analytics can do. But that's just the tip of the iceberg when it comes to making the monitoring and control of remote production facilities much more efficient and reliable. Connected to other systems, smart IoT sensors are a cornerstone for a wide range of applications to automate monitoring and control of oil and gas plant operations. 

Camera systems equipped with AI video analytics can analyze video data, for example, from continuous monitoring of the flaring or liquid levels of tanks and autonomously inform the crew via connected communication systems in case of anomalies. Similarly, smart video cameras can control barrier systems and automate access management in remote production facilities, using AI license plates and vehicle type recognition.

Key objectives (and benefits) of AI-driven automation in oil and gas production

• Adjust operations in real-time
• Improve worker safety
• Improved efficiency and output
• Minimize unplanned downtime of heavy equipment
• Predict problems and issues in advance
• Standardize field operations

As machine learning models are continuously trained and learn to understand processes in production and transportation facilities better, software apps and the devices on which they run using edge computing are becoming more powerful. At Azena, we bring the latest AI software to cutting-edge video cameras, so operators can get more out of their equipment as easily and with as little effort as possible.

Worth reading: Top technology trends that drive manufacturing 4.0

Analytics: data-driven decision making

Artificial intelligence also helps improve decision-making by processing and analyzing large amounts of data, even directly in IoT devices without having to share data between systems, using edge computing. Operators use data analytics to monitor, e.g. which components in a plant are showing wear and predict when they need to be serviced or replaced. 

An important application for data analytics is Digital Twins, digital representations of oil and gas production and transportation facilities, which help model and monitor operations by processing vast amounts of sensor data in real-time. For an introduction with examples of Digital Twins applications, please see our previous article on the Azena blog.  

Digital Twin of a Petroleum Refinery (Source: GE Ventures via ARC)

Example: Digital Twins at BP

BP developed a highly sophisticated simulation and surveillance system called APEX, designed to create virtual models of all its production systems. APEX allows BP to plan changes and interventions in the digital twin before employing them in the real world. As a surveillance tool, it identifies issues before they have major effects on production. (Source: GEP)

AI helps process vast amounts of data and gain valuable operational insights, either in real-time to control systems and detect faults, or over time to improve operations and support engineering. 5G enables fast data exchange between systems and lays the foundation for building smart ecosystems in oil and gas production and transportation.

Conclusion: AI is here to stay

92% of oil and gas companies are currently investing in some form of AI, or at least plan to in the next couple of years. The applications in the areas of monitoring, automation, and data analysis are almost limitless, and so are the technologies that are constantly evolving. However, AI does not have to be a large strategic project but can be implemented already in small value-adding applications. Azena, which brings AI to video cameras and makes them true business all-rounders, is the best proof of this.