automation-&-digitisation

Making Oil and Gas sector safer with Artificial Intelligence

March 2023, A&D India Magazine, Mr. Girish Dev, Head – Artificial Intelligence & Digital Transformation (AI&DT), Commtel Networks: 

Notwithstanding the endeavours and commitments to achieve net-zero emissions by 2050, the Oil and Gas sector is going to be in the energy system for decades and will play a significant role in the 2050 energy mix.

The production of oil and natural gas is often coupled, as the two are typically found together in nature. Nearly more than 80 countries currently produce oil and/or natural gas, and the two fuels are expected to maintain their importance across the energy sector. It is associated with environmental disasters such as oil spills. The sector has been demonised and operates in an environment of significant hostility. The prices of the two fuels, especially oil, are highly volatile, with fluctuations directly impacted by political and socioeconomic events. The world now needs as much oil as it did just before the pandemic, with demand bouncing back and disproving some theories from last year that global oil consumption would never return to pre-COVID levels. Global oil demand is set to rise by 1.9 mb/d in 2023, to a record 101.7 mb/d, with nearly half the gain coming from China following the lifting of its COVID restrictions. The industry is commonly divided into three main operational sectors:

Upstream: Exploration and extraction of crude oil and natural gas reserves. The upstream firms are often called E&P businesses.

Midstream: Transporting and storing the extracted products. Normally characterised by trucking, shipping, storing of raw materials, and pipelines.

Downstream: Processing facilities responsible for refining oil and gas (refineries) and turning them into the many finished products people rely on every day, like jet fuel, gasoline, asphalt, and heating oil.

Challenges faced by the industry

The energy industry is volatile. The oil and gas companies are under intense pressure to maintain profitable, sustainable, and safe operations.

– Must react faster than the market changes and pivot quickly

– Achieve net-zero commitments

– Use advanced technology to continually improve operations

Improving safety and efficiency in operations

Over 65 million people were employed in the energy and related sectors in 2019, accounting for almost 2% of formal employment worldwide. In the fuel supply, oil has the largest labour force, totalling almost eight million. This is followed by 6.3 million in coal supply and 3.9 million in gas supply. All drivers of energy employment are set to rise in 2022-2023. (Source: IEA) 

There are now comprehensive regulations covering all aspects of production, from drilling equipment to delivery to fuel stations and everything in between. Today, the American Petroleum Institute (API), Det Norske Veritas (DNV-GL), and International Standards Organisation (ISO) standards are probably the most widely recognised standards across the oil industry. In India, the OISD (Oil Industry Safety Directorate), a technical directorate under the Ministry of Petroleum and Natural Gas, formulates and co-ordinates the implementation of a series of self-regulatory measures aimed at enhancing safety in the Oil and Gas industry.

Safety performance indicators:

  • The IOGP incident reporting system covers worldwide E&P operations, both onshore and offshore, and includes incidents involving both member companies and their contractor employees. The 2021 report covered data from operations in 95 countries.
  • The International Association of Drilling Contractors (IADC), Rotary Rig Incidents Statistics Program (ISP) published for 2022 the total recordable (RCRD) incidents at 1103 (819 in 2021) with total 311 (224 in 2021) Lost Time Incidents (LTI), and 15 (7 in 2021) Fatalities (FTL).

An overview of safety in the complex setup of oil and gas:

  • E&P Safety – Onshore and Offshore
  • Transportation Safety – Pipeline, Marine, Rail
  • Refinery and Plant Safety – Fire, Occupational, Process
  • Consumer Safety

Digitally enabled processes, workplace and workforce

As observed through various studies, the oil and gas sector at large has invested proportionally less in digital and AI/ML related technologies in the last decade compared with other sectors, such as banking, automotive, health care, retail, and consumer products and software. In a report, EY noted that digital fluency and the use of AI/ML technologies have not been seen as core competencies in a sector that’s long been ruled by human operators—namely, mechanical, chemical, and electrical engineers. Early adopters are typically sectors that are comfortable with digital fluency and those that have access to standardised data sets. In simpler terms, AI helps augment the human decision making process by analysing a tremendous amount of data and presenting acceptable alternatives. For example, analytics automatically generate and evaluate multiple schedules and anticipated events within the complete supply chain, which allows the company to accomplish a day’s work in mere seconds.

Computer vision

A field of AI that enables computers and systems to derive meaningful information from digital images, videos, and other visual inputs — and take actions or make recommendations based on that information. If AI enables computers to think, computer vision enables them to see, observe and understand. (Definition: IBM). Some of the present-day CV applications that have found wide acceptance in the oil and gas industry:

  1. Flare monitoring and smoke detection
  • Flaring is the burning of natural gas associated with oil exploration. Thousands of gas flares at oil production sites worldwide burned approximately 144 billion cubic metres of gas in 2021
  • To operate these flares, the oil and gas companies must comply with specific regulations set in place by government entities such as the EPA
  • Some of the recent regulations state that they must monitor their flares 24/7 (real-time gas emissions monitoring) and report on potentially harmful flare behaviours and black smoke production
  1. Leak detection

Methane Leaks: Methane is a particularly potent greenhouse gas and frequent leaks from oil and gas pipelines is a waste product. Methane is the main component of natural gas, and it can leak anywhere along the supply chain – from the wellhead and processing plant, through pipelines and distribution lines, all the way to the home (stove).

Thermal imaging (infrared cameras) and AI

  • With thermal imaging cameras, the preset temperature levels measure the differences from a material’s normal emissivity to a temperature that is abnormal. This allows the cameras to immediately detect changes in ambient temperature in locations where they should not be.
  • In addition these thermal images are analysed with AI software to find growths, measure heat loss, and perform other tasks such as combining with maps to pinpoint exact location of any incident.
  • Deep Learning Model (ML) has been trained by the system to detect the shape of a methane gas leak when its spread is detected in gas industries.
  1. Monitoring storage tank levels
  • The monitoring of tanks is an essential practice at the oil and gas well pads, enabling operators to avoid overfilling/ overflow or when it is empty. It helps to calculate tank flow rates and identify possible interfaces between materials that may cause build-up (sludge) or emulsions.
  • There are multiple release points on a field storage tank viz. thief hatch, goose neck, PRV, level gauge assembly. Traditionally, operators have used guided wave radar for tank maintenance, which involves opening the ‘thief hatch’ at the top of the tank. This action is a large emitter of Volatile Organic Compounds (VOC), raising safety concerns for workers and damaging the environment.

Thermal imagers detect the liquid levels by observing the heat changes between the exterior walls of the tanks and through a virtual scale provides the level.

  • Raw data from cameras are also run through ML algorithms to identify VOC leaks
  • The ML models are built to detect accuracy within 1% of guided wave radar i.e., it mimics guided wave radar
  • The autonomous nature of the AI-driven system eliminates high risk practices, increasing worksite safety
  1. Worksite and worker safety

This has become one of the most important areas in the widescale adoption of AI, particularly powered by CV, assisting oil and gas companies across the entire scope of their operations. With a combination of CV and IoT (sensors), the AI solutions enable HSE personnel to address safety scenarios ranging from unsafe act and condition detection, poor housekeeping, PPE detection, body positioning assessments, access limits or area controls, vehicle controls and safety, evacuation, to HSE auditing and compliance.

Other AI-based methods

  1. Defect detection with precision

The equipment invariably operates at critical temperatures and pressures. Material degradation and corrosion are natural phenomena that can result in serious accidents and damages. Manually monitoring the equipment is error-prone. Together with the Internet of Things (IoT), a smart system that uses tiny sensors to monitor everything from individual equipment to entire production lines, the ML models monitor each individual element, identify potential risks, and provide the solutions.

  1. Maintaining Blowout Preventers (BOPs)

The blowout preventer is a large valve that is situated on top of a well, and it is generally used to close the well if the drilling crew for some reason loses control of well fluids. This piece of equipment prevents a rupture from happening at a drilling site. Using pattern recognition algorithms that leverage rig data from faults, alarms, and subsea control systems, the AI and ML solutions can provide the health of the BOPs in real-time.

  1. Red zone monitoring

These zones are high-risk areas on the rig floor where heavy drilling equipment operates and the risk of injury is highest but despite this, the drilling crew still needs to enter these zones to perform specific tasks. This leads to contradictory operational instructions. It involves an automated safety process to identify offshore workers in the Red Zone and eliminate human error. Combines Artificial Intelligence (AI), Laser Imaging, Detection and Ranging (LiDAR) and advanced edge computing technology to ensure safer and more efficient operations.

  1. Pipeline Intrusion Detection System (PIDS)

Pipeline companies have found great effectiveness in Pipeline Intrusion Detection System (PIDS), a highly sensitive, Distributed Acoustic Sensing (DAS) system utilising existing fibre optic cables along the crude oil and gas pipelines that provides the end users with an automated means of monitoring or measuring activity along long linear assets.

DAS works by effectively converting a Fibre Optic Cable (FOC) running alongside a pipeline into tens of thousands of individual-highly sensitive vibrational sensors. The system makes use of Artificial Intelligence (AI) and Machine Learning (ML). It not only detects risks to the pipelines but also determines the nature of the hazard, such as digging or excavation.

  1. Pipeline inspection

It is increasingly imperative to monitor and inspect the pipeline system, detect causes contributing to pipeline damage, and perform preventive maintenance in a timely manner. An integrated solution uses a drone equipped with a thermal camera that can monitor oil and gas pipelines to detect leakages and cracks in pipelines in remote and risky areas. The solution making use of deep learning technique is aimed to detect the targeted potential root problems—pipes out of alignment and deterioration of pipe support system— that can cause critical pipeline failures and predict the progress of the detected problems by collecting and analysing image data periodically. (IJDRS)

We’ve emailed you the access to the Whitepaper from

info@commtelnetworks.com

Kindly check your SPAM folder, if you have not received it.