Edge Computing in Drilling: Challenges and Solutions

Translated by | Liush

Service providers in the oil and gas industry face numerous challenges during drilling operations. Even well-designed oil and gas wells can encounter issues during drilling. Here are some significant challenges faced:

1. Downtime caused by equipment failures affects the entire drilling operation, leading to higher extraction costs.

2. Maintaining normal operating time of foundational assets.

3. Drilling data updates rapidly, requiring the processing of large amounts of data to derive results.

4. Transmitting real-time data from the well site to end-user workstations.

5. Time, materials, manpower, and equipment are needed to analyze drilling issues, significantly increasing drilling costs.

6. Environmental sustainability.

7. The bandwidth at the well site is limited, making it impossible to transmit large amounts of data beyond the site for further processing and monitoring.

The data collected from critical drilling equipment cannot be processed and analyzed at the well site. The vast amounts of data generated by drilling rigs must be transmitted from remote well sites to data centers. The transmission, analysis, and conclusion of data become a difficult and lengthy process. Therefore, currently, only a portion of the data is transmitted for analysis.

Offshore deep-water drilling facilities can incur daily costs of up to $600,000 to $800,000, and downtime can significantly increase costs. Such situations can occur multiple times throughout the year. Predicting downtime can save substantial costs through preventive maintenance. By analyzing this data, unexpected downtime, costs, injuries, and environmental impacts can be reduced.

With the current available edge computing and various automation solutions, data can be integrated and analyzed in real-time at the drilling site with low latency. Additionally, it can instantly process data streams collected by IoT (Internet of Things) sensors and perform real-time analysis. Automation solutions using artificial intelligence/machine learning models can handle data streams, conduct analyses, and provide recommendations to enhance drilling rig performance and efficiency in real-time.

One of the key goals of applications based on machine learning is to analyze data, make predictions, and provide recommendations. Continuously monitoring the accuracy of machine learning models is also crucial. Once predictive analytics obtain all necessary prediction points, insights can be provided. Furthermore, AI systems based on edge computing need to utilize event-driven architecture for data transmission, hence employing event-based messaging systems (like Kafka or RabbitMQ) to continuously send and receive required information. Predictive maintenance using monitored and analyzed information can minimize planned downtime and reduce machine failures.

Technical alliances such as the Open Subsurface Data Universe (OSDU^TM) are currently working on seamless integration of drilling site data and using the “Drilling Site Connection Framework” to transmit data to off-site environments. This data will be further transmitted to the “OSDU Edge” reference architecture for implementing drilling solutions used in the well construction phase.

The oil and gas industry is currently developing various automation solutions based on edge computing, such as alarm services, sending automatic setpoints to drilling control systems, and organizing automated operations for drilling rigs. Here are some cases:

1. Establishing a Universal Wellsite Monitoring and Control (UWC) solution on the edge platform for oil and gas wells and surface facilities, making wellsite control more open and interoperable. This allows supervision of various facility controls, such as monitoring tank liquid levels, chemical injection, flow control, etc.

2. Artificial Lift Optimization Solutions reduce the production pressure at the well bottom, leading to higher yields. Equipment such as Electric Submersible Pumps (ESP), Progressive Cavity Pumps (PCP), Rod Lift, Gas Lift, and Plunger Lift can also be similarly managed.

3. The friction torque between the drill string and the wellbore is a critical factor affecting the maximum drilling depth. Equipping the wellsite with a Friction Torque Solution will aid in automatically detecting drilling anomalies and assist in making efficient operational decisions. It can view and analyze the trends of weight on bit changes and use machine learning models for friction coefficient calculations. By comparing design and actual drilling results, engineers can perform real-time analysis and make informed decisions.

4. During drilling, collisions with adjacent wells can have catastrophic impacts on personnel and the environment. Collision Avoidance Technology is an effective means to mitigate such risks during drilling. This solution helps users observe and monitor the relative positions of the current well and adjacent wells to avoid entering restricted zones. Plotting the results on charts to show the distance to the nearest potential risks can help engineers make quick decisions.

Developing an edge solution that is recognized by regulatory authorities by considering the fundamental characteristics of the drilling site, temperature and other location factors, vibration, weather conditions, ecosystems, and equipment types will be key to efficient drilling.

Scalability and Sustainability: Besides simulated environments, the designed solutions must be evaluated on real devices to measure their reliability and flexibility. The solution must be designed to work across various platforms. It needs to be applicable to different models of equipment, different versions of software applications, and operating systems running on different devices.

Usability and Performance: Considering that most drilling platforms are located in remote areas, or worse, offshore, the designed edge solutions must be easy to deploy. This includes easy-to-install and troubleshoot software, plug-and-play models to simplify applications. Undoubtedly, performance must be within acceptable ranges to support real-time data processing.

Redundancy and Risk Mitigation: The solution must consider the necessity of continuous monitoring to discover any issues that may arise during system use. If any network failures occur during data transmission, the solution must respond promptly to retransmit data in case of failures. One of the key requirements of this design is the ability to perform remote troubleshooting and updates.

Defense: Edge solutions allow bidirectional communication between data source machines and target machines, presenting unique challenges for device security. Ensuring the network is protected from all threats is a primary concern for any operation. Security designs include device management, authorization, and most importantly, preventing misuse.

Main Goals of Edge Solutions:

1. Acquire necessary operational data from devices, analyze it, and transmit this data to centralized servers.

2. Utilize remote content management to provide updates for software general-purpose programmable controllers and profiles, timely updating patches for all devices.

3. Perform edge processing on devices, including remote and on-site monitoring of devices.

4. By carefully examining the advantages of each technology and the applicability of edge platforms, as well as the quantity and type of information, the most suitable method for specific operations can be determined.

Edge platforms are a collection of various solutions, algorithms, and devices. Advanced analytics and various predictive algorithms require different self-learning models. The prospects for edge computing development in the oil and gas industry are vast, with the key being to expand the application scope of edge computing. Its technological layout and components must be simple, inexpensive, energy-efficient, and adaptable to existing architectures.

However, the application of edge computing-based solutions in the oil and gas industry does come with some challenges. Given the current technological gaps and aging workforce, retraining of technical personnel is necessary to keep pace with technological advancements. To master future edge solutions, oil and gas companies, along with their service providers and suppliers, need to undergo advanced training. Considering the technology stack and skill enhancement requirements, these companies should reassess their processes, investments, value chains, and operational models to adapt to the new event-based edge analytics solutions.