Understanding edge technology complexities and choosing the right device

Users should make sure the edge technology they choose fits their application and can withstand modern manufacturing’s many challenges. Fourteen capabilities an ideal edge device should have are highlighted.

By Mario Torre July 3, 2024
Image courtesy: Brett Sayles

Edge device insights

  • High-performance edge devices are essential for autonomous operations in the oil and gas industry, offering real-time surveillance, control, automation and analytics at field level.
  • Selecting the right edge device involves ensuring a combination of hardware and software, real-time control capabilities, hardware toughness, high processing capacity and strong cybersecurity protection.

We’re in the era of high-performance intelligent electronic edge devices. These ready-to-be-installed high-tech machines can perform advanced and sophisticated autonomous operations at the field level. They are very much the must-have technology of today. Availability is part of a standard offering from many manufacturers and suppliers.

For the oil and gas industry, edge computing can provide real-time surveillance, control, automation, complex analytics and autonomy at the wellbore. This is alongside production surface equipment across long hydrocarbon pipelines and in carbon capture injection wells, among other uses.

The choice is immense and major computer manufacturers are offering advanced field computers as hardware ready to be programmed and installed in the field.

The question then becomes: How does the customer select the appropriate edge device that best aligns with their requirements, budget and long-term vision?

14 capabilities an edge device should have

Considering the vast offering, this can be quite a daunting task. To make the edge selection process easier, here are a set of capabilities and features an edge device should comply with to be successful:

  1. An edge device is more than just hardware. An edge device must offer a symbiotic combination of hardware and software, closely interlaced in providing a functional, safe, and robust solution. Getting bare metal hardware, selecting an operating system, and then building applications from there may be a long and expensive process that requires resources and time.

  2. Hardware toughness. Make sure any edge hardware can operate in harsh environments and on a wide operating range.

  3. Classic process controls, advanced analytics and autonomy – in one single device. Most edge offerings are not designed to deal with real-time, deterministic control. They often rely on a separate programmable logic controller (PLC) alongside the edge device. For the best performance, users should look for a device that provides PLC IEC 61131 programming capabilities for fine-grained real-time surveillance and control, combined with state-of-the-art analytical applications and built using high-level languages integrated with the PLC engine. The real-time engine also must be physically and logically detached from advanced applications. This means if one component fails, the other one continues to operate uninterrupted.

  4. On-board input/outputs (I/Os). It must include I/O points: Discrete, analog and control as well as a few high-frequency scan rate I/O suited for specific advanced applications.

  5. Hardware expandability. Expandability is key. This includes external I/O (through remote I/O modules and more data storage.

  6. Software immutability. The edge technology must guarantee critical software components (operating systems, drivers, libraries and applications) are protected and cannot be tampered with or modified.

  7. Software modularity. Software architecture should be highly modular, comprised of loosely coupled modules or containers that can be independently replaced or updated.

  8. High processing capacity. Any edge device must provide enough computing power to run several advanced applications. It should be built around major programming frameworks and languages and accept add-on hardware co-processor modules such as GPUs and TPUs.

  9. Seamless connectivity. Critical to usability, it will include several serial and network ports and wireless communication capabilities like cellular (LTE/5G), Wi-Fi and wireless I/O. Ideally, this kind of connectivity will be available to any application running on the edge device.

  10. Compatible with major communication protocols. De facto standard protocols (both server and client) must be available, like Modbus, OPC/UA and EtherNet/IP.

  11. Embedded user visualization capabilities. For plug-in-and-play capability, a quality edge device will be user interface readyin case local visualization is required using inexpensive screen hardware.

  12. Remote management is essential. Edge devices securely configured from a remote site will give ultimate flexibility — including edge management, updates, patches or new containerized applications — without interfering with ongoing edge operations.

  13. Strong cybersecurity protection. For remote management — and in general — data security is paramount. Any operation inside the device or in any interaction with external systems must adhere to all principal cybersecurity standards and fully adopt zero trust architecture (ZTA).

  14. Coded by the user. Building and deploying the applications without intervention or effort from equipment suppliers will give the user total flexibility. A software development kit (SDK) means anyone can build applications for deployment. Make sure developers can use the full extent of existing advanced languages, software frameworks and complex libraries, allowing them to create new programs that can run in real-time along with the rest of the edge components.

Edge devices and controllers need to be flexible

This list covers many essential considerations and the key takeaway is flexibility. Each operation comes with nuanced demands and users will need an edge controller to help guide them.

Getting the edge controller decision right optimizes performance, reliability and security — now and long into the future. While the market offers a wide range of options, choosing a device that meets the capabilities and features we’ve discussed will provide a valuable platform for moving the operation to the next level.

Mario Torre is digital architect, IoT, at Sensia. Edited by Chris Vavra, senior editor, Control Engineering, WTWH Media LLC, cvavra@wtwhmedia.com.


Author Bio: Mario Torre is digital architect, IoT, at Sensia (a JV of Rockwell Automation and SLB).