The impact of new technologies on automation and digitalization system architectures

Digital telecommunication evolution has been rapid over the past several years and is providing near real-time communication.

By Mario Torre March 10, 2023
Courtesy: Sensia

Digital communication insights

  • Digital communication has exploded in the last few decades and the trend will continue as hardware and software advances continue happening.
  • Machine learning (ML), deep learning, cryptography and other advanced algorithms can be implemented into so many devices.

Digital technologies have evolved in the past 20 years, and things that were unconceivable a couple of decades ago, right now are being adopted and taken as common standards. Telecommunications, computing hardware, integrated circuits miniaturization, cloud computing, machine learning, advanced analytics, cryptography and other related technologies are evolving, providing more functionality at affordable costs.

Digital telecommunication evolution has been rapid. Today, 5G wireless communication provides us super-high speeds and huge bandwidths on mobile or static devices. Multifrequency fiber optic cables already installed everywhere provide ultrahigh-speed data highways across the globe, which provides near-real-time video communications across nations including the ability to stream high-definition movies. This expansion of the telecommunications system will continue.

Hardware miniaturization and new chip manufacturing techniques are bringing us cheaper, faster, smaller and power-efficient computing devices that we may deploy pretty much everywhere—in appliances, in the field, or clustered in high-performance cloud computing arrangements. Along with the telecommunication explosion, computing capabilities are elastic and ubiquitous and becoming almost limitless.

Machine learning (ML), deep learning, cryptography and other advanced algorithms can be implemented today in computing systems and also into small intelligent devices, providing immense insights and value. All of this is possible due to the evolution of telecommunications networks and computing hardware.

Digital technology challenges

While digital technology has exploded, major new challenges are arising. People are more connected and demand more information to be delivered in real time. Industries run analytics, markets are very fluid and split-second business decisions must be taken. There is a very close relationship between technological evolution and upcoming challenges.

The automation world has been taking advantage of these technological bursts. We see more sophisticated devices, intelligent wireless instruments, powerful programmable logic controllers (PLCs), edge computing devices, supervisory control and data acquisition (SCADA) and distributed control systems (DCS) and historians, running in huge servers or in cloud computing environments, serving users on workstations, tablets and other portable devices.

Regardless of the advances in telecom and computing technologies, automation architectures have not changed much. While many components have advanced, their underlying architectures have largely remained the same.

For instance, the oil and gas industry has traditionally been conservative in terms of technology adoption, due also to the fact that it faces specific challenges that are different from other industries. Nevertheless, as technology evolved, operations have become more complex, markets are more volatile, competition is fierce, and environment and sustainability plans and issues must be addressed.  This industry is fully realizing the value of technology in the current operating environment. Hydrocarbons are getting harder to produce, operating margins are shrinking, more restrictive environmental regulations are in place, and safety policies demand less exposure of personnel to field conditions. All these developments are pushing this business not only to broadly implementing more automation and digitalization solutions, but also to adopt new advanced system architectures that best solve the actual and upcoming operational and business challenges in this industry.

Six architectural paradigms

Classic automation architectures are failing to keep up with the new challenges that the industry is facing. New system architectures for automation should be adopted that better fit with new industry operational and business challenges.

Identified below are several system architectural paradigms that have been successful in other computational architectures but that have been overlooked in the automation world.

1. Flatter architecture: Even when new automation systems are currently using updated technology, their architectures are too vertical. Field devices, connected through a telecom infrastructure to SCADA/DCS, then tied to a historian, then connected to applications and business intelligence systems, in a sort of a totem fashion, raise problems by having diverse user interfaces, many interfaces among layers and the use of multiple databases (Figure 1). New architectures must be flatter, with fewer database layers and less interfaces among levels. The entire system should be based on a unified digital platform that includes most of the functionality in fewer levels (Figure 2). This digital platform must also be flexible enough to leverage on existing instruments, automation devices, and systems as much as possible.

Field devices raise problems by having diverse user interfaces, many interfaces among layers and the use of multiple databases. Courtesy: Sensia

Field devices raise problems by having diverse user interfaces, many interfaces among layers and the use of multiple databases. Courtesy: Sensia

2. High availability and fault tolerance: Some automation systems still adopt the classic warm-standby configuration. Essentially, the user buys two systems just to use one, to cover the case when the system that is running fails. Automation architectures can improve by incorporating new fault-tolerant schemes, which automatically adapt in case of failure of one of the hardware components, with minimal or no failover time.

3. Elastic: In a corporation, business and operations scenarios may suddenly change and automation architectures must quickly adapt to those changes. Traditional architectures are rigid, resistant to changes. New architecture must quickly expand as business expands or shrink if business requires. It must be elastic to quickly expand or contract within minutes and without losing operativity.

4. Secure from the ground up: Cybersecurity must not be a layer installed on top of a classic automation system architecture. An automation architecture must incorporate cybersecurity mechanisms from its conception, adopting cybersecurity strategies like the zero-trust architectures (ZTA) approach from its early design stages. Nowadays, potential cyber-attacks over automation systems that are controlling critical assets (in either industry) represent a clear danger that must be properly addressed by the new architectures.

5. Single point of configuration: In current practice, as shown in Figure 1, adding a new field device or making a configuration change must be done on each one of the layers to maintain system integrity. New architectures must adopt a “configure-once” design. By installing a new device or changing its configuration, the changes must be automatically propagated throughout the entire system.

6. Push intelligence as low in the architecture as possible: “Dumb” remote terminal units (RTUs) or limited-function PLCs are part of the past. Data sponging, analytics, smart processing, machine learning (ML) and artificial intelligence (AI) must be done as closely as possible to the process. Edge devices can take on a good portion of the processing on new architectures.

The entire system should be based on a unified digital platform that includes most of the functionality in fewer levels and be flexible. Courtesy: Sensia

The entire system should be based on a unified digital platform that includes most of the functionality in fewer levels and be flexible. Courtesy: Sensia

While these are among the key paradigms that must be adopted by new automation and digitalization system architectures, there are many more that will help system designers and builders to best solve actual and upcoming challenges in the oil and gas industry.

Automation system design paradigms have changed. New technology is widely available, but new business and challenges have emerged.

This trend must lead people to think about new ways to build automation systems and solutions. The ultimate challenge is to design, develop, implement and deploy automation systems based on these new paradigms. This will help solve industry operational challenges with a faster and better return on investment (ROI).

Mario Torre is digital architect at Sensia (a JV of Rockwell Automation and SLB). Edited by Chris Vavra, web content manager, Control Engineering, CFE Media and Technology, cvavra@cfemedia.com.


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