Most of the conversation related to the Industrial Internet of Things (IIoT) has been about factory automation, but the ExxonMobil vision for Next Generation Open Automation System (NGOAS) and Open System Architecture (OSA) is bringing these same concepts to process automation. The good news is that thanks to the enterprise, we have open systems at the top—the cloud and associated storage/data transfer mechanisms—and at the bottom, the analog and digital networks are also well defined. The work that remains is the all-important middle, which just happens to be where all the proprietary systems, technology and information reside. This is what the NGOAS project hopes to separate from its hardware dependence.
The ExxonMobil team now has U.S. Patent 20160065656A1, “Method and System for Modular Interoperable Distributed Control,” which specifies in general terms their proposed system of the future. It describes a system with several elements, the key one being the Distributed Control Node (DCN) to interface the middle layer with the field layer.
The patent does not differentiate between the two types of DCNs defined in the document: the software or application DCN I call the application control node (ACN), and the hardware node that interfaces to the field devices I’ll call the device control node (DCN). Think of the DCN as similar to a combination single-loop controller/gateway with configurable I/O (analog, fieldbus, wireless), IPv6-addressable and suitable for mounting in Zone 2 (Class I, Div. 2, Group C/D)—in other words, an intelligent, single-signal I/O card that can be mounted anywhere.
The ACN, on the other hand, is similar to the Virtual Field Device in the Foundation fieldbus world, and can reside in any device with memory and microprocessor, from the DCN itself (thus enabling control in the field) to a redundant server on the control network. This effectively brings the cloud nearer to the edge, since the DCN is the fence on the edge of the control system and the devices themselves are the edge. Therefore, once the signal has been processed by the DCN, it’s simply another piece of data, likely in the form of an object or metatag that needs to be managed.
The patent indicates the desire to incorporate applications developed using anything from one of the IEC 61131-3 languages to imported code from products such as LabView, Matlab or GNU Octave, thus allowing anyone with knowledge of these widely deployed systems to work in the environment in which they’re most comfortable.
The public reason that ExxonMobil is pushing NGOAS is because they want to be untethered from the hardware, in part, so they can take advantage of the latest processors and migrate on demand without having to replace an entire system. A big concern they have is what happens if they pick the wrong existing standard, which is why all the discussions go through Open Group.
One requirement for wide adoption is that the results be published as standards. But standards require release of patents, which means that in order to succeed, (with apologies to the inventors), the ExxonMobil patent is more of a whitepaper or vision document that, because it is a patent, will be accessible and read.
One last thing I’m confident about is that SCADA-like technology will play a role in NGOAS and IIoT because IoT is really super-sized SCADA connecting lots of things to each other in real time. So I’m pleased that ISA is about to kick off the S112 committee, which will develop a new set of SCADA standards.
Yes, ExxonMobil has a good idea here, and many of the elements are available. However, the trick will be making it work and keeping it working because when it or something like it does happen, the IIoT will be “on or near the edge.”