The Industrial Internet of Things (IIoT) and cloud technologies show promise for the generation and use of “Big Data” in automation, but the unique demands of automation are not easily solved with commercial Internet of Things (IoT) solutions.
What are these demands? They can be summarized as:
· The need for real-time communications;
· Strict compliance to cybersecurity standards and practices;
· The ability to leverage new and evolving standards, such as Time-Sensitive Networking, OPC UA and Message Queue Telemetry Transport (MQTT);
· A path for device-to-cloud connections; and
· Adoption of cloud-based automation and IoT solutions.
In my column in the August issue of Automation World’s IIoT supplement (http://awgo.to/iiotnextgen), I discussed the first three demands listed above. Real-time networks, bullet-proof cybersecurity, and the adoption of new and emerging standards are critical for the transition to an “Industrial” Internet of Things. In this column, I’m taking a closer look at the fourth and fifth requirements, leveraging the cloud for next-gen automation systems.
In automation, the main source of primary data is the factory floor. Sensors were traditionally simple devices, providing little more than on/off signals. That is changing fast. The change began in complex devices such as drives, where cost levels are high enough to obscure the extra resources needed, but the pressure to get more from the floor is now intense.
Embedded chip solutions, such as Hilscher’s netX 90 and netX 4000 chips, have become more sophisticated too, allowing sensors to deliver more than basic on/off signals. For example, these chips can now deliver data about operating status (e.g., temperature, cycles) and health (e.g., vibrations).
But when a device becomes a thing on a network, a new set of questions arise. How should it interact with the higher-level data aggregation and service layers? How does it make itself known to the cloud: automatically or manually? How can a device explain what it offers? Are there consistent methods of data transfer? What about security? Which cloud manufacturers should we engage with?
There are several organizations working on standardized methodologies. Many are already working specifically on Device-to-Cloud (D2C) concepts. The Open Industry 4.0 Alliance, launched in late 2018, could be a major step towards defining those mechanisms, particularly at the field device level. Hilscher is one of its eight founding members.
Leveraging cloud-based automation
One vision of cloud-based automation is of domains of interest, beginning with the device management domain. In this scenario, IIoT, or entire fleets of them, are registered in the cloud at switch on. Profiles are created and managed according to type, location, or group. Simple data handling devices could be grouped and configured to exchange data locally or indirectly via the cloud. But generally, the real benefit of a cloud is to handle the big data with relevant applications.
In the best case, such applications may even be deployed into network-attached devices, such as Hilscher’s netIOT family of Edge devices, with some functions executed locally on-premise. Edge devices are becoming powerful computing platforms in their own right, allowing data flows to be kept restricted and saving computing resources elsewhere.
One of the best promises of the cloud lies in the analytics domain, where predictive maintenance and asset management can be undertaken to improve the management of plants. Or, data can be forwarded to a visualization domain for graphical preparation in web-based human-machine interface (HMI) solutions.
Device and system vendors must accept that they have hardly any influence on the cloud that the end customer chooses. So it would be wise for suppliers to support multiple platforms at once and offer connectivity to the top dogs.
Hilscher has already adopted this approach with its Industrial Raspberry Pi edge devices, making it extremely easy for a user to determine their optimum cloud solution. All versions are fully compliant with the Raspberry Pi standard. These industrial Pi devices are hardened for use in harsh environments and are IEC 62443 capable. Inside, netX is used as a companion chip for the standard Broadcom chipset to allow real-time network protocols such as Profinet, EtherNet/IP and EtherCAT to become part of the Industrial Raspberry Pi 3 ecosystem.
Hilscher’s Industrial Raspberry Pi 3 also supports Docker with “container images” for connecting to six leading cloud services providers (and counting) and which are readily available for download. Docker is complemented by Node-RED, a graphical user interfaces (GUI) data flow definer that makes configuring networks easy. No formal knowledge of programming is needed to set up an Industrial Pi 3 eco-system and start learning about the cloud.
Achieving nirvana: openness and transparency at last?
Will addressing these unique needs achieve long-awaited openness and transparency? Maybe! The challenges of IIoT are clear and many available technology solutions are well proven. Hardware, especially the chips needed for secure real-time communications, are available now. There is clarity about the importance of device-to-cloud data flows and how they can be achieved. Sensor companies especially recognize the key role they have to play. Openness will increasingly be available to all.
The future of automation will be profoundly different, but the changes will be evolutionary rather than revolutionary. Users who recognize the opportunities offered by cloud-based automation have the chance to migrate their systems step-by-step. By understanding and developing IIoT strategies now, users can stay ahead in an increasingly competitive world.