Linking PLCs to the cloud isn’t difficult, but technical questions still arise
More companies are looking at the Internet of Things and Industrie 4.0, prompting a growing number of industrial companies to investigate cloud computing. Utilizing the cloud for storage and some data processing can bring significant benefits to a broad range of companies, especially if lightweight protocols designed for IoT.
Today, there is a range of options for storing data gathered in automated facilities. SCADA systems and specialized off the shelf data collection options are commonly used. But they are all expensive and require advanced programming skills – the system must be configured and code must be written for the environment of each industrial site.
Storing data in the cloud has become a viable alternative, one that is fairly straightforward to implement. Programming is simple or non-existent, as are costs. Responses are also quick – cloud systems can provide near real time performance.
While it’s fairly easy to start utilizing the cloud, no industrial engineering configuration is without technical issues. Boris Cherkasskiy, principal engineer at DMC, addressed the concerns for one common Siemens platform at a recent Siemens conference. His presentation is entitled “Pushing Data from S7-1200 to Cloud.”
Cherkasskiy picked the Siemens S7-1200 PLC platform because it’s well known to automation engineers and widely used in the automation world. It’s also versatile, supporting a wide range of industrial protocols including Profinet, Modbus, Profibus and custom TCP/IP communications.
Unfortunately, none of these standard industrial protocols are currently supported by IoT infrastructure. Moreover, there aren’t many standard IoT protocols at all, IoT devices may use a range of communication options. But IoT systems commonly are based on Ethernet, Wi-Fi and other IP-based protocols.
A range of technologies can be used to overcome the incompatibilities that arise when protocols don’t fit the standard practices of today’s cloud computing systems. HTTP is perhaps the best known. SOAP (Simple Object Access Protocol) is another option, along with REST (Representational State Transfer).
However, these alternatives are heavy to implement, so they take a lot of computing power. That can be a big issue in PLC-based applications, where computing resources are limited because the system is designed to efficiently perform machine-control tasks, but not general web-based communication. Factories and other industrial facilities need protocols designed for the types of messages normally used by automated equipment.
Cherkasskiy suggests using Message Queuing Telemetry Transport. MQTT is a protocol created for machine to machine Internet of Things equipment. It is open, simple and lightweight, and it works on top of TCP-IP, so it’s PLC friendly.
The protocol has already seen extensive acceptance in the U.S. The list of the companies that have deployed the technology includes IBM, Facebook, Amazon and Microsoft. Simplicity is the key benefit of MQTT. That’s underscored by the documentation supplied to those who want to use it. The entire MQTT specification is less than 50 pages long.
MQTT is an event driven protocol built using brokers that handle messages. The standard employs a publisher- subscriber architecture in which the publishers send messages to the broker, which then distributes the messages to subscribers. Using a broker ensures that messages are sent out. The broker doesn’t care what type messages are being transmitted, it sends them all.
Because of the simplicity of the MQTT protocol, it is well suited to be implemented on Siemens S7-1200 PLC. An open implementation of the MQTT protocol demonstrated during the presentation is available for free download: https://github.com/boriz/MQTT-SN-Arduino/tree/master/S7-1200
While the protocol has a number of benefits, it also has shortcomings. One of the foremost issues is that encryption isn’t implemented in the standard. That can be a major issue given all the cybersecurity breaches that have occurred over the past few years.
Cherkasskiy says that users can overcome that issue by bridging the local MQTT broker to the Internet using a secured channel. For example, MQTT over SSL (Secure Sockets Layer) or VPN (Virtual Private Network) provide security that can help prevent intrusions. Brokers, which are quite simple applications, can easily bridge messages to another broker to help facilitate secure communications. Brokers in the cloud will provide access to remotely-stored data.
Once networks are set up, utilizing data is the obvious concern. Cloud-based brokers typically include a visualization platform that gives users a range of views so they can gain insight.
A range of cloud computing services can be used. Cherkasskiy noted that Microsoft Azure is a well-proven cloud platform that is very flexible, so it can easily be molded to fit the requirements of many different automated facilities. Some programming is required to handle this customization, he notes.
An emerging cloud platform, Adafruit, is very simple to implement, making it attractive to newcomers. Setting a few configuration items is all that’s needed to get going. However, the Alpha mode that’s currently available does not allow for storing much historical data.
IBM’s Watson continues to expand its reach, providing an advanced platform that can address complex requirements. However, it’s also a work in progress, and it can be confusing to configure. That confusion stems in part from the many options of the many options available to industrial users.
There’s also an Android platform that’s easy to configure. A few drag and drop connections are all that’s needed to get up and running. However, Cherkasskiy notes that the platform is somewhat limited in its functions.
Overall, the benefits that can be gained by moving to the cloud easily outweigh the challenges of making the transition. Ongoing improvements made by cloud service providers and standards bodies will continue to make it easier for industrial users to improve efficiency and simplify operations.
Have an Inquiry for Siemens about this article? Click Here >>