Testbed to integrate sensors into smart factories

16 January 2017

Daniel Walldorf and Michael Hilgner discuss the work which is being undertaken to connect existing sensors to Cloud-based IT systems, to allow them to become part of a smart factory.

Implementing applications for the Industrial Internet of Things (IIoT) requires the analysis of a great deal of production data (Operational Technology, OT) in IT systems. However, for many this is still a difficult task due to limited OT/IT connectivity. 

To help overcome this issue, TE Connectivity (TE) and three partners have set out to implement a testbed that shows how existing sensors can be connected to the cloud and machines upgraded for smart applications under the umbrella of the Industrial Internet Consortium (IIC). The core component is a smart I/O module, which TE will be presenting in prototype form at this year’s SPS IPC Drives event.

The IIoT, Industry 4.0 and Smart Factory are essentially based on a comprehensive digitalisation of factories. The aim of the IIC is to develop an IoT ecosystem in which appropriate solutions are discussed, described and tested. One of the key activities of this open organisation, which already has more than 250 members, is creating experimental platforms (testbeds) that will be used to put future technologies, products and applications through their paces to find out whether they are feasible and what benefits they offer to users.

One example of this is the previously mentioned testbed for connecting existing sensors to the cloud. TE, SAP, ifm and the OPC Foundation are using this to demonstrate that energy consumption can be sustainably reduced using an intelligent solution covering everything from capturing, transporting and processing the data, to optimising the process. The testbed includes a variety of software tools and a hardware component for simple integration into existing systems that are used to transfer information from the machine to the IT system via an OPC Unified Architecture (UA) interface. 

Although sensors in machines already record a great deal of information, only a fraction of it is available for analysis. TE's research indicates that, on average, controllers process only 5% of the data. By changing their programming, it would be possible to access the remaining 95%, but apart from the fact that not all components have sufficient computing power, this approach is very complicated.

In addition to the controllers, the data models for automation components - such as sensors, actuators and motors - which are usually made available via connected gateways, also need to be adapted. This needs to be done for each of the manufacturers machine types, resulting in high costs, as well as unpredictable risks for the uptime of the machines.

The alternative is to capture the data at sensor level, although this would require smart connectivity technology that gives intelligence to simple field devices. For this reason, TE has developed an I/O module for the IIC test bed, which, in addition to its I/O functionality, acts as an Edge computer. Usually, I/O modules capture data from sensors and actuators and forward it to the controllers via a higher-level bus system. With the smart variant, a further communication path can be set up in addition to this real-time route. This can be used to aggregate data for monitoring the machine's energy consumption and then transfer it to the IT system without impairing the performance of the control network. 

Because I/O modules are already available in machines today, they can easily be replaced without changing their architecture or interfaces. The semantics of the data models used to describe the remotely configurable I/O Link sensors of the IIC test bed are also defined in the open IIOD standard (I/O Device Description). This means that these sensors are not only very easy to integrate into the IT system, but all their data can also be managed there.

TE has now further developed the smart I/O module into an independent cyber-physical system that effectively bridges the gap between the world of automation and the world of IT. It is easy to retrofit and highly versatile, not merely for monitoring energy consumption but also for other applications that can be used to boost productivity, such as predictive maintenance.

The prototype of this system - IoT OmniGate - comprises hardware plus an administration environment and a data management tool. The hardware provides interfaces both to conventional sensors for factory automation, such as I/O-Link via M12, and for communication with the SPS via Ethernet – the prototype supports Profinet, but will also support other protocols in the future. There is also an Ethernet interface for connection to the IT/cloud. 

This interface can be used to set up the IT connection for the sensors and controllers and to generate IP interfaces for passing on the recorded data in structured form to other applications which means that the data no longer has to be requested from different locations. The data is also always in the same format, which makes work a lot easier.

Pilot projects are showing how factories can benefit from smart connectivity technology. These projects, which are geared to the classic value drivers that form the basis of operational excellence (resources, plants/processes, work, inventories, quality, safety and flexibility), have made it possible to improve overall equipment effectiveness (OEE), something that would previously have taken about a year a year has been achieved within three months. How the value drivers are weighted in each case varies from company to company, in this respect, digitalising the factory offers new technical possibilities that can be used to cost-effectively improve the classic value drivers.

The IIC test bed demonstrates how smart connectivity technology helps connect data from existing sensors to the cloud and then analyze it in an IT system in order to increase process efficiency – in this case, monitoring energy consumption. Existing infrastructures in particular need upgradable solutions that minimize downtime, save costs and increase productivity. Pilot projects at TE's own factories have shown that this approach also works in practice.

Daniel Walldorf is a Industrial IoT Platforms & Ventures TE Industrial; Michael Hilgner is a Manager Consortia and Standards TE Connectivity.

Contact Details and Archive...

Print this page | E-mail this page