DESIGN & PRODUCTS EDGE COMPUTING & AI
Industry 4.0: edge computing power is key
By Arkadiusz Grzegorek
Up to now, cloud computing has been one of the biggest
driving forces behind the growth in digital industrial
transformation – particularly expansion of the Industrial
Internet of Things (IIoT) – as well as increasingly high levels of
automation in a wide and diverse range of industry sectors.
However, today it is clear that using the cloud for data management
and storage no longer represents
the whole picture. Edge computing
used in conjunction with cloud computing
– that is, processing data locally then
sending it to the cloud – is becoming an
increasingly indispensable element of the
burgeoning Industry 4.0 concept.
As the number of smart devices
increases due to greater IIoT usage,
there has been a growing demand for
these devices to have the data that they
collect processed by units that are “on
the edge” – ie close to their users. With
this in mind, edge computing is capable
of minimising bandwidth usage and
increasing response times by enabling
data to be stored and managed close
to where it’s required, making it possible to use edge servers
within a much wider selection of applications.
Typical edge devices that many may be familiar with include
sensors for controlling lighting, heating and
other elements. These applications illustrate
the clear benefits to having such sensors as
close as possible to the computing power
being provided, not least the substantial
cost savings that can be achieved by cutting
Fig. 1: The Advantech ADAM-4520-EE robust
converter is capable of connecting easily to
devices.
down on the amount of data that has to
be transferred across long distances and/or
stored in the cloud.
Indeed, there are several potential pitfalls
for users who are looking to export data
from sensors directly on to the cloud. They
can include network latency, poor system
integration, high data management costs
and vulnerability in terms of IT security. For
these reasons, then, it is being acknowledged
that local is best and that taking on
edge computing can have hugely beneficial
impacts on the adoption of otherwise costly
Industry 4.0 systems.
Reducing risk
and increasing agility
In short, edge computing can reduce risks
when connecting devices and sharing data,
increase flexibility and agility within organisations,
reduce the amount of irrelevant
data that is processed, and ensure extremely
Fig. 2: The industrial gigabit ANTAIRA
LNP-0500G-24 PoE switch has a metal
housing and five access ports.
low and predictable latency levels. Finally, in a manufacturing
environment, where edge computing can integrate with
an enterprise resource planning (ERP) system and deliver data
processing in real-time, a company can speed up its automation
process significantly by keeping things local.
A key cornerstone of this new approach – and therefore an
influential factor in the widespread adoption of automation – is
having the ability to process data locally via robust, reliable,
customisable and easy-to-use solutions
that can be used in hitherto challenging
industrial environments. These solutions
include, for example, various types of
industrial controllers and communication
systems that can help to make the
installation and application of Industry
4.0 networks much more feasible to set
up and more straightforward to operate.
It is widely understood that Industry
4.0 requires the use of various tools that
not only facilitate automation but also
help to process and share data widely
and efficiently. In turn, this enables
manufacturing processes to be improved
and refined and end-products made to
precise customer specifications.
To make such a concept work in the real world, special interfaces
and devices are required. These include communication
cards compatible with various bus types,
industrial modules, connection leads, and
modern interfaces in the form of efficient
and lightweight panel computers. Indeed,
converters and industrial modules are now
essential elements in the move towards
Industry 4.0.
One of the solutions available to make
this possible is based around a range of
edge-based industrial controllers from
Brainboxes. These units are capable of
two-way transmission, where components
communicate with a department of engineers
who are able to interpret the data.
This kind of transmission requires communication
cards compatible with various types
of buses. One example is Brainboxes’ PCI
Express (PCIe) bus supported by the IX-100
card. Compatible with Windows, Linux and
other operating systems, which makes it
easier to integrate with existing set-ups, the
IX-100 also features overvoltage protection
and measures just 120x76x18mm. Finally,
the plug-and-play card provides one industry
standard 9-pin RS-232 serial COM port
in a single PCIe slot.
Robust and easy to connect
To use advanced communications technology in an IIoT environment,
it is important to use robust converters and in/out
modules that are capable of connecting easily to devices. For
example, the ADAM-4520-EE isolated RS-232 to RS-422/RS-
485 converter from Advantech (Figure 1), supplied in a durable
Arkadiusz Grzegorek is Head of Automation-Maintenance
Division at Transfer Multisort Elektronik – www.tme.eu
34 News April 2020 @eeNewsEurope www.eenewseurope.com
/eenewseurope
/www.tme.eu
/www.eenewseurope.com
