IEEE Electrification Magazine - September 2017 - 35

IIOT: An Enabling Technology
The IoT is the name given to the future of connected devices, as illustrated in Figure 2. Within the IoT, there are two
distinct subsets: the consumer IoT, which includes wearable computers, smart household devices, and network
appliances, and the IIoT, which includes smart systems
within processes such as power generation and distribution, manufacturing, medicine, and transportation. The
amount of data generated and distributed in a shipboard
IIoT application is rapidly increasing, approaching the definition of big data at a steady pace, with more sensors and
connected, integrated subsystems with IP stacks. In addition, the harvesting of thick data-qualitative data, such as
a human operator's experience-based knowledge-is also
increasing. This is a consequence of turning the system
development toward autonomous system solutions, where
smarter systems can compete with and replace human
decision making.
The rapidly increasing amount of available data in industrial (interconnected) systems calls for a total system architecture (or, more to the point, a system philosophy) that can
handle the large amount of qualitative and quantitative
data in different formats (protocols) and distribute that data
in standardized formats to subscribing system nodes. An
example of such a standardized data format is a DDS, which
is an object management group machine-to-machine (middleware) standard. With the large amount of data, despite
the availability of high bandwidth, it will be important to
reduce data traffic to put less strain on the system solution.
This can be achieved by choosing the right system architecture philosophy. As an example, instead of continuous distribution of data at a given frequency, data can be sent when
triggered by an event, often implemented as asynchronous

Field Equipment

Cables
HW: Analog and Digital
Serial: CAN, Modbus,
and so on

IIOT

Figure 2. An illustrative scope of the IIOT.

data communication. Such an event can, for instance, be
when a valve or bus-tie breaker opens or closes. To send a
continuous binary status update of a watertight door that is
usually open puts unnecessary strain on the bandwidth, not
to mention the subsystems subscribing to the data that
need to process all received data packages tagged with that
same topic. Figure 3 illustrates an example of a potential
IIoT system architecture designed for shipboard control and
information handling systems.
An IIoT-based system architecture provides a foundation to rethink system solutions where the traditional one-to-one communication structure is exchanged for
many-to-many communication through a fog-based data
distribution. As data are available to all who subscribe,
the potential of collaborative and coordinated control
between different subsystems to form intelligent and
configurable (dynamic) system solutions continues to

IO
IO Modules
and
Controller

Network
Control
Ethernet DDS Com. Control Logics
Publish/Subscribe
and Status
Switch A

IO IO Controller

User Interface
Remote Monitoring
and Control

Controller
Controller App
Com
Logics

Managed
SNMP

Config

Firmware IO App
API Logics Com
Config

GUI Station

Config

GUI App
Com
Logics

Managed
SNMP

Config

GUI

Switch B

Value-Oriented Data
<>

Object-Oriented Data
<>
Signals, Alarms, Component Data

Figure 3. An example of an IIoT architecture. The physical signals (value-oriented data), which are based on a vast range of protocols, are harvested
by an IO-controller that transforms them to object-oriented data and publishes that data to subscribing system nodes. API: Application programming
interface; App: application; Com: communication; Config: configuration; HW: hardware; GUI: graphical user interface; CAN: controller area network;
SNMP: Simple Network Management Protocol. (Images courtesy of Ulstein Power & Control AS, CAT, and Schneider Electric.)

IEEE Elec trific ation Magazine / S EP T EM BE R 2 0 1 7

Table of Contents for the Digital Edition of IEEE Electrification Magazine - September 2017