Next IIoT wave will be a population of digital twin. A digital twin is a real-time digital replica of a physical device. Developing an embedded digital twin allows superior device diagnostic and failure anticipation. Discover how to to implement an embedded digital twin using real-time monitoring, physical models, and machine learning

1. Milan, Italy
8.11.2018
Next IIoT wave:
embedded digital twin for
manufacturing by I.R.S

3. ni.com
Next IIoT wave:
embedded digital twin for manufacturing
Gianluca Bacchiega - I.R.S. srl
www.irsweb.it

4. IIoT solutions and trends
What’s next. Digital twin. What ?
Embedded Digital Twin
A twin using model technology 4.0
Conclusion
1
2
3
4
5

5. IIoT solutions and trends

6. Based on Moor Insights & Strategy's report "Segmenting the Internet of Things (IoT)"
SMART
Factory
Grid
Machine
City
Car
INDUSTRIAL
Internet of Things
SMART
Phone
Wearable
TV
Appliances
Home
CONSUMER
Internet of Things
Connectivity
Data Analytics

8. NI’s End-To-End Solution Architecture for IIoT
TheEdge
Sensors/Actuators
IT Infrastructure
(Big Data Analytics, Mining)
EdgeIT
(Local,Remote,Cloud)
Corporate/
FederatedIT
Data Acquisition, Control, and
Analysis Systems
(Test,Monitoring,Logging,Control)
MeasurementHardwareandFirmware
AnalysisandVisualizationSoftware
Decision Making and Control
Data Aggregation and Management

9. NI’s End-to-End Solution Architecture for IIoT

10. What’s next? Digital twin.
What ?

11. “Digital twins are becoming a business imperative, covering the entire lifecycle of an
asset or process and forming the foundation for connected products and services.
Companies that fail to respond will be left behind.”
Thomas Kaiser, SAP Senior Vice President of IoT
Ganesh Bell, chief digital officer and general manager of
Software & Analytics at GE Power & Water
“For every physical asset in the world, we have a virtual copy running in the cloud that
gets richer with every second of operational data
Digital twin Explosion:
billions of twins in next five years

12. A digital twin is a real-time digital replica of a physical device.

13. ModelSensors
A digital twin is a real-time digital replica of a physical device.
Sensors
digital twin

14. A digital twin is a real-time digital replica of a physical device.
chiller
digital twin
chiller
It’s more than a model

15. A simple digital replica ?
History
Log the device history
Sensors
digital twin
Future
Forecast device future

16. A bridge between the physical and digital world
Sensors
Physical device Data acquisition

17. Big data
Monitoring
Machine
Learning
& Models
A bridge between the physical and digital world

18. A bridge between the physical and digital world
Big data
Digital Twin
Monitoring
Machine
Learning
& Models
Sensors
Physical
devices
Data acquisition

19. A bridge between the physical and digital world
with Value and ROI embedded
Digital Twin
Maintain & log
the entire life of an asset
Understand
using a learning model
Warn
on health & efficiency
Predict
Failure and Optimize
Enhance
add virtual sensors

20. Embedded Digital Twin

21. Testing in unfeasable conditions
Shorter testing time Better accuracy and quality
Example of embedded digital twin for product testing

24. A platform ready for change

25. 1. Lifelong Device history
2. Real time model computed
virtual sensor
3. Real Time predictive alert
From monitoring to embedded digital twin

26. CompactRIO
Single-Board inside

27. A twin using model technology 4.0

28. Model technology 4.0
Physical Model
Fluid
properties
Components
Compressor
Heat
exchangers
Fans
Phenomena
Heat transfer
Mass transfer
DAQ
correction
embedded
digital twin
Machine learning

29. Physical Model
Physical Model
Fluid
properties
Components
Compressor
Heat
exchangers
Fans
Phenomena
Heat transfer
Mass
transfer
DAQ
correction
embedded
digital twin
The phenomenological model,
based on equations,
can identify the causes of
a possible malfunction

30. Machine learning
The machine learning approach needs no
detailed knowledge about machine operation.
It needs a learning phase to be able to
predict the system performance.
Feature extraction
Machine learning
algorithm
Unsupervised
data
New data
Predictive model
Supervised
data

31. Physical
Model
Fluid
properties
Components
Compressor
Heat
exchangers
Fans
Phenomena
Heat transfer
Mass
transfer
DAQ
correction
Sensor
Data
Physical
Model
Machine
Learning
Test system
Diagnostic detail and easy implementation

32. Merging model technology using NI platform
LabVIEW Machine
Learning Toolkit
embedded
digital twin

33. ▪ NI CompactRIO testing 4 appliance simultaneously
▪ Sensor optimization
▪ Digital twin for shortening testing time
Test systems for fridge testing on 100 % production
Test system for fridge : 100 % production

34. Conclusions
Lower complexity, reduce development time,
and add machine learning to your IIoTdevice
using NI platform and digital twin technology

35. info@irsweb.it
Thank you for your attention.
any question or inquiry