Cleanroom Optimisation Through Machine Learning

Cleanroom Optimisation Through Machine Learning
Summary

Objective

  • Analyse the underlying potential for energy reduction in cleanrooms.
  • The possibility of reducing current air change rate without affecting critical quality parameters were tested in four cleanrooms using fixed air change rate reduction or dynamic air change control.
  • Cleanroom and HVAC data was collected, a simulation model was developed replicating each of the four cleanrooms and its associated HVAC systems to test with different air change rates and analyse its implications.
  • A machine learning algorithm was developed to implement the dynamic control and was integrated with the cleanroom simulation model.
Results type(s)
Unfold all
/
Fold all
Attached files
File Type
IMR_steripack_stream.pdf PDF
More information & hyperlinks
Country: IE
Address: Unit A, Aerodrome Business Park, Rathcoole, Co. Dublin D24 WCO4
Geographical location(s)
Structured mapping
Unfold all
/
Fold all
Comment:

IMR IIoT Toolkit developed for factory deployment.  High voltage power components separated in custom enclosures.

It includes a range of sensors, IIoT edge components which perform data collection and aggregation at the edge of the network and which then sends the data to IMR’s IIoT Platform which can either be installed locally, be based in the cloud or even both. 

IIoT Toolkit software connectors allow interfaces to be established with operational technologies such as BMS over BACnet or EMS over HTTP.

 

Comment:

Over 700 data points from heating, cooling and ventilation systems are supplied to Building Management System via BACnet controllers.

IMR are using our IIoT Platform installed on-site to read this data from the BACnet controllers.  We supplement it with data from IMR sensors (cleanroom occupancy, particle counts, PIRs, door sensors).

This data is then sent in real-time to a containerised cloud-based IIoT Platform where it can be accessed by the Energy Team and the Data Analytics Team.