BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//pretalx.com//16thmodelicafmiconference//talk//QBWPWE
BEGIN:VTIMEZONE
TZID:CET
BEGIN:STANDARD
DTSTART:20001029T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
TZNAME:CET
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
END:STANDARD
BEGIN:DAYLIGHT
DTSTART:20000326T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
TZNAME:CEST
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
END:DAYLIGHT
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-16thmodelicafmiconference-QBWPWE@pretalx.com
DTSTART;TZID=CET:20250910T152500
DTEND;TZID=CET:20250910T155000
DESCRIPTION:This study presents a reduced-order model (ROM) of a large-scal
 e transcritical CO2 heat pump designed by MAN Energy Solutions\, currently
  operating in Esbjerg and under construction in Aalborg\, Denmark. The ori
 ginal high-fidelity model\, implemented in Modelon Impact and presented at
  Modelon Innovate\, captures the complex thermodynamic and control dynamic
 s of the heat pump but comes with high computational cost\, making it impr
 actical for real-time applications. To address this issue\, we trained a n
 eural network ROM using simulation data from the Modelica-based model\, en
 abling significant reduction in simulation time without compromising accur
 acy. We demonstrate how the ROM is integrated back into the Modelica frame
 work to facilitate fast simulations\, paving the way for real-time control
  scenarios and digital twin implementations. Performance evaluations show 
 that the ROM retains key dynamic behavior and delivers simulation speed-up
 s by more than two orders of magnitude. The model captures start-up and sh
 ut-down scenarios\, assessing system reliability and transient behavior\, 
 and evaluates the heat pump’s flexibility for grid balancing\, respondin
 g to variable load demands and grid needs. This approach not only accelera
 tes system-level studies but also supports online applications such as fau
 lt detection\, predictive control\, and system optimization. Our work high
 lights the potential of combining physics-based modeling with data-based m
 achine learning techniques to bridge the gap between simulation fidelity a
 nd computational efficiency\, crucial for the development of next-generati
 on digital twins in the energy domain.
DTSTAMP:20260519T213413Z
LOCATION:203
SUMMARY:Neural Network-Based Reduced-Order Model of a Large-Scale\nCO₂ He
 at Pump for Real-Time Simulation and Digital Twin\nApplications - Kamran I
 ranshahi\, Jakob Hermeler
URL:https://pretalx.com/16thmodelicafmiconference/talk/QBWPWE/
END:VEVENT
END:VCALENDAR