BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//modelica.simtek.cc//amfc2026//talk//GVDZN9
BEGIN:VTIMEZONE
TZID:CST
BEGIN:STANDARD
DTSTART:20000101T000000
RRULE:FREQ=YEARLY;BYMONTH=1
TZNAME:CST
TZOFFSETFROM:+0800
TZOFFSETTO:+0800
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:pretalx-amfc2026-GVDZN9@modelica.simtek.cc
DTSTART;TZID=CST:20260922T135000
DTEND;TZID=CST:20260922T141500
DESCRIPTION:Data center server rooms require thermal prediction components 
 that can be called by digital twins and system simulation platforms under 
 changing IT loads and cooling conditions. CFD provides detailed spatial in
 formation but is too expensive for step-wise online simulation. Machine-le
 arning surrogates are efficient\, but offline models often lack state mana
 gement\, rolling prediction\, and physical safeguards for continuous simul
 ation. This paper presents an FMI(Functional Mock-up Interface)compatible 
 fast thermal prediction FMU (Functional Mock-Up Unit) for data center serv
 er rooms. The FMU embeds a physics-guided graph neural thermal predictor a
 nd exposes standardized FMI inputs and outputs for supply-air temperature\
 , fan state\, IT load\, historical thermal states\, and predicted temperat
 ures. Each simulation step performs input validation\, historywindow updat
 e\, model inference\, output correction\, and FMI variable write-back. A C
 RAH M1-08 shutdown replay case shows an overall RMSE of 0.284 ° C and an 
 average step time of 0.0656 s\, supporting near-real-time use in thermal d
 igital twins.
DTSTAMP:20261004T074046Z
LOCATION:FMI & MBSE (R2001)
SUMMARY:An FMI-Based Fast Thermal Prediction FMU for Real-Time Digital Twin
 s of Data Center Server Rooms - Haobo Guo\, Siyuan Yang\, Chuangkang Yang\
 , Xiangmiao Hao\, Jie Wu\, Wenquan Tao
URL:https://modelica.simtek.cc/amfc2026/talk/GVDZN9/
END:VEVENT
END:VCALENDAR