BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//pretalx//modelica.simtek.cc//amfc2026//speaker//9G3BYT
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-TGUR3V@modelica.simtek.cc
DTSTART;TZID=CST:20260921T153500
DTEND;TZID=CST:20260921T160000
DESCRIPTION:In practical engineering\, equipment parameters are difficult t
 o obtain and measurement data is limited. This paper proposes a PINN-Model
 ica data-driven modeling method that couples a neural network with a Model
 ica model encapsulated as an FMU. Instead of directly constructing governi
 ng-equation residuals in the proposed method\, the FMU supplies a modular 
 physicsbased constraint during training. The neural network predicts both 
 terminal voltage and time-varying physical parameters\, while the FMU comp
 utes a physics-based voltage from the predicted parameters. Data and physi
 cs losses jointly constrain the neural-network voltage output\, and the se
 nsitivity of the physics loss to the FMU input parameters is evaluated num
 erically by central finite differences and combined with neural-network ba
 ckpropagation. A second-order Thevenin battery model is used as a proof-of
 -concept\, and coupled simulation is implemented through the Functional Mo
 ckup Interface (FMI) standard. The framework provides a reusable approach 
 for integrating Modelica physical models with data-driven parameter identi
 fication.
DTSTAMP:20261004T070734Z
LOCATION:Modelica Technology & AI (R1001)
SUMMARY:Research on PINN Data-Driven Modeling Method Based on Modelica Mode
 l - Kaixuan Zhang\, Yanfang Liu\, Yuan Sun\, Heng Wang\, Xudong Wang\, Zhi
 dong Zhang\, Wei Wu
URL:https://modelica.simtek.cc/amfc2026/talk/TGUR3V/
END:VEVENT
END:VCALENDAR