BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//EURODYN//Schedule//DE
CALSCALE:GREGORIAN
METHOD:PUBLISH
BEGIN:VTIMEZONE
TZID:Europe/Berlin
X-LIC-LOCATION:Europe/Berlin
BEGIN:DAYLIGHT
TZOFFSETFROM:+0100
TZOFFSETTO:+0200
TZNAME:CEST
DTSTART:19700329T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0100
TZNAME:CET
DTSTART:19701025T030000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
UID:487@eurodyn2026.org
DTSTAMP:20261026T144037Z
DTSTART;TZID=Europe/Berlin:20260930T175000
DTEND;TZID=Europe/Berlin:20260930T181000
SUMMARY:Bayesian synergistic filter: Model rectification\, noise paramete
 r estimation and state-space estimation with uncertainty quantification
LOCATION:Room F128
DESCRIPTION:MS 30\n\nPresenter: Si\, Yuning\n\nAuthors: Si\, Yuning\; Kuo
 k\, Sin-Chi\n\nChair: Yan\, Wang-Ji \n\nAbstract: Bayesian synergistic f
 ilter (BSF)\, a novel generic filter for model rectification\, noise par
 ameter estimation and state-space estimation with uncertainty quantifica
 tion\, is proposed. Model discrepancy between the actual system and the 
 computational model is one of the challenging issues for state-space est
 imation. Due to the model discrepancy\, the state-space estimation can b
 e seriously distorted and misleading. The proposed BSF evaluates the mod
 el discrepancy by fully exploiting the subtle information between the me
 asurements and the corresponding predictions. By compensating for the ev
 aluated model discrepancy\, the computational model can be rectified to 
 minimize its deviation from the actual system. Therefore\, the computati
 onal model and the model rectification are synergized to develop a relia
 ble state-space estimation. Another challenging issue is the proper sele
 ctions of the noise parameters. The proposed BSF estimates the noise par
 ameters in an online manner to release the stationarity constraint in co
 nventional state-space filters. Moreover\, by recursively adopting the o
 btained information between the model rectification and noise parameter 
 estimation\, the updated adjustment can be incorporated as the filter ev
 olves. Consequently\, the proposed BSF achieves reliable state-space est
 imation for general linear/nonlinear systems under stationary/nonstation
 ary situations. Furthermore\, by taking the benefit of Bayesian inferenc
 e\, the estimation uncertainties of all estimates can be quantified acco
 rdingly. To demonstrate the efficacy and applicability of the proposed B
 SF\, two simulated examples with various model discrepancies and station
 ary/nonstationary scenarios are discussed. Besides\, a case study of a 6
 00-meter-high television tower subjected to nonstationary earthquake exc
 itation is presented.
URL:https://eurodyn2026.org/submission/487
END:VEVENT
END:VCALENDAR