BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//TUC//Events//EN
CALSCALE:GREGORIAN
BEGIN:VTIMEZONE
TZID:Europe/Athens
TZNAME:EEST
DTSTART:19700329T030000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=3
BEGIN:STANDARD
TZOFFSETFROM:+0200
TZOFFSETTO:+0300
TZNAME:EET
DTSTART:19701025T040000
RRULE:FREQ=YEARLY;BYDAY=-1SU;BYMONTH=10
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CREATED:20260706T124216Z
LAST-MODIFIED:20260706T124216Z
DTSTAMP:20260813T093316Z
UID:1786602796@tuc.gr
SUMMARY:Παρουσίαση ΜΔΕ- ΣΠΗΛΙΟΠΟΥΛΟΣ ΓΡΗΓΟΡΙ
 ΟΣ -Σχολή ΧΗΜΗΠΕΡ
LOCATION:Κ2 - Κτίριο ΧΗΜΗΠΕΡ
DESCRIPTION:https://www.chenveng.tuc.gr/el/nea/i
 merologio-paroysiaseon?tx_tucevents2
 _tuceventsdisplay%5Baction%5D=show&t
 x_tucevents2_tuceventsdisplay%5Bcont
 roller%5D=Event&tx_tucevents2_tuceve
 ntsdisplay%5Bevent%5D=8542&cHash=1d0
 7cf76cb7c1088e62803b626cc4c24\nANNOU
 NCEMENT OF PRESENTATION OF POSTGRADU
 ATE THESIS\n \n First Name/Surname: 
 Grigorios Spiliopoulos\n Student Ide
 ntification Number: 2024057019\n Dat
 e: 08/07/2026\n Time: 10:00 (GMT +3)
 \n Room / Zoom Link: https://tucgr.z
 oom.us/j/91477145205?pwd=F3TMap3SobW
 a25XZHko4P8O8kla93b.1  \n Meeting ID
 : 914 7714 5205\n Password: 878737\n
  \n Title: “Algorithmic Synthesis an
 d Extrapolation of Wind Tunnel Aerod
 ynamic Coefficients for Single-Axis 
 Tracker Systems”\n \n Supervisor: Pr
 of. Theocharis Tsoutsos\n Three-memb
 er committee:\n 1. Prof. Theocharis 
 Tsoutsos\n 2. Prof. Apostolos Voulga
 rakis\n 3. Assoc. Prof. Alexandros S
 tefanakis \n \n Abstract:\n The reli
 ability of the structural integrity 
 of single-axis photovoltaic (PV) tra
 cking systems at a utilityscale, are
  highly dependent upon accurate aero
 dynamic pressure coefficients determ
 ined through \n wind tunnel testing.
  The physical scaling limitations pr
 esented by the prototype and the eco
 nomy of \n testing, restrict the abi
 lity to conduct wind tunnel tests fo
 r each row pitch, ground clearance a
 nd tilt \n angle. This, results in c
 ritical gaps in the existing aerodyn
 amic pressure coefficient data. This
  thesis\n presents a computational f
 ramework, referred to as WindFit, th
 at provides the means to synthesize 
 \n the transient aerodynamic loads f
 or the geometries of PV trackers tha
 t have not been tested.\n The method
 ology employs localized polynomial r
 egression to reconstruct the incompl
 ete azimuth and \n tilt data associa
 ted with tracker operation. Geometri
 c synthesis will be achieved by proj
 ecting \n structural parametric data
  onto a Cartesian coordinate system.
  Geometries that fall into an empiri
 cal \n convex hull will be resolved 
 using Barycentric Coordinate Mathema
 tics, while those that fall outside 
 \n the convex hull, will utilize the
  Virtual Anchor Vertex protocol with
 in the framework, to constrain \n th
 e extrapolated data. This topology-c
 onstrained boundary gradient approac
 h prevents the numerical \n divergen
 ce and reversed aerodynamic weightin
 g that are associated with tradition
 al unconstrained \n extrapolation me
 thods.\n The computational effective
 ness of the framework is verified by
  two benchmark approaches, which \n 
 use deterministic Inverse Distance W
 eighting (IDW) and stochastic Gaussi
 an Process Regression \n (GPR) model
 s. The comparative spatial validatio
 n of the models shows that IDW opera
 tes as a low \n pass filter, creatin
 g spatial blurring that greatly unde
 r-predicts the high-end boundary loa
 ds, as well \n as the Venturi amplif
 ications. The GPR model produces the
  same high-fidelity precision as Win
 dFit, \n but due to it's reliance on
  infinite mathematical smoothness, G
 PR removes sharp aerodynamic \n disc
 ontinuities and dramatically under-p
 redicts detached tip vortexes during
  extrapolation. \n Finally, WindFit 
 integrates local geometric interpola
 tion with boundary conditions based 
 upon \n physics to accurately model 
 complex flow separation phenomena. I
 n addition, it uses Barycentric \n a
 lgebraic weights to eliminate the ex
 tensive latency involved with perfor
 ming machine learning \n matrix inve
 rsions, thus making this approach hi
 ghly suited for real-time parametric
  structural design\n
STATUS:CONFIRMED
ORGANIZER;RSVP=FALSE;CN=TUC;CUTYPE=TUC:mailto:webmaster@tuc.gr
DTSTART:20260708T100000
DTEND:20260708T110000
TRANSP:OPAQUE
CLASS:DEFAULT
END:VEVENT
END:VCALENDAR