BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Chicago
X-LIC-LOCATION:America/Chicago
BEGIN:DAYLIGHT
TZOFFSETFROM:-0600
TZOFFSETTO:-0500
TZNAME:CDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0500
TZOFFSETTO:-0600
TZNAME:CST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20230124T170804Z
LOCATION:C1-2-3
DTSTART;TZID=America/Chicago:20221116T083000
DTEND;TZID=America/Chicago:20221116T170000
UID:submissions.supercomputing.org_SC22_sess274_rpost138@linklings.com
SUMMARY:Novel Multi Data Acquisition and Hybrid Neural Network for Pipe In
 spection and Imaging
DESCRIPTION:Posters, Research Posters\n\nNovel Multi Data Acquisition and 
 Hybrid Neural Network for Pipe Inspection and Imaging\n\nOoi, Ozakin, Most
 afa, Khater, Aljarro...\n\nA new machine learning-based non-destructive te
 sting (NDT) technique for the examination of conductive objects is present
 ed. NDT of objects behind barriers utilize the defect-induced distortions 
 on electromagnetic (EM) fields to detect flaws in the structure of inspect
 ed targets. Such distortions are highly non-linear, requiring significant 
 amounts of data for training neural networks. To this end, a massively par
 allelized data generation framework is proposed in conjunction with a mult
 i-frequency hybrid neural network (MF-HNN), to create a physics-informed i
 nversion AI model. The performance of the resulting inversion algorithm is
  applied on casings, where tubular pipes are inspected. For data generatio
 n, physics-based solvers are employed to simulate the EM field distributio
 n resulting from pipes with defects. The large-scale distribution of this 
 step leads to 43 times faster execution than a single CPU. This allows the
  MF-HNN to achieve significantly improved generalization performance and t
 o generate high-resolution cross-sectional images of the pipelines.\n\nReg
 istration Category: Tech Program Reg Pass, Exhibits Reg Pass
END:VEVENT
END:VCALENDAR