BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/New_York
X-LIC-LOCATION:America/New_York
BEGIN:DAYLIGHT
TZOFFSETFROM:-0500
TZOFFSETTO:-0400
TZNAME:EDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0400
TZOFFSETTO:-0500
TZNAME:EST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20210402T160100Z
LOCATION:Track 5
DTSTART;TZID=America/New_York:20201118T140000
DTEND;TZID=America/New_York:20201118T143000
UID:submissions.supercomputing.org_SC20_sess162_pap220@linklings.com
SUMMARY:Smart-PGSim: Using Neural Network to Accelerate AC-OPF Power Grid 
 Simulation
DESCRIPTION:Paper\n\nSmart-PGSim: Using Neural Network to Accelerate AC-OP
 F Power Grid Simulation\n\nDong, Xie, Kestor, Li\n\nThe optimal power flow
  (OPF) problem is one of the most important optimization problems for the 
 operation of the power grid. It calculates the optimum scheduling of the c
 ommitted generation units. In this paper, we develop a neural network appr
 oach to the problem of accelerating the current optimal power flow (AC-OPF
 ) by generating an intelligent initial solution.\n\nThe high quality of th
 e initial solution and guidance of other outputs generated by the neural n
 etwork enables faster convergence to the solution without losing optimalit
 y of final solution as computed by traditional methods. Smart-PGSim genera
 tes a novel multitask-learning neural network model to accelerate the AC-O
 PF simulation. Smart-PGSim also imposes the physical constraints of the si
 mulation on the neural network automatically.  Smart-PGSim brings an avera
 ge of 49.2% performance improvement (up to 91%) with respect to the origin
 al AC-OPF implementation, computed over 10,000 problem simulations, withou
 t losing the optimality of the final solution.\n\nTag: Machine Learning, D
 eep Learning and Artificial Intelligence, Requirements, Performance, and B
 enchmarks, Reliability and Resiliency\n\nRegistration Category: Tech Progr
 am Reg Pass
END:VEVENT
END:VCALENDAR

