BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//talks.cam.ac.uk//v3//EN
BEGIN:VTIMEZONE
TZID:Europe/London
BEGIN:DAYLIGHT
TZOFFSETFROM:+0000
TZOFFSETTO:+0100
TZNAME:BST
DTSTART:19700329T010000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=-1SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:+0100
TZOFFSETTO:+0000
TZNAME:GMT
DTSTART:19701025T020000
RRULE:FREQ=YEARLY;BYMONTH=10;BYDAY=-1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
CATEGORIES:Isaac Newton Institute Seminar Series
SUMMARY:Feedback-based online algorithms for time-varying
optimization: theory and applications in power sys
tems - Emiliano Dall'anese (University of Colorado
)
DTSTART;TZID=Europe/London:20190110T113000
DTEND;TZID=Europe/London:20190110T123000
UID:TALK116797AThttp://talks.cam.ac.uk
URL:http://talks.cam.ac.uk/talk/index/116797
DESCRIPTION:The talk focuses on the synthesis and analysis of
online algorithmic solutions to control systems or
networked systems based on performance objectives
and engineering constraints that may evolve over
time. Particular emphasis is given to applications
in power systems operations and control. The time
-varying optimization formalism is leveraged to mo
del optimal operational trajectories of the system
s\, as well as explicit local and network-level co
nstraints. The design of the algorithms then capit
alizes on an online implementation of primal-dual
projected-gradient methods\; the gradient steps ar
e\, however\, suitably modified to accommodate act
ionable feedback in the form of measurements from
the network -- hence\, the term feedback-based onl
ine optimization. By virtue of this approach\, the
resultant running algorithms can cope with model
mismatches in the algebraic representation of the
system states and outputs\, they avoid pervasive m
easurements of exogenous inputs\, and they natural
ly lend themselves to a distributed implementation
. Under suitable assumptions\, Q-linear convergenc
e to optimal solutions of a time-varying convex pr
oblem is shown. On the other hand\, under a genera
lization of the Mangasarian-Fromovitz constraint q
ualification\, sufficient conditions are derived f
or the running algorithm to track a Karush-Kuhn-Tu
cker point of a time-varying nonconvex problem. Ex
amples of applications in power systems will be pr
ovided.

Joint work with: A. Simonetto\,
Y. Tang\, A. Bernstein\, and S. Low.
LOCATION:Seminar Room 1\, Newton Institute
CONTACT:INI IT
END:VEVENT
END:VCALENDAR