BEGIN:VCALENDAR VERSION:2.0 PRODID:-//jEvents 2.0 for Joomla//EN CALSCALE:GREGORIAN METHOD:PUBLISH 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 UID:df7c61016503aafedb3e093c3ed4e7b36 CATEGORIES:CIGRE Academy Webinar SUMMARY:Session 2 | Accelerating AI on the Grid: PMU Fundamentals & Intro to AI DESCRIPTION:
Instructors p>
Alexandra “Sascha” von Meier, University of California, Berkeley
Kevin Jones, Dominion Energy
Sean Murphy, PingThings
Laurel Dunn,
University of California, Berkeley
Mohini Bariya, University of Calif
ornia, Berkeley
Miles Rusch, University of California, Berkeley
Abstract
This two-part virtual tutorial is geared at
training practitioners on how to use AI to analyze PMU (synchrophasor) data
. The first session (PMU fundamentals) will provide a foundation for unders
tanding and interpreting PMU measurement data and applications to practitio
ners at electric utilities. The second session (Intro to AI) will provide a
n introduction to Artificial Intelligence (AI), and will give attendees han
ds-on experience using AI to analyze PMU data in Python. The course will di
scuss opportunities for PMU data analytics to change best-practices in the
industry, and participants will become practiced at using tools that can st
reamline workflows for digesting and visualizing time series data at scale.
Day 2 | October 28, 2020 | Intro to AI
D
ay 2 will provide an introductory training for practitioners to start using
AI in their own work. The course will begin by covering fundamental concep
ts related to AI and big data, and will motivate the need for practitioners
in energy to become well-versed in AI tools. The course will step through
interactive coding exercises using the National Infrastructure for AI on th
e Grid (NI4AI) Python API to access publicly hosted PMU data. NI4AI is buil
t on PingThings’ PredictiveGridTM platform, a state-of-the-art tool optimiz
ed to support big data visualization and analysis workflows. Participants a
re requested to come prepared with a login to ni4ai.org and with Python ins
talled (both are free). Exercises will assume some familiarity with Python,
though participants with no programming experience will benefit from expos
ure to the concepts and tools presented.
Intro to AI ( Day 2)
1. Big data analytics and prediction
(Sean Murphy, 20 min)
2. Interfacing with PMU data i
n Python
(Laurel Dunn, 25 min)
2.1. Accessing the
NI4AI API
2.2. Phasor visualization
2.3. “Unwrapping” phase angl
e
3. Use cases for PMU data
3.1. Detecting voltage
sags
(Mohini Bariya, 30 min)
3.2. Analyzing frequency
(Miles Rusch, 30 min)
4. Closing remarks and outloo
k for AI on the grid
(Sean Murphy, 15 min)