Enumerate Customers Experiencing Outages with Advanced Analytics

Virtual Breakout Session

Tuesday, June 9, 2020
12:00pm EDT
Loading Sessions

Register now!

This is a consume-at-your-own-pace event, it will take place 100% online.  All of the content has been pre-recorded and will be released at its scheduled time. If you can’t listen or watch a session when it’s released, remember that you can access it at ANY TIME in the next 30 days.

  • Listen to podcasts
  • View webinars
  • Watch videos
  • Access articles, white papers, case studies and research reports
  • Chat with speakers and attendees via discussion forums
  • Visit sponsor profile pages

DTE will discuss how they have overcome the typical utility struggle to report real-time customers experiencing outages, which occurs primarily during inclement weather days where thousands of customers are waiting to be restored. Samia, a Sr. Analyst at DTE, shares how utilizing AMI and Cloud data along with OMS to accurately capture the current number of customers experiencing outages, helps them to meet the growing demand for real-time information.

Feel free to ask questions in the discussion forum below - speakers will be responding as quickly as they can. View this webinar by the end of the day on June 9 and get entered to win a $100 Amazon Gift Card.

Speakers:
Sponsored by:

Discussion

13
Leave a Reply

4 Comment threads
9 Thread replies
0 Followers
 
Most reacted comment
Hottest comment thread
8 Comment authors
Minjie YuCarol BogaczJames MackJohn Carpenterchris bowman Recent comment authors
newest oldest most voted
Joselle Barnett
Member
Joselle Barnett

What products/tools does GE use in order to perform their advanced analytics? What does GE’s advanced analytics and support team (DBA, data modelers, etc.) look like?

chris bowman
Member
chris bowman

Hi Joselle! Under the hood of GE’s advanced analytics platform there are many technologies, for data integration, storage, compute, then on up to the UI layers. Data base is the most scalable relational DB with a timeseries plugin so it can hold both. Storage is a pre-defined IEC CIM based model similar to what many electric Utilitities do. AI/ML compute occurs in a clustered Spark environment. So our data scientists, write the AI/ML code typically in Python – to execute in Spark – sourcing input data from our internal IEC CIM storage service. Hope this helps!

chris bowman
Member
chris bowman

I saw your additional question of what our support/product team looks like? We offer a managed service to host the Grid Analytic products, so we have cloud ops and application support staff to handle L1-L4 support issues, and to do routine patching and upgrades. If you were interested in our Product Development team – then our teams consist of about 25% data scientists, of which we source typically people with advanced degrees first in Power Engineering but who also have Utility experience and more recently Data Science advanced degrees. 75% of the team are professional software engineers – who are… Read more »

Chris Fischer
Member
Chris Fischer

How did you address the problem of not receiving AMI PONs from some/many meters during a larger outage event?

Minjie Yu
Member
Speaker
Minjie Yu

We looked mainly at the AMI 5 minute voltages data and would take the most recent read. During storms we would increase the number of interrogations, so we have more recent data being pulled. We would pull 15 minute interval data as well during catastrophic storms.

Kevin Praet
Member
Kevin Praet

DTE team, great information – what are some of the tools you used to develop your algorithm? Can you also share any challenges/issues you had when developing the algorithm?

Minjie Yu
Member
Speaker
Minjie Yu

We use Alteryx, similar to SPSS, to combine datasets and store our data. For databases, we use Cloudera Impala, and Oracle SQL. The challenge is that there was a lot of data during inclement days, and ensuring that we were accurately capturing the right number of customers. Fortunately we were able to validate against our OMS system, noting that when dealing with big data, it’s important to try to simulate the process for historical data and ensure the data aligns.

John Carpenter
Member
John Carpenter

How are you liking Alteryx for processing your data sets? Also, what do you do with the final data sets in your Alteryx workflows? Do you put them back into the database or do you push them on to visualization tools or ?? Thank you.

Minjie Yu
Member
Speaker
Minjie Yu

The Alteryx is a powerful and neat tool to process data. Final data sets in our alteryx workflows get stored in an oracle database.

Carol Bogacz
Member
Speaker
Carol Bogacz

What visualization tool do you use with Alteryx? Power Bi? Tableau?

Minjie Yu
Member
Speaker
Minjie Yu

We use PowerBI to display data

James Mack
Member
James Mack

To DTE: Using the additional data, in what scenarios would the interrupted customer count increase as opposed to the examples that you were showing where the numbers were lower?

Minjie Yu
Member
Speaker
Minjie Yu

On top of several jobs opening up for customers and duplicate jobs. During catastrophic storms jobs are being worked on by foreign crews and don’t get closed timely to show accurately the number of customers currently experiencing outages.

Related sessions

No related sessions currently available