Detecting a Grid Anomaly from Regression Residuals

​

Grading

​

Nickname

Change your nickname in the Discord server to your full name, first and last. This is how we identify you.

​

Development Environment Setup

Please refer to the Development Environment Guide for setup instructions. After you clone or import the class github repo you need to provide proof that your development environment is correctly setup:

1. Complete this notebook to verify that Torch is running properly inside the docker container.. Note that you need to consult the repos Pull Requests if you are on a Apple M silicon platform.
2. Create an account on Hugging Face and provide a screenshot of your profile.

Screenshots must be stored as .png files, named dev-environment-docker-xyz.png, and located under the assignment submission folder in your cloned class repo.

​

Grid Anomaly Detection

In February 2021 Winter Storm Uri broke the Texas grid. Temperatures collapsed, generation failed, and ERCOT began shedding load in the early hours of 15 February. Around 4.5 million homes and businesses lost power and hundreds of people died. Background is in the 2021 Texas power crisis article. You will build a model of ordinary grid behaviour, freeze it, and decide whether its residuals separate that week from a normal week. The model is the ridge regression from the linear regression and SGD pages. Only the design matrix changes. You will add temperature, calendar indicators, and interactions between them, and fit it by stochastic gradient descent.

​

What you will need

The notebook reads real ERCOT data from the gridstatus.io API. You need a GRIDSTATUS_API_KEY in your environment. Ask on Discord if you do not have one.

​

Assignment Notebook

​

Submitting

Commit the notebook to your own private clone of the class repo, under assignments/assignment-1. Save all cell outputs before you commit. A notebook without outputs is graded as if the cells never ran. Then submit the complete GitHub URL of that directory, not of the repository root, to Canvas or Brightspace. The assignment submission guide covers the rest: setting up your clone, adding the TAs as collaborators, and the README each assignment folder needs.

---