What kind of electricity are you getting right now? The WattTime grid API collects real-time information on the current generation fuel mix from the major American electricity markets (MISO, PJM, CAISO, ERCOT, ISONE, SPP, and BPA) using the pyiso library and displays it in a convenient standardized format. The API also calculates a real-time carbon footprint of electricity use in a given place and time based on this fuel mix. See the code in action at http://api.watttime.org/!

All the Django apps are in watttime_grid_api/apps. These are

• gridentities: provides models for metadata and descriptions of geographic entities (BalancingAuthority and PowerPlant) and data types (FuelType); standalone
• griddata: provides models for time series data (DataPoint); depends on gridentities
• supply_demand: provides Generation, Load, and TieFlow models that associates generation and load observations to DataPoints, and management commands and celery tasks for pulling the data; depends on gridentities and griddata
• carbon: provides Carbon and FuelCarbonIntensity models that associate carbon intensity observations to DataPoints, and tasks for calculating the data; depends on gridentities, griddata, and supply_demand
• marginal: provides MOER, MOERAlgorithm, and StructuralModelSet models that associate marginal carbon intensity observations (MOER) to DataPoints, and tasks for calculating the data; depends on gridentities, griddata, and supply_demand
• api: implements a Django REST Framework API to all other apps; depends on gridentities, griddata, supply_demand, carbon, and marginal
• etl: implements a extract-transform-load job flow to supply data to all other apps; depends on gridentities, griddata, supply_demand, carbon, and marginal

Start a postgres server (eg http://postgresapp.com/) and create a database called mydbname. You'll also need libmemcached:

   brew install libmemcached

Also install the GeoDjango requirements. See the platform-specific instructions at https://docs.djangoproject.com/en/dev/ref/contrib/gis/install/, I used homebrew:

   brew install postgresql
   brew install postgis
   brew install libgeoip
   psql mydbname
   # CREATE EXTENSION postgis;

Clone this repo and install the requirements:

  cd watttime-grid-api
  mkvirtualenv watttime-grid-api
  pip install -r reqs/dev.txt

Set up the environment variables. You can put then in a .env file and use foreman (get it at https://github.com/ddollar/foreman or as part of the Heroku CLI) or put them in your bash profile or whatever. If you're not using foreman, remove the 'foreman run' from all following commands.

  echo DATABASE_URL=postgres://localhost/mydbname > .env

Create database tables for the models (create user when prompted):

  foreman run ./manage.py syncdb
  foreman run ./manage.py migrate

Test the site:

  foreman run ./manage.py test
  foreman run ./manage.py loaddata isos gentypes fuelcarbonintensities griddata
  foreman run ./manage.py shell
  >>> from apps.gridentities import load
  >>> load.run_balancing_authority()
  foreman run ./manage.py runserver

To run tasks with celery, run rabbitmq-server &, then celery -A watttime_grid_api worker -B -l info or foreman start (if you have newrelic-admin installed locally).