Skip to main content

An API for examing the EIA Monthly Energy Review datasets.

Project description

pyleiades

Build Status

Python Library for EIA Data Examination & Exhibition

A tool for creating visuals from historical energy data (e.g. the EIA monthly energy review).

This tool is designed to provide insightful, aesthetic and more flexible visualizations of the Energy Information Administration (EIA) monthly energy review datasets. The datasets contain information about the sources of energy Americans have relied on for power since the middle of the 20th century. The datasets begin in 1949 with annual energy production, consumption, import, and export values, and extend up until the present. Monthly energy datapoints are reported starting in 1973.

The basic energy sources are reported in the following groups:

Fossil Fuels
  • Coal
  • Natural Gas
  • Petroleum
Renewables
  • Wind
  • Solar
  • Hydroelectric
  • Geothermal
  • Biomass
Nuclear
  • Fission

The data is published monthly on the EIA's website, and as of March 31st, 2019 records were provided up through December 2018. This package also includes data up to date through the end of 2018, though more recent data can be downloaded using an included script.

All reported values are in units of quadrillion british thermal units (1.0E15 BTU). Be aware that the datasets may provide more precision than is published in the PDF reports.

Installation

pyleiades is hosted through the Python Package Index (PyPI) and can be easily installed using pip. From the command line, run

$ pip install pyleiades

The module requires a recent version of Python 3 (3.6 or greater), pandas, and matplotlib, among others. If you run into trouble running the package, try using the Anaconda environment provided in this repo. Install the environment using the command

$ conda env create -f environment.yml

and activate the environment by issuing the command

$ conda activate pyleiades

Updating

An archive of EIA Monthly Energy Review datasets is kept in the pyleiades data repository. This may not include the most up to date information, and so the package comes with a script to update the available data. Once the package is installed, run

$ update_eia_data.py

from the command line to download the most recent data from the EIA website.

Using the API

The API is built around two main object types—the Energy and Visual classes.

The Energy object

To access the EIA data directly for a certain energy type, use the Energy class. For example, the energy consumption data for all renewable energy sources can be accessed with:

>>> from pyleiades import Energy
>>> renewables = Energy('renewable')

The resulting renewables object stores the complete consumption history within the energy_data dataframe attribute.

>>> renewables.energy_data
     date_code      value
6220    194913   2.973984
6221    195013   2.977718
6222    195113   2.958464
6223    195213   2.940181
...

The date_code column gives the reporting date (in the format YYYYMM, where the month code 13 indicates a yearly total) and the value column gives the consumption amounts (in QBTU) for each date. In the example above, the first four entries of the energy_data dataframe are the renewable energy yearly consumption totals for 1949 through 1952.

Energy consumption values are the default, however the Energy objects can also be used to access production, import and export statistics. The type of statistic can be selected using the stat_type keyword argument.

>>> renewables = Energy('renewable', stat_type='production')
>>> renewables.energy_data
     date_code     value
6220    194913  1.549262
6221    195013  1.562307
6222    195113  1.534669
6223    195213  1.474369

Perhaps more interesting than the complete history, however, are more sophisticated features of the data, like interval specific totals and extremes.

Using the totals method of an Energy object allows the data to be totaled at a specified interval—either monthly, yearly, or cumulatively.

>>> renewables.totals('monthly')
           value
  date
197301  0.403981
197302  0.360900
197303  0.400161
197304  0.380470

Notice that here the monthly data only goes back as far as 1973 (though the energy_data attribute showed yearly data for renewable energy dating back to 1949). By default, the totals method selects the entire range of available data. This behavior can be overriden by providing start and end dates for some interval as keyword arguments. To only get monthly renewable energy data from 2000 to 2010, this would be:

>>> renewables.totals(freq='monthly', start_date='200001', end_date='200912')
           value
  date
200001  0.505523
200002  0.498993
200003  0.558474
200004  0.567147

To get extremes over a dataset interval, use the maxima or minima methods.

The Visual object

A Visual allows the package to create plots of several energy types. The initialization parameters for a Visual are similar to those for an Energy object. A Visual can accept a single energy type or a list of energy types, optionally followed by a type of statistic (consumption by default).

visual = Visual(['coal', 'nuclear', 'renewable'])

This visual object's methods can then be used to generate any of a variety of visuals. The syntax is again similar to that of the Energy object, however it includes a subject argument corresponding to a method of the Energy object. Here's an example that generates a line graph of energy totals:

visual.linegraph(subject='totals', freq='monthly', start_date='1970')

visual comparing coal, nuclear, and renewable energy consumption since 1970

Run the very simple installed script pyleiades-demo.py to see the package in action.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

pyleiades-0.0.0.dev10.tar.gz (2.5 MB view hashes)

Uploaded Source

Built Distribution

pyleiades-0.0.0.dev10-py3-none-any.whl (2.5 MB view hashes)

Uploaded Python 3

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page