v3.1.2

Minor patches for Pandas, typos, and naming conventions.

v3.1.1 Release

This release patches an incorrect specification in the pyproject.toml causing the installation to complete without installing the actual package data.

v3.1

This release makes some minor upgrades to the documentation, and fixes upcoming breakages in the Pandas API. Please see the changelog for details.

v3.0

Please see the changelog and/or the documentation for complete details.

v3.0rc2

Release: 3.0rc2 (#223)

* change version number to 3.0rc1

* bumpy scikit-learn version

* bumpy scipy version

* udpate changelog for brief update commentary

* bump version:

* fix unnecessary version changes

* fix inequality typo

* Closes: 224

* bump python version to latest for action

Release 3.0 RC 1

This is the first release candidate for the completely refactored v3.0! Please see the documentation for this release candidate, and future release candidates at the "develop_v3" version of our Read the Docs Page: https://openoa.readthedocs.io/en/develop_v3/

Highlights:

• PlantData is now simplified to streamline loading, validation, and, most importantly, the user experience
• The addition of PlantMetaData to house the metadata specifications for each of the required data types enables users to bring their own column naming convention and let OpenOA map them to our internal conventions
• Remixed examples to better guide users getting started with, or wanting to learn more about, how to run analyses, experiment with the code base, or develop more functionality
• Underlying utils (formerly toolkits) and analysis (formerly methods) have performance and usability improvements, particularly for more standardized and publication-ready plotting routines that have more user control

Release 2.3

• Replaced hard-coded reanalysis dates in plant analysis with automatic valid date selection and added optional user-defined end date argument. Fixed bug in normalization to 30-day months.
• Toolkit added for downloading reanalysis data using the PlanetOS API
• Added hourly resolution to AEP calculation
• Added wind farm plotting function to pandas_plotting toolkit using the Bokeh library
• Split the QC methods into a more generic WindToolKitQualityControlDiagnosticSuite class and WTK-specific subclass: WindToolKitQualityControlDiagnosticSuite.
• Updated filter algorithms in AEP calculation, now with a proper outlier filter

Release 2.2

• IAV incorporation in AEP calculation
• Set power to 0 for windspeeds above and below cutoff in IEC power curve function.
• Split unit tests from regression tests and updated CI pipeline to run the full regression tests weekly.
• Flake8 with Black code style implemented with git hook to run on commit
• Updated long-term loss calculations to weight by monthly/daily long-term gross energy
• Added wind turbine asset data to example ENGIE project
• Reduce amount of time it takes to run regression tests by decreasing number of monte carlo iterations. Reduce tolerance of float comparisons in plant analysis regression test. Linear regression on daily data is removed from test.
• Bugfixes, such as fixing an improper python version specifier in setup.py and replacing some straggling references to the master branch with main.

Release 2.1

• Modify bootstrapping approach for period of record sampling. Data is now sampled with replacement, across 100% of the POR data.
• Cleaned up dependencies for JOSS review. Adding peer-reviewed JOSS paper.
• Add Binder button to Readme which makes running the example notebooks easier.
• Set maximum python version to 3.8, due to an install issue for dependency Shapely on Mac with Python 3.9.

Release 2.0.1

This update resolves a few issues including:

• create workaround for unknow zsh issue
• modifying ENGIE project pathname in exmaples so auto data extract works
• replace GeoPandas with pyproj and Shapely