Skip to content

RAVENv2.2
Compare
Choose a tag to compare
@wangcj05 wangcj05 released this 16 Nov 03:13
· 451 commits to devel since this release
RAVENv2.2
4070e52
This commit was created on GitHub.com and signed with GitHub’s verified signature. The key has expired.
GPG key ID: 4AEE18F83AFDEB23
Expired
Learn about vigilant mode.

Official Release of the RAVEN code.
Version: 2.2

Importance Features:

  1. Parallel Improvement: Support parallel executions on Windows using Ray (#1919). Various changes to improve the parallel and debugging the parallel (#1825). Fix various cluster issues (#1807)
  2. RAVEN Python Interactive Communication: Allow running RAVEN run RAVEN workflows in python scripts or Jupyter Notebooks (#1816). Enable re-running RAVEN workflows in Python or Jupyter notebook environments (#1843)
  3. Validation PostProcessors: Adding Physics-guided Coverage Mapping to the validation algorithms (#1726). Adding dynamic system scaling validation algorithms (#1619, #1830).
  4. Code Interface: Added Serpent interface (#741). Restructured the load for the code interfaces, and
    added CodePluginBase class to enable the code interface as plugin (#2000).
  5. Reduce Order Model Update: Implementation of the DMDc method for state identification (#1693). Added a deep neural network regression rom (#1707). Developed randomized window decomposition algorithm for TSA module (#1640). Added a new exporting capability of ROMs with a derivative support into a Pyomo concrete model, activatable via Outstream system. This derivative support is based on numdiff library and is used both in the Pyomo model and, in the future, in the FMI/FMU exporter for Model exchange (#1482). Added ensemble rom to improve the generalizability/robustness over a single ROM (#1720). Significant improvements and simplifications on ROM APIs (#1607).
    Old ROM XML input:

image

New ROM XML input

image

Experimental Features:

  1. New installation process by “pip” (see #1784 and #1773):
    This RAVEN release now can be installed by pip. This is currently experimental, but we would like to hear back if it fails. The current pip packages are: raven-framework teal-ravenframework and heron-ravenframework (If you are developing a RAVEN plugin, and wish to create your own pip package, let us know and we can help you.) If you are running Python 3.7 or 3.8, these can be installed with pip. Example:
    pip3 install raven-framework
    These should work on Linux, Mac OS and Windows. After installation, the raven_framework command can be used to run raven (and for HERON, the heron command). Please see package pages for more details:
    https://www.pypi.org/project/raven-framework/
    https://www.pypi.org/project/teal-ravenframework/
    https://www.pypi.org/project/heron-ravenframework/

  2. Exporter for FMI/FMU (#1481)
    Addition of the contrib library "pythonFMU" (see https://github.com/NTNU-IHB/PythonFMU) that has been modified for RAVEN needs (e.g. Darwin support). Addition of RAVEN generic FMU exporter for ExternalModels and ROMs. Addition of the capability to serialize RAVEN ExternalModel.

Other Features:

  1. General RAVEN maintenance updates are tracked in issue #1806, some specific updates including: User manual update (#1817, #1818, #2008), workshop slides update (#1895), and some general improvements on RAVEN source code (#1930, #1877, #1975, python #1933).
  2. Updated Plugin User Manual to guide users on how to build their own plugins (#2008)
  3. Added a way to for Scikitlearn models to output model data and uses it for LinearRegression and MLPRegressor (#1988)
  4. Enable DMDc ROM to be loaded/unpickled by externalROMloader script (#1935)
  5. Enable externalROMloader script to modify/update RAVEN pickledROM (#1857)
  6. Workshop updates (#1909)
  7. Enable CustomSampler to accept DataSets (#1859)
  8. Libraries update (#1831, #1919, #1599, #1933, #1971, #1973, #1984)
  9. Enable VariableGroups to be used by RavenRunRaven workflow (#1823)
  10. Fix truncated lognorm distribution (#1815)
  11. Percentile calculation speed up, and standard error update in Basic Statistics Post Processor (#1780)
  12. Addition of spearman coefficients in Basic Statistics Post Processor (#1542)
  13. Added custom plot for optimization results (#1725)
  14. Addition of the capability to store long arrays of strings in HDF5. This adds also a version tag into the database (for future conversion if any) (#1703)
  15. Allow the GA to handle the model failure or crash (#1677)
  16. Allows a DataSet to be loaded from file and used as a training input to a ROM (#1658)
  17. Addition of the XSD 1.1 validator to allow for Conditional Type Assignment (#1501)
  18. Split of the Steps.py module in their individual steps for better maintainability (#1481)
  19. Enhancement in EconomicRatio PostProcessor (#1763)

Submodule Updates:
The updates for the submodules are tracked by issue #1114. In this release, there are significant updates in following Plugins. We recommend the users to check the following links for more details.
TEAL: https://github.com/idaholab/TEAL
HERON: https://github.com/idaholab/HERON
SR2ML: https://github.com/idaholab/SR2ML
LOGOS: https://github.com/idaholab/LOGOS
FARM: https://github.com/Argonne-National-Laboratory/FARM

Internal Developer:
We would like to thank all RAVEN internal developers for their significant contributions, including but not limit to: @mandd @PaulTalbot-INL @joshua-cogliati-inl @Jimmy-INL @dylanjm @dgarrett622 @JunyungKim @worseliz @yoshiurr-INL

External Contribution:
We would like to thank all RAVEN external contributors for their significant contributions, including but not limit to @aalfonsi @wanghy-anl @greenwoodms06, @FlanFlanagan, @AnthoneyGriffith @huang714 @mgarrouste @maldil @j-bryan @yenili

Addressed Defects:
#1990
#1962
#1934
#1932
#1887
#1881
#1874
#1855
#1851
#1822
#1814
#1809
#1793
#1785
#1771
#1767
#1758
#1749
#1746
#1740
#1724
#1771
#1684
#1676
#1053

Contributors

FlanFlanagan, PaulTalbot-INL, and 17 other contributors
Assets 2
Loading