pyLife’s change log file

In this file noteworthy changes of new releases of pyLife are documented since 2.0.0.

2.3.0

New features

  • Implement parts of FKM linear guideline

Improvements

  • Add missing function for five segment meanstress transformation for load histograms (#216)

Bug fixes

  • Fix rebinning of partially occupied histograms (#234)

2.2.1

Bug fixes

  • Fix crash when HCM algorithm was given a named multiindex load series (#210)

2.2.0

Breaking changes

  • Drop support for python-3.8 ast it is end of life

  • The Wöhler analysis slightly changed (#138)

New features

Support for some FKM-non-linear calculations

  • HCM rainflow algorithm

  • P_RAM and P_RAJ damage calculation

  • Notch approximation law (Neuber and Seeger-Beste)

Minor improvements

  • Connectivity information is now transferred from odbserver to odbclient as ints.

Bugfixes

  • Fix unnecessary and harmful copy of pandas object in Wöhler classes (#146)

  • Don’t assume failure_probability=0.5 in Wöhler calculations (#192)

pylife-2.1.4

Bugfixes

Fix gradient calculation for non continous nodes in mesh.

pylife-2.1.3

Improvements

  • Massive performance improvement when reading element connectivity by odbclient. Note that also odbserver needs to be updated in its environment. (#121)

  • Fix some warnings on import

pylife-2.1.2

New features

  • New method LoadCollective.histogram() (#107)

Improvements

  • Sanitize checks for Wöhler analysis (#108)

  • Error messages when odbclient gets unsupported element types (#64)

  • Improved documentation

Bugfixes

  • MeshSignal now allows for additional indeces (#111)

pylife-2.1.1

Breaking changes

  • Change fracture load levels used for slope k_1 and scatter T_N estimation. Now only fractures in the finite_zone are used for estimation and not all fractures (in finite_zone and infinite_zone). Change based on findings in DIN50100:2022-12. (#80, #101)

    (see this discussion)

  • Rename FatigueData.fatigue_limit to finite_infinite_transition

  • The Bayesian Wöhler analyzer has been shutdown (#74) (see this discussion)

New features

  • Add option to manually set fatigue_limit (now renamed) for Woehler curve estimation. (#73)

Improvements / bug fixes

  • Rainflow counters work with pd.Series of all index type (#69)

  • Improved documentation

  • Fixed confusing load matrix after mean stress transformation (#105)

  • Massive performance improvements of three point and four point rainflow counters.

  • The Wöhler analyzer now ignores irrelevant pure runout levels (#100)

  • Support numpy>=2.0.0

pylife-2.1.0

New features

  • History output for odbclient

  • Introduce WoehlerCurve.miner_original()

Breaking changes

  • Non-destructive miner modifiers of WoehlerCurve

    The methods WoehlerCurve.miner_elementary() and WoehlerCurve.miner_haibach() now return modified copies of the original WoehlerCurve object, rather than modifying the original.

pylife-2.0.4

Minor bugfix release

  • switch to vtk > 9

  • fix two index bugs in gradient.py and hotspot.py

  • support for python 3.11

  • pandas-2.0 compliant

pylife-2.0.3

A minor release, mostly dependency related updates and documentation improvements.

pylife-2.0.2

Minor bugfix release

  • Fix bug for detection of Abaqus binaries in odbserver

pylife-2.0.1

A minor release, mostly dependency related updates and documentation improvements.

New features

  • The LoadCollective accessor class honors the cycles column in the give DataFrame. This is useful for manually created load collectives.

  • Support for python 3.10

Bug fixes

  • Fix off-by-one error in meanstress transformation rebinning

  • Correct the index of the detected turns when NaNs have been dropped in rainflow counting

Other changes

  • Documentation improved

  • Switch from pymc3 to pymc version 4 for Bayesian Wöhler analysis. The extension is not installable via pip as the current bambi release pulls a numpy version that is incompatible with our demandas. If you need it, please install bambi manually from their current git repo and then pymc via pip.

  • Lift numpy version restrictions