febiosoftware/FEBio: FEBio 4.13

Résumé fourni par la source

New Features: Introduced the concept of a "study", which is similar to a task but uses standardized input format, called the "options file" and produces an FEBio report. Added timer for serialization. Several previously un-timed functions are now included in the Update timer. Implemented "max shear strain" log variable. Strain based element erosion criterion Added error to optimization report. Optimization task can now print a summary report of the results. Implemented "fluid coronary outflow" boundary condition Implemented "scaled Jacobian" mesh adaptor criterion. Added "tet-quality" mesh adaptor criterion and plot variable. Added range to "density" parameter. Added Tangent_ShearViscosity_StrainRate member to all visous fluid materials. Implemented some support for nlopt's optimization algorithms. Improvements: Performance improvements for contact algorithms that make use of octrees (e.g. sliding-elastic). The plot file now stores the filename that was used to create it. Modified the "parameter_sweep" task to a study. Options files for studies now start with febio_study. Had to override writeNodalProjectedElementValues to avoid copying of FEParamDouble. Moved construction of local element list in FEFluidFSITraction from Init() to Activate() so that it is reconstructed after remeshing. Ensured that surface element pointers to elements are correctly re-initialized during hex-refine2d remeshing. Node numbers are now validated when building edges from the input file. Updated Starling parameters' meta. [skip ci] This change is needed to make the reversion of a prior commit on thermofluids remain compatible with the rest of the code Minor changes to the optimization input parameter classes. Modified FEThermoViscousFluid to include up to six FEFunction1D scale factors for temperature-dependent properties in fluids. Modified FENewtonianThermoFluid to take advantage of this feature. Updated build files to include nlopt dependency Bug Fixes: A bug was fixed in the contact potential distance check Fixed missing header include for strcpy (#92) Fixed problem that was preventing initial velocities from working in explicit solid solver. Fixed problem with "truss stretch" plot variable. Fixed bug in FESlideLineConstraint::BuildMatrixProfile. Fixed bug in evaluating node data record that was assuming one-based indexing of nodes. Fixed bug when displaying relative fluid velocity plot variable in a CFD analysis Fixed inconsistent use and output of FERigidConnected::m_nID. Fixed bug in FECoreBase::FindProperty(const ParamString& prop). Now catching error if node indices in NodeSet are out of range. Fixed issue with FEMaterialPointData::ExtractData that was preventing to find the correct component for some complex material definitions. Fixed compilation errors related to std::filesystem on old compilers Made truss material parameters density and E FEParamDouble. Fixed bug in FEPlotParameter so that it can be used by other domain classes besides FESolidDomain. Fixed problem in xmltool when reading scalar valuator classes. Fixed error in updating tabs in XMLWriter. Fixed buffer overflow on Linux caused by snprintf with bad size arg. Fixed issue with reading nodal loads from old 2.0 format. Added additional checks to avoid crashes exporting log data. Added checks to FEElasticMaterialPoint extraction to prevent crash when exporting log data. Fixed bug in FEHexRefined2D related to material ID not set for elements. Fixed issue with assigning validator to enum parameters. Fixed range of fiber-entropy-chain's n_term parameter. Set initial value for m_N parameter and fixed range for m_term in FEFiberEntropyChainUC. Fixed bug in GetMeshParameter when node is references via fromId function. Documentation: Fixed typo in CONTRIBUTING.md (#128) Added explanation to user manual about warning generated by FEBio when Holmes-Mow permeability encounters negative pore volume Fixed incorrect specification of fluid flux examples in biphasic analyses in user manual.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.