Overview of TFEL, MFront and MTest

Tickets fixed

Ticket 38: Bad normalisation of the tensor when computing eigenvalues

Tensors are normalised when computing the eigenvalues. There was a mistake in the normalisation which can lead to severe numerical instabilities in some (rare) cases.

For more details, see: https://sourceforge.net/p/tfel/tickets/38/

Ticket #35: regression when using @CastemFiniteStrainStrategy or @CastemFiniteStrainStrategies in the castem interface

When defining finite strain strategies in the castem interface, a regression was introduced in version \(3.0.0\), if only one finite strain strategy is defined: in this specific case, one defines two behaviours (in fact the second is barely an alias for the first one):

• A first one having the base name of the behaviour (e.g. umatnorton).
• A second name having the base name of the behaviour followed by the finite strain strategy (e.g. umatnorton_ss).

The regression was that the ModellingHypotheses symbols were only defined for the second behaviour.

For more details, see: https://sourceforge.net/p/tfel/tickets/35/