Page 61 - Biaxial Multiaxial Fatigue and Fracture

46 E LABESSE-JIED ET AL

The approach adopted for the local calculation of fatigue damage under multiaxial random
loading is shown on the flowchart of Fig. 4. The calculation is performed on each weak point
of the surface of the structure. The inputs are the nodal plane stresses coming from a dynamic
calculation (stationary random or transient) or from a quasi-static calculation.

/

t ace5 1
E

Fatigue Criterion ah, (n, T", an)

- I'
D(n, of*, Basquin Law)

Damage = max D(n)
time

Fig. 4. Calculation of the fatigue damage by the local approach

The operational service stresses obtained through the latter analysis have to be corrected for
the effects of stress concentration resulting from geometric singularity overlooked during the
calculation. This may concern a fillet, a hole, or even the micro-geometry of the weld toe or the
weld root. The correction is performed with a two-dimensional stress concentration factor
matrix [K], the components Ki, of which are identified from a local calculation of the structural
detail. The local stress states are thus obtained from the nominal stress states as:

It is well recognised now that residual manufacturing stresses, combined with service
stresses, play an important role from the fatigue point of view. These stresses are taken into
account in the proposed approach in the form of initial stresses to which the service stresses are
added.