System Identification Technique and Neural Networks for Material Lifetime Assessment Application - Complex System Modelling and Control Through Intelligent Soft Computations - page 775

Fig. 9 S-N curves obtained
by the NN-NARX model and
the experimental data for
R = 0.5 of Material I

Fig. 10 S-N curves obtained
by the NN-NARX model and
the experimental data for
R = − 0.5 of Material I

Fig. 11 S-N curves obtained
by the NN-NARX model and
the experimental data for
R = − 1 of Material I

obtained were comparable with those obtained by the previous researchers inves-
tigating also the same material using NN. See, for instance, in Freire Junior et al.
( 2005 , 2007 ) in particular for the NN fatigue life prediction results of R = 0.5 and

0.7, which were further casted in the corresponding S-N curves in the papers.
Furthermore, Figs. 13 and 14 depict respectively the S-N curves obtained by
the NN-NARX model and the experimental data for stress ratios R = 0.5 and

−

− 1of

Material II-on axis and of Material II-45 ° . It can be also seen that in general the