System Identification Technique and Neural Networks for Material Lifetime Assessment Application - Complex System Modelling and Control Through Intelligent Soft Computations - page 778

Fig. 16 Fatigue lives
predicted by the
RBFNN-NARX model for the
tested sets: R = 0.5 and − 1of
Material II-on-axis (from left
to right)

Fig. 17 Fatigue lives
predicted by the
RBFNN-NARX model for the
tested sets: R = 0.5 and − 1of
Material II-45 ° (from left to
right)

The accuracy of the RBFNN model prediction compared to the experimental
data was measured in mean squared error (MSE). The MSE prediction values for
the Material I, Material II-on-axis and Material II-45 were 0.258, 0.285 and 0.088,

respectively. It is important to note here that the produced mean squared error
(MSE) values of fatigue life prediction results of the RBFNN-NARX model
competed favorably, even better, with those of the MLP-NARX model. The
comparison was shown in Table 4 .

°