PSI - Issue 68

Tea Marohnić et al. / Procedia Structural Integrity 68 (2025) 84 – 90 T. Marohnić and R. Basan / Structural Integrity Procedia 00 (2025) 000–000

88

5

Fig. 2. Experimental vs. estimated numbers of load reversals, log(2 N f ) of low-alloy steels determined with the ANNs for low strength (LS) and high strength (HS) materials in low-cycle (LCF) and high-cycle fatigue (HCF) lives ranges.

Table 1. Values of evaluation criteria E f ( s =3), ( E a ) total , ( E a ) Dset and Ē determined for ANNs method.

LOW-ALLOY STEELS

E f ( s = 3)

( E a ) total

( E a ) Dset

Ē

All strengths, All fatigue lives (ALLS – ALLF)

0.881

0.730

0.743

0.785

Low strength, Low-cycle fatigue (LS – LCF)

0.800

0.893

0.861

0.851

Low strength, High-cycle fatigue (LS – HCF)

0.870

-0.089

0.536

0.439

High strength, Low-cycle fatigue (HS – LCF)

0.932

0.774

0.807

0.838

High strength, High-cycle fatigue (HS – HCF)

0.893

0.712

0.568

0.725

When values in Table 1 are thoroughly examined, for LS – HCF subgroup, ( E a ) total criterion has extremely low, negative value which significantly impacts the end result for Ē . Since ( E a ) total represents the goodness of fit between