PSI - Issue 28

Oleh Yasniy et al. / Procedia Structural Integrity 28 (2020) 1392–1398 Oleh Yasniy et al. / Structural Integrity Procedia 00 (2019) 000–000

1396

5

to 1.6 MPa/s. The AMg6 aluminum alloy stress–strain diagram was predicted by the methods of machine learning according to the experimental data obtained in Fedak (2003). While training, the dataset was divided into two unequal parts, that is, training and test sets. The dataset consisted of 70 elements, 70% of which were chosen randomly for the training set and 30% were left for estimating the quality of predictions. The input parameters were the stress parameter  p (  i ), while the strain jump   (  i ) was chosen as the output parameter. The obtained results generally coincide with the experimental ones. 4. Results and discussion By using the method of machine learning, there were plotted the dependences of the experimental jump-like strain (   (  i ) true ) on the predicted values of   (  i ) (   (  i ) prediction ) (Fig. 2). a b

0,017

0,017

Exper/Pred

Exper/Pred

0,015

0,015

0,013

0,013

0,005  (  i ) prediction, mm/mm 0,007 0,009 0,011

0,011

0,005  (  i ) prediction, mm/mm 0,007 0,009

0,003

0,003

0,003

0,008

0,013

0,003 0,008 0,013

 (  i ) true, mm/mm

 (  i ) true, mm/mm

c

d

0,017

0,017

Exper/Pred

Exper/Pred

0,015

0,015

0,013

0,013

0,005  (  i ) prediction, mm/mm 0,007 0,009 0,011

0,005  (  i ) prediction, mm/mm 0,007 0,009 0,011

0,003

0,003

0,003

0,008

0,013

0,003

0,008

0,013

 (  i ) true, mm/mm

 (  i ) true, mm/mm

Fig. 2. Predicted (   (  i ) prediction ) and experimental (   (  i ) true ) jump-like strain obtained by the methods of (a) neural networks, (b) boosted trees, (c) support-vector machines, and (d) k -nearest neighbors. The dependences between the value of jump-like increments of strain and the corresponding tensile stress   (  i ) –  p (  i ) are shown in Fig. 3.