Table 1 Our newly synthesized alloys during the iterative process.

From: Accelerated discovery of high-strength aluminum alloys by machine learning

Alloy no.

Nominal elemental-content (wt.%)

 

UTS (MPa)

 

Zn

Mg

Cu

Y

Ce

Ti

EI value

Predicted

Experimental

1–1

7.67

1.65

2.38

0.40

0

0.04

19.037

454 ± 68

419 ± 6

1–2

7.85

1.68

2.38

0.38

0.06

0.05

15.767

446 ± 86

432 ± 8

1–3

7.57

1.68

2.38

0.38

0.01

0.07

10.008

411 ± 156

474 ± 12

2–1

6.89

1.70

2.38

0.35

0

0.08

11.261

477 ± 73

443 ± 2

2–2

7.36

1.69

2.38

0.36

0

0.08

9.854

482 ± 39

468 ± 2

2–3

6.20

1.71

2.37

0.35

0.01

0.08

8.741

456 ± 120

477 ± 8

3–1

6.08

1.72

2.35

0.38

0.01

0.08

28.260

502 ± 93

439 ± 8

3–2

7.13

1.67

2.35

0.41

0

0.08

24.060

492 ± 116

446 ± 2

3–3

7.43

1.78

2.36

0.41

0

0.09

20.584

486 ± 118

517 ± 7

4–1

8.20

1.82

2.37

0.40

0

0.10

26.822

543 ± 66

528 ± 1

4–2

7.84

1.86

2.37

0.40

0

0.10

22.741

538 ± 65

549 ± 13

4–3

7.47

1.93

2.37

0.39

0

0.09

10.529

509 ± 107

562 ± 18

  1. Notes: “1–2” represents the second alloy in the first iteration. The predictive confidence-interval is the prediction plus or minus three standard errors given by the machine-learning model.
  2. EI expected improvement function.
Back to article page