Mathematical simulation experiment based on optimisation of heat treatment process of aluminium alloy materials
Publié en ligne: 15 déc. 2021
Pages: 609 - 616
Reçu: 16 juin 2021
Accepté: 24 sept. 2021
© 2021 Jing Su et al., published by Sciendo
This work is licensed under the Creative Commons Attribution 4.0 International License.
Fig. 1
The structure of the neural network optimisation model.Fig. 2
Training performance curve of the model.Fig. 3
Predictive verification results of the model.Fig. 4
Wear surface morphology of specimens heat-treated with different processes.Fig. 5
SEM photos of the impact fractures of heat-treated samples of different processes.Process parameters of 2618 aluminium alloy for automobiles
Process parameters |
Production line tradition |
Neural network model optimisation |
Annealing temperature/°C |
360 |
330 |
Annealing time/h |
7 |
5.5 |
Solution temperature/°C |
530 |
510 |
Solution time/h |
4.5 |
6 |
Aging temperature/°C |
180 |
160 |
Aging time/h |
17 |
14 |
Artificial neural network prediction points and verification points
Sample |
Heat treatment process |
HRB hardness |
θ1/°C |
t1/h |
θ2/°C |
t2/h |
Actual value |
Predictive value |
Relative value/% |
1 |
460 |
0.5 |
110 |
20 |
41.2 |
40.9 |
0.71 |
2 |
460 |
1 |
120 |
48 |
55.7 |
55.6 |
0.18 |
3 |
460 |
1 |
130 |
75 |
61.2 |
61 |
0.32 |
4 |
460 |
1.167 |
110 |
100 |
60.3 |
60.5 |
−0.33 |
5 |
460 |
1.167 |
120 |
30 |
45.2 |
45.3 |
−0.22 |
6 |
460 |
1.167 |
130 |
50 |
51.3 |
51.3 |
0 |
7 |
470 |
0.5 |
120 |
100 |
62.5 |
62.4 |
0.16 |
8 |
470 |
0.5 |
130 |
125 |
54.3 |
53.9 |
0.73 |
9 |
470 |
1 |
110 |
20 |
38.7 |
37.8 |
2.3 |
10 |
470 |
1 |
120 |
48 |
55.7 |
55.9 |
−1.4 |
11 |
470 |
1 |
130 |
75 |
49.5 |
48.7 |
1.6 |
12 |
470 |
1.167 |
110 |
100 |
59.8 |
60.2 |
−0.67 |
13 |
470 |
1.167 |
120 |
30 |
41.7 |
42.1 |
−0.95 |
14 |
470 |
1.167 |
130 |
75 |
56.8 |
57 |
−0.35 |
15 |
480 |
0.5 |
110 |
45 |
53.9 |
54.2 |
−0.55 |
16 |
480 |
1 |
120 |
40 |
50.8 |
50.6 |
0.39 |
17 |
480 |
1 |
120 |
20 |
53.2 |
52.9 |
0.56 |
18 |
480 |
1.167 |
130 |
30 |
55.6 |
56.3 |
−1.23 |
191) |
460 |
1 |
120 |
55 |
57.5 |
57.9 |
−0.69 |
201) |
470 |
1 |
110 |
50 |
56.8 |
56.9 |
−1.7 |
211) |
470 |
1.617 |
110 |
55 |
58.3 |
58.3 |
0 |
221) |
470 |
1.617 |
120 |
50 |
62.5 |
61.9 |
0.96 |
231) |
480 |
1 |
120 |
45 |
60.1 |
59.8 |
0.49 |