Information Technology Reference
In-Depth Information
Table 1.
Weights and thresholds for each of the 3 BP neural networks
category
Fg
CO
H
2
category
Fg
CO
H
2
category
Fg
CO
H
2
w11
1.8635
4.9084
5.9429
w44
4.8276
1.7488
4.5775
w77
0.3240
0.4327
23.1049
w12
-0.1369
-9.6307
-13.5372
w45
-0.0668
9.9990
0.3142
w78
0.7294
-25.6689
20.2206
w13
-2.4829
-4.2817
-17.2574
w46
0.5664
-4.7937
-5.4420
w81
2.0962
24.6860
-10.0065
w14
-0.8504
-3.4585
2.0645
w47
0.0793
-3.0714
-26.4697
w82
-0.0461
-26.6032
37.3559
w15
1.6463
-7.3019
-1.5329
w48
0.7136
-0.7556
4.3870
w83
6.3219
17.1456
-8.0812
w16
1.5550
7.9417
1.4441
w51
-3.836
-25.859
22.3753
w84
4.6829
-14.7542
42.7188
w17
0.1801
4.3339
15.3921
w52
-7.4787
22.9181
8.9603
w85
-1.1910
2.6048
-18.3320
w18
2.6966
12.3964
-8.2909
w53
0.4707
10.7267
35.6822
w86
-1.8889
-17.7350
9.7700
w21
5.6230
-4.8583
23.8403
w54
0.3503
-0.9656
22.1330
w87
-1.2744
-4.3293
-0.5461
w22
-4.5553
-22.6512
-55.3841
w55
0.3606
-10.759
31.0260
w88
0.1974
11.3994
19.5003
w23
-1.7120
-1.1308
34.5353
w56
0.4855
22.9518
-17.9031
v1
-1.7597
-1.2467
9.2067
w24
-2.8146
-10.6592
-0.7447
w57
-3.8633
-12.909
0.7290
v2
2.6515
-1.2787
-1.2212
w25
-1.2761
42.0020
-22.4337
w58
1.1296
-8.6603
-19.1587
v3
-6.3735
12.3021
-15.3544
w26
-1.9847
-3.4065
-0.5483
w61
0.0474
0.3232
21.0601
v4
-4.4436
-2.8620
-1.2450
w27
-4.7681
1.4985
-18.7561
w62
1.7063
8.3583
-14.0994
v5
1.6059
-0.9401
-0.9907
w28
-0.2870
-26.2014
-14.2768
w63
0.9223
-3.8937
20.2876
v6
1.6943
1.0200
-1.1413
w31
-10.145
-17.5820
0.0019
w64
-1.0234
-7.8132
12.0555
v7
-3.0079
2.8675
1.0988
w32
1.7225
-22.9627
36.8186
w65
-0.7430
-10.303
23.5590
v8
1.5435
1.8162
1.3969
w33
4.1829
1.9811
9.9481
w66
-0.6885
-7.1607
9.1791
θ1
-1.8709
11.6277
4.1325
w34
-3.3798
0.8228
28.6763
w67
-0.2844
-3.9438
7.9657
θ2
1.1129
-17.6772
-21.5644
w35
1.4201
-11.6556
-25.4485
w68
-0.0763
19.6221
0.0733
θ3
-2.2206
-3.8848
-45.9344
w36
1.2433
-21.7318
6.4054
w71
-4.4433
-2.9771
13.8629
θ4
1.1723
-2.5505
9.0688
w37
4.5140
-15.2801
32.9958
w72
0.3643
16.5940
49.8957
θ5
1.8759
-19.0972
5.7394
w38
0.6301
28.0138
-3.5233
w73
7.0671
27.9733
29.1182
θ6
0.7607
14.6930
2.4959
w41
-3.1632
-19.4046
23.8918
w74
-1.3045
34.7179
49.9074
θ7
1.0704
10.2298
28.9257
w42
6.0327
-8.0793
-1.1947
w75
-0.7360
-13.746
31.4887
θ8
0.4504
17.4301
-15.3954
w43
2.9478
4.6934
33.4183
w76
-1.0396
26.2516
11.1552
r
1.9946
1.6019
-6.4239
Table 2.
Optimal results between 3LM-CDE algorithm and MO-3LM-CDE algorithm
Optimize value of the 3 control parameters
Optimize results of the thr
ee outputs
Effective gas yield
(m
3
/t)
Algorithm
x
Fo
(m
3
/h)
x
Foc
(m
3
/h)
x
Fw
(m
3
/h)
Fg(m
3
/h)
CO(V%)
H
2
(V%)
30297.852
4447.998
392.323
222782.338
40.555
39.673
3683.318
3LM-CDE
MO-3LM-
CDE
34860.535
3622.009
374.148
212362.374
43.654
41.031
3706.047
Application-main objective
method
Application-linear weighted
method
MO-3LM-CDE
MO-DE
MO-3LM-CDE
MO-DE
1
0.95
0.9
0.9
0.85
0.8
0.8
0.7
0.75
0.7
0.6
0.65
1
0.5
1
0.9
0.8
0.8
0.8
0.6
0.6
0.8
0.4
0.6
0.4
0.7
0.2
0.2
Fig. 27.
Main objective method of application
Fig. 28.
Linear weighted method of
application
0
0.4
0
CO
CO
Fg
Fg
Application-max-min met
hod
Application-ideal point met
hod
MO-3LM-CDE
MO-DE
MO-3LM-CDE
MO-DE
1
1
0.9
0.9
0.8
0.8
0.7
0.7
0.6
0.6
1
0.5
1
0.5
0.8
0.8
0.8
0.8
0.6
0.6
0.4
0.4
0.6
0.6
0.2
0.2
0.4
0.4
CO
0
CO
0
Fg
Fg
Fig. 29.
Max-min method of application
Fig. 30.
Ideal point method of application
Search WWH ::
Custom Search