数据处理作业分析解析Word格式文档下载.docx
《数据处理作业分析解析Word格式文档下载.docx》由会员分享,可在线阅读,更多相关《数据处理作业分析解析Word格式文档下载.docx(15页珍藏版)》请在冰豆网上搜索。
b
c
aa
bb
cc
ab
ac
bc
120
100
20
7.3372
-120
-100
-43.3333
14400
10000
1877.78
12000
5200
4333.33
200
60
8.5979
-3.3333
11.11
400
300
110
8.6156
46.6667
2177.78
-12000
-5600
4666.67
1.1005
-40
1600
4000
-1866.67
-4666.67
9.8174
1733.33
10.0736
-4000
133.33
-333.33
2.0318
160
25600
-16000
-533.33
333.33
3.4031
7466.67
8.893
16000
-6933.33
-4333.33
田口设计
田口正交表设计
L9(3**3)
因子:
3
试验次数:
9
列L9(3**4)阵列
124
逐步回归:
r与a,b,c,aa,bb,cc,ab,ac,bc
入选用Alpha:
0.15删除用Alpha:
0.15
响应为9个自变量上的r,N=9
步骤1234
常量6.6526.6526.6526.652
b0.02850.02850.02850.0316
T值2.653.769.3314.11
P值0.0330.0090.0000.000
c-0.0480-0.0480-0.0480
T值-2.86-7.08-11.01
P值0.0290.0010.000
a-0.0120-0.0120
T值-5.64-8.78
P值0.0020.001
ac0.00012
T值2.85
P值0.046
S2.641.860.7490.481
R-Sq50.0278.8197.1299.05
R-Sq(调整)42.8871.7595.4098.10
在mimitab软件中,减去均值,设置响应为R,预测变量为a,b,c,aa,bb,cc,ab,ac,bc,进行逐步回归。
得到方程:
R=6.652+0.0316*b-0.0480*c-0.0120*a+0.00012*ac
(a=x1-240,b=x2-200,c=x3-63.333)
故正交方程
R=6.652+0.0316*(x2-200)-0.0480*(x3-63.333)-0.0120*(x1-240)+0.00012*(x1-240)*(x3-63.333)
1.2随机取10组数据
表2实验结果表
正交结果
正交设计方误
270
175
108
3.465552
3.518799
0.002835
280
225
5.097366
5.377999
0.078755
315
195
95
4.190496
4.358999
0.028393
115
1.304328
1.323998
0.000387
255
220
105
4.982996
5.178998
0.038417
350
185
85
4.016286
4.103999
0.007694
310
1.677378
1.633999
0.001882
290
190
98
4.117086
4.279999
0.026541
250
4.268067
4.347998
0.006389
260
130
1.870393
1.843998
0.000697
0.191988
正交设计均方误
0.019199
2.1在不同试验量条件下,设计两组不同的设计:
正交9组L9(3*3)
表3正交设计L9(3*3)实验方案及实验结果表
2.2正交16组L16(4*3)
表4正交设计L16(4*3)实验方案及实验结果表
150
80
10
6.22725
-150
-97.5
-47.5
22500
9506.3
2256.25
14625
7125
4631.25
40
7.37789
-27.5
-17.5
756.3
306.25
4125
2625
481.25
70
7.2948
22.5
12.5
506.3
156.25
-3375
-1875
281.25
7.60357
102.5
52.5
10506.3
2756.25
-15375
-7875
5381.25
1.49286
-50
2500
4875
-625
-1218.75
2.72535
1375
-2625
-1443.75
9.37603
-1125
2375
-1068.75
9.78991
-5125
875
-1793.75
0.20443
50
-4875
-5118.75
3.50455
-1375
625
-343.75
6.11529
1125
-875
-393.75
9.69676
5125
-2375
-4868.75
450
1.80306
-14625
1706.25
4.93339
-4125
-7125
1306.25
3.18781
3375
7875
1181.25
6.0415
15375
1875
1281.25
响应为9个自变量上的r,N=16
步骤123456
常量5.4615.4615.4615.4615.4615.650
b0.02960.02960.02960.03040.03100.0312
T值4.056.2816.9119.0525.0930.05
P值0.0010.0000.0000.0000.0000.000
c-0.0423-0.0423-0.0423-0.0443-0.0442
T值-4.54-12.23-13.79-18.18-21.69
P值0.0010.0000.0000.0000.000
a-0.01037-0.01037-0.01037-0.01037
T值-9.07-10.22-13.39-15.99
P值0.0000.0000.0000.000
ac0.000060.000110.00011
T值2.063.974.63
P值0.0640.0030.001
ab-0.00004-0.00004
T值-2.98-3.49
P值0.0140.007
bb-0.00004
T值-2.29
P值0.047
S2.131.380.5110.4540.3460.290
R-Sq53.9782.2197.7498.3799.1399.45
R-Sq(调整)50.6879.4797.1797.7798.7099.09
MallowsCp2372.1911.4109.378.640.826.3
步骤789
常量5.4805.3585.222
b0.031190.031190.03116
T值37.0742.8853.19
P值0.0000.0000.000
c-0.0450-0.0450-0.0450
T值-26.76-30.96-38.46
a-0.01037-0.01037-0.01065
T值-19.74-22.84-27.46
ac0.000110.000110.00010
T值5.646.537.92
ab-0.00004-0.00004-0.00004
T值-4.18-4.84-6.05
P值0.0030.0020.001
bb-0.00004-0.00004-0.00004
T值-2.84-3.28-4.08
P值0.0220.0130.006
cc0.000120.000120.00012
T值2.392.763.43
P值0.0440.0280.014
aa0.000010.00002
T值1.933.21
P值0.0960.018
bc-0.00005
T值-2.19
P值0.071
S0.2350.2030.164
R-Sq99.6899.7999.88
R-Sq(调整)99.4099.5599.71
MallowsCp16.512.810.0
R=5.480+0.03119*b-0.045*c-0.01037*a+0.00011*ac-0.00004ab-0.00004bb+0.00012cc
(a=x1-300,b=x2-177.5,c=x3-57.5)
R=5.480+0.03119*(x2-177.5)-0.045*(x3-57.5)-0.01037*(x1-300)+0.00011*(x1-300)*(x3-57.5)-0.00004*(x1-300)*(x2-177.5)-0.00004*(x2-177.5)*(x2-177.5)+0.00012*(x3-57.5)*(x3-57.5)
2.3随机选取10组数,比较两种设计方法的均方误
表5两种设计方法均方误比较
正交9组结果
正交设计9组方误
正交16组结果
210
30
10.71853
10.00798
0.504876
9.983925
0.539639
7.224056
7.537986
0.098552
7.477425
0.064196
170
5.455955
5.547981
0.008469
5.397675
0.003397
1.369033
1.453982
0.007216
1.130125
0.057077
6.562285
7.074985
0.262861
6.806
0.059397
4.103988
0.007692
4.182675
0.027685
1.633987
0.001883
1.659125
0.000333
4.279986
0.026536
4.302105
0.034232
380
15
6.886535
6.47999
0.165279
7.015175
0.016548
1.39942
0.741992
0.432211
1.395725
1.36511E-05
1.515574
0.802518
正交9组设计均方误
0.151557
正交设计16组均方误
0.080252
3对电脑仿真实验结果加入不同误差,分别建立恰当的数学模型。
3.1加入不同误差后对模型精度的影响
表6误差为±
0.05时MSE值
Reactionrate
R
MSE
±
0.05
MSE±
7.635979
0.089269
7.336684
2.67E-07
8.299979
0.088757
8.434034
0.026852
8.339979
0.075967
9.01037
0.155843
1.507982
0.166042
1.063197
0.001391
9.419982
0.157941
9.853403
0.001296
10.46798
0.155537
10.41235
0.114749
1.66799
0.132357
2.083598
0.002683
3.38799
0.000228
3.363194
0.001593
9.13999
0.061004
9.204032
0.096741
0.037148
表7误差为±
0.04时MSE值
0.04
7.627588
0.084324983
8.273336
0.105341474
8.719737
0.010844463
1.079535
0.000439544
9.612391
0.04202852
10.0