Prev: "sensitivity" command
Next: decimate() vs downsample()
From: Chamil on 4 Mar 2010 06:51
Hi,


I used sigmoidnet function to obtain a nlarx model. But, I'm not sure how to obtain the final modl equation. The what i get in matlab command window is shown below. If anybody knows to get the final equation of the nlarx model please let me know.

IDNLARX model with 1 output and 6 inputs
Input names: u1, u2, u3, u4, u5, u6
Output name: y1
Standard regressors corresponding to the orders:
na = [0]
nb = [1 1 1 1 1 1]
nk = [0 0 0 0 0 0]
No custom regressor
Nonlinear regressors:
u1(t)
u2(t)
u3(t)
u4(t)
u5(t)
u6(t)
Nonlinearity estimator: sigmoidnet with 13 units
Loss function: 10.9293
Sampling interval: 1
Estimated model (NLARX)
SearchMethod: 'Auto'
Criterion: 'det'
Weighting: 1
MaxIter: 20
Tolerance: 1.0000e-005
LimitError: 0
Display: 'on'
MaxSize: 250000
IterWavenet: 'auto'
Advanced: [1x1 struct]

Name: ''
Ts: 1
TimeUnit: 's'
TimeVariable: 't'
InputName: {6x1 cell}
InputUnit: {6x1 cell}
OutputName: {'y1'}
OutputUnit: {''}
na: 0
nb: [1 1 1 1 1 1]
nk: [0 0 0 0 0 0]
CustomRegressors: {}
NonlinearRegressors: [1 2 3 4 5 6]
Nonlinearity: [1x1 sigmoidnet]
Focus: 'Prediction'
Algorithm: [1x1 struct]
EstimationInfo: [1x1 struct]
NoiseVariance: 11
Notes: {}
UserData: []

ans =

Status: 'Estimated model (NLARX)'
Method: 'NLARX using SearchMethod = Auto'
LossFcn: 10.9293
FPE: 11.1578
DataName: 'data'
DataLength: 6600
DataTs: 1
DataDomain: 'time'
DataInterSample: {'zoh'}
WhyStop: 'Maximum number of iterations reached.'
UpdateNorm: 55.9410
LastImprovement: 3.3189
Iterations: 20
InitialState: []
Warning: ''
InitRandnState: []
EstimationTime: 9.6253


ans =

RegressorMean: [46.5188 0 141 171.5000 186.5000 202]
NonLinearSubspace: [6x3 double]
LinearSubspace: [6x3 double]
LinearCoef: [3x1 double]
Dilation: [3x13 double]
Translation: [8.5415 -11.8708 -4.0892 -1.4355 0.9680 -1.2880 0.5828 -0.2302 -0.5916 6.7346 4.1635 -0.0640 -6.9024]
OutputCoef: [13x1 double]
OutputOffset: 183.0830


ans =

1.0015
17.7674
6.6782


ans =

Columns 1 through 12

-1.9587 1.9238 1.8344 1.1216 2.4819 1.9029 0.7461 -3.0806 -0.2872 2.5524 3.4486 -1.3884
-0.0245 -0.1964 -1.8547 1.2147 0.0420 -2.9483 -0.2924 1.7971 0.0755 -0.5179 -0.9031 3.2111
5.8305 8.3523 0.6112 3.2744 -2.7473 0.0366 6.1359 0.1158 3.4826 -3.0737 0.2752 -0.9748

Column 13

-2.6254
0.5402
0.0715


ans =

26.6365
-32.0844
12.4778
-0.8223
-7.2964
5.3226
21.5580
-1.7478
-44.6054
6.7544
26.8001
-2.9696
11.9433
 | 
Pages: 1
Prev: "sensitivity" command
Next: decimate() vs downsample()