Page 1 of 27

1

Benchmark Problems

- Coupled Electric Drives

- F16 Ground Vibration

- Cascaded Tanks System

- Wiener-Hammerstein Process Noise System

- Bouc-Wen System

- Parallel Wiener Hammersterin

- Wiener Hammerstein

- SilverBox

Wiener Approaches

- Wiener Neural Identification [1]

- Iterative Wiener Identification [2]

[1] MM Arefi, A Montazeri, J Poshtan, MR Jahed Motlagh, “Wiener-neural identification

and predictive control of a more realistic plug-flow tubular reactor,” Chemical

Engineering Journal, vol 138, No 1-3, pp. 274-282, 2008.

[2] H Kazemi, MM Arefi, “A fast iterative recursive least squares algorithm for Wiener

model identification of highly nonlinear system,” ISA Transactions, vol. 67, pp. 382-388,

2017.

Page 2 of 27

An Investigation of the Wiener

Approach for Nonlinear System

Identification Benchmarks

Allahyar Montazeri

Mohammad Mehdi Arefi, Mehdi Kazemi

Lancaster University

Faculty of Science and Technology

E-mail: a.montazeri@lancaster.ac.uk

Page 3 of 27

Block Oreinted Approach

3

 Physically insightful

 Lower number of parameters

× The model is not linear in parameters

× Difficulty in finding a good initial condition

G q( )

f ( ) x

Wiener

u n( ) z n( ) y n( ) model structure

 Can approximate almost any nonlinear system

with high accuracy

 No specific assumption on the spectrum input

or static nonlinearity

Page 4 of 27

Wiener Neural Technique

NN x( ) u n( ) z n( ) y n( )

Step 1 Estimating the linear part using the given input/output data

Estimating the nonlinear part (Neural Network)

using the estimated linear model and the

measured output

Step 2

Step 3

Parametrising the estimated models in the

previous steps and optimising the overall

parameters of the Wiener model

Page 5 of 27

Wiener Neural Technique

 Assume no non-linear part, input-output data

 Calculate , ,

Step 1 NN x( ) u n( ) z n( ) y n( )

(,,,) ABCD

x(0)

n

Step 2

1 1

( ( )) ( , ) ( , , ) ( ) ( , ) ( , 1) ( )

l

j s

i j

NN z n s i s i j z n b s i b s n

  

 

   

               

2

1

1 1

1 1

1

( ) ( , ) ( , , )ˆ ( ) ( , ) ( , 1)

( ) (1, ) (1, , )ˆ ( ) (1, ) (1, 1)

min

 

 

 

 

 

 

N

k

i

l

j

l j

i

l

j

j

y

k l i l i j

z

k

b l i b l

y

k i i j

z

k

b i b

θ

(

s,i

)

 (

s,i, j

)

b

(

s,i)

b

(

s,

1

) ((

2) 1)  l l 

θ

R

u n( ), ( ) : 1 y nn N 

Page 6 of 27

Wiener Neural Technique

 Parametrisation of the linear state-space model

 Optimising the whole parameters of the system

Step 3 NN x( ) u n( ) z n( ) y n( )

Definition- The pair is in the output normal form if (,) A C

n

T

T

A

A

C

C

I

,

nn ln   A C  

Page 9 of 27

Benchmark Criteria

 

2

1

1 () () ˆ

N

rms

t

e

y

k

y

k

N

  

ˆ

(1 ) 100 y y FIT

y y

  

 

1

1 () () ˆ

N

mean

t

e

y

k

y

k

N

  

 A plot with the modelled output and the simulation error in

time domain

 A plot of simulation error in frequency domain

J. Schoukens, J. Suykens, L. Ljung, “Wiener-Hammerstein Benchmark,” 15th IFAC Symposium on

System Identification (SYSID 2009), July 6-8, St. Malo, France, 2009.

Page 12 of 27

Winer-Hammerstein

System

3rd order Chebyshev filter with cut

off 4.4 kHz

3rd order inverse Chebyshev filter

with cut off 5kHz

 Transmission zero in the frequency

band of interest

1 G q( )

f ( ) x u n( ) x n( ) w n( )

2 G q( ) y n( )

Input/Output test data

 Filtered Gaussian signal with cut off

10kHz

 Sampling frequency 51.2kHz

 SNR 70dB

Estimation

data

Test data

Page 14 of 27

Winer-Hammerstein

Frequency response plot of the

linear part

-1 -0.5 0 0.5 1 1.5 2 2.5

-1

-0.8

-0.6

-0.4

-0.2

0

0.2

0.4

0.6

0.8

1

Pole-Zero Map

Real Axis

Imaginary Axis

Pole-Zero map

of the linear part

NOTE: transmission zero of the second

filter is captured

Mag(dB)

P1= 0.89 േ i0.17

P2=0.7

േ i0.4

Z1= 0.6 േ i0.99

Z2= 0.8

േ i0.6

Page 23 of 27

Winer-Hammerstein

- Delay

- Transfer function order

- Static nonlinearity = exponential

2,

a

n

2

d

n

0 200 400 600 800 1000 1200 1400 1600 1800 2000

-1

-0.5

0

0.5

Test

FIT = 98.1439 %

Measured Output

Model Output

12345678

Samples 104

-0.01

0

0.01

Residu

rms= 0.0044405, mean = -4.884e-05

residu (dB)

2

b

n

Residual error spectrum