© The Institution of Engineering and Technology
A consequent and consistent continuoustime approach to system parameter estimation is introduced. Estimation algorithms, the underlying quality criteria and models of identified systems are described in the continuoustime domain, while suitable discretising operations are performed solely for the purpose of ultimate numerical realisation of estimation procedures. The considered indices of estimation quality take the form of integrals of absolute prediction errors rather than a common form of integrals or sums of square errors. In order to overcome the problem of analytical minimisation of such nondifferentiable criteria, an approximate method is derived and applied in practical implementation of the resultant estimation schemes. Specific weighting mechanisms utilised in the algorithms allow for tracking the timevariant parameters of nonstationary systems, while with the employed instrumental variable the accuracy of estimates gets improved by means of suppression of the asymptotic bias. Following the socalled direct approach, an auxiliary discretetime model that retains ‘physical’ parameterisation is obtained based on ‘finitehorizon’ splinebased integration of both sides of the presumed differential equation. In this aspect, application of splines makes the respective discretetime processing resistant to cumulation of numerical errors. The attached numerical examples demonstrate the performance of the discussed estimation routines.
References


1)

H. Garnier ,
L. Wang
.
(2008)
Identification of continuoustime models from sampled data.

2)

R. Johansson
.
Identification of continuoustime models.
IEEE Trans. Signal Process.
,
887 
897

3)

Z. Kowalczuk
.
Discretetime approximation of continuoustime systems – a survey.
Proc. IEE G
,
264 
278

4)

Kowalczuk, Z.: `Online continuoustime system identification based on normal integrating operators', Proc. 1st Int. Symp. on Methods and Models in Automation and Robotics, September 1994, Miedzyzdroje, Poland, p. 325–330.

5)

Z. Kowalczuk
.
Discretetime realisation of online continuoustime estimation algorithms.
Control Comput.
,
33 
37

6)

Z. Kowalczuk
.
Adaptive discretetime identification of continuoustime systems using adjusted integration.
Appl. Mathematics Comput. Sci.
,
41 
75

7)

Kowalczuk, Z., Kozlowski, J.: `Continuoustime identification of continuoustime systems', Proc. 11th IFAC Symp. on System Identification, July 1997, Kitakyushu, Japan, p. 1343–1348.

8)

Kowalczuk, Z., Kozlowski, J.: `Integrated squared error and integrated absolute error in recursive identification of continuoustime plants', Proc. Int. Conf. on Control, September 1998, Swansea, UK, p. 693–698.

9)

Z. Kowalczuk ,
J. Kozlowski
.
Continuoustime approaches to identification of continuoustime systems.
Automatica
,
1229 
1236

10)

R.H. Middleton ,
G.C. Goodwin
.
(1990)
Digital control and estimation. A unified approach.

11)

S. Sagara ,
Z.Y. Zhao
.
Numerical integration approach to online identification of continuoustime systems.
Automatica
,
63 
74

12)

Söderström, T., Fan, H., Carlsson, B., Mossberg, M.: `Some approaches on how to use the delta operator when identifying continuoustime processes', Proc. 36th IEEE Conf. on Decision and Control, December 1997, San Diego, USA, p. 890–895.

13)

Söderström, T., Fan, H., Mossberg, M., Carlsson, B.: `A biascompensation scheme for estimating continuous time AR process parameters', Proc. 11th IFAC Symp. on System Identification, July 1997, Kitakyushu, Japan, p. 1337–1342.

14)

H. Unbehauen ,
G.P. Rao
.
Continuoustime approaches to system identification – a survey.
Automatica
,
23 
35

15)

Unbehauen, H., Rao, G.P.: `Identification of continuoustime systems: A tutorial', Proc. 11th IFAC Symp. on System Identification, July 1997, Kitakyushu, Japan, p. 1023–1049.

16)

Z.J. Yang ,
T. Tsuji ,
S. Sagara
.
Integral equation approach to parameter identification without consideration of initial conditions.
Int. J. Systems Sci.
,
763 
774

17)

P. Young
.
Parameter estimation for continuoustime models – a survey.
Automatica
,
23 
39

18)

Sagara, S., Yang, Z.J., Wada, K.: `Identification of continuous systems from noisy sampled inputoutput data', Proc. 9th IFAC/IFORS Symp. on Identification and System Parameter Estimation, July 1991, Budapest, Hungary, p. 603–608.

19)

Chen, T., Miller, D.E.: `Reconstruction of continuoustime systems from their discretizations', Proc. 14th IFAC World Congress, July 1999, Beijing, People's Republic of China, p. 187–192.

20)

Janiszowski, K.B.: `To estimation in sense of the least sum of absolute errors', Proc. 5th Int. Symp. on Methods and Models in Automation and Robotics, August 1998, Miedzyzdroje, Poland, p. 583–588.

21)

Kozlowski, J.: `Nonquadratic quality indices in estimation, approximation and control', Proc. 9th IEEE Int. Conf. on Methods and Models in Automation and Robotics, August 2003, Miedzyzdroje, Poland, p. 277–282.

22)

L. Ljung ,
T. Soderstrom
.
(1983)
Theory and practice of recursive identification.

23)

E.D. Fiodorov
.
Least absolute values estimation: computational aspects.
IEEE Trans. Autom. Control
,
626 
630

24)

T.C. Hsia
.
(1977)
System identification.

25)

S. Sagara ,
Z.J. Yang ,
K. Wada
.
Identification of continuous systems using digital lowpass filters.
Int. J. Syst. Sci.
,
1159 
1176

26)

B.M. Ninness ,
G.C. Goodwin ,
N.K. Sinha ,
G.P. Rao
.
(1991)
The relationship between discrete time and continuous time linear estimation.

27)

Y.C. Chao ,
C.L. Chen ,
H.P. Huang
.
Recursive parameter estimation of transfer function matrix models via Simpson's integrating rules.
Int. J. Syst. Sci.
,
901 
911

28)

B. Ninness ,
H. Hjalmarsson ,
F. Gustafsson
.
The fundamental role of general orthonormal bases in system identification.
IEEE Trans. Autom. Control
,
1384 
1406

29)

Garnier, H., Sibille, P., Nguyen, H.L., Spott, T.: `A biascompensating leastsquares method for continuoustime system identification via Poisson moment functionals', Proc. 10th IFAC Symp. on System Identification, July 1994, Copenhagen, Denmark, p. 675–680.

30)

T. Söderström ,
P. Stoica
.
Comparison of some instrumental variable methods – consistency and accuracy aspects.
Automatica
,
101 
115

31)

H. Dai ,
N.K. Sinha ,
N.K. Sinha ,
G.P. Rao
.
(1991)
Use of numerical integration methods.

32)

Schoukens, J., Pintelon, R., Guillaume, P.: `On the advantages of periodic excitation in system identification', Proc. 10th IFAC Symp. on System Identification, July 1994, Copenhagen, Denmark, p. 153–158.

33)

K.J. Åström ,
B. Wittenmark
.
(1984)
Computercontrolled systems. Theory and design.

34)

Kozlowski, J., Kowalczuk, Z.: `Iterative and recursive algorithms derived from absoluteerror measures', Proc. eighth IEEE Int. Conf. on Methods and Models in Automation and Robotics, September 2002, Szczecin, Poland, p. 1291–1296.

35)

J. Kozlowski ,
Z. Kowalczuk ,
Z. Kowalczuk ,
B. Wiszniewski
.
(2007)
Robust to measurement faults parameter estimation algorithms in issues on systems diagnosis.
http://iet.metastore.ingenta.com/content/journals/10.1049/ietcta.2010.0310
Related content
content/journals/10.1049/ietcta.2010.0310
pub_keyword,iet_inspecKeyword,pub_concept
6
6