Joint Parameter and Time-Delay Identification Algorithm and Its Convergence Analysis for Wiener Time-Delay Systems
- 59 Downloads
New developments for parameter and time-delay identification are presented for discrete nonlinear systems with delayed input. The proposed approach is based on overparametrization approach which involves subsuming the delay term into an extended numerator polynomial of the linear block of Wiener time-delay system. On this basis, the parameter identification problem can be then solved using recursive least squares-based optimization techniques and then, the delay is calculated directly based on the extended numerator polynomial identified: For a noise-free system, all extended numerator parameters are equal to zero. However in the noisy-output case, it is necessary to introduce an upper bound and the extended parameters whose values are smaller than a threshold level should be identified as zero. Then, the delay is determined as the first number of null extended parameter values. In addition, the convergence of the identified parameter vector is studied. The performances of the proposed identification algorithms are illustrated through simulation examples.
KeywordsIdentification Wiener systems Time-delay estimation Parameter estimations Recursive least squares method Convergence analysis
This work was supported by the ministry of Higher Education and Scientific Research in Tunisia.
- 1.A. Atitallah, S. Bedoui, K. Abderrahim, Identification of Wiener time delay systems based on hierarchical gradient approach, in 8th Vienna International Conference on Mathematical Modelling, Vienna, Austria, vol. 48(1), pp. 403–408 (2015)Google Scholar
- 4.A. Atitallah, S. Bedoui, K. Abderrahim, New results on Wiener time delay system identification, in The 15th Annual European Control Conference ECC, pp. 1637–1642 (2016)Google Scholar
- 30.A. O’Dwyer, Time delay estimation in signal processing applications: an overview, in IT & T Conference, October, pp. 1–6 (2002)Google Scholar
- 34.M. Sbarciog, R. De Keyser, S. Cristea, C. De Prada, Nonlinear predictive control of processes with variable time delay. A temperature control case study, in 17th IEEE International Conference on Control Applications, pp. 1001–1006 (2008)Google Scholar
- 36.A. Srinivasan, P. Lakshmi, Wiener model based real-time identification and control of heat exchanger process. J. Automat. Syst. Eng. 2(1) (2008)Google Scholar