Premium
Identification of linear systems with time delay and unknown order
Author(s) -
Harada Ken'ichi,
Kobayashi Yasuhide,
Okita Tsuyoshi
Publication year - 2003
Publication title -
electrical engineering in japan
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.136
H-Index - 28
eISSN - 1520-6416
pISSN - 0424-7760
DOI - 10.1002/eej.10166
Subject(s) - a priori and a posteriori , identification (biology) , maximum a posteriori estimation , nonlinear system , bayes' theorem , system identification , computer science , algorithm , mathematics , maximum likelihood , mathematical optimization , control theory (sociology) , artificial intelligence , data mining , statistics , bayesian probability , physics , botany , control (management) , epistemology , quantum mechanics , biology , measure (data warehouse) , philosophy
Abstract This paper proposes a system identification method for linear systems with time delay and unknown order. We hypothesize a number of candidate models, which have different order, because system order is unknown. In each model, we estimate the model parameters based on the maximum likelihood method by using nonlinear optimization technique. Then both local optimization technique and global search method are used because the estimated parameters may fall into a local minimum. After all candidate model's parameters are estimated, one model is selected among these models to estimate the system order by using a posteriori probability based on Bayes's theorem. Finally, the validity of the proposed method is demonstrated by numerical examples. © 2003 Wiley Periodicals, Inc. Electr Eng Jpn, 145(3): 61–68, 2003; Published online in Wiley InterScience ( www.interscience.wiley.com ). DOI 10.1002/eej.10166