Title :
Estimation of nonlinear ARX models
Author_Institution :
Control Syst. Sect., Eindhoven Univ. of Technol., Netherlands
Abstract :
Estimation of some nonlinear ARX (NARX) models is studied. First the model structures are discussed and their numerical tractability examined. Three NARX models axe proposed. Parameter estimation is solved for the three models. A relaxation iteration scheme is proposed by using the model structures in which the error is bilinear-in-parameters. The convergence and consistency of the method are established. Simulation study will be used to illustrate the method.
Keywords :
autoregressive processes; computational complexity; convergence of numerical methods; iterative methods; parameter estimation; relaxation theory; NARX models; bilinear-in-parameters error; consistency; convergence; model structures; nonlinear ARX model estimation; numerical tractability; parameter estimation; relaxation iteration scheme; Algorithm design and analysis; Control systems; Convergence; Ear; Linear systems; Linearity; Nonlinear equations; Nonlinear systems; Parameter estimation; Recursive estimation;
Conference_Titel :
Decision and Control, 2002, Proceedings of the 41st IEEE Conference on
Print_ISBN :
0-7803-7516-5
DOI :
10.1109/CDC.2002.1184860
