Volume 12, Issue 2 (2012)
MJEE 2012, 12(2): 36-41 |
Back to browse issues page
Download citation:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
BibTeX | RIS | EndNote | Medlars | ProCite | Reference Manager | RefWorks
Send citation to:
Sojoodi M, Soleymani F. System identification by a generalized interior point algorithm and nonlinear optimization methods considering ARMA model. MJEE 2012; 12 (2) :36-41
URL: http://mjee.modares.ac.ir/article-17-10942-en.html
URL: http://mjee.modares.ac.ir/article-17-10942-en.html
1- Advanced Control Systems Laboratory, School of Electrical and Computer Engineering, Tarbiat Modares University, Tehran,
2- Advanced Control Systems Laboratory, School of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran
2- Advanced Control Systems Laboratory, School of Electrical and Computer Engineering, Tarbiat Modares University, Tehran, Iran
Abstract: (4721 Views)
In this paper, we describe our implementation of an interior point algorithm for large scale systems. First we identify system with small and medium methods convex optimization, then we use interior point method for identification. Finally we offer an interior point method that uses nonlinear cost function and see that we achieve a good trade-off between error and CPU time. Actually, in this paper, we are looking for a method that can identify large scale systems with low model order, error and CPU time of solution of simulation. Previous articles didn’t check the order of the computed model, and the relationship between the error and CPU time. We assume that the model of our simulation is ARMA. We are going to identify a large scale system and compute the error and CPU time and compare the relationships. Examined data in this paper is related to cutaneous potential recordings of a pregnant woman. These data are pendulous and have a large standard deviation; therefore, it can’t be fitted with ordinary curve fittings, so we use the smoothing spline for computing the order of the model. Finally, we checked the influence of the number of data on error and CPU time and order of model
Received: 2015/02/23 | Accepted: 2015/06/17 | Published: 2015/11/22
| Rights and permissions | |
 |
This work is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License. |