Pemodelan Sistem Dengan Metoda Identifikasi Parameter Menggunakan Pendekatan Model ARX (Autoregressive Exogenous)
Abstract – Systems that exist in nature are a combination of linear and nonlinear elements, and there are still many other factors that affect the linearity of the system. System models are very difficult to obtain with mathematical calculations, and also many neglected elements, especially nonlinea...
| Main Author: | |
|---|---|
| Format: | Article |
| Language: | Indonesian |
| Published: |
Department of Electrical Engineering, Faculty of Engineering, Tanjungpura University
2019-10-01
|
| Series: | Elkha: Jurnal Teknik Elektro |
| Subjects: | |
| Online Access: | http://jurnal.untan.ac.id/index.php/Elkha/article/view/30970 |
| Summary: | Abstract – Systems that exist in nature are a combination of linear and nonlinear elements, and there are still many other factors that affect the linearity of the system. System models are very difficult to obtain with mathematical calculations, and also many neglected elements, especially nonlinear elements. So that the resulting model becomes less effective to be applied directly in the field. In addition to these methods, there are other modeling methods, namely the identification method. This identification method is carried out on the basis of the input signal and output of the system to be modeled. All of that is considered to be an integrated system (black box). The model structure approach that will be used is ARX (Autoregressive Exogenous), which is a linear model structure, and parameter estimation algorithm uses the least square. Based on the analysis that has been done, the system is linear, there is no noise, has a time delay with off-line data pairs. From the modeling process carried out, order of the system and model parameters are obtained, so that a good system model is obtained. The model obtained, then can be used for other purposes, for example for the purposes of control, analysis needs, and for other simulation purposes. |
|---|---|
| ISSN: | 1858-1463 2580-6807 |