Deterministic Mutation Algorithm As A Winner Over Forward Selection Procedure
System identification is a field of study involving the derivation of a mathematical model to explain the dynamical behaviour of a system. One of the steps in system identification is model structure selection which involves the selection of variables and terms of a model. Several important criteri...
Saved in:
Main Author: | |
---|---|
Format: | Article |
Language: | English |
Published: |
Asian Research Publishing Network (ARPN)
2016
|
Subjects: | |
Online Access: | http://eprints.utem.edu.my/id/eprint/17003/2/jeas_0416_4078%20DMA%20vs%20FSP.pdf http://eprints.utem.edu.my/id/eprint/17003/ http://www.arpnjournals.org/jeas/research_papers/rp_2016/jeas_0416_4078.pdf |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Institution: | Universiti Teknikal Malaysia Melaka |
Language: | English |
Summary: | System identification is a field of study involving the derivation of a mathematical model to explain the dynamical
behaviour of a system. One of the steps in system identification is model structure selection which involves the selection of variables and terms of a model. Several important criteria for a desirable model structure include its accuracy in future prediction and model parsimony. A parsimonious model structure is desirable in enabling easy control design. Two methods of model structure selection are closely looked into and these are deterministic mutation algorithm (DMA) and forward selection procedure (FSP). The DMA is known to be originated from evolutionary computation whereas FSP may be listed under the study of regression. They have close similarities in characteristics, more specifically known as forward search in model structure selection. However, both also function in a population-based optimization and statistical approaches, respectively. Due to the closeness, this research attempts to clarify the advantages and disadvantages of both methods through model structure selection of difference equation model in system identification. Simulated and real data were used. To allow for fair comparison, DMA was altered so as to equalize its strength, where applicable, to that of FSP. In the real data simulation, both methods obtained the same model structure whereas in simulated data modelling, only DMA was able to select the correct model structure. This concludes that DMA not only has the advantage of simpler procedure but it also superseded the performance of FSP, even with a handicapped alteration. |
---|