Publication detail

Using AI Methods to Find a Non-Linear Regression Model with a Coupling Condition

MATOUŠEK, R.

Czech title

Užití metod umělé inteligence pro nalezení nelineárního regresního modelu s vazební podmínkou

English title

Using AI Methods to Find a Non-Linear Regression Model with a Coupling Condition

Type

journal article - other

Language

en

Original abstract

In the real-life engineering practice, non-linear regression models have to be designed rather often. To ensure their technical or physical feasibility, such models may, in addition, require another coupling condition. This paper describes two procedures for designing a specific non-linear model using AI methods. A Radial Basis Functions (RBF) based optimization is presented of the model using Genetic Algorithms (GA).

Czech abstract

V reálných aplikacích se v otázce regrese často uplatňují podmínky dané fyzikální podstatou (např. neze uvažovat teplotu pod danou mez apod.). V tomto příspěvku je řešen problém návrhu nelineárního regresniho modelu s vazební podmínkou nezápornosti. Příklad je uveden na reálných technologických datech. K řešení je využito RBF neuronové sítě a pokročilého genetického algoritmu.

English abstract

In the real-life engineering practice, non-linear regression models have to be designed rather often. To ensure their technical or physical feasibility, such models may, in addition, require another coupling condition. This paper describes two procedures for designing a specific non-linear model using AI methods. A Radial Basis Functions (RBF) based optimization is presented of the model using Genetic Algorithms (GA).

Keywords in Czech

Regrese, RBF, Neuronová síť, Genetický algoritmus

Keywords in English

Regression, RBF, Neural netvork, Genetic algorithm

RIV year

2011

Released

05.01.2011

Publisher

Pavel Heriban

Location

Brno

ISSN

1802-1484

Journal

Engineering Mechanics

Volume

17

Number

5/6

Pages from–to

419–431

Pages count

13

BIBTEX


@article{BUT50585,
  author="Radomil {Matoušek},
  title="Using AI Methods to Find a Non-Linear Regression Model with a Coupling Condition",
  journal="Engineering Mechanics",
  year="2011",
  volume="17",
  number="5/6",
  month="January",
  pages="419--431",
  publisher="Pavel Heriban",
  address="Brno",
  issn="1802-1484"
}