Evolutionary Optimization of Catalysts Assisted by Neural-Network Learning

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F67985807%3A_____%2F10%3A00351863" target="_blank" >RIV/67985807:_____/10:00351863 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Evolutionary Optimization of Catalysts Assisted by Neural-Network Learning

Popis výsledku v původním jazyce

This paper presents an important real-world application of both evolutionary computation and learning, an application to the search for optimal catalytic materials. In this area, evolutionary and especially genetic algorithms are encountered most frequently. However, their application is far from any standard methodology, due to problems with mixed optimization and constraints. The paper describes how these difficulties are dealt with in the evolutionary optimization system GENACAT, recently developed for searching optimal catalysts. It also recalls that the costly evaluation of objective functions in this application area can be tackled through learning suitable regression models of those functions, called surrogate models. Ongoing integration of neural-networks-based surrogate modelling with GENACAT is illustrated on two brief examples.

Název v anglickém jazyce

Evolutionary Optimization of Catalysts Assisted by Neural-Network Learning

Popis výsledku anglicky

This paper presents an important real-world application of both evolutionary computation and learning, an application to the search for optimal catalytic materials. In this area, evolutionary and especially genetic algorithms are encountered most frequently. However, their application is far from any standard methodology, due to problems with mixed optimization and constraints. The paper describes how these difficulties are dealt with in the evolutionary optimization system GENACAT, recently developed for searching optimal catalysts. It also recalls that the costly evaluation of objective functions in this application area can be tackled through learning suitable regression models of those functions, called surrogate models. Ongoing integration of neural-networks-based surrogate modelling with GENACAT is illustrated on two brief examples.

Druh

D - Stať ve sborníku

CEP obor

IN - Informatika

OECD FORD obor

—

Projekt

Výsledek vznikl pri realizaci vícero projektů. Více informací v záložce Projekty.

Návaznosti

Z - Vyzkumny zamer (s odkazem do CEZ)

Rok uplatnění

2010

Kód důvěrnosti údajů

S - Úplné a pravdivé údaje o projektu nepodléhají ochraně podle zvláštních právních předpisů

Název statě ve sborníku

Simulated Evolution and Learning

ISBN

978-3-642-17297-7

ISSN

—

e-ISSN

—

Počet stran výsledku

10

Strana od-do

—

Název nakladatele

Springer

Místo vydání

Berlin

Místo konání akce

Kanpur

Datum konání akce

1. 12. 2010

Typ akce podle státní příslušnosti

WRD - Celosvětová akce

Kód UT WoS článku

000289185200023