Prediction of the compressive strength of concrete using selected machine learNing regression models

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F68407700%3A21110%2F21%3A00356426" target="_blank" >RIV/68407700:21110/21:00356426 - isvavai.cz</a>

Výsledek na webu

—

DOI - Digital Object Identifier

—

Jazyk výsledku

angličtina

Název v původním jazyce

Prediction of the compressive strength of concrete using selected machine learNing regression models

Popis výsledku v původním jazyce

Prediction of mechanical properties of cementitious composites is a topic of great concern as it could minimize the need for costly and laborious laboratory tests. In this paper, several machine learning models (Linear, Ridge, Lasso, and Support Vector Machine regression) are trained and evaluated on a publicly available dataset containing various concrete compositions and their compressive strength measured at different ages from casting. In this study, Support Vector Machine regression showed the highest accuracy when testing on the public dataset (mean absolute error 3.63 MPa). The trained models were also subsequently applied on additional more current data. Unfortunately, none of the models proved to be suitable which might be due to the low representativeness of the older public dataset for the currently used mixtures.

Název v anglickém jazyce

Prediction of the compressive strength of concrete using selected machine learNing regression models

Popis výsledku anglicky

Prediction of mechanical properties of cementitious composites is a topic of great concern as it could minimize the need for costly and laborious laboratory tests. In this paper, several machine learning models (Linear, Ridge, Lasso, and Support Vector Machine regression) are trained and evaluated on a publicly available dataset containing various concrete compositions and their compressive strength measured at different ages from casting. In this study, Support Vector Machine regression showed the highest accuracy when testing on the public dataset (mean absolute error 3.63 MPa). The trained models were also subsequently applied on additional more current data. Unfortunately, none of the models proved to be suitable which might be due to the low representativeness of the older public dataset for the currently used mixtures.

Druh

D - Stať ve sborníku

CEP obor

—

OECD FORD obor

20101 - Civil engineering

Projekt

—

Návaznosti

S - Specificky vyzkum na vysokych skolach

Rok uplatnění

2021

Kód důvěrnosti údajů

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

Název statě ve sborníku

Proceedings of PhD Workshop, Department of Concrete and Masonry Structures 2021

ISBN

978-80-01-06842-7

ISSN

—

e-ISSN

—

Počet stran výsledku

6

Strana od-do

—

Název nakladatele

katedra betonových a zděných konstrukcí

Místo vydání

Praha

Místo konání akce

Praha

Datum konání akce

21. 5. 2021

Typ akce podle státní příslušnosti

EUR - Evropská akce

Kód UT WoS článku

—