Fiber-reinforced cementitious composite: sensitivity analysis and parameter identification

Kód výsledku v IS VaVaI

<a href="https://www.isvavai.cz/riv?ss=detail&h=RIV%2F28399269%3A_____%2F20%3AN0000018" target="_blank" >RIV/28399269:_____/20:N0000018 - isvavai.cz</a>

Nalezeny alternativní kódy

RIV/00216305:26110/20:PU138154

Výsledek na webu

<a href="https://www.vbripress.com/aml/pdf/1493" target="_blank" >https://www.vbripress.com/aml/pdf/1493</a>

DOI - Digital Object Identifier

<a href="http://dx.doi.org/10.5185/amlett.2020.031488" target="_blank" >10.5185/amlett.2020.031488</a>

Jazyk výsledku

angličtina

Název v původním jazyce

Fiber-reinforced cementitious composite: sensitivity analysis and parameter identification

Popis výsledku v původním jazyce

Advanced assessment of fracture–mechanical properties is primary important for subsequent numerical simulations of components/structures made fibre-reinforced concrete (FRC). The variability of experimental results using specimens made of quasi-brittle materials such as FRC high due to the heterogeneity of this materials. Then an assessment of fracture–mechanical parameters is much more difficult and problematic. To remain at deterministic level is therefore impossible and without virtual statistical approach, simulation and probabilistic result assessment the consequent practical design of quasi-brittle material-based structures can be risky. A key parameter of nonlinear fracture mechanics modelling is certainly fracture energy of concrete and its variability. The aim of the performed research is to deepen knowledge about the complexity of the behaviour of this advanced and progressive composite material especially in relation to its resistance to crack propagation. The obtained knowledge is a prerequisite for efficient design of this composite and the consequent expansion of its applicability for increase of sustainability of constructed elements, structures and buildings. The research is focused on several topics: First, development of suitable constitutive law for FRC cementitious composites under the framework of nonlinear fracture mechanics software ATENA (Červenka et al. 2002) is of primary importance. The computational model has been verified using data from company DURA technology company, Malaysia. Sensitivity analysis which showed importance of parameters of computational model have been performed. Finally, material parameters identification was performed based on the combination of the efficient statistical simulation (Novák et al. 2014) and training of the artificial neural network (Lehký et al. 2014). The inverse analysis – identification technique – is based on the combination of statistical simulation and training of the neural network. The material model parameters are considered as random variables described by a probability distribution, rectangular distribution is a “natural choice” as the lower and upper limits represent the bounded range of physical existence. The variables are then simulated randomly based on the Monte Carlo type simulation, the small-sample simulation LHS is utilized. Multiple calculation of the deterministic computational model using random realizations of material model parameters is performed and a statistical set of the virtual structural response is obtained. Random realizations and the corresponding responses from the computational model serve as the basis for the training of an appropriate neural network (Novák & Lehký, 2006). After the training the neural network is ready to solve the opposite task: To select the best material parameters in order the numerical simulation will result in the best agreement with experiment. This is performed by means of the simulation of network using measured response as an input. It results in a set of identified material model parameters. The last step is results verification – calculation of the computational model the using identified parameters. The proposed methodology and software is based on experimental and computational methods falling within the field of fracture mechanics, soft computing and reliability theory.

Název v anglickém jazyce

Fiber-reinforced cementitious composite: sensitivity analysis and parameter identification

Popis výsledku anglicky

Advanced assessment of fracture–mechanical properties is primary important for subsequent numerical simulations of components/structures made fibre-reinforced concrete (FRC). The variability of experimental results using specimens made of quasi-brittle materials such as FRC high due to the heterogeneity of this materials. Then an assessment of fracture–mechanical parameters is much more difficult and problematic. To remain at deterministic level is therefore impossible and without virtual statistical approach, simulation and probabilistic result assessment the consequent practical design of quasi-brittle material-based structures can be risky. A key parameter of nonlinear fracture mechanics modelling is certainly fracture energy of concrete and its variability. The aim of the performed research is to deepen knowledge about the complexity of the behaviour of this advanced and progressive composite material especially in relation to its resistance to crack propagation. The obtained knowledge is a prerequisite for efficient design of this composite and the consequent expansion of its applicability for increase of sustainability of constructed elements, structures and buildings. The research is focused on several topics: First, development of suitable constitutive law for FRC cementitious composites under the framework of nonlinear fracture mechanics software ATENA (Červenka et al. 2002) is of primary importance. The computational model has been verified using data from company DURA technology company, Malaysia. Sensitivity analysis which showed importance of parameters of computational model have been performed. Finally, material parameters identification was performed based on the combination of the efficient statistical simulation (Novák et al. 2014) and training of the artificial neural network (Lehký et al. 2014). The inverse analysis – identification technique – is based on the combination of statistical simulation and training of the neural network. The material model parameters are considered as random variables described by a probability distribution, rectangular distribution is a “natural choice” as the lower and upper limits represent the bounded range of physical existence. The variables are then simulated randomly based on the Monte Carlo type simulation, the small-sample simulation LHS is utilized. Multiple calculation of the deterministic computational model using random realizations of material model parameters is performed and a statistical set of the virtual structural response is obtained. Random realizations and the corresponding responses from the computational model serve as the basis for the training of an appropriate neural network (Novák & Lehký, 2006). After the training the neural network is ready to solve the opposite task: To select the best material parameters in order the numerical simulation will result in the best agreement with experiment. This is performed by means of the simulation of network using measured response as an input. It results in a set of identified material model parameters. The last step is results verification – calculation of the computational model the using identified parameters. The proposed methodology and software is based on experimental and computational methods falling within the field of fracture mechanics, soft computing and reliability theory.

Druh

J_ost - Ostatní články v recenzovaných periodicích

CEP obor

—

OECD FORD obor

20101 - Civil engineering

Projekt

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

Návaznosti

P - Projekt vyzkumu a vyvoje financovany z verejnych zdroju (s odkazem do CEP)

Rok uplatnění

2020

Kód důvěrnosti údajů

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

Název periodika

Advanced Materials Letters

ISSN

0976-3961

e-ISSN

—

Svazek periodika

11

Číslo periodika v rámci svazku

3

Stát vydavatele periodika

SE - Švédské království

Počet stran výsledku

5

Strana od-do

20031488-20031488

Kód UT WoS článku

—

EID výsledku v databázi Scopus

—