Multiobjective and multiscale optimization of composite materials by means of evolutionary computations

Beluch, Witold; Długosz, Adam

doi:10.15632/jtam-pl.54.2.397

Cited by 13 publications

(5 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The FE based modeling of the deformation behavior of the microstructure can provide very good accuracy however it should be noted that it may lead to very long times of computations due to many To solve the described optimization problem a global optimization method like for example a genetic algorithm can be applied [35][36][37][38][39]. In the case of the theoretical model, naturally the FE model which reflects the actual microstructure can be used.…”

Section: Inverse Identification Problemmentioning

confidence: 99%

Analysis of Strain Field Heterogeneity at the Microstructure Level and Inverse Identification of Composite Constituents by Means of Digital Image Correlation

Ogierman

Kokot

2020

Materials

View full text Add to dashboard Cite

The present paper is devoted to the theoretical study on the estimation of the full-field strain at the microstructural level of composite materials by means of Digital Image Correlation (DIC). The main aim of the paper is to investigate the influence of speckle size on the accuracy of the strain field measurement at the microscale. The DIC analysis was conducted based on artificial speckle patterns generated numerically and the deformation behavior of the composites was simulated by using the finite element method (FEM). This approach gives the opportunity to compare the results of the DIC in terms of speckle size with the reference FEM solution. Moreover, the paper focuses on the inverse identification of the material constants of the composite constituents by using information associated with the measured strain field. The inverse problem is solved by using a novel two-step optimization procedure, which reduces the problem complexity. The feasibility and accuracy of the proposed approach are presented by analysis of two exemplary microgeometries representing the microstructures of fiber reinforced composites.

show abstract

Section: Inverse Identification Problemmentioning

confidence: 99%

Analysis of Strain Field Heterogeneity at the Microstructure Level and Inverse Identification of Composite Constituents by Means of Digital Image Correlation

Ogierman

Kokot

2020

Materials

View full text Add to dashboard Cite

show abstract

“…The objective function is defined as follows: where H ij GIVEN is the reference (known) fourth order orientation tensor and H ij FOUND is the fourth order orientation tensor determined in terms of variables which are spherical angles and weights of the pseudo-grains (contracted index notation for the orientation tensor has been applied). Solution of the optimization problem can be found using the evolutionary algorithm, which is widely used in engineering applications involving multi-scale modelling (Beluch and Długosz, 2016; Burczyński et al , 2010). The following constraints have been applied on the variables: …”

Section: Homogenization Proceduresmentioning

confidence: 99%

A two-stage homogenization for modelling of elastic-plastic functionally graded composites

Ogierman

2020

View full text Add to dashboard Cite

Purpose The purpose of this study is to develop a homogenization approach that ensures both high accuracy and time-efficient solution for elastic-plastic functionally graded composites. Design/methodology/approach The paper presents a novel two-stage hybrid homogenization approach that combines advantages of the mean field homogenization and homogenization based on the finite element method (FEM). The groundbreaking nature of the developed approach is associated with division of the hybrid homogenization procedure into two stages, which allows to very efficiently determine the solution for arbitrary volume fraction of the reinforcement. This paper concerns also on modelling of composites with randomly distributed prolate and oblate particles. For this purpose, the hybrid homogenization was implemented in the framework of the discrete orientation averaging procedure involving pseudo-grain discretization method. Findings Agreement between the results obtained using the proposed approach and the standard FEM-based homogenization is very good (up to the volume fraction of 0.3). Originality/value The proposed two-stage homogenization approach allows to obtain the solution for materials with arbitrary volume fraction of the reinforcement very efficiently; therefore, it is highly beneficial for the two-scale modeling of nonlinear functionally graded materials and structures.

show abstract

“…The individuals exchange genetic material between each other (by crossover operators) and random changes occur in the genes of individuals (mutation operators) -this way new individuals are created. From the potential solutions, the biggest chance to survive have those mostly adapted to the environment (Beluch and Długosz, 2016;Jin, 2011;Mrozek et al, 2015).…”

Section: Evolutionary Optimization Approachmentioning

confidence: 99%

A new nondeterministic method for optimal selection of master degrees of freedom for dynamic condensation based on evolutionary optimization

Mucha¹

2020

Journal of Theoretical and Applied Mechanics

View full text Add to dashboard Cite

The following paper presents a new method for choosing a set of master degrees of freedom for the process of dynamic condensation in order to reduce a finite element model. The general rule is that the more degrees of freedom are eliminated, the more accurate the reduced model is. However, eliminating different subsets (of equal sizes) of degrees of freedom may influence the accuracy differently. Therefore, choosing an optimal subset is crucial. The presented method is based on multicriterial evolutionary optimization which makes it the first nondeterministic approach based on computational optimization technique for this application.

show abstract

Multiobjective and multiscale optimization of composite materials by means of evolutionary computations

Cited by 13 publications

References 29 publications

Analysis of Strain Field Heterogeneity at the Microstructure Level and Inverse Identification of Composite Constituents by Means of Digital Image Correlation

Analysis of Strain Field Heterogeneity at the Microstructure Level and Inverse Identification of Composite Constituents by Means of Digital Image Correlation

A two-stage homogenization for modelling of elastic-plastic functionally graded composites

A new nondeterministic method for optimal selection of master degrees of freedom for dynamic condensation based on evolutionary optimization

Contact Info

Product

Resources

About