User login


You are here

Mirkhalaf's blog

Mirkhalaf's picture

Deep-learning model using a small dataset

One the main challenges of developing data-driven models is the data-hungry nature of Artificial Neural Networks (ANNs). In our recent paper, we introduced a data augmentation approach to expand a small original dataset without conducting extra expensive high-fidelity simulations. We then used the original and augmented datasets for developing ANN models for non-linear path dependent composites. The obtained results showed the great impact of the data augmentation approach on the accuracy of the data-driven models.

Mirkhalaf's picture

ECCOMAS 2024: Mini symposium on "Advances in Machine Learning for Composite Materials"

We are organizing a mini symposium at ECCOMAS 2024 ( entitled "Advances in Machine Learning for Composite Materials". The conference will be held in Lisbon, Portugal on 3-7 June 2024.

Mirkhalaf's picture

Multiscale deep learning for woven composites

Woven composites exhibit a complex hierarchical structure with multiple heterogeneous sub-scales, stemming from microscale fiber arrangements and mesoscale interlacing patterns, necessitating sophisticated modeling approaches to accurately capture their intricate multiscale nature.

Mirkhalaf's picture

PhD positions for a MSCA-DN project (DurAMat)

PhD positions are available for DurAMat which is a Marie Skłodowska-Curie Actions - Doctoral Network (MSCA-DN) project.  A total number of 11 PhD students will be hired for DurAMat. One PhD student will join us at the department of Physics of the University of Gothenburg in Sweden. The project details and the application process are described in the project website:

Mirkhalaf's picture

[Polymers] Special Issue: Mechanics of 3D-Printed Polymers and Polymer Composites

A special issue on "Mechanics of 3D-Printed Polymers and Polymer Composites", is recently launched in the journal "Polymers". If your research is related to the topic, I would like to invite you to submit your latest research developments to the special issue. For more details and submitting your manuscript, please see:


Mirkhalaf's picture

A micromechanics-based deep learning model for short fiber composites

If you are curious about application of machine learning techniques in mechanics problems, our latest paper is probably interesting for you. In this paper, we are proposing a micromechanics-based artificial neural networks model for short fiber composites. You can find the paper here: 

Mirkhalaf's picture

PhD position on machine learning enhanced multi-scale modelling of textile composites at the University of Gothenburg

We have an open PhD position on machine learning enhanced multi-scale modelling of textile composites. The following link provides more information about the project, and the details of the application process. Please keep in mind that only applications sent through the online application system will be evaluated.

Description of the PhD project, and how to apply


Mirkhalaf's picture

Post yield response of amorphous polymers under different stress states

In this contribution, an elasto-viscoplastic constitutive model based on the single mode EGP (Eindhoven Glassy Polymer) model is proposed to describe the deformation behaviour of solid polymers subjected to finite deformations under different stress states. The material properties of the original model are determined and calibrated from a uniaxial compression-loading test. Then, several numerical examples under different stress states are presented to illustrate the limitations.

Mirkhalaf's picture

RVE size determination

The definition of the size of the Representative Volume Element (RVE) is extremely important for the mechanics and physics of heterogeneous materials since it should statistically represent the micro-structure of the material. In the present contribution, a methodology based on statistical analysis and numerical experiments is proposed to determine the size of the RVE for heterogeneous amorphous polymers subjected to finite deformations.

Mirkhalaf's picture

Polymer constitutive modelling: Formulation and computational aspects

  • This work formulates an elasto-viscoplastic model at finite strains.
  • A particularly efficient numerical integration algorithm is presented.
  • A closed-form analytical expression is derived for the consistent tangent operator.
  • The non-linear behaviour of polymers is captured in the numerical examples.
Mirkhalaf's picture

EUROMECH Conference

13 - 16 September 2016, Porto, Portugal


Plenary Speakers:

Prof. Wing K. Liu, Northwestern University

Mirkhalaf's picture

EUROMECH Conference on Multi-uncertainty and Multi-scale Methods and Applications


13 September – 16 September 2016, Portugal

The conference aims to:

(a) present the state-of-the-art in this field by showing the most recent developments by leading experts, and

Mirkhalaf's picture




Mirkhalaf's picture

Multiscale Material Modelling

Computational modelling of materials behaviour
is becoming a reliable tool to underpin scientific investigations and to
complement traditional theoretical and experimental approaches. In cases where
an understanding of the dual nature of the structure of matter (continuous when
viewed at large length scales and discrete when viewed at smaller length scales) and
its interdependences are crucial, multiscale materials modelling (MMM)

Subscribe to RSS - Mirkhalaf's blog

Recent comments

More comments


Subscribe to Syndicate