Artykuły (CN-cmms)

Permanent URI for this collectionhttps://repo.agh.edu.pl/handle/AGH/102779

Artykuły czasopisma Computer Methods in Materials Science

Browse

Now showing 1 - 20 of 93

Access status: Open Access ,
A case study on tundish fluid flow with electromagnetic stirring
(Wydawnictwa AGH, 2024) Zielińska, Monika; Yang, Hongliang; Madej, Łukasz; Malinowski, Łukasz
Tundish is a crucial component just before casting and plays a pivotal role in enhancing the cleanliness and overall homogeneity of the final steel composition. The paper deals with the development of an advanced Computational Fluid Dynamics (CFD) model, specifically focusing on the molten steel flow within the tundish to numerically support its further improvements. A noteworthy addition to the model is the consideration of an electromagnetic stirring device. This device significantly influences steel cleanliness and composition, thereby affecting the final properties of the formed metallic parts in subsequent processing stages. The current investigation presents a comprehensive analysis of flow patterns and stirring energy distributions in relation to active and dead zones within the tundish. Through the developed coupled electromagnetic/fluid dynamic model, the paper demonstrates the feasibility of optimizing mixing processes to control the properties of the final product.
Access status: Open Access ,
A dedicated sensitivity analysis and optimization application for industrial processes
(Wydawnictwa AGH, 2021) Myczkowska, Kamila; Szeliga, Danuta
The paper describes the architecture and the use case of the developed Modelbox system for sensitivity analysis (SA), uncertainty analysis (UA) and the subsequent optimization of industrial processes. The proposed solution addresses the most common practical and technical problems encountered by researchers and engineers when performing sensitivity analysis. It combines the functions from the numerical toolbox with a simulation management system. Maintaining usability and a good user experience while managing complex investigations of time-consuming industrial process simulations is a very important feature of the system. Several improvements were introduced to optimize the computation time of analysis/modelling tasks, including the automatization of distributed calculations, persistent, transparent caching of simulation data and duration estimations from collected statistics. The system has the ability to perform remote, parallel, asynchronous computations of both analytic algorithms and numerical simulations. The system is dynamically scalable horizontally by using serverless computing endpoints and thus it can be easily adapted to the user's current needs in a flexible way. Modelbox provides web-based access to analysis/modelling tasks from sampling, SA/UA, optimization to metamodelling. It is extended with numerous interactive visualization components for effective results control. In addition, to access data from the completed analysis, the system supports convergence tracking for SA estimates and intermediate optimization results. The process of controlled cooling of rails was considered as a case study. The formulated optimization task was to find a combination of process parameters that ensures a minimum volume fraction of bainite along with required interlamellar spacing and optimal homogeneity of hardness. Different sensitivity analysis methods were used to evaluate the significance of all variables with respect to their influence on the model output.
Access status: Open Access ,
A Digital Twin for temperature prediction in the laser hardening process of NC10 steel
(Wydawnictwa AGH, 2026) Lacki, Piotr; Derlatka, Anna; Lacki, Michał; Lachs, Kuba
In this study, Artificial Neural Networks (ANN) were created to develop a Digital Twin (DT) for temperature prediction in the laser hardening process of NC10 steel. The ANN were trained to predict temperature on the top layer during the laser hardening process of NC10 steel samples with different thicknesses and with various laser power and laser scanning speeds. The prediction developed during the project work was based on a parametric numerical model of the laser hardening process for a sample of NC10 steel, using the Finite Element Method (FEM) within the ADINA software. Numerical simulations enabled a detailed analysis of the temperature produced on the surface of each sample, as well as a visualization of the structural changes made to the sample according to the laser hardening process. It is crucial to create data that reflects reality as closely as possible to assess the best setting for each process. A well created DT allows to make automatically important changes along laser hardening process. To obtain a set of the most efficient parameters for the desired result, Genetic Algorithms (GA) were integrated with the developed ANN. As a result, the authors developed an effective and efficient tool to predict the temperature produced along the laser hardening process.
Access status: Open Access ,
A finite-strain model for a superelastic NiTi shape memory alloy
(Wydawnictwa AGH, 2022) Jiang, Dongjie; Xiao, Yao
A finite-strain constitutive model of a superelastic NiTi shape memory alloy is proposed in this paper. Via backward Euler implicit integration scheme and the incorporation of material softening, the model is implemented into finite element code to reproduce a Lüders like deformation of a superelastic NiTi. The simulation results are in agreement with the experimental results, indicating that the constitutive model can reasonably predict the mechanical behavior of a superelastic NiTi. A parametric study further verifies that the magnitude of softening modulus has a significant effect on the stress-strain response and Lüders-like deformation of a superelastic NiTi.
Access status: Open Access ,
A hybrid statistical approach for texture images classification based on scale invariant features and mixture gamma distribution
(2020) Benlakhdar, Said; Rziza, Mohammed; Oulad Haj Thami, Rachid
Image classification refers to an important process in computer vision. The purpose of this paper is to propose a novel approach named GGD-GMM and based on statistical modeling in the wavelet domain to describe textured images and rely on a number of principles that give its internal coherence and originality. Firstly, we propose arobust algorithm based on the combination of the wavelet transform and Scale Invariant Feature Transform. Secondly, we implement the aforementioned algorithm and fit the result using the finite mixture gamma distribution (GMM). The results, obtained for two benchmark datasets show that the proposed algorithm has a good relevance as it provides higher classification accuracy than some other well-known models (Kohavi, 1995). Moreover, it shows other advantages relied upon Noise-resistant and rotation invariant.
Access status: Open Access ,
A model for changing the technological process for the growth of epitaxial layers by means of the heating of the growth zone
(Wydawnictwa AGH, 2021) Pankratov, Evgenij Leonidovič
The nonstationary transfer of heat during epitaxial layer growth in gas phase epitaxy reactors is analyzed within the work. Based on this analysis, several recommendations on the organization of the heating of the growth zone to increase the homogeneity of the epitaxial layers were formulated. An approach to analyze the transfer of heat during epitaxial layer growth from the gas phase is also introduced. The approach leads to the possibility of simultaneously accounting for heat transfer nonlinearity and changes of parameters of heat transfer in both space and time.
Access status: Open Access ,
A new BEM for modeling and simulation of 3T MDD laser-generated ultrasound stress waves in FGA smart materials
(Wydawnictwa AGH, 2021) Fahmy, Mohamed Abdelsabour
The goal of this study is to present a new theory known as the three-temperature memory-dependent derivative (MDD) of ultrasound stress waves in functionally graded anisotropic (FGA) smart materials. It is extremely difficult to address the difficulties related to this theory analytically due to its severe nonlinearity. As a result, we suggest a new boundary element method (BEM) to solve such equations. The suggested BEM technique incorporates the benefits of both continuous and discrete descriptions. The numerical results are visually represented to demonstrate the impacts of MDD three temperatures and anisotropy on the ultrasound stress waves in FGA smart materials. The numerical findings verify the proposed methodology's validity and accuracy. We may conclude that the offered results are useful for comprehending the FGA smart materials. As a result, our findings contribute to the advancement of the industrial applications of FGA smart materials.
Access status: Open Access ,
A numerical simulation study of mold filling in the injection molding process
(Wydawnictwa AGH, 2021) Baum, Markus; Anders, Denis
Injection molding can undoubtedly be regarded as one of the most widely used manufacturing processes for polymers (Guevara-Morales & Figueroa-Lopez, 2014). Furthermore, injection molding has found its way into various branches of industry (Fernandez et al., 2018) since it has several essential advantages over other processing techniques in terms of good surface finish, the ability to process complex parts without the need for secondary operations, and low cost for mass production. In order to find optimal process settings, it is necessary to gain a deeper insight into the filling process and the underlying physical phenomena, as well as a thorough understanding of the complex material behavior. In this context, the numerical simulation of the injection molding process is increasingly important. Therefore, the current contribution is dedicated to present a thorough comparative numerical study for the mold filling of an exemplary thin-walled mold geometry, including a realistic non-Newtonian viscosity model for the polymer melt. For the numerical simulation, the authors employ the commercial CFD software packages Cadmould 3D-F and ANSYS CFX. While ANSYS CFX is a well-established CFD software for numerical modelling of multiphysical phenomena, Cadmould 3D-F is a highly specialized and computationally efficient alternative suitable for certain geometric configurations in the context of injection molding. The present study is new in the sense that it demonstrates the equivalence of the considered software packages for the simulation of the injection molding process in thin-walled mold geometries.
Access status: Open Access ,
A repeatability study of artificial neural network predictions in flow stress model development for a magnesium alloy
(Wydawnictwa AGH, 2021) Siewior, Hubert; Madej, Łukasz
This work is devoted to an evaluation of the capabilities of artificial neural networks (ANN) in terms of developing a flow stress model for magnesium ZE20. The learning procedure is based on experimental flow-stress data following inverse analysis. Two types of artificial neural networks are investigated: a simple feedforward version and a recursive one. Issues related to the quality of input data and the size of the training dataset are presented and discussed. The work confirms the general ability of feedforward neural networks in flow stress data predictions. It also highlights that slightly better quality predictions are obtained using recursive neural networks.
Access status: Open Access ,
A universal convolutional neural network for the pixel-level detection and monitoring of weld beads
(Wydawnictwa AGH, 2024) Wang, Zhuo; Kayitmazbatir, Metin; Banu, Mihaela
In weld-based manufacturing processes such as welding and metal deposition additive manufacturing (AM), the weld bead is a direct indicator of manufacturing quality. For example, the geometry of the weld bead was optimized to a net shape which outperformed conventional geometries. Automatic monitoring of weld bead is thus of prime importance for welding process control and quality assurance. This paper develops a general-purpose convolutional neural network (CNN) for pixel-level detection and monitoring of beads, regardless of welding materials, machine, manufacturing conditions, etc. To achieve the generality, we collected a great variety of welding images containing 2677 single-line beads from 231 research articles, followed by pixel-wise hand-annotation. Consequently, the trained CNN can recognize different beads from various backgrounds at a pixel level. Case studies show that compared to the image-level classification in prior research, its pixel-level labeling permits real-time, complete characterization of weld beads (e.g., detailed morphology, discontinuity, spatter, and uniformity) for more informed process control. This research represents a significant step towards developing a truly human-like monitoring system with low-level scene understanding ability and general applicability.
Access status: Open Access ,
Accounting for the random character of nucleation in the modelling of phase transformations in steels
(Wydawnictwa AGH, 2023) Poloczek, Łukasz; Kuziak, Roman; Foryś, Jakub; Szeliga, Danuta; Pietrzyk, Maciej
In our earlier work, a stochastic model of multi-stage deformation at elevated temperatures was developed. The model was applied to calculate histograms of dislocation density and grain size at the onset of phase transformation. The histograms were used as input data for the simulation of phase transitions using the traditional deterministic model. Following this approach, microstructural inhomogeneity was predicted for different cooling conditions. The results obtained, showing the effect of dislocation density and inhomogeneity of austenite grain size on the microstructural inhomogeneity of the final product, can be considered reliable as they are based on material models determined in previous publications and validated experimentally. The aim of the present work was to extend the model by taking into account the stochastic nature of nucleation during phase transitions. The analysis of existing stochastic models of nucleation was performed, and a model for ferritic transformation in steels was proposed. Simulations for constant cooling rates as well as for industrial cooling processes of steel rods were performed. In the latter case, uncertainties in defining the boundary conditions and segregation of elements were also considered. The reduction of the computing costs is an important advantage of the model, which is much faster when compared to full field models with explicit microstructure representation.
Access status: Open Access ,
Aircraft wing structural design application in MATLAB App Designer
(Wydawnictwa AGH, 2021) Bilal, Ahmed; Siddiqui, Muhammad Faisal; Abbas, Messam; Mansoor, Mohtashim
This study focuses on developing an automated application in the MATLAB® App Designer module, based on basic structural members and different theories for various loading cases, providing ballpark values of bending and torsional stiffness and sizing of the load-carrying structural member at a span wise location. All the code developed on MathWorks (2018) is automated using the App Designer module. In this approach, governing equations of structural members under different types of loading are solved in MATLAB IDE with the assumption that in the preliminary phase MATLAB App Designer provides an easy drag and drop type application developer that can be easily subsumed in any mathematical automation process.
Access status: Open Access ,
An accuracy analysis of the cascaded lattice Boltzmann method for the 1D advection-diffusion equation
(Wydawnictwa AGH, 2020) Straka, Robert; Sharma, Keerti Vardhan
We analyze higher order error terms in a modified partial differential equation of a cascaded lattice Boltzmann method (CLBM) for one conservation law - the advection-diffusion equation. To inspect the behavior of the error terms we derived an equivalent finite difference equation (EFDE), this approach is different from other techniques like the Chapman-Engskog expansion, equivalent partial differential equations or the Maxwell iteration used in the literature. The resulting EFDE is obtained from the recurrence formulas of the lattice Boltzmann equations for the CLBM and is subsequently analyzed by standard analytical techniques. We have found relations of the LBM parameters which could cancel some of the higher order terms, making the method more accurate. The detailed derivation of the EFDE and higher order terms' pre-factors is the main result of this paper. The resulting explicit form of the error terms are derived and presented.
Access status: Open Access ,
An analytical model for the tool center point placement in Robotic Roller Forming
(Wydawnictwa AGH, 2024) Stewens, Thomas; Liu, Yi; Wang, Ling; Min, Junying
Robotic Roller Forming (RRF) is a novel process using an articulated robotic manipulator that can bend Ultra-High Strength materials into thin-walled profiles. For high strength or difficult-to-form sheet materials, a laser can be employed to synchronously heat and soften the local material during RRF. The aim of RRF is to establish itself as a highly flexible process for rapid prototyping as well as for small batch production. However, in finished parts formed with different materials, a new defect that shapes the profile like that of a hook was observed. To overcome this defect and to improve the adaptability of the process, a new analytical model is suggested for the automatic calculation of the tool center point based on the given process parameters. The model was compared to the previous state, where the hook defect was noticeably reduced. Additionally, the control of the bend radius was studied, and the resulting bend radius diverged from the target radius by 0.04 mm (2.45%). Further, when examining the reproducibility, the same bend angles could be achieved as in previous experiments using the constant laser power density. Finally, the development of the bend allowance was studied in various experiments. The analytical model for RRF is a promising method for calculating tool placement and controlling the bend radius in a freeform environment.
Access status: Open Access ,
An efficient Monte Carlo Potts method for the grain growth simulation of single-phase systems
(2020) Maazi, Noureddine; Lezzar, Balahouane
The choice of the lattice sites to be reoriented in the Monte Carlo Potts algorithm for grain growth simulation is repeated in a non-homogeneous way. Therefore, some grains are favorably growing than others. This fact may seriously affect the simulationresults. Soa modified MC method is presented. Lattice sites are selected for reorientation one by one following their positions in the matrix in each Monte Carlo step (mcs). This approach ensures that the various selections of one lattice site within every mcs are eliminated, and no favorable growth of grains at the expense of others. The calculation time considerably decreases. The effect of real-time and physical temperature on the grain growth kinetics is discussed.
Access status: Open Access ,
An evaluation of discrepancies between CPFE simulations and mean-field approximations for dual phase materials
(Wydawnictwa AGH, 2025) Mirhosseini, Shahrzad; Atzema, Eisso H.; van den Boogaard, Antonius H.
This paper explores the discrepancies observed between 2D and 3D crystal plasticity finite element (CPFE) simulations and mean-field approximations in terms of macroscopic flow curves. Two hypotheses are proposed to address the discrepancies: (1) the type of yield function in the mean-field approach, and (2) differences in stress states between the two methodologies. Based on the first hypothesis, the type of yield function may influence the stress-strain partitioning in the mean-field approach. Consequently, the von Mises criterion is replaced with the Hershey yield function. To test the second hypothesis, CPFE simulations are extended to 3D to achieve comparable stress states in both methods. This analysis reveals that the exact shape of the yield function has a marginal impact on the discrepancies, whereas the proper 3D stress distribution significantly reduces them. This comprehensive study also uncovers a limitation of the mean-field approach in terms of accuracy in the prediction of macroscopic material response and stress partitioning for a two-phase polycrystalline material.
Access status: Open Access ,
An evaluation of the capabilities of image-based metal component defect recognition with deep learning techniques
(Wydawnictwa AGH, 2024) Wójcik, Michał P.; Pawlikowski, Kacper; Madej, Łukasz
In the era of Industry 4.0, deploying highly specialised machine learning models trained on unique and often scarce datasets is an attractive solution for advancing automated quality control and minimising production costs. Therefore, the main aim of this research is to evaluate the capabilities of three deep learning models (ResNet-18, ResNet-50 and SE-ResNeXt-101 (32 × 4d)) in the identification of surface defects in forged products. Leveraging advanced photography techniques, including studio lighting and a shadowless box, high-quality images of complex product surfaces were acquired for the training data set. Given the relatively small size of the image dataset, aggressive data augmentation techniques were introduced during the training and evaluation process to ensure robust model generalisation ability. The results obtained demonstrate the significant impact of data augmentation on model performance, highlighting its importance in training and evaluating deep learning models with limited data. This research also emphasises the need for innovative data pre-processing strategies in an efficient and robust machine learning model delivery to the industrial environment.
Access status: Open Access ,
An evaluation of the mechanical properties of 13MnSiCr7 steel by digital image correlation
(Wydawnictwa AGH, 2021) Kempny, Marcin; Rozmus, Radosław
"In this paper, the possibility of replacing tensile extensometers with a non-contacting device for measuring elongation has been analyzed. An example of a non-contacting device is a Digital Image Correlation System (DIC). Such systems are widely used in various areas, for example, biology or modern engineering. DIC systems have several advantages that seem to be promising for testing modern materials. The most important is the fact that there is no physical contact between the sample and the DIC and therefore no additional force is applied during the experiment. On the other hand, a lack of contact with the sample can cause large measurement inaccuracies. Another advantage would be that a DIC can measure strain on the whole surface of the sample in all directions, instead of measuring part of the surface in one direction like in other extensometers. Because of these abilities, the environment impact on test bench (DIC + load device), and differences between conducted experiment with normalized tensile test needed to be investigated. The testing machine was replaced by a DIC system cooperating with a tension-compression module. The proposed method was used to monitor and record the images to determine the basic properties of 13MnSiCr7 grade steel. Twelve tests were performed. The analysis was done by comparing the values of mechanical properties obtained in a static tensile test, such as yield strength, tensile strength, Young's modulus, elongation of the material"," with the values of these properties determined experimentally. For each sample, stress-strain curves were evaluated. To check if the results were correct, a Q-Dixon test was performed in each case, confidence intervals were also calculated. Finally, the obtained properties were compared with those from the standard tensile test acquired from the manufacturer's material card."
Access status: Open Access ,
Analysis of modification of the evolutionary algorithm for sequencing production tasks
(Wydawnictwa AGH, 2022) Ciepliński, Piotr; Golak, Sławomir; Wieczorek, Tadeusz
Evolutionary algorithms are one of the heuristic techniques used to solve task sequencing problems. An important example of such a problem is the issue of sequencing production tasks. The combinatorial optimization of task sequences allows the minimization of the cost or time of a set of production tasks by reducing the components of these values which are present in the transitions between tasks. This paper aims to analyze the influence of the production nature expressed by a set of production task parameters and a definition of the task transition cost on the effectiveness of the modification of the evolutionary algorithm based on new directed stochastic mutation operators. The research carried out included the influence of the space dimension of the task parameters, the number of levels of the value of the cost function, and a definition of this function. The results obtained allow us to assess the effectiveness of the directed mutation in task sequencing for productions of various natures.
Access status: Open Access ,
Analysis of the decay time and bound-states energies of a particle in a specific structure GaMnAs/GaAs quantum well
(Wydawnictwa AGH, 2025) Ali, Alaa Y.; Ali, Hassan H.; Ali, Mustafa Y.
The bound states and decay time in a certain quantum well structure (GaMnAs/GaAs) were analysed and identified at the minimum decay time. Through the analysis of quantum mathematical equations, we derived specific formulas for energies that significantly amplify the numerical solutions of equations throughout all dimensions of confinement. Without altering the parameters utilized, the quantification, barriers, and well width were predominantly influenced by the spatial dimension parameters, such as the barrier height and well width. The principal bound state and lowest decay time were determined at a well width of 40 Å and a barrier thickness of 46.27 Å. This work revealed a novel characteristic known as interfacial tunnelling, which refers to the phenomenon where an electron establishes a tunnelling state between two interfaces. This tunnelling process is significantly influenced by the characteristics of the materials used, as well as the dimensions of the wells and barriers.