Mariia Shapovalova

Ms. Mariia Shapovalova

orcid

Research Focus & Metrics

Academic Career Length

8Years (2019-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
1 July 2019
Verified
Conference Paper
Closed access
MSMariia Shapovalova
Mariia Shapovalova
OVOleksii Vodka
Oleksii Vodka
The search for new methods of materials quality control, based on the study of their microstructure, is one of the urgent tasks in the modern world. Development in the field of information technologies and machine learning allowed analyzing the images obtained during a nondestructive control method. In this work the architecture of convolution neural network for image analysis of steel microstructure has been offered. The network gives the predicted composition of heterogeneous materials. Training and testing of the neural network operability have passed on images set of steel microstructures. The system predicted the percentage of chemical elements entering the studied metal. Good results with accuracy of more than 83% are received. Such results demonstrate the expediency of convolutional neural networks application to study the materials microstructure.

3 January 2020
Verified
Book Chapter
Closed access
PDF file
AKAndrii Kelin
Andrii Kelin
OLOleksiy Larin
Oleksiy Larin
RNRaisa Naryzhna
Raisa Naryzhna
OTOleksandr Trubayev
Oleksandr Trubayev
Oleksii VodkaOleksii Vodka
Oleksii Vodka
MSMariia Shapovalova
Mariia Shapovalova

1 September 2019
Verified
Conference Paper
Closed access
AKAndrii Kelin
Andrii Kelin
OLOleksiy Larin
Oleksiy Larin
RNRaisa Naryzhna
Raisa Naryzhna
OTOleksandr Trubayev
Oleksandr Trubayev
OVOleksii Vodka
Oleksii Vodka
MSMariia Shapovalova
Mariia Shapovalova
With long-term operation of pumping and other equipment, there is often a need to extend the life and assess the residual life. Modern life-time assessment methods are based on three-dimensional modelling and the finite element method. Within the framework of these methods, geometric models are constructed that take into account the thinning of the walls of the pumps as a result of operation. By constructing geometric models, finite element models are constructed. These models are made taking into account the conditions of loading of the structure, tightening of bolted connections, technological and temperature loads. The constructed models allow to assess the strength of the pump casing and the main bolted connections during normal operation and gyro testing. The paper assesses the cyclic strength of these structural elements. The results of the work confirmed a sufficient residual life-time for safety pump operation in the next 15 years.

1 January 2023 in Computer Methods in Material Science
Verified
Journal Article
Open access
Oleksii VodkaOleksii Vodka
Oleksii Vodka
MSMariia Shapovalova
Mariia Shapovalova
This paper delves into the expansive world of cellular automata (CA), abstract models of computation comprised of cells that interact based on predefined rules. Originating from John von Neumann’s work in the 1940s, CA has evolved into a multidisciplinary field with applications ranging from mathematical concepts to complex simulations of biological, physical, computer science, material science, and social systems. The paper reviews its historical development, emphasizing John Conway’s influential Game of Life and Burk’s seminar collection. The authors categorize and explore a myriad of CA topics, including self-replicating automata, the universality of computation, compromises in CA, variants, applications in biological systems, fault-tolerant computation, pattern recognition, CA games, fractals, dynamic properties, complexity, image processing, cryptography, bioinformatics, materials modeling, probabilistic automata, and contemporary research. The significance of cellular automata for materials modeling cannot be overstated and considerable attention has been devoted to the issues of modeling nucleation and recrystallization. The review aims to provide a comprehensive resource for both beginners and experts in the field, shedding light on cellular automata’s dynamic and diverse applications in various aspects of life and scientific inquiry.


13 September 2021
Verified
Conference Paper
Closed access
MSMariia Shapovalova
Mariia Shapovalova
OVOleksii Vodka
Oleksii Vodka
Automated intelligent decision-making systems, working with the use of mathematical methods of data processing, can reduce the influence of the human factor in the analysis, reduce the time spent on research, improve the accuracy and reliability of the results. They help automate the quality control process and enable association material properties to its microstructure. Numerical and experimental studies of the material microstructure can be implemented using hybrid methods with the introduction of computer simulation methods. The paper proposes to use the OpenCV technology for image analysis of cast iron with spherical graphite inclusions, followed by classification of the recognized material. The optimal parameters of image preprocessing and edge recognition are set. The result of the presented work is the distribution function of inclusions depending on their concentration on the plane.