2nd International Scientific and Technical Conference 'Infocommunication Systems and Artificial Intelligence Technologies' (ICSAIT 2024)
December 4-5, 2024, AzTU, Baku

Selected peer-reviewed, extended articles based on abstracts presented at the 2nd International Scientific

and Technical Conference “Infocommunication Systems and Artificial Intelligence Technologies” (ICSAIT 2024)

 

 

AXIOMATIC MODEL FOR ENSURING STABILITY OF COMPLEX INTELLIGENT SYSTEMS
Mehman Hasanov, Vitaly Evdokimov, Anatoliy Navrotskiy, Hasan Najafov, Ilham Suleymanov

https://doi.org/10.24412/1932-2321-2025-783-13-21

This paper presents a review analysis of an axiomatic model aimed at ensuring stability, integration, and adaptability of complex intelligent systems within the context of Industry 4.0 and Industry 5.0 technologies. The model encompasses 15 axioms that describe key principles for the development and synthesis of independent subsystems into a unified ecosystem. Special attention is paid to the issue of balanced interaction between adaptability, resilience, resource optimization, and the capacity for evolution and scaling, which is crucial for stable functioning and long-term development of smart city infrastructure and Industry 4.0 and 5.0 systems in general. The article is intended for researchers and specialists in the fields of systems analysis, telecommunications, and control, as well as those involved in the design and implementation of complex adaptive intelligent systems.

MODELS OF CONTROL SYSTEMS WITH PARAMETRIC AND ADDITIVE-PARAMETRIC FEEDBACK
Sergey Yablochnikov, Tofig Mansurov, Irina Yablochnikova, Rahman Mammadov

https://doi.org/10.24412/1932-2321-2025-783-22-28 

For processes implemented in object management systems of various nature, quite often the input signal can be formalized as one of a limited set of some, so to speak, standard functions. In particular, such a function may correspond to an increasing exponential. The functioning of control systems for these processes should be ensured in such a way that the reaction of the system itself to such highly probable signals is adequate and successfully formed in accordance with the criterion of minimum error. One of the ways to solve this problem is to form a parametric or additive–parametric feedback circuit in the control system, as well as integrate components with parameters changing according to a certain law into the structure of the system. At the same time, the error should not depend on the magnitude of the parameters noted above and, in fact, is minimized. The authors synthesized and substantiated models and algorithms for the functioning of control systems with parametric feedback by transmission coefficient, which make it possible to eliminate control errors and ensure the noise immunity of the system. It is also proposed to use other means that ensure the high efficiency of such systems that implement the management of objects and processes of various types.

 

FORMATION OF AN INFORMATION LEAKAGE CHANNEL FROM AN OPTICAL FIBER BY THERMAL ACTION
Andrey Zenevich, Tatiana Matkovskaia, Nurana Camalzadeh, Javid Namazov

https://doi.org/10.24412/1932-2321-2025-783-29-35 

The article evaluates the possibility of forming a channel for information leakage from a defect in optical fiber created by thermal action. The properties of optical fiber inhomogeneities caused by such action are currently practically unstudied, which determines the relevance of the research. The paper shows that local temperature action makes it possible to form a defect in optical fiber that allows part of the optical radiation to be removed beyond this fiber, that is, to create a channel for unauthorized data retrieval. It is shown that with an increase in the wavelength of optical radiation propagating along the fiber, the loss of radiation power on the defect formed by thermal action on the optical fiber increases. It is found that with the same loss of power on the defect formed by thermal action, the optical radiation power removed from such a defect has the greatest value when using G.652 optical fiber, and the least when using G.657 fiber. The results presented in the article can be used in designing systems for protecting information transmitted over fiber-optic communication lines.

COMPARATIVE ANALYSIS OF LANET AND FE-NET IN IMAGE SEGMENTATION
Zhao Di, Alevtina Gourinovitch, Shadiye Sultanova, Baloghlan Najafov, Gizilgul Israfilova

https://doi.org/10.24412/1932-2321-2025-783-36-43 

Image segmentation in infocommunications helps analyze infrastructure, enhance video quality, manage networks to improve efficiency and service quality. Biomedical image segmentation is very important for the medical imaging. It requires precise delineation of anatomical structures and pathological regions for various diagnostic tasks (for example, tumor detection and treatment planning) in improving clinical outcomes. In recent decades, the automatic medical segmentation methods which are based on deep learning (DL) models, such as convolutional neural network (CNN) architectures, have facilitated the automatic delineation of organ and lesion boundaries. This changed traditional manual segmentation approaches and improved efficiency and accuracy in clinical practice. This paper presents a comparative analysis of two DL-based models: LANet and FE-Net. LANet employs an Efficient Fusion Attention (EFA) module and an Adaptive Feature Fusion Decoder (AFF) module to improve segmentation efficiency and precision. In contrast, FE-Net integrates a Feature Awareness Module (FAM) for enhances its features to capture multi-scale and process segmentation details. The advantages of both methods in handling different scales, details, and edges were investigated and experimentally evaluated on various public datasets, and their performance in specific scenarios was assessed. The experiment showed that LANet is superior in computational efficiency and feature refinement. FE-Net shows superior performance in handling complex variations and edge details. The source code of LANet and FE-Net can be found on GitHub at [https://github.com/tyjcbzd/LANet] and [https://github.com/tyjcbzd/FE-Net].
 

TWO-LAYER CHARCOAL-CONTAINING MICROWAVE ABSORBERS WITH A RELIEF SURFACE FOR SERVER EQUIPMENT PROTECTION FROM INTERFERENCE
Olga Boiprav, Vadim Bogush, Mehman Hasanov, Vyacheslav Mokerov, Elena Belousova

https://doi.org/10.24412/1932-2321-2025-783-44-50 

The article presents the electromagnetic radiation reflection and transmission characteristics in the frequency range of 0.7–17.0 GHz of multilayer microwave absorbers. These absorbers consist of modules with a relief surface made of a mixture of powdered activated charcoal and a binder (polyvinyl acetate dispersion aqueous solution or polyurethane mastic). According to the presented characteristics, electromagnetic radiation reflection and transmission coefficients values in the frequency range of 0.7–17.0 GHz of the specified absorbers vary, respectively, within the limits from –2.0 to –18.0 dB and from –10.0 to –40.0 dB. The studied absorbers can be used for wall cladding or creating internal partitions in server rooms. Using such absorbers, it is possible to solve the practical problem of ensuring electromagnetic compatibility of server equipment and other information processing equipment.

MODELING THE TOPOLOGY OF FSON USING DIJKSTRA'S ALGORITHM
Agil Movsumov, Shadiya Sultanova, Emin Payizov, Turana Rasullu, Nurana Camalzadeh, Javid Namazov, Mala Dutta

https://doi.org/10.24412/1932-2321-2025-783-51-57 

This paper focuses on modeling the topology of Free Space Optical Networks (FSON) using Dijkstra's algorithm, demonstrating its potential for efficient and reliable data transmission in distributed communication systems. The work highlights the integration of Li-Fi technology and graph theory to optimize routing and minimize time costs, ensuring adaptability and scalability for real-world applications. By addressinFg challenges like interference and line-of-sight constraints, the proposed methodology enhances network performance for smart cities, IoT systems, and space communications. The results confirm the effectiveness of the approach in improving bandwidth efficiency, reducing delays, and dynamically adapting to changing network conditions.

COMPREHENSIVE EVALUATION AND PERFORMANCE ANALYSIS OF A DEEP LEARNING MODEL WITH HYPERPARAMETER TUNING FOR LUMPY SKIN DISEASE CLASSIFICATION IN DAIRY COWS
Gunikhan Sonowal, Soraisam Gobinkumar Singh, Prasanta Bairagi, Utpal Barman, Dulumani Das, Mammadov Iltimas, Gulnar Gurbanova

https://doi.org/10.24412/1932-2321-2025-783-58-65 

This work attempts to classify lumpy skin conditions using CNN and hyperparameter tuning. This model is comprised of many procedures, including selecting a pre-trained model, altering the architecture, and training the model on a specific dataset. During tweaking, the proposed model attained a validation accuracy of 89.73 percent. The model’s generalisation performance was confirmed with an accuracy of 80.68% in the final test set evaluation. It significantly increased the timeliness of LSD identification, making it a valuable tool for farmers and veterinarians. Furthermore, a Receiver Operating Characteristic (ROC) curve with an Area Under the Curve (AUC) of 0.88 indicates that our binary classifier performed satisfactorily.

 

DEVELOPMENT PROSPECTS AND MATHEMATICAL SOLUTION METHODS FOR INTEGRATING BEACON SYSTEMS INTO UAVS
Elshan Hashimov, Elkhan Sabziev, Samad Muradov

https://doi.org/10.24412/1932-2321-2025-783-66-73 

The paper describes radio beacon systems and presents a mathematical solution for locating an unmanned aerial vehicle (UAV) equipped with a direction-finding device. This system provides continuous flight and allows you to determine the exact coordinates of targets regardless of satellite signals. It enables the successful execution of combat missions in adverse weather conditions and when using radio electronic warfare systems based on signals received from radio beacons. Thus, due to the recent development of unmanned aerial vehicles, many countries are improving their radio electronic warfare systems. Because in a real war, destroying cheap UAVs with expensive Air Defensive Systems missiles does not benefit any country financially. Radio electronic warfare systems are being developed rapidly because they are more effective in this respect. Taking this into account, it is possible to increase the resistance against radio electronic warfare systems by developing the issue of integration of the beacon systems into the UAV proposed in the article. Thus, by setting up beacon systems, it is possible to perform a UAV flight during radio electronic warfare application in any conditions.

IMPROVING THE PERFORMANCE OF ASSOCIATION RULES HIDING USING HYBRID PARTICLE SWARM OPTIMIZATION ALGORITHM
Eirene Barua, Mala Dutta ,Zafar Jafarov

https://doi.org/10.24412/1932-2321-2025-783-74-80 

In today’s digitized world, data can be taken from many sources like e-marketing sites, social platforms, social networking sites etc. in bulk volumes for usage. Privacy Preserving is a very delicate issue to be looked upon. Hence it becomes necessary to focus on the important privacy preserving parameters. Algorithms for optimization plays an important role in reducing non-sensitive rules in association rule hiding. This paper speaks about a hybrid Particle Swarm Optimization algorithm that requires the properties of all the algorithms which are used for hiding Association Rules and it also highlights the usage of less time.

RESEARCH SPECTRAL EFFICIENCY OF NEW MODULATION FORMATS IN FIBER-OPTIC NETWORKS
Bayram Ibrahimov, Asmar Nabieva, Aygun Hamidova

https://doi.org/10.24412/1932-2321-2025-783-81-89 

The rapid growth of the use optical technologies requires the development of methods and means to improve the spectral efficiency and noise immunity of fiber-optic networks when using wavelength multiplexing. On the basis of the research a new approach to the construction of the method of calculation spectral efficiency indices of new modulation formats in fiber-optic networks has been developed. The proposed method calculation of indicators takes into account the efficiency indicators fiber-optic transmission systems, the algorithm demodulator synthesis and effective methods modulation formats such as M-ary Quadrature Amplitude Modulation, Differential Phase Shift Keying and M-ary pulse position modulation. On the basis of the calculation method, important analytical expressions evaluating the characteristics of line capacity and noise immunity of optical signal reception are obtained.

RESEARCH AND ANALYSIS QOS AND QOE INDICATORS IN MULTISERVICE TELECOMMUNICATION NETWORKS
Vyacheslav Shuvalov, Sevinc Ismayilova, Samina Rustamova

https://doi.org/10.24412/1932-2321-2025-783-90-98 

The rapid evolution multiservice telecommunication networks and the increasing demand for multimedia services have heightened the importance of ensuring Quality of Service (QoS) and Quality of Experience (QoE) for user satisfaction. This study explores QoS and QoE metrics in multiservice networks, leveraging insights from the "Network 2030" project, advanced digital technologies like SDN, NFV, 5G, and emerging 6G frameworks. With a focus on modern, end-to-end digital architectures, the paper proposes a mathematical model for analyzing and optimizing performance indicators, considering parameters such as network throughput, reliability, delay, and packet loss. The model aims to support real-time QoE assessments and enhance service quality by identifying performance thresholds across applications, from voice and video to M2M traffic. Metrics are examined through ITU-T recommendations, incorporating probabilistic-temporal characteristics and subjective user perceptions measured by the MOS (Mean Opinion Score) and R-factor ratings. Findings suggest that dynamic QoE monitoring, aligned with the convergence of NGN and FN architectures, offers significant potential to address rising user expectations in content quality and service reliability, guiding future research toward efficient QoS and QoE assessment in multiservice environments.

SECURITY CONTROL SYSTEM FOR INFORMATION EXCHANGE IN TELECOMMUNICATION NETWORKS
Elvin Abaszade, Zafar Ismayilov, Almaz Mehdiyeva, Huseyn Qasimov

https://doi.org/10.24412/1932-2321-2025-783-99-105 

The principles of information security and the problems faced were investigated. Cyber security issues are explored. Also, technical requirements for ensuring information security and means ensuring the security of Information systems were studied. Based on Visual Studio and MySQL programs, the issue of web page security and user data protection has been resolved. In this application, Users can access their personal cabinet by typing username and password on the web page. If a user is not registered on this page, then that user will encounter a problem, which means that the database does not have that user's information. The proposed method ensures security.

ELECTRICAL DISCHARGE IN DIELECTRIC STRUCTURES
Hikmet Aliyev, Hikmet Fattayev, Naib Hajiyev

https://doi.org/10.24412/1932-2321-2025-783-106-111 

Based on investigations on partial discharges under different physical conditions and when using different dielectric barriers, it was found that the discharge in the gas gap confined by dielectrics has a pulse character under all conditions. The discharge current is shown to consist of discrete pulses whose shape and frequency depend on the dielectric parameters, the type of gas in the gap and the pressure. One of the main reasons for the deterioration of the electrophysical properties of dielectrics is the development of electric discharges in dielectric structures consisting of dielectrics with significantly different properties. Therefore, the study of the influence of the properties of dielectrics and the electrophysical parameters of these dielectrics on the mechanism of development of the discharge under conditions of electric discharge is an urgent issue [1-3].
 

METHOD FOR INCREASING THE ACCURACY OF INFORMATION EXCHANGE IN COMMUNICATION NETWORKS
Gunay Khasmammadova, Sevinj Bakhshaliyeva, Javanshir Zeynalov

https://doi.org/10.24412/1932-2321-2025-783-112-119 

The presented article is devoted to the issue reducing losses in the information exchange channel. Satisfactory results have been obtained from the conducted research: calculations show that the implementation of the evolution of wireless communication network control center increases the performance of the network. It is based on G technology and its generations. Research shows that the LTE (4G) architecture is the way to improve mobile devices and data based on GSM/EDGE and UMTS/HSPA standards in the field of telecommunications. Ensuring the accuracy of information in wireless networks was thoroughly investigated, and for this, corrective filtering was proposed for receiving data in the processing process, and a positive result was obtained without filtering. Modeling of LTE was carried out in the OPNET modeling software package. During the research, simulation modeling was carried out in the Matlab environment, and satisfactory results were obtained.

A NEW APPROACH TO NUMERICAL CALCULATION OF NON-STATIC PROCESSES IN COMPLEX MAIN GAS PIPELINES

Cherkez Yusubov, Habib Abbasov, Ankur Pan Saikia

https://doi.org/10.24412/1932-2321-2025-783-120-126 

The indicators of the main gas pipeline complexes are analyzed and studied, and a new approach to the numerical calculation of non-static processes in the system is proposed. Issues of unsteady movement liquid and gas in pipes are significant importance for both the design and operation pipelines. Based on the new approach, a method for calculating the characteristics of non-static processes in main gas pipeline complexes has been constructed. Based on the calculation methods, some important analytical expressions for assessing the performance of main gas pipeline complexes have been obtained.

DETECTION OF SMALL-SCALE UNMANNED AERIAL VEHICLES USING ACOUSTIC RECONNAISSANCE TOOLS
Aliqismat Mehdiyev, Seriyye Qasımova

https://doi.org/10.24412/1932-2321-2025-783-127-133 

This paper reviews the methods and tools for detecting small-scale unmanned aerial vehicles (UAVs) by acoustic reconnaissance tools. Information is provided on the methods for detecting small UAVs using acoustic waves across various types of sound ranges.

EXPLORING THE IMPACT OF AI ON PRIVACY AND ETHICAL CONSIDERATIONS: ANALYSING THE LEGAL AND REGULATORY FRAMEWORKS

Ankur Pan Saikia, Ananya Kalita, Parvana Movsumova

https://doi.org/10.24412/1932-2321-2025-783-134-147 

The rapid advancement of artificial intelligence (AI) has brought about significant implications for privacy and ethical considerations. This research paper aims to explore the impact of AI on privacy and ethical concerns by analysing the existing legal and regulatory frameworks. The paper reviews relevant literature, research papers, case studies, and laws to identify key concepts, theories, and gaps in the current understanding of AI's impact on privacy. Additionally, it examines the strengths and weaknesses of existing legal frameworks and regulations related to AI and privacy. The analysis reveals that AI poses challenges to personal data privacy, including discrimination, privacy breaches, biased decision-making, and lack of transparency. It underscores the need for stronger data protection laws, algorithmic fairness, and transparency in AI systems. Furthermore, the paper discusses the ethical implications of AI in various contexts, such as healthcare, surveillance, and hiring processes. Based on the findings, the research paper proposes a strategic framework to enhance the legal and regulatory frameworks for AI and privacy. The framework emphasizes stakeholder engagement, ethical principles, data protection, algorithmic transparency, industry accountability, international collaboration, and public awareness. Moreover, the paper provides recommendations for policymakers, industry stakeholders, and researchers to guide their actions in addressing the legal, ethical, and privacy challenges posed by AI. In conclusion, this research paper highlights the urgent need to strengthen legal and regulatory frameworks to address the evolving impact of AI on privacy and ethical considerations. By adopting the proposed strategic framework and implementing the recommendations, stakeholders can work towards a responsible and privacy-conscious AI ecosystem that balances innovation with individual rights and societal well-being.

 

EVALUATING THE PREDICTION OF COPD USING DATA ANALYSIS AND ENSEMBLE MACHINE LEARNING TECHNIQUES
Arpita Nath Boruah, Mrinal Goswami, Elchin Rzayev

https://doi.org/10.24412/1932-2321-2025-783-148-154 

Chronic Obstructive Pulmonary Disease (COPD) is a progressive and debilitating respiratory condition characterized by persistent airflow limitation, typically associated with chronic bronchitis and emphysema. COPD represents a significant global health burden, affecting millions of individuals worldwide, with increasing prevalence and mortality rates. The primary risk factor for COPD is tobacco smoking, although other factors such as occupational exposure to pollutants, genetic predisposition, and respiratory infections also contribute to its development. Chronic inflammation, oxidative stress, and protease-antiprotease imbalance play pivotal roles in the pathogenesis of COPD, leading to structural changes in the airways and alveoli, progressive airflow limitation, and impaired gas exchange. In recent years, there has been growing interest in applying Machine Learning (ML) techniques to various aspects of COPD management, including diagnosis, prognosis, treatment optimization, and exacerbation prediction. So also data analysis plays an important part in the performance the ML techniques. This work investigates the performance of different machine learning classifiers used in COPD prediction, especially in single and ensemble classification. A detailed performance comparison among all the classifiers is also done, considering accuracy, precision, recall, and F1 score.

MODERN ECOLOGICAL STATE OF THE SOILS OF ABSHERON PENINSULA AND WAYS SOLUTIONS
Leyla Ibrahimova

https://doi.org/10.24412/1932-2321-2025-783-155-161 

The interaction between ecosystems and people began from the very beginning of their existence. Over time, this process began to enlarge. The development of science and the emergence of new inventions in technology have led to the further intensification and gradual sharpening of the interaction. The expansion and intensification various sectors of the economy, the constant increase in the number of people, the emergence of megacities, etc. have also had an impact on the geographical environment. It has revealed difficult to solve ecological problems for the environment surrounding us and has set before us the task of their restoration and re-circulation. The increasing scope of the various ecological problems that have arisen has made their solution inevitable, which has become one of the topical issues of our time. Currently, the development of methods for assessing anthropogenic impact is of particular importance for all components of the environment - soil, vegetation, fauna, water bodies, air, etc. Problems related to the need to monitor the real situation with anthropogenic pollution of soils necessitate the application of new approaches in the fight against the threat toxicants, in addition to chemical analysis. It is necessary to assess the integral toxicity of the soil, reflecting the impact of a complex of all factors. In terms technogenic pollution of soils in our republic, the Absheron Peninsula stands out in particular, and therefore the presented article stands out with its relevance.

ARTIFICIAL INTELLIGENCE FOR AUTOMATED ENERGY LOSS SEARCH
Sergey Bogatenkov, Dmitry Bogatenkov, Elkhan Mammadov, Konul Shammadova, Gunay Dadashova, Eirene Barua

https://doi.org/10.24412/1932-2321-2025-783-162-168 

Automated search for energy losses is difficult due to the lack of artificial intelligence methods to minimize the risks associated with staff errors. The aim of the work is to develop an artificial intelligence method for automated search for energy losses. Based on the application of the heuristic method of artificial intelligence and digital twin technologies, algorithms for automated energy loss search have been developed. The results of the study were implemented using the complex of technical means (CTM) "Energy" using examples of searching for losses of electricity and energy carriers at energy-intensive enterprises in Russia and Azerbaijan. The application of the research results in automated energy metering systems minimizes the risks associated with personnel errors.

PERFORMANCE COMPARISON OF K-MEANS, PARALLEL K-MEANS AND K-MEANS++
Ramiz Aliguliyev, Shalala F. Tahirzada

https://doi.org/10.24412/1932-2321-2025-783-169-176 

K-means clustering is a fundamental unsupervised machine learning technique widely applied in various domains such as data analysis, pattern recognition, and clustering-based tasks. However, its efficiency and scalability can be challenged, particularly when dealing with large-scale datasets and complex data structures. This thesis explores strategies to improve the performance of the K-means clustering algorithm through parallelism and iterative techniques. Parallelism leverages modern parallel computing architectures, including multi-core processors and distributed frameworks like Apache Spark, to enhance computational efficiency and scalability. On the other hand, an iterative approach involves refining clustering results through multiple iterations, adjusting cluster centroids, and optimizing convergence criteria. It delves into the design frameworks of these approaches, highlighting their respective advantages and limitations. Comparative analyses are conducted to evaluate the effectiveness of parallelism and iterative techniques in terms of execution time, scalability, clustering accuracy, and convergence speed. The findings contribute to advancing the understanding of how parallelism and iterative strategies can significantly improve K-means clustering performance, especially in the context of big data and complex datasets. By comparatively analyzing parallelism and iterative approaches, this paper aims to contribute to the development of more efficient and scalable clustering algorithms in the Big Data context.

EXPLORING BIG DATA CLUSTERING: APPROACHES, ALGORITHMS, AND PLATFORMS
Ramiz Aliguliyev, Tural Badalov

https://doi.org/10.24412/1932-2321-2025-783-177-187 

Clustering as the problem of discovering natural grouping in data has gotten a lot of attention due to its wide range of applications in health care, customer segmentation, image processing & transformation, market and recommendation systems, social network analysis, etc. It is an unsupervised learning task used to discover similar objects in a large dataset without relying on any prior information and gathering them into the same group. With the rapid growth of big data as result of data sets acquired by mobile devises, cameras, various sensors and other sources has necessitated research into extracting valuable information from enormous data sets. In this paper, we looked at different big data clustering approaches in the context of general clustering methods. In addition, we discussed several similarity measures as well as key clustering challenges such as cluster tendency assessment and cluster validity.

ARCHITECTURAL-TECHNOLOGICAL PRINCIPLES OF A PATIENT CENTERED DIGITAL TWIN AND ITS VISUALIZATION ALGORITHM IN CLINICAL PRACTICE 
Masuma Mammadova, Zarifa Jabrayilova, Aytan Ahmadova

https://doi.org/10.24412/1932-2321-2025-783-188-196 

This article presents the essence and architectural-technological principles of digital twin technology, and presents an algorithm for creating a digital twin based on the visualization of medical data through Power BI. The digital twin supports the early prediction of diseases and the analysis of personalized medical indicators by simulating the human body and enabling real-time monitoring of the condition. The article proposes a generalized architecture for building a digital twin of the human body at the level of organs and relevant diseases, the stages of its formation, and develops a schematic description of a virtual object. The Power BI platform is chosen to visually present the virtual object update simultaneously with any changes occurring in the physical object, which is the main architectural component of the medical digital twin. As a real source of information, it is referred to the available national database where doctors periodically collect information about patients in traditional clinical practice. Dashboards are developed on the Power BI platform to form the trajectory of change of medical indicators based on their values given in a time series. The digital twin created based on this approach enables the real-time monitoring of dynamic changes in test results, which supports the acceleration of physician decision-making related to health condition management, the improvement of service quality, and the provision of more prompt and personalized patient care.

OPTIMIZED CNN-BASED APPROACH FOR ALZHEIMER’S DISEASE BY TACKLING CLASS IMBALANCE IN MRI CLASSIFICATION
Soraisam Gobinkumar Singh, Dulumani Das, Utpal Barman, Hasan Huseynov

https://doi.org/10.24412/1932-2321-2025-783-197-207  

Accurate and early diagnosis of Alzheimer’s Disease (AD) is crucial for effective intervention and treatment. This study presents a Convolutional Neural Network (CNN)-based approach for the classification of brain MRI images into four categories: Mild Demented, Moderate Demented, Non-Demented, and Very Mild Demented. To address the challenges of class imbalance inherent in the dataset, we employed class weighting and focal loss during training. Class weighting ensured that underrepresented classes received adequate attention, while focal loss emphasized harder-to-classify examples, resulting in improved model performance on minority classes. The model achieved remarkable results, with an accuracy of 97.66%, precision of 97.66%, recall of 97.66%, F1-score of 97.66%, specificity of 98.98%, and Cohen's Kappa of 96.14%, indicating a robust performance across all metrics. A comparative analysis with state-of-the-art methods demonstrated that our approach outperformed many existing models, including Siamese CNNs, 3D DenseNet ensembles, and other transfer-learning-based techniques. The ROC-AUC analysis further highlighted the model's ability to distinguish between classes with near-perfect curves for all categories. These results underscore the effectiveness of combining CNN architectures with class imbalance-handling strategies for medical image classification. The proposed method holds promise for improving diagnostic accuracy and early detection in AD, thereby supporting clinical decision-making.

EFFECTIVE USE OF ARTIFICIAL INTELLIGENCE METHODS FOR THE IMPLEMENTATION OF PREREQUISITES IN CURRICULUMS
Zafar Jafarov, Vahid Garuslu, Atif Namazov

https://doi.org/10.24412/1932-2321-2025-783-208-215 

Prerequisites play a critical role in the construction of a curriculum for courses where student success heavily relies on previously acquired knowledge or skills. The application of artificial intelligence (AI) methods for the implementation of prerequisites in curriculums can significantly enhance the efficiency and effectiveness of educational planning. The article argues that association rules can be effectively applied in the implementation of prerequisites in the curriculum.

GAME THEORY-BASED OPTIMIZATION FOR SUSTAINABLE ENERGY TRADING SYSTEMS
Elbey Rustemzade, Nurali Yusifbayli

https://doi.org/10.24412/1932-2321-2025-783-216-223 

This article explores the transition from traditional energy networks to Microgrid networks, highlighting the implications of Energy 4.0 and the adoption of modern innovative approaches in energy systems. As energy demands evolve, decentralized energy trading systems are emerging as crucial mechanisms that empower consumers and enhance grid resilience. The paper presents a technical solution to the challenges posed by this transition, focusing on optimizing energy trade through Corporate Game Theory methods. By employing these strategic frameworks, the study aims to improve decision-making processes among market participants, ultimately leading to more efficient energy transactions. The results demonstrate significant advancements in optimizing energy trade, showcasing the potential for increased efficiency and sustainability in decentralized energy markets. This research contributes to a deeper understanding of how innovative strategies can facilitate the effective integration of Microgrid networks within the broader context of Energy 4.0.

 

SECURING THE FUTURE OF ENERGY TRADING: ENHANCING CYBERSECURITY WITH BLOCKCHAIN IN WEB 3.0
Elbey Rustemzade, Nurali Yusifbayli

https://doi.org/10.24412/1932-2321-2025-783-224-232 

Cybersecurity is becoming increasingly important in energy trading, especially as the sector becomes more interconnected and reliant on digital technologies. With the rise of distributed energy resources and peer-to-peer (P2P) trading, new cyber threats are emerging, making it essential to protect data and transactions to maintain trust and stability in energy systems. Web 3.0 represents a significant shift towards decentralization and user empowerment, primarily driven by blockchain technology. This innovative approach allows for secure transactions without intermediaries, enabling transparent exchanges between consumers and producers in the energy market. The integration of Ethereum platforms is crucial for enhancing cybersecurity in energy trading. Ethereum’s advanced smart contract capabilities facilitate automated and secure transactions, greatly reducing the risks of human error and fraud. Provides a robust environment for testing these smart contracts, allowing developers to identify vulnerabilities before deployment. For Azerbaijan, which is modernizing its energy infrastructure and diversifying its energy sources, adopting these technologies can significantly improve security and efficiency in its energy markets. By focusing on cybersecurity measures within Ethereum frameworks, stakeholders can foster a safer, more efficient energy market.

RISKS OF CASCADING FAILURES IN CRITICAL INFORMATION INFRASTRUCTURE
Arzu Babayeva, Yadigar Imamverdiyev

https://doi.org/10.24412/1932-2321-2025-783-233-238 

The article presents an analysis of risk research related to cascading failures in critical information infrastructure. An example of a systems approach implemented in the areas of critical information infrastructure (CII) activity is considered. A model for the spread of cascading failures between interdependent CII objects is proposed, using a weighted graph model. Based on this model, formulas for calculating reliability dependencies and assessing risks between CII objects are presented. Additionally, a risk assessment method accounting for cascading effects for CII objects has been developed and applied.

INVESTIGATION OF SURFACE ROUGHNESS IN HYDROABRASIVE MACHINING DEPENDING ON CHANGES IN ABRASIVE GRAIN SIZE AND PRESSURE
Sylvio Simon, Nizami Yusubov, Samir Amirli

https://doi.org/10.24412/1932-2321-2025-783-239-245 

The article investigates the surface roughness generated on HARDOX-500 chromium-nickel steel blanks in hydroabrasive machining as a function of changes in abrasive grain sizes. The study examines the intervals of roughness variation based on various technological, kinematic, structural, and processing environment factors, and identifies optimal roughness values. Experimental results indicate that as the granularity of abrasive grains increases, the surface roughness on the cut surface of the blanks also increases. However, as the pressure of the water-abrasive mixture and the consumption of abrasive grains rise, the height of the resulting surface roughness decreases.

EFFECT OF HIGH-SPEED SINTERING ON THE STRUCTURE AND PROPERTIES OF MOLYBDENUM POWDER STEELS
Subhan Namazov, Shahin Mashayev, Taleh Taghiyev

https://doi.org/10.24412/1932-2321-2025-783-246-251 

Due to the fact that the technology of sintering powder steels is carried out at different tem-peratures and conditions, the formation of their structure and properties is obtained some what dif-ferently. Proper preparation of the sintering technology allows to obtain high density and properties of smooth steel. In most cases, the reason for the reduction of the properties of powder steels is that the diffusion process, which occurs due to the low sintering temperature and sintering time, is weak or not at all. As we know, the density and many physical and mechanical properties of the product increase due to pore diffusion during sintering.

IMPROVING THE EFFICIENCY OF MACHINING OPPOSITELY DIRECTED CONICAL SURFACES BY MANAGING DYNAMIC TECHNOLOGICAL RELATIONSHIPS
Nariman Rasulov, Ugurlu Nadirov, Irada Abbasova

https://doi.org/10.24412/1932-2321-2025-783-252-258 

This paper presents a technology for sequential-parallel machining of internal cylindrical and conical surfaces of oil field couplings, identifies a unique system of forces acting on the boring bar during parallel machining, and since all three corresponding components of the cutting forces are directed in opposite directions, a sharp decrease in the elastic deformations of the elements occurs in the technological system, due to the control of dynamic technological relationships and main angles in plan of the cutters during parallel machining, the values of the cutting force components change in a favorable direction, ensuring a decrease in the range of elastic deformations of the elements of the technological system, reducing their impact on the accuracy of machining, provides the results of studies of elastic deformations by modeling in ANSYS, it was recommended to apply the developed technology and technological measures that ensure the processing of responsible surfaces of couplings with high accuracy and productivity.

ISSUES OF INCREASING THE EFFICIENCY OF CYLINDRICAL GEAR GRINDING USING COPYING METHODS THROUGH A SYSTEMATIC APPROACH
Nariman Rasulov, Arastun Mammadov, Mursal Alakbarov, Elgun Shabiyev, Yusif Huseynov

https://doi.org/10.24412/1932-2321-2025-783-259-266 

The paper presents a system analysis of gear grinding with copying of cylindrical gears, subsystems associated with the gear grinding system, as well as their inputs and outputs connections; both direct and indirect ways of increasing the efficiency of gear grinding are identified, based on the management of connections by hierarchical sequence vertically and by sources of quality indicators; includes the results of an indirect increase in efficiency due to the adoption of a reasonable value for the allowance for grinding, a direct increase due to a reduction in the number of working passes while ensuring the required quality during gear grinding and also the use of a methodology for forming threads by plastic deformation; the developed methodology is recommended for use in solving similar problems when forming surfaces using other methods.

APPLICATION AREAS OF CURVES AND SURFACES IN ENGINEERING
Ahmed Imanov, Zakir Galandarov

https://doi.org/10.24412/1932-2321-2025-783-267-273 

The general classification of curves widely used in constructing surfaces for rapid movements along curved surfaces in engineering is presented. Key parameters ensuring the smoothness of curved lines are analyzed. The potential positive or negative effects of the curvature of these lines on moving objects, profiled surfaces, and similar applications are examined. The study considers the fields of application of curved lines and surfaces in engineering. Surfaces obtained using the formula representing the curved line, resolved as an inverse problem based on curvature, are investigated. The formula derived from curvature and its modified version are applied in analyzing the relationship between smoothness and potential energy, further confirming the accuracy of this connection. Types of surfaces are shown, and examples are provided of surfaces generated based on guiding curves and frames. The efficiency of surface modeling using computer technologies is also examined.

MODELING AND SIMULATION OF DUCTILE-IRON BLANK CASTING PROCESSES FOR AN ELECTROHYDRAULIC POWER AMPLIFIER BODY
Vitaly Dubrovin, Boris Kulakov, Andrey Karpinsky, Dmitry Ardashev, Anastasiya Degtyareva-kashutina, Ramil Dadashov

https://doi.org/10.24412/1932-2321-2025-783-274-280 

The paper describes the process of modeling a casting mold for a body blank of an electrohydraulic power amplifier and the technology of ductile iron melting in small furnaces.

FLUID FLOW MODELING IN THE SPOOL AND SLEEVE OF AN ELECTRO-HYDRAULIC POWER AMPLIFIER
Darya Khabarova, Sergey Bitiutckikh, Alexander Ismagilov, Dmitry Ardashev, Mukhaddin Samadov, Heyran Abbasova

https://doi.org/10.24412/1932-2321-2025-783-281-287 

This article deals with the initial data for modeling a spool valve. It describes the volumetric solid model of the flowing part of the high precision spool pair layouts in the electro-hydraulic power amplifier; the generation of the finite-element mesh of the solid model; the physical and mathematical model of the fluid flow process; and the estimations and their analysis. The pressure characteristics of the spool valve under different load are calculated. 3D modeling of fluid flow through the annular clearance between the spool and sleeve in the electro-hydraulic power amplifier (fluid leak modeling) is performed.

 

INVESTIGATION OF THE INFLUENCES ON ENERGY CONSUMPTION DURING TURNING AND ITS MODELLING
Ilgar Abbasov, Rezo Aliyev, Arastun Mammadov, Mahabbat Suleymanov, Huseyn Mammadov

https://doi.org/10.24412/1932-2321-2025-783-288-296 

This article examines the interrelationships between technological parameters, including cutting speed, feed rate, and depth of cut, and their influence on average power consumption. A statistical design of experiments was employed with the objective of developing a model that would enable a quantitative and qualitative description of the interactions in question. The insights yielded by these analyses facilitate the optimisation of energy consumption during the turning process, thereby reducing the environmental impact of the manufacturing process. In order to identify the optimal process parameters, preliminary experiments were conducted under three distinct conditions: dry, coolant lubricants, and idle. Furthermore, restrictions were defined with regard to rough and finish turning. At the same time the study examines the influence of minimum quantity lubrication on energy consumption, thus providing further insights into the domain of energy-efficient machining.

WEAR PROPERTIES OF CAMSHAFT CAMS AND IMPROVEMENT OF THEIR WEAR RESISTANCE
Vaqif Abbasov, Fariz Amirov, Azad Karimov

https://doi.org/10.24412/1932-2321-2025-783-297-303 

In the article, the problems of increasing the durability and wear resistance of the cam of gas camshaft by using the ion implantation method were considered. In operation, since the cams of camshaft work the working profile under high pressure, the wear is fast. As a result, the regular operation of the engine is disturbed and the engine does not provide the necessary power and force at the output. It is known that the gas camshaft mechanism requires complex and quite economic costs in the repair of engines. In this regard, increasing the wear resistance of camshaft using the ion implantation method is appropriate. To conduct the research, samples of steel 40X, 50X and X6B3MTiC used in the production of camshaft were prepared and researches were conducted.

PROCESSING OF HIGH-SPEED STEELS BY PULSED LASER RADIATION
Igor Savin, Isag Khankishiyev, Asim Mirzayev, Jeyhun Rahimov, Rufat Abbasov, Gunay Dadashova

https://doi.org/10.24412/1932-2321-2025-783-304-309 

The paper considers the issues of improving the performance of cutting tools made of high-speed steels. The analysis of modern methods of changing the properties of the surface layers of tools according to the specified parameters is carried out. Currently, about a hundred types of hardening technologies of cutting tools are known to have been developed. Almost all performance criteria are determined by the properties of the surface layers. It is shown that the achievement of a given set of properties of surface layers is possible not only by conventional methods, but also by laser treatment. Various options for laser surface treatment of cutting tools made of high-speed steel are considered. It is determined that pulsed laser radiation has a number of advantages for such processing compared to conventional processing. Based on the results of the study, optimal modes of pulsed laser processing of high-speed steels were established, carried out without melting the surface.

THE INFLUENCE GAS JET ON THE QUALITY LASER CUTTING METALS
Alexander Shaparev, Asim Mirzayev, Malik Qarayev, Sadaqat Mehdiyeva, Rufat Abbasov, Yusif Huseynov

https://doi.org/10.24412/1932-2321-2025-783-310-319 

The methods using gas jet in laser cutting are considered. The criteria for the quality laser cutting, such as the absence craters, burrs or ridges on the cutting surface, depend not only on the power the laser radiation, but also on the optimal use the gas jet. An analysis the pressure losses the gas jet in the gap between the nozzle and the metal surface is performed. The passage the gas jet in the cutting zone is significantly affected by the shock wave formed in the gap between the nozzle and the surface the cut metal. Double-jet nozzles provide increased efficiency in removing the liquid phase from the surface the cutting zone.

 
TECHNOLOGICAL FEATURES LASER CUTTING COPPER AND BRASS
Alexander Shaparev, Iliya Avvakumov, Vagif Movlazade, Ugurlu Nadirov, Jeyhun Rahimov, Lachin Babayev

https://doi.org/10.24412/1932-2321-2025-783-320-327 

The article discusses the features laser cutting copper and brass, certain requirements for equipment and technology. The advantages using fiber lasers for cutting copper and brass are formulated, such as better absorption laser radiation, high precision cutting small parts, clean and smooth cut without burrs, high cutting speed. The features cutting copper and brass blanks with lasers with a power 1 and 2 kW are studied. Recommendations for laser cutting copper and brass are developed. Copper and brass laser cutting is possible, but much more difficult than other metals. This is partly due to the fact that copper is a highly reflective material. Copper's reflective properties make it difficult for the laser's infrared light to be absorbed, slowing down the cutting process. To get the most out laser cutting copper, you need to consider speed, power, reflectivity, and focal point.

ON THE ISSUE OF ALLOY CRYSTALLIZATION DURING CASTING INTO METAL MOLDS
Igor Savin, Renat Gavariev, Mukhaddin Samadov, Elgun Shabiyev, Fazil Orujov

https://doi.org/10.24412/1932-2321-2025-783-328-334 

The modern scientific and technological revolution, along with social progress, has ensured tremendous scales of production development in our country. When obtaining many machine parts, instruments, and structures, cast blanks are the most cost-effective option, and in the case of particularly complex parts or when using alloys that cannot be processed by pressure, casting technology becomes the only possible method. At the same time, a significant number of specific defects occur in castings; surface roughness and dimensional accuracy often do not meet the requirements set for products; the properties of alloys in their cast state are generally lower than those in their deformed state. Therefore, the problem of further improving the quality of cast blanks has become critically important today. To address this issue, a specialist must understand the mechanisms of casting processes and be able to control them. The article discusses the physical and mechanical processes that occur during the pouring of molten metal into a metal mold. Various scenarios of premature formation of heterogeneous nucleation sites are considered. The features of nucleus formation on the solid surface of the mold are described depending on its geometric characteristics. Additionally, physical models of crystallization of heterogeneous nuclei are presented, describing the mechanism of appearance of a widespread defect in castings in the form of shrinkage cavities.

DEVELOPMENT OF A PARAMETRIC MODEL FOR CALCULATING CUTTING FORCES IN EXTERNAL CYLINDRICAL TURNING OF 20CRMN STEEL (1.7147) USING AN SNMG 15 06 16-PR 4425 INSERT
Igor Balabanov, Vagif Movlazade, Nizami Yusubov, Heyran Abbasova, Ramil Dadashov, Rasul Huseynov

https://doi.org/10.24412/1932-2321-2025-783-335-342 

This article develops and presents a mathematical model for calculating cutting forces during the machining of 20CrMn steel (1.7147) using an SNMG 15 06 16-PR 4425 T-Max® P insert for turning. We conducted experimental research on a specially designed test rig based on the 16D25 lathe. This setup measures spindle speed, feed rate, cutting depth, and the cutting forces generated during the machining process with high precision. We used the LTR-EU-8 workstation for data acquisition and analysis, equipped with galvanic isolated modules and a synchronized data transmission interface to ensure accurate measurements. The system transmitted real-time data to a computer for further processing, which helped verify the theoretical model. The results showed a high correlation with actual measurements: the deviation between calculated and experimental values did not exceed 5.68%, proving the model’s accuracy in predicting cutting forces. This accuracy plays a key role in optimizing machining processes, reducing tool wear, and lowering energy consumption. The study also found that cutting forces provided by major tool manufacturers are often overestimated. In some cases, the discrepancies between calculated and actual forces reached 17.8%, potentially affecting the accuracy of process planning and the choice of optimal cutting parameters. Additionally, the study revealed that the cutting forces typically provided in calculations by leading tool manufacturers are often overestimated. In some cases, discrepancies between calculated and actual force values reached up to 17.8%, which can impact the accuracy of process planning and the selection of optimal cutting parameters.

DEVELOPMENT OF AN ALGORITHM FOR AUTOMATIC CUTTING TOOL SELECTION
Leonid Shipulin, Egor Shulezhko, Sadaqat Mehdiyeva, Konul Shammadova

https://doi.org/10.24412/1932-2321-2025-783-343-349 

An inexperienced technologist will have to spend a lot of time to select a cutting tool, so a neural network for cutting tool selection is needed. To create a neural network, we need to understand the algorithm, how to select the tool now. The purpose of this paper is to analyse domestic and foreign sources in the selection of cutting tools. The methodology in this study is to find an algorithm from all possible sources. Each source found is analysed to find the algorithm. A block diagram has been developed, and this is the algorithm itself to create an automatic selection of cutting tools. These publications have shown that work in the direction of automatic selection of cutting tools is ongoing and are of scientific and practical interest. But the majority of works have declarative character with the absence of such important information as: criteria of criteria of cutting tool selection, results of approbation, influence of selection results on the cost of technological operation. on the cost of technological operation.

 

STUDY OF DYNAMIC CHARACTERISTICS OF THE ROTARY HONING PROCESS IN THE PROCESSING OF NON-RIGID THIN-WALLED PARTS.
Aydin Gafarov, Isag Khankishiyev, Alihuseyn Haziyev, Irada Abbasova

https://doi.org/10.24412/1932-2321-2025-783-350-357 

The article studies the dynamic characteristics of the rotary honing process when machining highprecision non-rigid thin-walled parts. The process is modeled and optimized to determine its rational parameters, providing the lowest cutting forces.

 

RESEARCH OF MAIN DIMENSIONS OF NEW GENERATION SUBSEA CONSTRUCTION VESSELS AND INVESTIGATION OF INFLUENCE OF CHOICE OF DIVING COMPLEX AND REMOTELY OPERATION VEHICLES ON CONCEPT DESIGN STAGE
Rasim Bashirov, Alexander Egorov, Oyrad Abdullayev, Zaur Jafarov, Rahim Abdullayev

https://doi.org/10.24412/1932-2321-2025-783-358-364 

The mobile diving complexes and remotely operated vehicles available to companies operating offshore in the Caspian Sea are described, along with their proposed installation on subsea construction vessels. Based on research, methods for selecting the main dimensions of subsea construction vessels are presented, utilizing a database of vessels with similar functions and taking into account the installation of the described mobile diving complexes and equipment. The optimization of the main dimensions of subsea construction vessels is carried out using various methods, considering the parameters of the diving complexes and remotely operated vehicles. The main dimensions of the proposed subsea construction vessel are determined according to the parameters of the installed diving complexes and remotely operated vehicles.

DEVELOPMENT OF HIGH-STRENGTH DEEP-WELL PUMP RODS FOR OIL PRODUCTION BASED ON INNOVATIVE METALLURGICAL TECHNOLOGIES
Rahim Shukyurov, Naila Mirbabayeva, Lala Azimova

https://doi.org/10.24412/1932-2321-2025-783-365-371 

In the world, including the Republic of Azerbaijan, oil production from 75-85% of wells is carried out by deep-well pumps. A deep-well pump located at the bottom of the well is connected to a pumping unit at the wellhead with rods. The pump brings oil to the surface of the wells either due to the reciprocating motion of the rod string or due to its rotation in one direction. In the first case, the rod string operates under a repetitive, variable tensile load with an asymmetric cycle, and in the second - under a torsional load. The chemical composition and design of the rods used in both cases are the same. Pump rods (Fig. 1) are long (8 m long) products of small diameter (16-25 mm) and are considered the weakest link in oil production by pumping. Shutdown of oil wells in most cases occurs as a result of rod breakage. Therefore, the level of oil production by this method depends on the strength and reliability of the pump rod. The aim of the research work is to obtain high-strength rods for deep-well pumps by using innovative metallurgical technologies. This is a pressing scientific and technical problem.

THE GENERAL REGULARITY OF SURFACE LAYER WORK HARDENING IN THE HONING OPERATION OF MEDIUM CARBON STEELS
Sarvan Aziz Shirvan

https://doi.org/10.24412/1932-2321-2025-783-372-378 

This paper discusses the general issues of controlling the formation of work hardening in the internal cylindrical surfaces during the honing process using technological methods. Based on the general characteristics of the honing operation of medium carbon steel materials, and considering surface quality and productivity, the technological regularity of work hardening has been identified. Experimental research results on the influence of the main processing parameters-such as the forward-backward (Vb-f) and rotational (Vr) speeds of the honing head, specific pressure (Ps.p), abrasive grit size (Z), and processing time (T) on the degree of work hardening are presented. The functional dependencies of the main input parameters of the process on work hardening, which is taken as the main output parameter for surface quality, have been graphically determined. As a result, the technological foundations for selecting optimal processing regimes have been established

DETERMINATION OF DEFORMATION AND MACHINING ALLOWANCE OF PRECISION PARTS HARDENED BY LASER METHOD
Alakbar Huseynov, Ilgar Nazarov, Farid Huseynli, Mirzabay Safarov

https://doi.org/10.24412/1932-2321-2025-783-379-385 

In laser diffusion metallization, precision parts of machines and equipment are heated to a temperature of 1100-1250 °C. Deformation of parts occurs at this temperature. Another reason for the deformation of parts when increasing the strength of the surface with laser technology can be the normalization of the internal stresses of the parts. Deformation of products under conditions of exposure to high temperatures can also arise from its own weight, which is not possible during diffusion processes in densely packed powder environment, but is possible with other saturation methods (gas, vapor vacuum, liquid). The characteristic of the variation of the value of the deformation depending on the thickness of the diffusion zone. The value of the deformation changes with the change of Poisson's ratio. The article is devoted to the determination of the change value of bending during nanodiffusion metallization.

INVESTIGATING THE APPLICATIONS AND IMPACTS OF LARGE LANGUAGE MODELS IN CHEMISTRY EDUCATION
Canan Kocak Altundag, Sencer Yucel, Faxraddin Yusubov

https://doi.org/10.24412/1932-2321-2025-783-386-392 

Organic chemistry, which is an experimental science, deals with the acquisition and characterization of pure organic compounds from natural or synthetic pathways by appropriate methods. In addition to a good textbook in chemistry education, the laboratory of this course is also of great importance. For the realization of the processes, equipment made of various materials and their efficient use are required. This study has been prepared in order to make this laboratory more understandable and more enjoyable for students to recognize organic synthesis and see its products. In this article, Iodoform Synthesis experiment is taken as the basic experiment to explain the system. A virtual experimental environment powered by a Large Language Model (LLM) was created. The system presents students with a multi-step experiment which they are asked to complete correctly. The innovative side of this environment is how it interacts with the student when they make a mistake. This study was also conducted in an Organic Laboratory class, and thus created a research question: How did the idea of using Large Language Model in organic chemistry course affect student’s mental workload?

MATRIX MODEL OF ACCURACY IN MACHINING CONICAL SURFACES ON CNC LATHES
Nizami Yusubov, Heyran Abbasova, Ramil Dadashov

https://doi.org/10.24412/1932-2321-2025-783-393-400 

The article presents the development of a matrix model for accuracy in machining conical surfaces on CNC lathes. Spatial computational schemes of elastic displacements of technological subsystems were constructed based on the balance of force interaction between the cutting tool and the workpiece. These schemes cover both single-tool setups with a rotating carriage and single-tool two-coordinate setups. Additionally, models of dimensional distortions were developed, taking into account the characteristics of the setups, and were analyzed based on cutting conditions, cone angle, and the comprehensive compliance characteristics of the technological systems. It is also possible to calculate setup parameters using the proposed models. The developed models can be applied in computer-aided design (CAD) systems for machining conical surfaces on CNC lathes, as well as for studying and predicting the influence of various parameters on the dimensional accuracy achieved during conical surface machining..

 
INVESTIGATION OF THE POSSIBILITY OF CONTROLLING THE DEFORMATION OF THE CENTER AXIS OF A WORKPIECE PROCESSED BY TURNING DUE TO THE CUTTING FORCE WITH A DIGITAL PROGRAM
Agasi Agayev

https://doi.org/10.24412/1932-2321-2025-783-401-409 

The article considers the issue of compensating the axial deformation caused by the cutting force during the cantilevered machining of rod-shaped and thin-walled parts processed by lathe operation with a digital program. The deformation of the axis of the part due to the cutting force causes a change in the cutting depth, as a result, the machining accuracy is not ensured, and after machining, shape errors appear in the cross section and also in the longitudinal section. This deformation takes a maximum value at the end of the cantilevered part, and as a result, the shape of the part is obtained according to the hyperbolic curve as it approaches the end. As a solution to this, the article proposes to use an equidistant to compensate for this deformation. The contour of this equidistant was taken into account in the digital program and the deformation of the center axis of the cantilevered parts due to the cutting force during the machining of experimental parts was minimized. The application of the equidistant in the numerical control program was performed in the turning operation of cantilever-mounted rod-shaped parts with a diameter of 10 mm and hollow parts with a diameter of 18 mm.

 

STUDY OF MECHANICAL PROPERTIES OF THE COMPOSITION BASED ON LOW-DENSITY POLYETHYLENE MODIFIED WITH CARBON NANOTUBES
Matanat Mehrabova, Elbay Babayev, Farhad Kerimov, Musa Asadov, Niyazi Hasanov, Fail Shamilov, Yashar Musayev

https://doi.org/10.24412/1932-2321-2025-783-410-419 

The composition and concentration of carbon nanotubes leading to a significant improvement in the electro-physical properties of low-density polyethylene (LDPE) of 10803-020 grade have been experimentally determined. Modified multi-wall carbon nanotubes (MWCN) were used as modifying additives. The difference of developed LDPE nanocomposite was that for the first time a small amount of carbon nanotubes was introduced into the LDPE, which varied within the range of 0.01–0.1 % wt. It was found that the content of 0.05 % wt. of carbon nanotubes in the composition of low-density polyethylene (LDPE + 0.05 % wt. of NC) significantly increased its electric strength. The temperature dependences of dielectric loss tangent (tgσ), specific volumetric resistivity (ρV) and permittivity (ε), as well as the kinetics of physical and mechanical changes in them under the influence of electric discharges in the air and UV irradiation were studied. Specific volumetric resistivity significantly increases in the optimal LDPE nanocomposite and dielectric loss tangent is reduced, i.e. a good correlation is observed between these characteristics. It is shown that the addition of carbon nanotubes in an amount of 0.05 % wt. significantly increases its resistance to the effects of electric discharges and UV irradiation. Based on the experiments, it was concluded that the observed improvement in the dielectric properties of the LDPE film after the introduction of 0.05 % wt. of carbon nanotubes into its composition is associated with a change in the supramolecular structure of the LDPE.

DESIGN AND PRODUCTION TECHNOLOGY OF SPECIAL FRICTION CLUTCH INSIDE INNOVATIVE REDUCERS OF RAILROAD SWITCHES
Ayaz Abdullaev, Isa Khalilov, Goshgar Rasulov

https://doi.org/10.24412/1932-2321-2025-783-420-427 

The article discusses the development of innovative technologies aimed at increasing the speed, efficiency, productivity and safety of railway transport systems based on ensuring modern requirements for dimensions, the number of components and the reliability of manufactured switch devices. The design of a special three-disk friction clutch is designed, which is placed between the gears of a double-crown gear block of the second stage of a package innovative reducer intended for mechanical transmission systems of railroad switch devices, based on their design and functional features. The design of the friction clutch is quite compact, provides for a decrease in the dimensions and weight of the entire transmission mechanism, an increase in its reliability and technical indicators, and protection of the system from overloads. Considering the advantages and technical level of the reducer kit, with a specially designed friction clutch, it is recommended for use on switch drives of the "СП" brand in the railway transport systems of the CIS countries.

STRESS DISTRIBUTION IN THE MATRICES OF END FRICTION SEALS UNDER LOADING
Afet Jafarova, Farid Jafarov, Fuad Jafarli

https://doi.org/10.24412/1932-2321-2025-783-428-434 

The distribution of stresses generated in antifriction seal materials is considered. A photo of isochromes is presented when loading an isotropic material and a material containing solid inclusions. Isochromes when loading a material consist of solid inclusions of various shapes under a load of 120 kg. Photo drawings of the distribution in depth from a concentrated load of tangential stresses in an isotropic material are also presented. In a material consisting of solid inclusions of a round shape, the distribution of tangential stresses depends on the introduction of a "stamp" into the material. Isochromes are obtained when loading a material containing many inclusions.

 
FEATURES OF OPTIMIZATION OF PRESSING MODES OF POWDER MATERIALS FOR PARTS OF SHIP MECHANISMS
Nizami İsmayilov, Fazil Orujov, Elkhan Mammadov, Lachin Babayev, Aga Shixseyidov

https://doi.org/10.24412/1932-2321-2025-783-435-441 

The article considers the features of optimization of modes of pressing powder materials for ship machinery parts using mathematical models. It is established that in order to obtain an adequate mathematical model, it is necessary to more accurately specify the rheological properties of powder materials, which largely determine the kinetics of compaction during HIP. It is indicated that the solution of this problem by the finite element method for products of complex configuration with nonlinearity of the used relations and non-stationary nature of deformation requires the use of iterative procedures in the presence of a large number of finite elements at each time step. It is determined that the upper-level model does not allow solving the problems of optimization and optimal control of the HIP pressing process. To overcome these difficulties, it is possible to use the lower level of modeling, including zero-dimensional and one-dimensional HIP models. A system of technological modes of hot isostatic pressing of powder materials is proposed, which provides for discrete and continuous mathematical modeling of HIP. Discrete and continuous representation of the HIP technological process design system structurally includes the interaction of mathematical models of the upper and lower levels. The use of lower-level models with the use of an optimization apparatus allows us to seriously narrow the search area for technological solutions, thereby obtaining the most reliable and accelerated information on the modes of pressing powder materials for parts of ship mechanisms.

SENSOR-INTEGRATED MACHINE ELEMENTS - A BRIEF OVERVIEW OF PRODUCTS AND CURRENT TECHNICAL DEVELOPMENTS
Andreas Bürger, Sylvio Simon, Pascal Dirk Fritzsche, Shalala Hasanli

https://doi.org/10.24412/1932-2321-2025-783-442-449 

Digitalisation offers enormous potential for significant innovations. In order to utilise this potential, comprehensive and reliable data on the operating status of machines is required. Almost every machine contains standardised machine elements (screws, bearings, gears, seals, etc.) in the immediate vicinity of the process and thus offer the possibility of obtaining and evaluating processrelated measurement data by integrating sensor systems into these machine elements. A brief overview of commercially available sensor-integrated machine elements and a presentation of some current developments will be given. In addition, solutions for the power supply and data transmission of sensor-integrating machine elements are also briefly presented.

SURFACE TOPOGRAPHY IMPROVEMENT OF 18CRNIMO7-6 STEEL USING THE TAGUCHI TECHNIQUE
Andrzej Dzierwa, Anita Ptak, Ahmedov Beyali, Anar Hajiyev

https://doi.org/10.24412/1932-2321-2025-783-450-458 

The ball burnishing process provides a fast, cost-effective, and straightforward method to enhance the physical-mechanical properties and surface integrity of industrially manufactured parts. In this study, ball burnishing was applied to improve the surface topography of 18CrNiMo7-6 steel, with milling used as a pre-burnishing treatment. A Taguchi L9 orthogonal array was employed to conduct hydrostatic ball burnishing investigations and optimal values of the process parameters have been identified. The positive effect of the burnishing process on a number of surface topography parameters was also demonstrated.

MATHEMATICAL MODEL OF THE SINTERING PROCESS OF A MIXTURE OF MESOSCALE WC-CO POWDERS
Arif Mamedov, Aqil Babayev, Mukhtar Huseynov, Beture Musurzayeva

https://doi.org/10.24412/1932-2321-2025-783-459-467 

The paper proposes models of the non-isothermal (solid-phase) sintering stage and isothermal (liquid-phase) sintering. When constructing the sintering process model, assumptions were made about the structure of the solid alloy, the mechanisms of its compaction and grain growth. Based on literature data, a system of equations for the mathematical description of non-isothermal solid-phase sintering of WC-Co was constructed. The equations of the compaction kinetics at the jm temperature stage and at the first 3.5% shrinkage, as well as the equations of the grain growth kinetics and temperature increase, are presented. The coefficients of volume grain boundary and surface diffusion were calculated. A system of equations for the mathematical description of liquid-phase (isothermal) sintering of WC-Co under pressure in a vacuum is proposed. The proposed system considers the following equations: compaction kinetics and rheological model of a porous material, as well as the equation of the alloy grain growth kinetics. The relationships of the quality index which depends on the residual porosity, average grain size and density of the sintered hard alloy are derived. The general equation of compaction kinetics j-m in the temperature mode, as well as the control equation of the product quality index and the characteristics of the WC-Co synthesis process with a mesostructure are obtained.

 
INTELLIGENT ADAPTIVE SYSTEMS FOR PERSONALIZED EDUCATION: A NEURO-FUZZY APPROACH
Tokhirov Ezozbek, Subhan Namazov, Bakhtiyar Badalov

https://doi.org/10.24412/1932-2321-2025-783-468-474 

Neuro-fuzzy models, integrating the adaptive learning capabilities of neural networks with the interpretability of fuzzy logic systems, have emerged as powerful tools in educational data mining. This research explores the application of neuro-fuzzy models in education, focusing on their role in predicting student performance, classifying academic outcomes, and enhancing personalized learning experiences. By analyzing various case studies and methodologies, this study highlights the effectiveness of neuro-fuzzy systems in handling the inherent uncertainties and complexities of educational data.

 
THE ROLE OF ARTIFICIAL INTELLIGENCE IN ENGINEERING STUDIES: A COMPARATIVE PERSPECTIVE FROM GLOBAL UNIVERSITIES AND AZERBAIJAN TECHNICAL UNIVERSITY
Bakhtiyar Badalov, Parvana Movsumova

https://doi.org/10.24412/1932-2321-2025-783-475-481 

The advance of AI creates a field for innovation, efficiency, and creativity in many fields, not passing beyond engineering education. This study examines the integration of AI technologies into engineering curricula, highlighting practices at globally leading universities, including Stanford University, the Technical University of Munich (TUM), and the National University of Singapore (NUS), and comparing them with initiatives undertaken at Azerbaijan Technical University (AzTU). Using the narrative review approach, the article analyzes the role of AI in modeling, virtual laboratories, generative design and personalized learning. In addition, it discusses the critically important skills that modern engineers must acquire, and considers the ethical issues associated with the implementation of AI. The results show that although AI significantly enhances the educational experience and expands students’ technical skills, strategic planning and oversight are critical to its evaluation. The article concludes with recommendations for future AI advocates in engineering education, particularly in new academic contexts.