ABSTRACT. This paper provides an overview of emerging trends in electric machine technology for hybrid electric vehicles (HEVs) and presents a comparative study of different machine configurations. The paper begins by examining the propulsion system requirements for HEVs and the different configurations, including series, parallel, and dual-mode HEVs. The requirements for motor characteristics in an electric vehicle propulsion system are then described, including efficiency, torque, speed range, power density, and reliability. The paper then looks at a comparative study of induction and synchronous machines. For induction machines, the design and advantages are discussed, as well as their limitations and techniques for improving performance, such as vector control and multiphase drives. Synchronous machines, including surface permanent magnet (SPM), internal permanent magnet (IPM), and insert and spoke (BLDC) motors, are also reviewed, highlighting their advantages, disadvantages, and design considerations. New trends are highlighted by presenting permanent magnet hybrid motors as a promising solution for hybrid electric vehicles. Finally, a comparative analysis of these types of electric machines provides a better understanding of their suitability for different vehicle applications. Power density, efficiency, controllability, reliability, torque characteristics, and cost-effectiveness are among the factors taken into account in the assessment. In particular, the study highlights the importance of adapting the choice of electric motor to the specific needs and design priorities of hybrid electric vehicles. Overall, this paper provides valuable information on the latest trends and advances in electric machine technology for hybrid electric vehicles.

ABSTRACT. The control of wind power is crucial for the efficient and sustainable utilization of this renewable energy source. Doubly Fed Induction Generators (DFIGs) are commonly used to harness wind energy efficiently while maintaining grid stability. Predictive control strategies have become increasingly important to improve the performance of these systems. In this context, this paper contributes to the application of deadbeat predictive control to a DFIG-based wind system. First, a comprehensive mathematical model incorporating the DFIG is presented, followed by a detailed explanation and application of the deadbeat predictive control principle to both rotor and grid side converters. The whole system is then simulated using MATLAB/Simulink software and the results are discussed. The control algorithm anticipates the future behavior of the system and adjusts the control inputs to achieve the desired performance in terms of tracking and response speed during dynamic operation. This approach has significant potential to improve the performance and reliability of DFIG wind turbines.

ABSTRACT. This article reports the new approach in the diagnostics of Electronic Parking Brake (EPB) Module, the electromechanic part that is used as literally electronic brake and in a hill-holder system. Each module contains the gear system, that consists of 4 rotating elements: pinion gear, worm gearbox, electric motor, and spur gear that can suffer the damage during its operation. In order to precisely define the defective element, we focus on its diagnostics with piezoelectric sensors, i.e. Micro Fiber Composite (MFC). Each of mounted piezoelectric sensors on the EPB’s housing is sensitive to the vibrations during its operation. Recorded voltage time-series are used for the quantitative estimation of the module’s condition with help of several statistical indicators. Such approach allows to state, which of elements generates high amplitudes and should be replaced. The proposed approach is the alternative to the diagnostics method applied in the company dealing with the regeneration of the EPB’s modules.

ABSTRACT. This paper presents an experimental stand designed to demonstrate the influence of the elastic deformation of a metallic cantilever under an external loading force. An automatic position-adjusting system of the cantilever was built and configured to compensate for the elastic deformation computed into the automation management software using the known material resistance theories. The stand can help estimate the position errors due to the mechanical and constructive tolerances of the equipment it is built on.

ABSTRACT. The autonomous movement of a car in a structured environment is possible based on a map, with the location of the vehicle by GPS and a destination specified by a user or by the vehicle management system [1]. LIDAR sensors are essential for autonomous vehicles to detect other vehicles or obstacles on the road. Controlling the movement of self-driving vehicles is an important aspect of the development of intelligent transportation systems (ITS), for people and goods safety [2], [3]. The estimation of distances is very similar to the perception of distances through binocular vision in humans and using artificial intelligence (AI) techniques can optimize it [4]. In this paper, we simulate controlling the movement of an autonomous vehicle on a lane of the road using MATLAB program. We discuss the results for different time values when obstacles occurs in front of the vehicle.

ABSTRACT. This study presents a comparative study between the Finite Control Set Model Predictive Control (FCS-MPC) and the multi-loop control strategy based on a Proportional Resonant (PR) Controller for a three-phase four-leg inverter to generate various grid disturbances under various load conditions. It is shown that MPC has a better transient response in all cases of grid disturbances. Regarding the steady-state performances, the output voltage has a lower THD when using MPC, especially for inductive loads, while the relative error on the output voltage is relatively the same for both control schemes.

ABSTRACT. Industry 4.0 represents a significant evolution in the field of automation and production, imposing the need to develop advanced educational platforms to train the next generation of engineers and researchers. In this article, is present an innovative didactic platform for the study of Industry 4.0, focused on the implementation of an automated pick and place system. Our platform integrates advanced hardware and software components, including pneumatic cylinders, vacuum grippers, IO-Link sensors, IO-Link master and IO-Link solenoid valve group, HMI (Human-Machine Interface) and control buttons. These components are interconnected to create a modular and flexible system that is easy to use for educational and research purposes. The article will detail the architecture and operation of the system, as well as the software development process, including the programming languages and environments used. It will also discuss the implemented communication protocols (Profinet, OPC UA, MQTT) and the possibility of activating a web server for interfacing with other systems. The results obtained demonstrate the efficiency and applicability of our platform in the educational environment, providing students and researchers with an interactive and practical environment for understanding and applying Industry 4.0 concepts. This platform is a valuable resource for training future generations of automation and advanced technology specialists.

ABSTRACT. Industry 5.0 marks a new stage in industrial production, with a focus on close collaboration between humans and machines. This paper presents the key technologies that define Industry 5.0, including artificial intelligence (AI), collaborative robotics, Internet of Things (IoT), Digital Twins, and advanced data analytics. At the same time, current applications of these technologies are examined, highlighting their impact on productivity and operational efficiency. The main challenges in implementing these technologies and proposed solutions to overcome them are also discussed. Finally, the paper provides an insight into emerging trends and future innovations that will shape Industry 5.0, highlighting the importance of continuous research and innovation to ensure a sustainable and efficient development of the industrial sector.

ABSTRACT. A deep learning-based framework for automatic vehicle detection, classification and counting is introduced in this research work. Two convolutional neural networks (CNN) are created, trained, validated and tested for the detection and recognition tasks. First, a YOLO V2 detection network is built by modifying a MobileNet-v2 deep network and using it for the high-level feature extraction. Then, the obtained CNN model is trained and validated on a voluminous traffic video dataset. The vehicles whose bounding boxes are detected applying the trained CNN-based detector on the movie frames are then extracted using a level-set based active contour model. They are then recognized with another transfer learning network, created by modifying a GoogLeNet classification model and trained on a large vehicle dataset. Next, the detected and classified vehicles are tracked within traffic video applying a matching process using an intersection over union (IoU)–based metric and some conditions.

ABSTRACT. Ensuring road safety is becoming more and more significant for researchers and for all the society. One of the main issues in this area considers driver drowsiness. Therefore, the focus of this paper is the comprehensive analysis and comparison of four distinct machine learning models: Logistic Regression, Random Forest, Support Vector Machine (SVM), and Gradient Boosting. The ultimate goal is to gain insights into the intricate factors that shape driver behavior. When assessing the effectiveness of each model, four key metrics are taken into account: precision, recall, F1 score, and accuracy. In this study, we used a “TRYOUT” dataset. The study yielded several significant findings regarding the performance of the machine learning models employed. Remarkably, the accuracy scores for all models were notably high, falling within a range of 0.8651 to 0.9999. Among all the models, the Random Forest model stood out as the most effective, displaying an impressive accuracy score of 0.9999 on the merged data. The SVM model, while not as proficient as the Random Forest, still managed a respectable performance, particularly on the MPM39 dataset where it achieved an accuracy of 75.06%. The Logistic Regression model demonstrated its strength on the CVD08 dataset, although its performance was comparatively weaker on the other datasets. Lastly, the Gradient Boosting model demonstrated consistent competency across all datasets, underscoring its reliability in a range of circumstances.

ABSTRACT. Trading in the financial markets is driven by both fundamental and technical analysis. These methods have however been shown to have limitations owing to the inefficiency of the markets attributed to human behavior and psychology. Algorithm led methods have been highlighted as a suitable alternative which for years has banked on simple heuristics and rules. Deep Reinforcement Learning presents the next frontier in these automated methods owing to its ability to develop agents that can master any given environment like financial markets. In this paper, these agents are tested by trading three financial instruments. While developing these agents, simple prediction models are first implemented using a variety of deep learning algorithms. Sequence to sequence models outperform all other algorithms in the prediction of the three instruments. As for the agents, their performance is not consistent across the three instruments but Gold is traded with better results that is, better profit and loss measures. In the end, the paper recommends the use of better trade management heuristics to determine quality of trades taken and the refinement of the action space to incorporate risk management strategies.

ABSTRACT. Monitoring a computer network and protecting the resources within it is a topic of great interest for a wide range of applications such as those used in the military, finance, healthcare, and many others. Also, in addition to effective monitoring, a continuous development of cyber security tools is also necessary because attacks are becoming more frequent and attackers are also developing more and more mechanisms to steal information and affect services which, at first glance, would be of extremely low interest to anyone. Therefore, a system that constantly monitors and detects intrusions into a computer network is an increasingly common requirement in most institutions and companies. This system can be realized using both hardware and software resources and solutions. Within the work, a classification of pre-tagged network traffic will be performed to detect malware applications. The experiments will comprise two distinct approaches: a binary classification approach, to simply distinguish between normal and malicious traffic, and a multi-class classification approach, to detect different types of malicious traffic. Afterwards, a comparison will be realized, in terms of efficiency and accuracy, between the Machine Learning algorithms used in the experiments.

ABSTRACT. Cow's milk is very sensitive to spoilage processes that could negatively affect its quality. Several methods have been developed for the analysis of milk spoilage, and in recent times solutions using e-nose are more and more common. This paper analyzes the possibility of using the Bosch BME688 (BioMedical Engineering) sensor, which can detect different gases by analyzing their different electronic signatures. Samples of fresh, pasteurized and UHT (Ultra High Temperature) sterilized milk, which were at room temperature (22°C) during the 4-day experiment, were tested. The obtained data for every day were used to train nine models based on different classification algorithms (K Nearest Neighbor, Logistic Regression, Support Vector Machine, Naive Bayes, Decision Tree, Random Forest, Gradient Boosted Trees, Probabilistic Neural Network) with the aim of distinguishing the type of milk with known time after opening, i.e. milking. The focus of the research was on the analysis of classification algorithms that can be used for the classification of milk with regard to the standing time at room temperature. The manufacturers of the sensor itself enabled the use of neural networks exclusively in the BME AI-Studio program, which is used for configuration and training of the model, and their decision is correct in most cases.

ABSTRACT. Within the modern industry, the analysis and critical evaluation of existing techniques for the detection, prevention, and monitoring of industrial equipment are essential aspects for improving performance and operational efficiency. This study investigates innovative methods designed to enhance the reliability and accuracy of defect detection, with the main goal of optimizing maintenance programs and preventing failures within production processes. The analysis will focus on a diverse range of techniques, including methods based on signal acquisition and analysis, artificial intelligence, audio analyses, and image processing, as well as vibration analyses and other relevant techniques. Through comparison and critical evaluation of these techniques, this paper aims to provide a deep understanding of the advantages and limitations of each approach, as well as the context in which they are applicable in industrial practice. Through this study, a series of methods and technologies are pursued that contribute to the strengthening of knowledge in the field of monitoring industrial equipment and identifying future directions for research and development, aimed at the continuous improvement of performance and reliability.

ABSTRACT. This study proposes methods that can be used to examine and interpret comments that users have made after watching videos on YouTube on a particular topic. YouTube tutorials are very popular among young people. They have become an important pillar in informal education, thus contributing to the rapid acquisition of skills and knowledge. Under these circumstances, we were interested in analyzing the YouTube videos comments, knowing that the platform has a continuous increase in popularity, which is also due to the oportunities of sharing them. Since ChatGPT-themed YouTube videos have seen a significant surge in popularity since 2022, we were interested to analise some videos that approache the topics of "ChatGPT, AI clone, AI robot and Deep Learning."

ABSTRACT. This paper proposes a new approach to improve weak supervised classification of deep CNN architectures based on data augmentation by integrating Grey Wolf Optimizer (GWO) with Wasserstein Generative Adversarial Network (WGAN). The model is applied for hyperspectral image classification (HSI). We have chosen a weak supervised CNN classifier, where only 5% of hyperspectral training data is authentic, the remaining 95% ones being virtually generated using the model GWO-WGAN. The performance of the proposed model is evaluated using Indian Pines, Pavia University, and Salinas hyperspectral image datasets.

ABSTRACT. Our study focuses on the detection of breast cancer using medical image analysis. The researchers explore the effectiveness of various oversampling methods in improving the performance of deep learning models for breast cancer detection. The dataset used in the study has a severe class imbalance with a disproportionate number of cancerous and non-cancerous examples. Six oversampling methods are evaluated in this study. Each oversampling method is applied to the dataset, and the augmented data is used to train deep learning models. The performance of each oversampling method is evaluated using metrics such as accuracy, precision, recall, and F1-score. The results demonstrate that oversampling methods significantly enhance the performance of deep learning models for breast cancer detection. SVM-SMOTE and ADASYN consistently outperform other methods, achieving the highest F1 scores on both ResNet-50 and AlexNet architectures. The findings also suggest that the choice of oversampling method has a substantial impact on model performance, emphasizing the importance of selecting an appropriate oversampling technique for imbalanced data. Overall, this study highlights the significance of addressing class imbalance in medical image analysis and provides valuable insights into the effectiveness of different oversampling methods in improving the performance of deep learning models for breast cancer detection.

ABSTRACT. The detection of potholes in asphalt has been a concern in the field of deep learning to identify patterns with the best possible accuracy. In the smart city concept, autonomous cars are used more and more to take traffic images and transmit them to a specialized center, which processes and distributes the information to all drivers in real time. In this study, we modified the structure of the standard ResNet-50 model of a convolutional neural network (CNN) architecture used to train deep neural networks. Using a data set previously taken from the traffic with images categorized into 2 classes: with pits and without pits, to be used in training using the Matlab Deep Network Designer. Using different training epochs, we verified and synthesized the validation accuracy for different training epochs with the aim of identifying an improvement in performance compared to the classical residual neural network model with 50 deep layers processed in Matlab.

ABSTRACT. In the works, the last generation technologies in the field of Smart fire management systems are debated. Paper presents fire prevention and control technologies in areas such as: forestry areas, cities, buildings and in transport. The processed data are extracted from databases such as Google, WOS and Scopus. The paper presents the advantages and disadvantages of the technologies used in fire prevention and control, projects and architectures. The work also analyzes, using the Prisma Smart fire management systems method, the summary of graphs revealed in the field of interest. The conclusions of the work are that it is a vast and intensively researched field in continuous expansion and remodeling in trend with the latest technologies.The work is composed of three chapters, so chapter 1 presents the analyzed technologies, chapter 2 presents the Prisma analysis and chapter 3 presents the conclusions.

ABSTRACT. Bird populations have recorded important changes in number and distribution since the 1970s, due mainly to land management policy and climate changes. So, bird populations monitoring and scientific research became important for environmental preservation. Automatic bird sound recognition starts receiving more attention since the mid-nineties due to technological advances and the importance of the discipline, as many bird species are often better detectable by their sounds than by vision. The underlying methodology takes advantage of similarities with speaker recognition and has evolved on the tracks of speech-based technologies. With the With the evolution of the deep learning systems, many attempts using these methods were made. The Danube Delta and its close geographical areas are home to 327 bird species and have received over time national and international recognition regarding nature protection. The paper presents the research in the field of bird sound recognition applied to some bird species specific to the Danube Delta. We have tested several settings for the Gaussian Mixture Modelling, Convolutional Neural Networks and Feed Forward Neural Networks. The experiments consider five to fifteen bird sound classes.

ABSTRACT. This paper discusses the use of a set of modeling concepts used in the field of systems and software engineering known as IDEF, which stands for "Integration Definition". Specifically, IDEF0 is used for the purpose of representing the many activities, actions, and choices that occur within an organization. Applying IDEF0 modeling to collaborative design in the context of integrating local and area service process management information systems can provide stakeholders with the ability to visually represent and understand the processes that are involved in integrating local and area service process management information systems. Benefits include the use of a clear visualisation helps to improve communication between stakeholders, which in turn helps to build a shared understanding of the collaborative design process. In conclusion, the use of IDEF0 modelling in collaborative design for information systems integration brings clarity, efficiency and strategic alignment to the process. It provides companies with a structured framework that enables them to traverse the complexities of managing local and area service processes. In addition, it encourages collaboration between stakeholders and optimises integration efforts to achieve better results.

ABSTRACT. Horizontal models of community collaboration involve the creation of partnerships and cooperation between various government departments, agencies and organizations that provide public services at the local and area level. This collaborative strategy strives to improve efficiency, reduce silos, and ensure seamless service delivery. A model of horizontal collaboration among government departments, agencies and organizations ensures seamless public services for local and area interests. Implementing this entire paradigm requires dedication, effective communication, and a shared understanding of the benefits of collaboration. Regular reviews and revisions based on lessons learned will help ensure the longevity and effectiveness of this horizontal community cooperation approach

ABSTRACT. The paper presents an exhaustive analysis of security in embedded systems, with a focus on identifying and managing vulnerabilities in communication channels. The introductory chapter highlights the importance of a comprehensive analysis of these vulnerabilities, highlighting the basics of cryptography used in these systems. The paper explores the basic principles of cryptography and its role in securing communications, detailing specific cryptographic technologies and protocols. It also analyses potential weaknesses in communication channels in depth, providing concrete examples of vulnerabilities and approaches to detect them through machine learning. An enlightening chapter of the paper presents the various attack methods used to compromise security in embedded systems, including case studies of notorious attacks and ways to protect against them. Finally, the paper draws the conclusions obtained and indicates directions for further research in the field.

ABSTRACT. This study presents a realistic and comprehensive framework for creating food traceability and safety from producer to consumer and adds to the expanding body of knowledge on Internet of Things, blockchain and Artificial Intelligence applications. More than ever recently, quality controls of various foods have been carried out. These revealed a series of problems from the point of view of their contamination with a larger than normal amount of harmful elements for the body. The lack of possibility and quick information of the consumer who wants to purchase a certain product was found. This research aims to offer the consumer, through the use of an application, the opportunity to find out about the origin, composition and method of obtaining each product before it is purchased. When obtaining these performances, the new technologies such as the Internet of Things, Blockchain and Artificial Intelligence appear, to be tested in certain research areas, and then to be used on a large scale. The combined use of the Internet of Things, Blockchain and Artificial Intelligence have contributed to the realization of a new architecture and the respective creation of a new operating platform in commercial spaces that offer special products to consumers. The proposed platform is not only useful for the consumer, but also for the authentic producers and all the intermediaries that intervene in the circuit of the products from the producer to the consumer.

ABSTRACT. In this article, blockchain technology is proposed for use in the personal document archiving process. This area is less covered in the literature/specialist applications in the field. The aim was to obtain benefits in terms of security, privacy and accessibility of personal data. The architecture of such a system is debated/justified to determine the justification of the importance of applying this research to the implementation of a: digital identity platform; personal health management system; digital portfolio of education and qualifications; financial identity management system; archiving of legal documents; blockchain for wills and estate planning; intellectual property management system; platforms for managing insurance documents; professional career management system; platforms for travel and immigration documents; And so on

ABSTRACT. This paper investigates potential weaknesses in embedded systems communication channels, with a focus on identifying and analyzing specific vulnerabilities. The problem of securing communication in embedded systems is critical due to the increasing reliance on these systems in various industries, from automotive to healthcare, where breaches can lead to significant safety and security risks. To address this issue, the study employs a multi-faceted approach using machine learning and statistical methods for vulnerability detection. Machine learning techniques, such as supervised learning algorithms, were utilized to identify patterns indicative of potential security breaches. Additionally, Markov Chains were applied to model and analyze the probabilistic behavior of communication sequences, identifying anomalies that could signify vulnerabilities. A practical application was developed using virtual instrumentation in LabVIEW, providing a user-friendly interface for real-time vulnerability detection and analysis. The results demonstrate that the combined use of machine learning and statistical methods significantly improves the accuracy and efficiency of detecting vulnerabilities in embedded systems communication channels. This research highlights the importance of integrated security measures and presents a robust framework for enhancing the security of embedded systems.

ABSTRACT. Image encryption is vital for safeguarding sensitive visual data as it moves over different networks or gets stored in digital formats. Recently, cellular automata (CA) have emerged as promising tools for this task, thanks to their capability to produce intricate and chaotic patterns. This study introduces an encryption algorithm based on cellular automata rules 90 and 150. These rules are chosen for their distinct attributes in creating pseudo-random sequences, which serve for scrambling the pixel values of the input image. Experiments carried out on the results, including comparing histograms between the original and encrypted images, along with statistical analyses, illustrate the effectiveness and efficiency of the proposed encryption approach in ensuring security.

ABSTRACT. This paper addresses the problem of ASR (Automatic Speech Recognition), a crucial field for the advancement of HCI (Human-Computer Interaction) technologies and applications in voice assistants and machine translation. The proposed methodology employs CNNs (Convolutional Neural Networks) to predict spoken commands from voice signals by extracting essential features such as Spectrograms, ZCR (Zero Crossing Rate), Spectral Centroid, Spectral Rolloff, and MFCCs (Mel-Frequency Cepstral Coefficients). A filter bank was also implemented to enhance signal processing. To evaluate the system's performance, comparisons of accuracy across different processing methods were conducted. Additionally, a hardware implementation using the Simulink graphical programming environment and a Raspberry Pi microcontroller was developed, demonstrating the system's practicality and efficiency in real-world contexts. The results indicate that CNNs, coupled with these specific features, provide a better accuracy in speech recognition compared to traditional methods. The implementation on the Raspberry Pi confirms the potential for deploying this system in practical applications, paving the way for the development of cost-effective and efficient ASR solutions.

ABSTRACT. To help with diagnosis, a novel MRI-based approach has been developed. Advances in deep learning networks have the potential to reliably and early detect heart muscle fibrosis. This paper presents FibrosisNet, a novel type of deep network. When present, it can recognize and classify fibrosis. It combines several distinct series levels to achieve the objective of fibrosis identification. The recently deployed classification system is evaluated and trained to produce the best possible performance results. Additionally, deep transfer-learning models are used to different popular convolution neural networks to obtain new designs for fibrosis detection. The Fibrosis Net design performs well, with an accuracy of 96.05%, sensitivity of 33 97.56%, and F1-Score of 96.54%.

ABSTRACT. Air quality is a major public health issue globally. Air pollution is responsible for millions of premature deaths annually and can cause a range of health problems, including respiratory diseases. The need to develop emergent materials and new technologies to increase air quality offers a series of opportunities due to the benefits offered. The emerging materials can be used to create more efficient air filters, air quality monitoring sensors and catalysts to reduce emissions. The article summarizes the main legislative regulations for reducing pollution, as well as emerging materials with a role in increasing air purity, by reducing pollutants.

ABSTRACT. Nanomaterials have become important materials used in various fields due to their small size (<50 nm). Since the 1990s, nanotechnology has become a widespread term as it has various industrial applications. Nanomaterials research is a rapidly evolving field with new discoveries and applications emerging all the time. Currently, nanoparticles are indispensable for many industries such as pharmaceuticals, electrical, automotive, nautical. The article aims to present the current trend in nanomaterials elaboration. Every type of method highlights the technological parameters and advantages.

ABSTRACT. Current research is focused on obtaining zinc oxide nanoparticles through environmentally friendly syntheses. The interest in developing zinc oxide is due to its applicability in electronics, optoelectronics, medicine, cosmetics, food industry, etc. Along with the increase in applicability, the need to synthesize ZnO in environmentally friendly conditions also increases. In this work, the biogenic synthesis of zinc oxide consisted of elaboration of zinc oxide by using ovoalbumin and zinc nitrate as precursors. The optimal reaction conditions are room temperature, basic pH, time and 1200Cfor degradation of ovoalbumin. Samples were analyzed by ATR-FTIR to determine when ZnO formed and completely eliminated the protein. This synthesis proves to be an easy method to obtain zinc oxide starting from natural resources.

ABSTRACT. Micro-arc oxidation (MAO) coatings have several advantages for example: expanded surface zone, important process mechanical properties, optical properties, biomedical gadgets, or aviation components. MAO can be applied for surface treatment which can be used to obtain useful coatings on metals and alloys. The process consists in applying a high voltage electrical trigger to the substrate. This makes a plasma that connected with the surface of the substrate, making a thick coating. MAO coatings are commonly composed of a blend of oxides, hydroxides, and other compounds. Coating structure is affected by several components, tallying voltage, current characteristic, electrolyte composition and planning time. MAO coatings can be used to create surfaces with wear resistance and optical properties such as anti-reflective properties, while also being anti-corrosion and self-cleaning. This makes MAO a promising innovation for a huge range of applications. Our article presents the experimental studies involving MAO with different electrolytes at certain parameters. The obtained layers present very good corrosion properties were obtained on aluminum metals but also on their alloys that can also be applied in the automotive industry.

ABSTRACT. There are a variety types of heat treatment furnaces available, each with its own advantages and disadvantages. Choosing the right type of furnace, depends on a number of factors, such as: the type of material to be treated, the temperature and the duration of the heat treatment required. In this paper, a laboratory solution for automation an electric heat treatment furnace is presented. The proposed solution offers the possibility to configure operating parameters: temperature and time. The novelty of the proposed solution is represented by the possibility of implementing (with minimal costs and structural changes) a digital control on any type of electric heat treatment furnace. By using two temperature sensors, we want to obtain a better temperature monitoring, eliminating the effects caused by the malfunction of one of the sensors. At the same time, the digital temperature control, with a resolution of 0.25ºC, results in an electricity saving compared to the classic bimetal sensor control.

ABSTRACT. After a network breakdown, modern networks offer partial data recovery. Multiprotocol Label Switching (MPLS) incorporates methods for quick troubleshooting and secure link management. This paper investigates a current issue affecting email traffic in MPLS networks using Active Label Switch Routers and quality of service guarantees. To underscore the effectiveness of active LSRs, we conducted a comparison between two types of networks: one with all LSRs active and the other with a mixed network configuration.

ABSTRACT. The security of IoT devices has become imperative given their exponential proliferation worldwide. This underscores the critical need for ensuring their security, especially considering their connectivity through diverse types of networks. This paper introduces a device I have designed for encrypting data in wireless networks. I will illustrate a scenario commonly encountered in organizations. Our aim was to create a cost-effective solution for encrypting data in Wi-Fi environments, taking into consideration the prevailing cyber security landscape in 2024.

ABSTRACT. The notion of Internet of Things (IoT) has revolutionized whole industries, simply because it has provided constant connectivity between devices that hitherto existed in isolation. This has made Low Power Wide Area Networks (LPWAN) the network of choice when it comes to implementing an IoT, because they guarantee long-range communication and low power consumption. Great emphasis has been put on LoRaWAN and NB-IoT among the LPWAN technologies. This paper presents a comprehensive assessment of the performance metrics of LoRaWAN and NB-IoT under varied payload data sizes. The findings are substantially notable in two methods: LoRaWAN and NB-IoT. In comparing NB-IoT with LoRaWAN, it is evident that it has an overall trend to more problems in the transmission with an increase in the size of the payload. Albeit with higher bit error rates than LoRaWAN, NB-IoT generally is more susceptible to the risk of suffering from communication-related issues. This implies that LoRaWAN will always be in a position to achieve faster speeds of connectivity as compared to NB-IoT because its transmission time is smaller. LoRaWAN is always good for the whole time in terms of energy efficiency compared to NB-IoT; it uses less energy with different sizes of payload. Furthermore, LoRaWAN provides constant resource use, whereas with NB-IoT, a lesser amount of resource is used to transmit the same amount of data.

ABSTRACT. This article presents an overview of security considerations in the context of smart cities, highlighting key areas of connectivity of IoT devices in a smart city, vulnerabilities and strategies for risk mitigation. It explores possible vulnerabilities associated with interconnected IoT devices, the potential impact of cyberattacks on essential services, and the importance of implementing robust security measures to mitigate risks.

ABSTRACT. This paper presents a comprehensive study on the precision of Ultra-Wideband (UWB) technology for indoor positioning in controlled environments. In this study, we evaluate the positioning accuracy of Decawave DWM1001 sensors based on the IEEE 802.15.4 standard within two different spatial configurations: a 3x3 meter area and a 3x12 meter area. Our method involved deploying a network of four anchors and one tag in these two setups and conducting a series of empirical measurements to assess the technology’s performance. Our findings demonstrate the high accuracy of UWB localiza- tion, particularly in smaller spaces. However, our experiments uncovered challenges in larger areas, where environmental factors play a significant role in influencing the accuracy and reliability of positioning measurements, compared to the high accuracy promised by the UWB technology positioning and shown by our simulations in ideal conditions. We experimented a high level of accuracy in the smaller 3x3 meter area, with an average error of 6.950 cm at a 1-meter height. The larger 3x12 meter area exhibited increased error rates, with an average error of 9.339 cm at the same height.

ABSTRACT. Modern maritime surveillance systems integrate various RF sensors to construct regional awareness pictures. Such sensors include the Automatic Identification System, coastal and over-the-horizon radars, or passive signal analysis systems. However, these technologies exhibit inherent limitations due to variable propagation conditions. This paper introduces a novel data-driven approach that combines both meteorological and AIS detection range datasets. Standard and recurrent neural networks were implemented to predict detection ranges for AIS maritime receivers. By leveraging 72 hours of data, the models forecast the AIS discovery performance for the subsequent 24 hours with an accuracy of almost 89%. The proposed methodology offers benefits such as detecting AIS spoofing, jamming, and hidden activities, thus enabling efficient resource allocation and optimizing data acquisition strategies. This research establishes an innovative AI-driven analysis for predicting maritime sensor detection ranges, contributing to enhanced maritime security and operational effectiveness.

ABSTRACT. The Digital evidence is derived from diverse sources like computers and their volatile and non-volatile storage, portable digital devices, network traffic, social networks, records and logs from internet service providers, websites, databases, and information systems on both global and local levels. This study focuses on assessing the effectiveness and appropriateness of different security approaches in safeguarding the integrity of digital evidence. The evaluation results in a series of suggestions that should be taken into account when choosing the appropriate algorithm to ensure the integrity of digital evidence in a broad sense. This research examines the cryptographic methods for safeguarding privacy in digital forensics and their classification them based on their support for trustworthy third parties, numerous investigators, and multi-keyword searches have been done. The Author outlines the limitations of using cryptography-based techniques in safeguarding privacy of digital forensics and propose possible remedies for these deficiencies.

ABSTRACT. The economy is currently experiencing a substantial and uncertain transformation propelled by recent advancements in artificial intelligence. Companies that embrace calculated risks and proactively position themselves ahead of the curve will be the ones poised to capitalize on the substantial growth and value-creation opportunities that artificial intelligence offers across nearly every industry. To achieve this, leaders must acknowledge AI's extensive potential as the all-encompassing technology of the twenty-first century. The purpose of this study is to comprehend how artificial intelligence is currently being used in leadership-related areas like hiring, training, and career development. This paper examines the advantages and disadvantages of both using and not using AI in leadership. The transformation of leadership brought about by AI is highlighted in this paper.

ABSTRACT. The fourth industrial revolution has ushered in a transformative era for the industrial sector, marked by the integration of advanced technologies that increase production efficiency and quality. Among these innovations is Smart Scheduling, which aims to optimize production processes while minimizing costs and meeting manufacturing requirements. This article explores the development and application of a Hybrid Genetic Algorithm, optimized by Tabu Search, to meet the complex challenge of multi-objective industrial production planning. The proposed framework introduces a Genetic Algorithm designed for intelligent, multi-objective industrial production planning, demonstrating its effectiveness in generating high-quality planning. It significantly reduces production times while maintaining quality and business needs, as validated in a real-world case study. The algorithm improves on traditional planning methods by dynamically adapting to manufacturing priorities. It also describes potential future adaptations to a wider range of industrial contexts and sectors. This research highlights the fundamental role of genetic algorithms in advancing smart manufacturing, offering a scalable and adaptable solution to the challenges of Industry 4.0.

ABSTRACT. This paper presents a new metaheuristics optimization algorithm inspired by one of the most intelligent birds, Harries Hawks, called Harris Hawks Optimization (HHO) for solving optimal power flow (OPF) problems to achieve technical benefits. Managing active and reactive power in transmission networks significantly influences their performance. To improve the performance of transmission networks, the mechanisms for determining the location of renewable energy source (RES) and capacitor shunt (CS) are most popular and important. The voltage stability index represents the indicator used to determine the optimal placement. The main goal to use this algorithm HHO is select optimal sizing of RES and CS. This paper has presented four scenarios: the initial case (only power flow calculation), optimal power flow using HHO, optimal power flow using HHO after installing CS, and optimal power flow using HHO after installing RES. In these scenarios, the three most popular objective functions (OFs) were proposed, which are the real power losses, voltage deviation, and voltage stability index of whole system. Integrating a local search scheme improves the search space's capability and enhances the exploration rate, facilitating the discovery of a global solution. IEEE 30-bus is the standard network that was applied. The numerical and simulation results demonstrate the performance of HHO in searching for global solutions and preventing local solutions. The optimal results of OFs obtained by HHO will be compared the optimal results obtained by other optimization techniques to confirm the superiority and efficiency of HHO.

ABSTRACT. : This research paper reviews the technologies for optimizing automotive air conditioning systems that would ensure passenger comfort and well-being during travel, especially as the automotive industry moves toward electrification and sustainability. Critical analysis of proposed solutions for modern vehicle air conditioning systems focuses on advanced technologies that increase efficiency and sustainability while maintaining optimal performance. The study aims to show that the integration of air conditioning materials and components can improve engine performance to achieve high levels of energy efficiency and reduce fuel consumption. As the demand for environmentally friendly and efficient fossil fuel vehicles is increasing, recent research has focused on studying advanced engineering solutions and revolutionary technologies that will change and streamline the functions of classic car air conditioning systems in correlation with imposed energy efficiency performances and environmental care restrictions, proposing solutions for a sustainable and efficient future. Critically analyzing the research in the thermal management of motor vehicles carried out so far, the work exposes improvements from a technological point of view in the evolution of the cooling systems of the car interior that align with global climate changes and the need for more economical and sustainable transport solutions.

ABSTRACT. The overall performance of a fuel cell electric vehicle (FCEV) is specifically related to the total fuel consumption during a load cycle (eg NEDC, EUDC, WLTP, FTP, etc.). This reveals the importance of fuel economy strategies in defining the system performance. In approaching a performance strategy for fuel economy, both fuel and air flow will be considered in order to operate the fuel cell system (FCS) efficiently. The system is based on a Load-Following (LFW) controller for load tracking and also a Global Extremum Seeking (GES) algorithm used to minimize fuel consumption. The load following control will ensure the charging sustained mode for the Energy Storage System (ESS), improving their lifetime. The objective of this research paper is to analyze the performance of an FCEV and to implement an energy management strategy for fuel economy in a European Extra-Urban Drive Cycle (EUDC) with the aim of extending the lifetime of the ESS and maximizing of the vehicle's overall efficiency.

ABSTRACT. This paper presents a comparative study on the efficacy of machine learning algorithms in enhancing fault diagnosis in electric motors. Through detailed experimentation and analysis, our research highlights the superior performance of the Extra Trees Classifier in accurately classifying various motor faults. With an impressive accuracy rate of 84%, the Extra Trees Classifier emerges as the top performer among the algorithms tested. Additionally, the Random Forest Classifier demonstrates considerable effectiveness with an accuracy of 82%. These findings underscore the potential of machine learning techniques to significantly improve fault diagnosis and prognostication in industrial machinery. Notably, the analysis showcases the successful classification of diverse fault types, including Normal, Horizontal Misalignment, Imbalance, and various other fault conditions. Moreover, our study emphasizes the importance of ongoing research and development efforts to further refine these algorithms and address any remaining challenges in fault diagnosis within motor systems.

ABSTRACT. In the domain of medical imaging, ensuring the security of confidential patient data stands as a paramount concern. The imperative for robust encryption to safeguard medical images against unauthorized access has driven the widespread adoption of the RSA algorithm, celebrated for its efficacy in data security assurance. This study specifically centers on the deployment of RSA-1024 cryptography on an Artix-7 FPGA, targeting applications in medical imaging. Acknowledging the demand for heightened security and accelerated processing, we have engineered RSA crypto-accelerators as hardware implementations of the RSA algorithm. We present a comprehensive methodology for implementing the 1024-bit RSA encryption/decryption algorithm, leveraging the square and multiply technique for modular exponentiation. Our algorithms are meticulously instantiated through VHDL code, tailored for execution on the Artix-7 xc7a35tcsg324-1 FPGA manufactured by Xilinx. Rigorous testing using the Vivado 2022.2 tool validates the functionality and dependability of our design. This research underscores the pivotal role of advanced cryptographic methodologies in safeguarding medical images and underscores the strides made in FPGA-based security solutions.

ABSTRACT. Wind turbines are currently one of the world's most commonly employed renewable energy generation technologies. The selection of an optimal wind farm site might be challenging considering the variety of site possibilities in the same region. Using the second scenario of a novel subjective and goal-oriented weighting technique called the Vital-Immaterial and Mediocre MCDM Method, this study presents an optimal wind farm site selection methodology, considering three main objectives: biological conservation, safety during production, and durability, which are weighted by a pair-wise comparison based on the opinion of decision-makers. Moreover, this paper gathers eight criteria, including earthquakes and seismic activity. This study also examines four wind farm site possibilities using the TOPSIS MCDM approach. This investigation found identical ranking outcomes between the TOPSIS and COPRAS ranking methods.

ABSTRACT. Sometimes parking practices lead to inefficiency and confusion, for example traffic congestion, drivers searching for vacant spaces, and insufficient revenue management for parking operators. Here the role of smart parking systems is highlighted in overcoming these challenges by providing live updates about the presence of unoccupied parking spots, improving the use of all parking spaces, and completing a convenient payment process. This paper introduces a parking system that incorporates cloud technology for seamless integration (Internet of Things IoT). Proposed system involves deploying an IoT module on-site to monitor and indicate the system offers up-to-date information on the presence of each parking space, and it also offers a mobile application for end users to conveniently check parking space availability and make reservations as needed. The paper also provides an overview of the system architecture at a high level. Furthermore, the paper includes a use that demonstrates the operational functionality of the model being proposed. This use case serves as evidence to validate the accuracy and effectiveness of the system. Overall, this research contributes to addressing issues like traffic congestion, restricted parking options, and roadway safety through the implementation of IoT technology in a parking system that is integrated with cloud technology. This system strives to improve the utilization of parking spaces and elevate the parking experience by providing real time data and convenient booking choices.

ABSTRACT. The increasing adoption of rooftop photovoltaic (PV) power generation systems in residential areas necessitates accurate monitoring and disaggregation of behind-the-meter (BTM) load and PV power. Despite recent advancements, existing BTM disaggregation approaches suffer from three major drawbacks: neglecting task-relevant spatiotemporal features, overfitting, and lack of a sparse neural architecture which leads to high sample complexity. This paper addresses them by introducing a deep sparse attention graph recurrent framework. This framework conceptualizes a set of neighboring residential units as a graph where the nodes are the net load values of the units and the edges show the mutual information (MI) of these measurements. We develop an Attention Gated Recurrent Unit (AGRU) to capture enhanced temporal characteristics of the net load. We employ a novel low-rank Dictionary Learning (DL) method to discern spatiotemporal features of these measurements and further utilize a Rectified Linear Unit (ReLU) neural network that incorporates an MI-based dropout to provide a sparse model for the estimation of the BTM load and PV. Experimental results validate the effectiveness of our proposed model, exhibiting superior performance on the Ausgrid dataset in BTM load and PV power estimation compared to state-of-the-art methods.

ABSTRACT. Content-based recommender systems and collaborative filtering recommender systems might be useful in recommending items that might be preferred by users based on their preferences from the past, these systems rely on users’ preferences, items’ ratings, and items’ descriptions. Thus, these systems might be less useful for items for which users’ preferences and ratings cannot be easily collected. On the other hand, a constraint-based recommender system can suggest items that fulfil users’ requirements (or constraints) by relying on users’ requirements and knowledge about the domain of the items for giving a recommendation. In the case where, no item can be found, the system will proceed with the constraints of a relaxation phase to search for possible modifications in the constraints, give advice to the user, and enable them to edit their requirements to obtain more results. In this study, we propose a new approach called a mixed-based approach. The approach is based on an integration of soft and hard relaxation-based approaches. The proposed approach is implemented as a notebook recommender system. We conducted experiments on the proposed approach using collected data from e-commerce websites. The results also show that the proposed mixed-based approach is able to control constraints, and gives better performance in giving recommendations to users.

ABSTRACT. In modern applications, performing a large number of activities and exchange of information, the architectural approach of microservices is increasingly required. If we make a more detailed analysis of them, they are nothing more than standalone applications that are synchronized in their communication between them, they are interchangeable in certain situations, and they are also divided by functionality. In addition to microservices, the so-called monolithic applications also work with the reception and processing of requests, which architecturally translates into an even greater concentration of functionality in one place and this leads to difficulties in their maintenance. In modern microservice architectures, different methods of communication are often used: microservices often communicate via HTTP with RESTful API design. URLs are used to detect and access microservices. Message Brokers: the use of communications brokers such as RabbitMQ or Apache Kafka provides an effective way of asynchronous communication. Whether we are talking about monolithic or microservice applications, it always comes the moment with the growing code, complex service structure and maintenance. In the current development, we focus on the implementation of a new service in a microservice architecture. The goal of the development is to create an application architecture using design patterns and optimization solutions that lead to fast and efficient operations of the business layer, which remains stable in its performance even with a large amount of data to process.

ABSTRACT. In terms of the digital transformation of the business, the success of the organization largely depends on the integration and construction of a digital ecosystem and digital platform, which are key structures of the digital economy. The focus of the study is on the digital platform as the main technical tool for the formation of conditions for the creation of new products and services by participants in the ecosystem. The article aims to define the positive aspects of digital platforms and to provide recommendations and good practices for their integration into the activities of companies. The study also addresses some gaps in the scientific literature regarding the digital platform ecosystem. The main method of research is the systematic review of the scientific literature on the subject. By applying a structural approach and critical analysis of publications on the topic, conclusions and summaries are made about the potential of the ecosystem approach for business development, the new opportunities they receive from the use of digital platforms, and possible risks and threats that they face. An attempt has been made to identify key steps and tools for the successful implementation of the strategy for the integration of the digital platform, considering the requirements of the dynamic business environment. The research is directed to those companies that are planning a transition to an ecosystem model of development and construction or inclusion in a digital platform. The author’s future research is related to the study of the potential of the second component of the digital ecosystem - the application programming interface.

ABSTRACT. Parkinson's disease (PD) presents significant challenges in early detection, often relying on subjective clinical assessments and costly imaging procedures, leading to low accuracy rates, especially in detecting PD at its early stages. This research investigates the transformative potential of machine learning and deep learning algorithms in PD detection and diagnosis. Leveraging a diverse dataset comprising vocal fundamental frequency measurements, nonlinear dynamical complexity measures, and image data, we explore various ML/DL models. Our findings reveal promising accuracies across different models, with DenseNet121 achieving a maximum accuracy of 96.67% for image data and hybrid architectures like 1D CNNGRU achieving 94.87% accuracy for vocal data. Furthermore, the study identifies key limitations and challenges associated with each model, emphasizing the need for continued refinement and innovation. This research contributes to advancing PD diagnosis, offering potential improvements in patient outcomes and paving the way for more accessible and effective diagnostic solutions.

ABSTRACT. Data Acquisition system using Ethernet for Zynq 7000 SoC FPGA, In autonomous vehicles, data transmission is essential for being reliable and for the efficient operation of the vehicle. In this project we are trying to implement the 10G Ethernet MAC for the data transmission. The Advanced Driver Assistance System (ADAS) in autonomous vehicles now uses CAN networks, which are not very reliable in today's faster environment. The Ethernet is much faster than the CAN. Therefore, Ethernet is the protocol we're employing to achieve this fast data transmission. Analog data is transformed into digital data and stored in FIFO in this data acquisition system. Ethernet is used to transfer the data that is stored for analysis and visualization. The simulation is the primary emphasis of this project. This simulation's output is organized according to the IEEE 802.3 Ethernet standard. This Ethernet output will be of use in ADAS system for increasing the rate of transmission and reliability.

ABSTRACT. In light of the current energy crisis and the environmental damage that fossil fuels have wrought, the generation of electrical power from renewable sources like photovoltaics is emerging as a viable alternative to traditional approaches. Due to the photovoltaic system's low and changeable output voltage, utilizing DC-DC converters with high voltage and great efficiency is recommended. As a result of these converters' nonlinear features, the device becomes highly sensitive to modifications in the circuit since the output voltage varies frequently. Consequently, the nonlinear structure and rapid dynamics of these converters add to the complexity of their control process. In this paper, a two-degree-of-freedom proportional-integral- derivative (2DOF PID) controller is utilized as an enhanced method for controlling an extended high step-up DC-DC converter. Moreover, employing the marine predators' algorithm (MPA) allows the controller parameters to be implemented optimally. This system controls the voltage at the output in accordance with variations in both the input and the load. The proposed controller coefficients are designed to minimize the time-domain performance indices of the closed-loop system, with the ISTAE index acting as the objective function. Finally, the efficacy of the suggested controller has been demonstrated via comparisons with the commonly used PID controller.

ABSTRACT. The increasing demand for operational efficiency and redundancy in electric distribution networks necessitates continuous advancements in power distribution network expansion planning methodologies. Traditional approaches are no longer sufficient due to the integration of distributed generation resources, capacitor banks, and the proliferation of heavy-duty electric trucks. This paper presents a novel approach that integrates dedicated heavy-duty electric truck urban charging stations into modern electric distribution networks. The proposed framework employs joint expansion planning for both electric distribution networks and dedicated heavy-duty electric truck urban charging stations, encompassing constructed elements such as substation construction, circuit reinforcement, distributed generation allocation, and capacitor bank installation. To address computational complexities, the study leverages a novel meta-heuristic algorithm called corona-virus search optimization, inspired by the dynamics of the COVID-19 virus. The model's efficiency and scalability are demonstrated through evaluations using an 18-bus and a 54-bus system. The analysis reveals that depot charging of HDETs presents a more challenging scenario with higher costs. However, integrating dedicated heavy-duty electric truck urban charging stations in Scenario II leads to significant cost reductions (4.34% and 7.56% for two systems). The model's adaptability make it a valuable solution for integrating modern multi-energy electric distribution networks with dedicated charging urban infrastructure.

ABSTRACT. The growing complexity and evolving requirements of Modern Electric Distribution Networks (MEDNs) necessitate a paradigm shift in Expansion Planning (EP) methodologies. Conventional Expansion Planning (CEP) approaches primarily focused on Substation Construction/ Reinforcement (SCR) and Circuits Expansion/Reinforcement (CER) often fall short in addressing the challenges of MAEDNs . This pioneering study introduces a novel Holistic Expansion Planning (HEP) framework moving beyond CEP, marking a significant advancement in EP for MAEDNs. The proposed HEP framework integrates SCR, CER, Capacitor Banks (CBs), Voltage Regulators (VRs), and Distributed Generators (DGs) into a unified optimization process. This comprehensive approach leverages the power of the Osprey Optimization Algorithm (OOA), inspired by the hunting behavior of ospreys, to efficiently solve the non-linear optimization problem associated with optimal EP strategies. The OOA's unique capabilities contribute to the innovative nature of the HEP framework. The HEP aims to minimize total costs while optimizing EP strategies for EDNs. It considers both installation costs of SCR, CER, CBs, VRs, and DGs, along with operational costs associated with power losses. The effectiveness of HEP is rigorously evaluated through a series of case studies conducted on a 24-bus system. The results demonstrate the superiority of HEP compared to conventional SCR and CER approaches. HEP consistently identifies more cost-effective and efficient EP strategies while achieving improved voltage regulation and reduced power losses. This groundbreaking approach to HEP empowers engineers to identify optimal EP strategies for MEDNs that meet system requirements, optimize resource allocation, and enhance overall network performance. HEP paves the way for a more efficient and sustainable future for MEDNs .

ABSTRACT. The energy system that relies on wind turbines contains many disadvantages, the most prominent of which are energy ripples and the quality of the current. The latter is used to produce a doubly-powered induction generator (DPIG). To overcome these problems, it is proposed in this work to use fractional-order third-order sliding mode control as a suitable solution due to its durability and outstanding performance. Ease of implementation and simplicity are the most prominent features of this designed control. In this proposed strategy, internal loops are not used, as the power estimation is used to calculate the power error. The MATLAB software has been used to validate the designed control for multi-rotor wind power (MRWP) systems. The performance of the proposed control is compared with traditional power control of the 1.5 MW DPIG under the reference tracking and variation of parameters. The fractional-order third-order sliding mode control compared with the traditional power control and the numerical and graphical results show the superior robustness of the fractional-order third-order sliding mode control.

ABSTRACT. This paper outlines a maximum power point tracking (MPPT) method based on the tip speed ratio (TSR) of a singular wind power system, operated by a doubly-fed induction generator (DFIG). The conventional MPPT approach using a proportional-integral (PI) regulator faces multiple challenges, including those related to maintaining the system’s stability and performance during steady state. To realize a more efficient variable-speed wind turbine system, MPPT strategies rooted in artificial intelligence, namely type 1 fuzzy logic control (T1-FLC) and type 2 fuzzy logic control (T2-FLC). The designed MPPT techniques effectiveness is validated using numerical simulations executed via MATLAB. The derived results demonstrate the efficacy of our proposed MPPT strategies, with a notable preference for T2-FLC over T1-FLC and the traditional PI control.

ABSTRACT. Microgrids offer a promising solution to the mounting challenges of surging energy demand and environmental issues. This article explores the crucial role of microgrids in addressing contemporary energy needs while mitigating pollution risks. Microgrid operations are intricately influenced by constraints falling into two fundamental categories: those intrinsic to the microgrid and those imposed by the external environment. These constraints, stemming from component limitations, environmental factors, and grid connections, exert significant influence over the microgrid's operational capabilities. Particularly important is the three-tiered control framework, comprising primary, secondary, and energy management controls, ensuring optimal microgrid function by regulating power quality, frequency, and voltage within defined parameters. The energy management control, the third tier, deserves thorough examination, revealing the art of fine-tuning microgrid components and integrating them seamlessly with their environment to streamline energy flow and ensure uninterrupted operation. In essence, this article scrutinizes the intricate interplay between microgrid constraints and energy management parameters, underscoring how subtle adjustments to these parameters are instrumental in achieving the twin goals of cost reduction and CO2 emission minimization, thereby molding a more sustainable and environmentally conscious energy landscape.

ABSTRACT. The proposed circuits are robust for designing energy-efficient full adders, crucial in approximate computing to enhance circuit efficiency. Full adders serve as essential components in arithmetic circuits, crucial for diverse computational devices. Circuits utilizing Metal Oxide Semiconductor Field Effect Transistor (MOSFET) technology are crucial in digital signal processing applications due to their significant role. The Cadence Virtuoso software is utilized to evaluate new approximate adders designed for 45nm CMOS processing with different supply voltages. The assessment focuses on metrics such as power delay product, power consumption, and transistor count.

ABSTRACT. The research explores the hiring and selection processes of Information Technology (IT) firms in Hyderabad, with a focus on state-of-the-art technologies such as Artificial Intelligence (AI), chat platforms, social media, and virtual reality. It seeks to comprehend the strategies employed in talent acquisition amidst Hyderabad's ascent as an IT center. Artificial Intelligence has notably expedited candidate identification and screening, while chat boards and social media platforms facilitate community-building and engagement with potential hires. Virtual Reality adds an immersive dimension to recruitment experiences. Employing a quantitative survey methodology, the study aims to assess the impact of these technologies on talent acquisition within Hyderabad's IT industry. The findings aspire to offer valuable insights for strategic decision-making in the competitive IT arena. Emphasizing the importance of recruitment processes in acquiring skilled personnel, the research examines prevalent practices and challenges encountered by IT companies in Hyderabad. It concludes that effective recruitment practices are widespread in the region. Furthermore, the study sheds light on how leading IT enterprises leverage AI, Chabot’s, Social Media, and Virtual Reality to revamp their recruitment and selection approaches, evaluating the drivers, obstacles, and efficacy of these technologies in attracting top talent and enhancing the candidate experience amidst Hyderabad's evolving IT landscape.

ABSTRACT. In recent years, the Internet of Things (IoT) has significantly impacted the digital economy. This article provides a comprehensive examination and thorough analysis of the architectural framework that supports the IoT ecosystem, tracing its evolution from a complex and fragmented system to a network of interconnected devices with limitless potential. This study explores the various applications of IoT across different sectors, including manufacturing, smart enterprises, city management, healthcare, agriculture, and transportation. It highlights the significant impact of IoT on enhancing efficiency, safety, sustainability, accessibility, and overall quality of life. Although IoT presents promising opportunities, there are persistent challenges, particularly in the areas of distributed processing, data analysis, interoperability, security, and privacy. These challenges demand innovative solutions and efforts across industries and academia. This article aims to provide insights into the ongoing transformative journey driven by IoT technologies, by examining their benefits and drawbacks. The article emphasizes the potential for unprecedented interconnectedness, innovation, and societal advancement on a global scale. The focus is on the use of the IoT solution developed in Python 3 and the server implementation on the Kali Linux operating system.

ABSTRACT. Maximum power point tracking (MPPT) methods for photovoltaic (PV) systems have evolved from traditional techniques such as incremental conductivity (IC) to more advanced artificial intelligence-based methods such as fuzzy logic controllers (FLC). This study systematically compares the efficiency of two MPPT methods, IC and FLC, under different temperature and solar irradiation conditions. The FLC dynamically adjusts the DC-DC converter's duty cycle to ensure accurate MPP tracking, while the IC uses traditional techniques to adjust the system operating point. Analysis in MATLAB/Simulink shows that FLCs consistently outperform ICs in energy efficiency, achieving efficiencies as high as 98% under certain illumination conditions. The results highlight the importance of fuzzy logic-based methods for optimizing photovoltaic system performance, especially in the context of climate fluctuations. This approach provides a fast, oscillation-free response and improves the overall efficiency of MPP tracking in dynamic photovoltaic environments.

ABSTRACT. The healthcare practitioners, especially in India, traditionally rely on manual paper logbooks for maintaining patients' records. Our study underscores the crucial need of transitioning from manual to digital patient health records management. Also in emergency conditions, accessing all the patient's health records becomes challenging since the paper records might be stored in various locations, making it uncertain that all the necessary files are readily available. Additionally, the risk of losing paper records further highlights the vulnerability of traditional record-keeping methods. When consulting different doctors, it is essential to have an easy track on patients’ health records. To address these challenges, we have developed DrFilEz, a web application to streamline and enhance the management of medical records. Our system will enable doctors to store and consolidate health records on a single platform and allow credentialed doctors to access patients' health information seamlessly. On our platform, different doctors and patients can connect to seek advice and share medical knowledge. After completion of our web application, all the functions were tested, and its performance was evaluated. Overall, the evaluation's findings were according to expectations.

ABSTRACT. Since a growing number of parameters in deep learning model occurred, the overhead of inference performance is comparable to training, which promotes to various deep learning frameworks continually optimized for providing better inference performance. As one of the prevalent model tasks in recent years, super-resolution plays an essential role in the development of computer vision. Our study aimed to evaluate the inference performances between two most popular frameworks, TensorFlow and PyTorch. We selected eight representative super-resolution models in deep learning and calculated three metrics, inference latency, Peak Signal-to-Noise Ratio (PSNR), and Structural Similarity Index Measure(SSIM). For enhancing the reproducibility of our experiments, Open neural network exchange (ONNX) was introduced to convert trained models between TensorFlow and PyTorch. We also conducted experiments on both local and cloud platforms to ensure the reliable results. The results on local platform showed that the overall inference latency differences between two frameworks were up to 4%, and the differences in PSNR and SSIM were within 1%. However, noticeable differences persisted between the two frameworks in separate eight models and four datasets. The results of local platform were reproduced on cloud platform. Our findings can provide proposals for end-users when choosing frameworks between TensorFlow and PyTorch.

ABSTRACT. A major step in lung cancer diagnosis is the classification of nodule malignancy, but in early stages, benign and malignant nodules appear very similar, leading to frequent misdiagnoses. This study developed a novel multimodal image-based CNN (MIB-CNN) model architecture to classify pulmonary nodules as either benign or malignant, performing multimodal learning on only computed tomography (CT) images, without the need for other clinical data like genomic tests. MIB-CNN takes in CT images of nodules, convolutionally extracts chosen semantic features from the images to obtain numeric data, and integrates it with the image data using a novel method, improving model performance and uncovering the mechanisms of the “black box” of this deep learning task. The results showed that the MIB-CNN model achieved 0.94 AUC on the LIDC-IDRI dataset compared to 0.90 AUC with a basic image CNN, and 0.91 specificity in comparison to 0.86 specificity of the basic model, indicating a significant decrease in the number of false positives. This study also identifies the primary causes of inaccurate predictions: small airways and other thoracic organs cause noise in the image data and decrease visibility of small nodules. Furthermore, the premise of MIB-CNN is not limited to this lung nodule malignancy classification task, as this methodology can be applied to other medical image-based deep learning tasks to overcome the challenge of limited multimodal data availability.

ABSTRACT. As the usage of photovoltaic is emerging to palliate the degradation of the atmosphere and the natural environment. Photovoltaic (PV) systems appear to be the key for green energy production for households and industrial sectors. However, due to the dynamic change of load for a PV system or the variations of weather conditions, most PV systems are equipped with a maximum power point tracker (MPPT) to operate at their optimum capabilities. However, the efficiency of the MPPT controller varies from the techniques and algorithms used thus affecting the operations of the PV system. This paper proposes a hybrid technique based on Artificial neural network (ANN) combines with perturbation and observation (P&O) or incremental conductance (IC) algorithms using MATLAB Simulink for tracking the ideal maximum power point under uniform and sudden change of weather conditions. A comparison is depicted with conventional techniques, the findings suggest that the hybrid ANN-IC technique has an improved accuracy of 98% and response time 0,154 seconds to MPP under uniform weather. The model presents a fast-tracking response under rapidly changing conditions for a convergence time of 1,013 us.

ABSTRACT. Electricity load forecasting plays a crucial role in the management of electricity power grids, enhancing operational efficiency, ensuring network reliability, facilitating infrastructure planning, and promoting energy sustainability. As the complexity of energy consumption patterns increases, traditional forecasting techniques struggle to accommodate the intricate and nonlinear temporal dynamics characteristics present within the data. This paper introduces a novel hybrid model called RLIDT that merges reinforcement learning (RL) with the deep transformer architecture to address the complicated challenges associated with load forecasting effectively. The integration of RL for hyperparameter optimization within the transformer framework not only utilizes their respective advantages but also provides a dynamic, adaptive model that exhibits versatility, robustness, and enhanced predictive accuracy through continuous learning. The experimental investigations conducted on real-world datasets have clearly demonstrated the remarkable advantage of the proposed RLIDT model when compared to traditional methods.

ABSTRACT. Abstract: As 3D printing technology revolutionizes manufacturing processes, the quest for improving print quality remains paramount. This research paper explores novel avenues for enhancing infill quality in 3D printing by applying advanced machine-learning techniques. Traditional infill patterns often face structural integrity and overall print performance challenges. Leveraging machine learning algorithms, we propose a comprehensive approach to optimize infill patterns based on critical parameters such as printing speed, Acceleration, and material utilization. This study underscores the transformative potential of artificial intelligence in the realm of 3D printing, paving the way for more innovative and sustainable manufacturing practices. The research establishes algorithms to support quality control for optimizing 3D printing parameters, predicting mechanical performance, and evaluating 3D-printed products. We allow the end-user and practitioner to reduce the calibration (ramp-up) time for obtaining the part with good infill quality and improve the productivity of part production.

ABSTRACT. Neurofeedback training (NFT), as a method of self-regulating brain activity, may be beneficial in improving subjective and objective attentional performance or attentional network efficiency. In this pilot study, we investigated how sensorimotor (SMR)/theta NFT or placebo-feedback training (PFT) can improve the attentional performance and the electrical activity of the brain. An experimental program of attention training based on the Electroencephalogram (EEG) neurofeedback method, 5 sessions of neurofeedback training, and a total of 5 weeks was designed. We use response time of attention network test (ANT) and completion time of Schulte grids to measure objective attentional performance, and three attention scales to measure subjective attentional performance. The results showed that subjects in the NFT group had a significant improvement in SMR/theta and SMR power, but not theta power. Whereas PFT did not improve these three rhythms Subjective attention scores improved in the NFT and PFT groups, with significant decreases in response time for both Schulte's square and ANT. The NFT group showed a statistically significant improvement in executive control network whereas PFT did not. NFT could be seen as a useful non-invasive, non-pharmacological tool to strengthen the executive control network, enabling people to improve cognitive flexibility.

ABSTRACT. Abstract— A novel lightweight convolutional neural network architecture is proposed, denotes as VRES-CNN. Unlike most of the actual lightweight architectures VRES-CNN provides a versatile set of hyper-parameters associated with variable number of macro-blocks. Moreover, the macro-block definition is rather simple and allows very good accuracies by emulating nonlinear convolutions. The use of residual connections, dropout, batch normalization and separable convolutions ensure very good accuracies with a relatively small complexity of the model. With a proper choice of the hyper-parameters, it is demonstrated that models with less than 100 kilo parameters and near state-of-the-art accuracy can be identified for a wide range of datasets (with low or high input image resolutions). In terms of accuracy performance, it is shown that VRES-CNN performs in the state-of-the-art range and much better than a recently proposed lightweight architecture (EtinyNet) for the same or smaller model size.

ABSTRACT. A novel circuit for the design of Second Generation Current Controlled Conveyor (CCCII) based Comparator and Binary Amplitude Shift Keying (BASK) modulator are proposed. Comparators find applications in almost every field related to the electronics and BASK find applications in the areas of wireless communications, RFID systems, Infrared (IR) remote controls, Digital Audio Broadcasting, Data Communication over Power Lines, Testing and Prototyping, Control Systems and Industrial Automation. The two topologies are designed using single CCCII and is tested using Cadence Virtuoso simulator tool with 45nm technology files and found satisfactory. Delay and power dissipation are also calculated for the proposed topologies

ABSTRACT. In this article, Harris Hawks Optimization (HHO) is the optimization technique that was proposed to solve economic dispatch (ED) problems. This optimization technique was inspired by one of the most intelligent birds, Harries Hawks. Managing active and reactive power in transmission networks significantly influences their performance. To improve the performance of transmission networks, the mechanisms for determining the location of renewable energy sources (RES) and capacitor shunts (CS) are the most popular and important. Voltage stability index represents the indicator used to select the optimal placement. The optimal sizing of RES and CS has been performed using the HHO algorithm. This paper has presented four scenarios: the initial case (only power flow), optimal power flow using HHO, optimal power flow using HHO after installing CS source, and optimal power flow after installing RES. In these scenarios, two of the most popular objective functions (OFs) was proposed, which are the total fuel cost of generators and emissions issued by fossil fuels to achieve economic and environmental benefits. Integrating a local search scheme improves the search space's capability and enhances the exploration rate, facilitating the discovery of a global solution. To prove the efficiency of HHO performances, the IEEE 30-bus is the standard network that has been applied. The numerical results and simulation confirmed the performance of HHO in searching for global solutions and preventing local solutions. The comparison among the results obtained by HHO and meta-heuristic optimization techniques confirmed the superiority and efficiency of HHO algorithm.

ABSTRACT. High-speed, High-resolution ADCs are increasingly used in communication applications, where the need for dynamic comparator arises. This paper proposes a low power high speed inverter based differential input cascode cross-coupled dynamic comparator. By using the inverter based differential input concept to the existing cascode architecture low power consumption was achieved. Due to increased transconductance the design works at low voltages too. The proposed design and existing designs are simulated in 45nm technology, illustrating that the proposed comparator has a significant benefit in power consumption. It produces a low power of 15.89 pW.

ABSTRACT. Noor 1 is a parabolic trough concentrating solar power (CSP) plant in Morocco, about 10 km to the north-east of Ouarzazate. It concerns the first and largest cylindrical-parabolic mirror complex in Africa. With a nominal capacity of 160 Mwe, the plant covers 480 hectares of solar field. Masen is operating the project, which cost over 7,000 million MAD, while Acwa Power, a Saudi group, owns and operates it on a BOOT basis. Comparative analysis reveals that Noor 1 costs 1.62 dirhams per kWh. Although officially placed at the end of 2015, the plant's commercial operation date is January 26, 2016. In this study two configurations are employed for the simulations, leveraging the System Advisor Model tool to assess each model’s performance. A critical review of these configurations was conducted by comparing the input-output correlations against actual data reported from the Noor 1 plant. This review process allows evaluation of the accuracy and reliability of each model, ensuring the simulated outputs closely match real-world operational data. This correlation analysis provided insights into the models’ robustness, facilitating informed adjustments and optimizations for accurate techno-economic evaluations.

ABSTRACT. In general data science is the prioritization on large size data provisioning multidisciplinary approach towards extraction of meaningful insights in general as well as business applications. In this context, general purpose programming languages have already been thought of in which the basic importance has been provisioned towards code readability with transparency. As the direct involvement may be on multiple programming paradigms, the interpreters associated with the conceptualized programming environment can be provisioned with many operating systems. Somehow due to structured programming approach, the linkages can be associated towards functional programming as well as aspect oriented and logic programming. In many situations, it has been observed that the dynamic implementation can be initiated provisioning specified sequenced references in memory management. Accordingly, the programming languages associated with the features of binding methods can be properly sequenced during program execution. As per maximum extensibilities of some specific programming languages, the programmable interfaces towards existing applications can enhance the feasibilities with the existing applications. In practical situation, analysis of the huge amount of heterogeneous data implementing learning principles is possible to focus on meaningful intelligent applications to maximize the outcomes from each and every stage of application. As it is understood that modeling data with predictive analysis is quite feasible using intelligent agents as well as meta-heuristic approaches, accordingly in this work,the meta-heuristic technique has been prioritized to focus on causal inference with intensification data.

ABSTRACT. The combination of artificial intelligence (AI) and cloud computing has resulted in significant alterations to the admissibility rules for electronic evidence in modern legal situations. Due to cloud-based services, data processing, storage, and access have completely changed, presenting both new opportunities and challenges for the area of electronic evidence. New standards are needed as cloud platforms become more widespread in order to handle issues with data integrity, chain of custody, and authentication in a virtualized and decentralized context. The swift advancement of computer technology and Internet technology in our nation coincides with a growing prevalence of computer network crime. This paper addresses how artificial intelligence and cloud computing are combined and how it affects the admissibility of electronic evidence.

ABSTRACT. FinFET device represents one of the options to settle and to overcome the Short Schannel Effect (SCE). In this paper, a layout based FinFET design approach have been presented at 7 nm node technology using Spice model. The electrical performance and characteristics of both N and P device has been studied to realize the CMOS circuits such as CMOS NOT gate, CMOS NOR and CMOS NAND. An impact parameters of load capacitance (CL), supply voltage (VDD) and gate width (Wgate) has been study on the total power to have optimal CMOS circuit. The simulation results show that the minimum total power can reach with Wgate=16 nm and VDD=0.65 V for CMOS NOR gate. The optimal load capacitance is 0.5 fF for CMOS NOT Gate.

ABSTRACT. In today's technological era, as AI systems become more integral to critical decision-making, the importance of Explainable Artificial Intelligence (XAI) has become more pronounced. It addresses the challenge of understanding complex machine learning and deep learning models, ensuring transparency, interpretability, and accountability. This research paper provides a comprehensive analysis of XAI, focusing on its significance, methodologies, challenges, and future prospects. Theoretical foundations of XAI are elucidated, clarifying key concepts such as interpretability, transparency, and accountability. We differentiate between model-agnostic and model-specific XAI methods, outlining their strengths and limitations. A range of recent XAI techniques, including Local Interpretable Model-agnostic Explanations (LIME), SHapley Additive exPlanations (SHAP), and Gradient-weighted Class Activation Mapping (Grad-CAM), are scrutinized. Through case studies in Healthcare (Pneumonia Classification), Finance (Stock Price Prediction), and Entertainment (Sentiment Analysis), we demonstrate how XAI enhances the understandability and trustworthiness of AI systems. Additionally, a comparative study of all three methods on all three case studies has been conducted, and the results are compared. Challenges such as scalability issues and ethical considerations, including biases and fairness, are discussed. Looking ahead, we offer insights into future XAI research trajectories, aiming to foster public trust and shape a future where AI systems are both intelligent and comprehensible.

ABSTRACT. The cost of establishing renewable power plants is challenging for both public and private entities, as it will have an adverse effect on the economy. This includes the price of procuring and installing the required quantity of solar panels for electric vehicle charging stations with varying capacities, as well as the cost of obtaining electricity from a common network. In the proposed article, the techno-economic feasibility of establishing three different-sized solar power facilities in the parking area of Ufa International Airport for charging electric vehicles was analyzed. The study recommends that prospective investors make cost-effective decisions based on the attractiveness of the acquired financial metrics.

ABSTRACT. As educational institutions increasingly integrate remote (online) laboratories into their systems, the need for secure and efficient access to lab resources becomes paramount. While traditional Virtual Private Networks (VPNs) have historically served as the primary solution for remote access, they are not without limitations. In this paper, we explore how zero-trust networking access solutions (ZTNA) effectively address these limitations. We examine the advantages of ZTNA-based solutions compared to traditional VPNs, discuss the underlying principles of zero-trust, and illustrate a practical application of ZTNA in educational online labs through the utilization of Twingate, a novel ZTNA-based commercial solution.

ABSTRACT. In recent years, the pervasive dissemination of misinformation and deliberately falsified content, commonly referred to as "fake news," has become a critical challenge in the realm of information dissemination. With the exponential growth of digital media and online platforms, the ease and speed at which misleading information can be generated and spread have increased substantially. Addressing the formidable task of detecting fake news has thus become an imperative within the field of information technology. The supervised model gives promising results. However, the availability of labelled data is critical in the realm of fake news detection. Leveraging unlabeled data alongside labelled data is an effective approach, but it requires careful handling. This paper explores the application of Semi-Supervised Learning for fake news detection by integrating Label Propagation and various ML models as base classifiers. By leveraging both labelled and unlabeled data, Label Propagation with CatBoost Classifier seeks to overcome traditional limitations, offering a promising avenue for bolstering the reliability of automated systems in distinguishing between authentic and deceptive information in the vast landscape of digital content.

ABSTRACT. Abstract— In today's modern world, Automatic Number Plate Recognition (ANPR) systems play a pivotal role in various applications, including law enforcement, traffic management, and security. This paper presents a comprehensive ANPR system specifically tailored for recognizing Uzbekistan automobile plate numbers. The developed model integrates advanced image processing and Optical Character Recognition (OCR) techniques to achieve accurate and efficient license plate recognition specifically the Uzbekistan automobile plate numbers. The system's versatility is demonstrated through successful testing on static images and live video feeds, showcasing its potential for widespread deployment.

ABSTRACT. The increased dominance of Internet of Things (IoT) devices has paved the way of improved connectivity and comfort. Since IoT and other technology types are mostly using the Internet as the connectivity medium, the new challenges of cyber security have also emerged. This review paper highlights the significance of Intrusion Detection Systems (IDS) by focusing on those implementations that secure IoT devices and related networks. The paper describes how the IDS have evolved over time and have become a sophisticated tool of securing the network traffic and computer applications. The paper describes the unique challenges of the IoT environment and its distinct features that add additional complexities in building and using IDS. The paper presents a comprehensive and comparative analysis of the techniques of intrusion detection. Different datasets and evaluation metrics have also been discussed and how the edge computing can optimize the system bandwidth and data routing. The paper highlights how the machine learning techniques and the concepts of artificial intelligence have assisted in developing effective IDS and how these technologies can provide an effective threat defense mechanism by implementing anomaly detection and deep learning techniques.

ABSTRACT. We show how to create Artificial Neural Network based models for performing time series exponential-like smoothing and the well-known Holt time series analysis. Our work fares well compared to the well known Holt time series analysis and prediction method, while avoiding the burden of searching for the parameters of the model. We present the theoretical justification of the connection between the two models and experimental results showing the similarities of these models.

ABSTRACT. IoT systems are widely used in modern times. However, they present a weak link in any network, and any infected device can seriously damage organizational networks. The security threats for IoT devices are increasing substantially with the advancement of processing power and the rise of AI. Different approaches are used to detect attacks, including machine learning approaches like SVM and RandomForest. However, the traffic volume, the accuracy, and the complexity of such approaches limit the power of detecting potential malware promptly. In this paper, we propose a CNN-LSTM approach to detect a list of known attacks as well as unknown ones. Traffic behavior is a great source of information for detecting infected devices in a network. We used the IOT-23 dataset, which is an open dataset that contains more than 230 million records labeled with various attacks and normal traffic. The model proposed reached 96\% accuracy. The proposed method ensures faster detection of infected devices, thereby limiting the effect of infected devices on the network.