ABSTRACT. Facial Recognition (FR) plays a crucial role in a variety of fields, such as surveillance, healthcare, security, and robotics. Although several facial recognition algorithms are available, many still struggle to deliver high accuracy and real-time performance due to inherent limitations. Recent breakthroughs in deep learning, especially with the implementation of advanced neural network architectures like Convolutional Neural Networks (CNNs), have significantly enhanced the effectiveness of facial recognition systems. This paper presents a novel system that integrates the Viola-Jones face detection algorithm, the Daubechies Discrete Wavelet Transform (DWT), and CNNs. This integrated approach significantly enhances recognition accuracy, achieving a 98\% success rate and outperforming previous methods that relied on individual techniques for specific image sets.
Grasp Planning in Manufacturing with NAO Robot Using Reinforcement Learning
ABSTRACT. Robotic manipulation plays a pivotal role in modern
manufacturing, where efficient grasp planning is crucial for
automation tasks. This paper presents a novel approach to
grasp planning in manufacturing using the NAO robot and
reinforcement learning (RL). Leveraging RL algorithms, the NAO
robot learns optimal grasp strategies for manipulating various
objects commonly encountered in manufacturing scenarios. The
methodology involves simulating diverse manufacturing tasks,
training the robot in a virtual environment, and evaluating its
performance in both simulated and real-world settings. Results
demonstrate the effectiveness of the proposed approach in achiev-
ing high grasp success rates and efficient manipulation. The
proposed method was successfully tested on a real NAO robot
within the FESTO MPS Station, further validating its applica-
bility to real-world manufacturing environments. This research
contributes to advancing robotic manipulation capabilities in
manufacturing and opens avenues for further exploration in the
integration of RL techniques with robotic systems for industrial
automation.
ABSTRACT. This paper deals with the fractional control of an RC circuit using the non-convex optimization method. The RC circuit is identified by a commensurate non integer order system and then the controller parametrs are tuned taking into account the temporal specifications. The temporal performances are fixed using the comparaison beetwen the caracterstic equation and a target model wich is determined by the caracteristic ratio assignment CRA method
ABSTRACT. In recent years, the benefits of Cloud Computing have placed it at the heart of the architectures proposed for the Internet of Things (IoT). The homogeneous, predictable, and efficient infrastructure has made Cloud an adequate solution for the processing and analysis of data from IoT objects. However, the benefits of using Cloud proved to be problematic for latency sensitive IoT systems that require geographic distribution, consideration of the local environment, and mobility of objects. Fog Computing is an emergent concept aimed at extending Cloud at the edge of the IoT. Thus, it envisions a layer of nodes (Fogs), making it possible to provide connected objects with support for communication management and data persistence.
This paper is motivated by the promising opportunities for the concept of Fog computing. We propose a Fog Computing Framework based on SSN ontology for IoT applications, by applying it on an intelligent parking system which, on the one hand, exploits IoT objects with services, and on the other hand increases the representation and data consumption with local analyzes at the Fog nodes.
For demonstrating, the effectiveness of our implemented framework, simulations have been performed in SUMO, a simulation tool used for urban traffic. Simulation results exhibit that the proposed fog-based implementation reduces the parameters the latency and usage of network compared with that of the cloud-based deployment of the smart car parking system.
A Comparative Analysis of Stereoscopic Vision and Voxel-Based Reconstruction in 3D Object Detection*
ABSTRACT. This research article presents a comprehensive comparison between two distinct methodologies for determining the three-dimensional (3D) axes of objects: stereovision and Voxel-based reconstruction The study aims to explore the strengths and limitations of each approach and evaluate their respective performances in terms of accuracy, efficiency, and applicability.
Embracing the Future: The Evolution of Flexible Reconfigurable Manufacturing Systems.
ABSTRACT. Flexible Reconfigurable Manufacturing Systems (FRMS) are a radical approach to modern manufacturing, focusing on adaptability and efficiency to respond to dynamic market demands. The fundamental principles and advantages of FRMS are discussed in this article. These systems offer a versatile solution to traditional, inflexible manufacturing setups thanks to their modularity, automation, reconfigurability and data-driven decision-making. The key benefits include the ability to adapt to changing production requirements, cost efficiency, improved product quality, and resource optimization. FRMS are found in various industries, such as automotive, electronics, aerospace, and pharmaceuticals, and they meet the needs of customization, precision, and rapid production shifts. FRMS's future looks promising, as emerging trends like AI integration, sustainability, customization, and interconnectivity are poised to further enhance their capabilities. Manufacturers can excel in an ever-evolving industrial landscape thanks to FRMS's position in meeting the challenges of an increasingly dynamic global market as they continue to evolve.
New study of Electro-Magnetic Effects of the power line on the mobile phone handsets
ABSTRACT. In this paper we studied the analytical and experimental effect of the electromagnetic fields produced by the power lines on the mobile phone circuits. Indeed to study this EMC problem two main techniques’ are used; the first one is the image theory which treats the problem from the electrostatic point of view and the second one is inspired from the faraday law to explain the problem as a magnetic circuit leading to a mutual inductances between the electromagnetic fields produced by the power line and the mobile circuit . For these, we introduce, the analytical transmission line modeling of the mobile telecommunication system located in a disturbed EM environment. The effect of this environment on the characteristics of these circuits is systematically clarified in permanent mode. The obtained results, shows the correlation between the presence of the electrical power line system and the electrostatic and magneto static effect on the mobile phone circuits, and to develop the principal characteristics in term of the quasi-static potential, and the electromotive force induced in the multi-conductor transmission line, and we used a plate form of measurement Nemo Outdoor which enabled us to raise a group of indicators: Mobile Station Power (MSP), Received signal power level (RxLevl), Received signal quality (RxQual).
Temperature Control System Enhanced by LSTM Predictive Modeling for Real-Time Monitoring and Adaptation
ABSTRACT. In the ever-evolving landscape of scientific research
and innovation, laboratories serve as the cornerstone for generat-
ing reliable experimental data. Maintaining precise and consistent
temperature control in these environments is crucial for ensuring
the accuracy of research outcomes. However, conventional man-
ual temperature monitoring methods face challenges such as hu-
man error and a lack of real-time insights. To address these issues,
our study introduces the Automated Temperature Monitoring
System (ATMS), a tailored solution designed specifically for
laboratory settings. The ATMS seamlessly combines cutting-edge
sensors, advanced data acquisition mechanisms, and intelligent
algorithms, including Long Short-Term Memory (LSTM) based
machine learning, to provide real-time monitoring, automated
temperature adjustments, and instant notifications. This integra-
tion significantly reduces the reliance on manual intervention,
optimizes laboratory processes, and enhances efficiency and
precision, thereby showcasing the harmonious blend of machine
learning with IoT objectives. Extensive evaluations demonstrate
the ATMS’s exceptional ability to maintain stable temperature
conditions, thereby fortifying data accuracy, elevating laboratory
safety standards, and reducing energy consumption. Further-
more, the ATMS prioritizes cost-effectiveness and streamlined
data management, making it a versatile tool with applications
in healthcare, agriculture, and environmental monitoring. In an
era where laboratories increasingly adopt automation and cloud-
based solutions, the ATMS emerges as a catalyst for scientific
discovery, ushering in a more efficient, data-driven future for
research and innovation by integrating with IoT systems and
incorporating robust LSTM models to enhance temperature
control and predictive capabilities.
Design of efficient QCA module gate with power dissipation analysis used for Optimized Feynman gate
ABSTRACT. Quantum-dot Cellular Automata (QCA) nanotechnology is a new emerging Nano scale circuit using for computing devices. This new concept leads to higher device density, lower power consumption, a faster-switching speed, and multi-function with different re-configuration. In this article we present an efficient module that can be used as three-logic gates at the same time, one with two-input XOR/XNOR gate and the other with three-input XNOR gate. This module uses only 10 cells in an area of 0.006 µm2 with delay of 0.25 Clock. Whereas the three-input Exclusive-OR (XOR) gate consumes 0.0084 µm2 of area with only 12 cells. The proposed circuits gate dissipate less energy and provides a 50.2% consumption gain in comparison with others conventional QCA gate, using QCA Pro tool. Moreover, a novel architecture of QCA design of Feynman gate is realized by using proposed Exclusive-OR with the lowest number of QCA cells and minimal clock delay. This novel design of Feynman gate can improve a gain of about 16% in complexity, 9% in area and 85% in latency, in comparison with best previous work. To validate the accuracy of the proposed circuit, the well-known simulator QCADesigner-E version 2.0.3 was used.
Markov Chains-Based Approach for Unavailability Modeling in Multi-State Systems
ABSTRACT. afety Instrumented Systems (SIS) are critical for ensuring the safe operation of industrial processes. Evaluating their performance, particularly the average Probability of Failure on Demand (PFDavg ), is essential for system reliability and safety. This paper models the performance of a multi-state SIS subjected to proof tests using the multiphase Markov chain method. A generic approach based on Markov chains is employed
for the assessment of the unavailability of multi-state systems.
The unavailability of the 1oo1 structure is used to determine the performance of different subsystems of the complete SIS. A case study is presented demonstrating the impact of proof test parameters, such as the proof test coverage rate and the
probability of failure due to the test, on the overall reliability.
Detection of Authors’ Personalities: Application to Detecting Psychological Disorders on Social Media Based on RNN
ABSTRACT. The proliferation of social media offers a unique opportunity to analyze user behaviors and personality traits. This paper explores the application of deep learning for detecting author personality traits, with a specific focus on identifying psychological disorders. We present a methodology that integrates advanced natural language processing (NLP) techniques and deep neural networks to analyze texts posted on social media. Using annotated datasets, we train and evaluate various deep learning models to predict indicators of psychological disorders such as depression, anxiety, and bipolar disorder. Our results demonstrate that deep learning approaches outperform traditional text classification methods, offering promising tools for large-scale mental health monitoring. This study highlights the ethical and practical implications of automated psychological disorder detection, while proposing ways to improve the accuracy and reliability of predictive models.
ABSTRACT. Diffusion models are a recent topic in generative modeling, with remarkable results in the field of computer vision. The idea of modeling the joint probability distribution of input and output data is the basis of generative diffusion models. They generate new data using random samples at each step using an iterative process. This paper is a survey on diffusion models. It introduces the mathematical background behind diffusion models especially the Denoising Diffusion Probabilistic Models and the different categories of diffusion models.
ABSTRACT. Wildfires pose a significant threat, necessitating robust fire detection systems. Deep learning-based techniques, particularly those utilizing object detection models like YOLOv8, offer a promising alternative for fire and smoke detection. In this paper, we present a YOLOv8-based method for fire and smoke detection in images. We constructed a comprehensive dataset and trained a YOLOv8 model to achieve high accuracy in detecting fire and smoke. The results suggest the effectiveness of one of the recent models of YOLO.
Early Prediction of Depression on Social Networks using BERT Model
ABSTRACT. This research aims to build an advanced system that
leverages deep learning to predict signs of depression based on
social media posts. By utilizing BERT, a powerful neural network
model, our system can identify subtle indicators of depression
in online content. This method seeks to significantly enhance
the accuracy of mental health assessments, providing timely and
effective support to those who may be at risk. Our work aspires
to contribute to the field of mental health technology, offering
essential tools for clinicians, researchers, and mental health
professionals. By applying the latest deep learning techniques,
we aim to bridge the gap between technology and mental health,
creating new ways to detect and address mental health issues
early.
Multi-Scale Convolutional Neural Network with Attention Mechanism for COVID-19 Detection from CT Images
ABSTRACT. The COVID-19 pandemic has spurred urgent developments in medical imaging technologies for accurate and timely diagnosis. Computed Tomography (CT) imaging, in particular, offers detailed insights into lung abnormalities associated with COVID-19 infections. This paper introduces a novel Multi-Scale Convolutional Neural Network with Attention Mechanism (MS-CNN-Att) designed to enhance COVID-19 detection accuracy from CT images. The model integrates multi-scale feature extraction capabilities with attention mechanisms to highlight relevant patterns indicative of COVID-19 pathology. Experimental results demonstrate the efficacy of the proposed approach in achieving high diagnostic accuracy, thereby supporting clinical decision-making and public health management strategies during the ongoing pandemic.
Aerodynamic and static study of flows around a 3 kWc photovoltaic station
ABSTRACT. A numerical analysis using Computational Fluid Dynamics methodology was carried out to study the effect of different horizontal spacing distance between the modules on wind loads of a solar panel. The wind flow is simulated as turbulent, incompressible flow with a free stream velocity reaching speeds of up 33.3 m/s and a wind direction of 180 degrees. The different spacing; λ including the values {0, 2, 4, 6, 8, 10 mm}, is used to determine the optimal spacing between the solar panels horizontally. The finite-volume method is employed to solve the governing equations, coupled with the k─ε turbulence model with standard-wall treatment. The results indicate that velocity vector and the wind force on the panels decrease with the horizontal spacing increase.
Experimental investigation of the thermal performance of a panel type solar cooker using a finned cooking vessel
ABSTRACT. In this work, a new panel solar cooker prototype was manufactured and tested. The cooker has a multifaceted reflector to concentrate solar radiation on a cooking pot that is placed inside a plexiglass enclosure to create a greenhouse effect. The prototype allows both an azimuthal and zenithal manual orientation. The benefit of adding copper fins to the cooking vessels in a solar cooker is investigated in this experimental study.
A test bench used to characterize the cooker performance is then described. Experimental tests without load were carried out to evaluate the maximum cooker temperature and to determine the first of figure F1. Tests with load, conducted using cooper vessel containing 3 loads of water (0.5kg, 1kg and 3kg) were accomplished. A characterization of the performance of this cooker including: the time duration required for achieving load temperature from T_1=40°C to T_2=90°C , second figure of merit and overall average efficiency is established. The effect of the load is evaluated in terms of second figure of merit and overall thermal efficiency.
Investigating the impact of sensible and latent heat storage materials on a single solpe solar still's performance
ABSTRACT. The classic solar still has relatively little production in terms of distilled water. Heat storage materials offer great opportunities to improve their performance. Brackish water is desalinated using a passive solar still and a heat storage material.
A heat storage system combined with a solar distiller is a viable desalination technology for arid and semi-arid locations that face acute freshwater scarcity while having an abundance of brackish water.
Numerical study: Impact of Collector Fin Quantity on Wind Capture System Performance
ABSTRACT. Wind power is a renewable energy source that
harnesses natural airflow to produce electricity. Advanced
wind capture systems, such as the one studied here, aim to
improve efficiency to maximize energy conversion. This
paper presents a numerical study of the influence of the
number of inlets or the number of fins of the wind collector
on the performance of the system, focusing on maximizing
the wind speed in the cylindrical part. We will start by
simulating different configurations with varying numbers
of fins, analyzing how this affects the wind capture ability
of the system. We will then examine how this wind capture
capacity relates to the amplification coefficient, which
measures the increase in wind speed through the system. By
conducting this study, we will determine the optimal
number of blades to maximize wind capture efficiency and
wind speed amplification compared to the baseline system.
This will allow us to optimize the system design for optimal
performance in various wind conditions
An enhanced evolutionnary algorithm for engineering design
ABSTRACT. This study deals with an enhanced version of harris’ hawks (HHA) approach, named EHHA. The proposed EHHA has been developed and investigated to resolve high dimensional and constrained engineering design problems. A new method for the transition stage, which is called enhanced transition stage, is implemented to improve the efficiency of the algorithm to enhance the global search capability and to find the optimal solution. In order to assess the efficiency of EHHA, spring design, welded beam optimization problems and high dimensional test functions are used. The results illustrate that for both engineering applications and high dimensional test functions, the proposed EHHA outperforms other known algorithms in terms of exploitation capability.
Numerical study of geometrical parameter’s effect on the optical performance of parabolic dish system
ABSTRACT. In this study, a Monte Carlo ray-tracing method is applied to predict the radiation flux distribution on the receiver periphery of a parabolic dish. This software allows us to develop heat flux map for every configuration of solar systems.
The impact of several parameters is studied in order to define the most efficient configuration of the parabolic dish.
The effect of the receiver placements at different distances from the dish vertex has been simulated. Six different distances D are tested: D= 1.8 m; D=1.81 m; D=1.82; D=1.84 m; D=1.85 m and D=1.86 m, respectively. It has been found that the solar heat flux increases with decreasing of the receiver placements.
Furthermore, the effect of the diameter of the receiver on the radiation flux distribution has been investigated. The parameter is studied in the following ranges: diameter of the receiver from 0.5 to 3 m. The results show that the solar heat flux increases with decreasing of the diameter of the receiver.
Internal Model Control for Non-Minimum Phase Systems with Uncertainties Using Kharitonov’s Theorem
ABSTRACT. This paper addresses the tracking control problem of uncertain continuous systems. We introduce an innovative robust extension of internal model control. The resulting intelligent control scheme is termed genetic algorithm-based internal model control. To assess its effectiveness and robustness, we conducted several scenarios under various operating conditions on the coupled-tank system.
A Spacecraft Attitude Stabilization Using State Feedback Control
ABSTRACT. This paper presents a comprehensive approach to the design and implementation of state feedback controllers for the stabilization of spacecraft attitude. We begin by formulating the spacecraft's dynamic and kinematic models, which describe the evolution of attitude angles (roll, pitch, yaw) and their corresponding angular velocities. The design of the state feedback controller is achieved using the pole placement method, where the gain matrix K is computed to position the closed-loop poles at desired locations in the complex plane. This approach ensures robust stability and precise control of the spacecraft’s attitude.
The effectiveness of the proposed control strategy is demonstrated through simulations, which show that the attitude angles and angular velocities converge to desired values efficiently. Our results confirm that state feedback control is a powerful tool for achieving stable and controlled spacecraft orientation, highlighting its practical applicability in space missions.
A new elliptical shape antenna design for the 5GHz wireless communication application
ABSTRACT. In this paper, introduces new patch antenna geometry for 5G wireless communication application. This antenna was studied by the CST simulator for the 3.5 GHz frequency band. The prototype was placed respectively on different FR4, RO4003C and RT5880 substrate with having relative permittivity of 4.3, 3.55 and 2.2. The simulation results present the electrical and radiation characteristics of the proposed antenna. The reflection coefficient obtained reaches a maximum value of -45dB with a gain obtained up to 5.94dB. The comparative study of the different permittivity’s showed that the effectiveness of that the antenna designed by Roger RT5880 offered a better percentage of order 96%, compared to those given by FR4 and RO4003C which were equal to 87% and 89%.
Numerical investigation of a noval 10.5 kWc Staircase-Shaped PV Structure
ABSTRACT. Wind-induced forces and vortices pose a significant
challenge to the structural integrity and efficiency of photovoltaic
(PV) panels. This study presents a novel PV structure designed to
mitigate these wind effects. Computational analysis is employed to
evaluate the effectiveness of the proposed structure compared to
conventional designs. The results demonstrate a significant
reduction in wind load and a decrease in vortex formation around
the panels. This innovative structure offers a promising solution to
improve the durability and performance of PV systems in windy
environments.
Optimal design of lower limb knee articulation mechanism
ABSTRACT. The purpose of this study is to optimize the design of a crossed four-bar knee mechanism capable of replicating the complex flexion and extension movements of the knee joint in the sagittal plane. An optimization method is proposed, and the results demonstrate the feasibility of achieving minimal tracking error. These findings highlight the effectiveness of the suggested optimization process for the optimal synthesis of a crossed four-bar knee mechanism, outperforming methods found in the literature.
ABSTRACT. The implementation and diagnosis of industrial systems are essential for optimizing production. They involve integrating advanced technologies and quickly resolving issues to maintain high efficiency and quality output. This article entitled "Implementation and Diagnostic of a Production Line " aims to improve existing systems and at the same time to seek solutions to actual problems. This system is an industrial production line dedicated to chemical products that contains a diagnostic system for fault detection.
Investigation of Green Hydrogen Production Chain Efficiency Based on Field Experience
ABSTRACT. Algeria emphasizes producing and utilizing low-carbon and renewable hydrogen as part of its economic recovery plan, driven by its abundant energy resources, particularly solar potential, and its commitment to achieving energy transition goals. This work analyses an experimental investigation into the synthesis of solar hydrogen through the indirect connection of solar PV modules and PEM electrolyzer HG-60 in the Ouargla region, Algeria. Initially, we examined the performance of the system elements and then evaluated the Solar-hydrogen production chain efficiency. Our main findings indicated that during the experimental periods from 08:00 to 17:00, the PV energy production attained 2668.3 Wh with an efficiency of about 11.6 %. Moreover, the electrolyzer production rate varied from 19.2 to 37.5 L/h, resulting in a hydrogen volume of around 273.8 litres. Hence, the total power management unit efficiency attained 64 %, while the electrolyzer and integrated rectifier efficiency was 36.3 %
Real-Time Pieces Detection and Recognition by the NAO Robot Using Tiny-YOLOv4/3L Algorithm
ABSTRACT. This research paper presents a system for object detection using the Tiny-YOLOv4/3L model, with a focus on a humanoid robot NAO. The system is designed to detect specific objects under various conditions. The efficacy of the model was evaluated based on its performance under standard conditions and under common real-world scenarios such as occlusion and low light. The results demonstrate that the model can provide reliable detection with commendable accuracy under standard conditions and shows a degree of resilience against minor occlusion and low light. However, significant occlusion and dim light conditions pose considerable challenges to the performance of the model.
Detecting and Mapping Forest Loss In TABARKA and AIN DRAHAM Using Satellite Image Classification and Spatial Analysis
ABSTRACT. This paper examines the detection and mapping of forest loss in the Tabarka and Ain Draham regions using satellite image classification techniques and spatial analysis methods.The study spans from 2016 to 2022,validating classification results based on Recall,UAccuracy,and Kappa index.We used the Apriori algorithm and the elbow method to identify relevant spatial patterns contributing to deforestation, such as slope, proximity to water bodies,dams,roads,and villages.The use of Geographic Information Systems (GIS) was crucial for visualizing and analyzing geospatial data.This paper proposes perspectives for improving classification techniques, real-time monitoring, and interdisciplinary collaboration to strengthen forest conservation efforts
ABSTRACT. This paper proposes an arrhythmia classification system that combines 1D Convolutional Neural Network with three selected temporal features: QRS duration, and the current and next RR intervals. Our aim is to leverage the advantages of deep learning, which allows for automatic feature extraction, while also utilizing some crucial hand-crafted features. We test the performance of this model with the MIT-BIH Arrhythmia Dataset, adhering to the guidelines provided by the Association for the Advancement of Medical Instrumentation. The proposed hybrid approach demonstrates good performance in terms of accuracy, sensitivity, and specificity, while maintaining a simpler structure
An Extensive Survey of Self-Adaptation of Web Service composition based on Middleware Architectures
ABSTRACT. Web service Compositions are rapidly gaining acceptance as a fundamental technology in the web field. They are becoming the cutting edge of communication between the different applications all over the web. With the need for the ubiquitous computing and the pervasive use of mobile devices, the self-adaptation web service composition becomes a hot topic. This latter aims to adapt the web service composition behavior to the user‘s context such as his specific work environment, language, type of Internet connection, devices and preferences. This paper provides a survey of Self-Adaptation of Web Service composition based on Middleware. An overview of each middleware is provided, along with the description of the main features. Based on the conducted survey, this paper compares and contrasts the various characteristics of context-aware middleware architectures. We present the analysis of the middleware architectures based on several parameters including fault tolerance, adaptability, interoperability, architectural style, discoverability, location transparency and aspect oriented composition.