Keywords: Image processing, Modeling and simulation, Control algorithms implementation
Abstract: Skin cancer is one of the most widespread and deadly types of cancer. Dermatologists primarily diagnose this disease visually. The classification of skin cancer into multiple categories is challenging due to the precise variation in the appearance of different diagnostic categories. On the other hand, recent studies have shown that convolutional neural networks outperform dermatologists in classifying multi-category skin cancer. In this work, a straight forward methodology was adopted with the aim of attaining high performance at a low cost. The methodology encompassed three stages: image resizing, normalization, and, finally, the classification of the seven types of skin cancer. Through training the convolutional artificial neural network on the HAM10000 dataset and subsequently subjecting it to testing, a performance rate of 78% was achieved by the model.

Keywords: Image processing, Optimization
Abstract: Computer-aided diagnosis (CAD) stands out as a prominent area of research within diagnostic radiology and medical imaging. This paper introduces an optimized CAD system employing the Multi-Objective Optimization (MOO) approach for the classification of brain Magnetic Resonance Images (MRIs). The proposed methodology leverages an Optimized Kernel Principal Component Analysis (OKPCA) combined with an Artificial Neural Network (ANN), referred to as the OKPCA-ANN scheme, to discern between pathological and normal brain images. The study unfolds across three key stages: initial data acquisition and preprocessing, followed by the dimensionality reduction phase, and concluding with the classification stage. To enhance the generalizability of the scheme, we incorporate the k-fold cross-validation technique, facilitating adaptation to diverse benchmarks. Our dataset encompasses images representing seven brain diseases, namely Huntington's disease, sarcoma, glioma, Alzheimer's disease, Alzheimer's disease plus visual agnosia, meningioma, and Pick's disease, all sourced from the 'Harvard Medical School' website. The experimental results underscore the exceptional efficiency of the proposed technique within CAD systems, affirming its robustness and potential for application in various medical imaging scenarios.

Keywords: Image processing
Abstract: Alzheimer’s disease is a debilitating neuro-logical condition impacting millions globally. Therefore, an accurate diagnosis plays a major role in treating or managing it effectively. Convolutional neural networks (CNNs), popular deep learning algorithms applied to image processing tasks, offer such an opportunity in this study where a CNN model for classifying Alzheimer’s patients was proposed. The research yielded impressive results: recall and precision scores as high as 0.9958 which indicate trustworthy identification of true positives while maintaining few false positives; test accuracy exceeding 99% confirming desirable generalization capabilities from the training dataset to live scenarios; ROC AUC score at an astronomical height of 0.9999 signifying great potential in distinguishing between afflicted individuals from their non-affected counterparts accurately. The proposed network achieved a good classification accuracy. Comparing the proposed model alongside two state-of-the-art models VGG19 and ResNet50 resulted in superior performance for the proposed model demonstrating CNNs can increase diagnostic precision and effectiveness of Alzheimer’s disease, leading to early detection, enhanced treatment plans, and enriching the quality of life for those affected.

Keywords: Image processing
Abstract: Breast cancer stands as the second leading cause of cancer-related fatalities among women and is the most prevalent form of noncutaneous malignancy worldwide, affecting more than one in ten women globally. Ongoing efforts in the development of novel approaches and strategies are dedicated to enhancing breast cancer prevention, elevating survival rates, and reducing mortality. This study introduces an innovative technique for breast cancer detection and classification known as Tabu Search and Kernel Principal Component Analysis (TKPCA). In implementing TKPCA, a Recurrent Neural Network (RNN) is utilized to construct a classifier named TKPCA-RNN, proficient in distinguishing between benign and malignant tissue within medical images. The Wisconsin Breast Cancer Dataset (WBCD) and Wisconsin Diagnosis Breast Cancer (WDBC) are chosen as fundamental breast cancer databases sourced from the UCI benchmark repository. Notably, experimental findings validate the superiority of our proposed TKPCA method over established classifiers, representing a significant breakthrough in breast cancer research.

Keywords: Image processing
Abstract: Handwritten digit recognition is a significant challenge in the field of machine learning, particularly for pattern recognition and computer vision applications. It has found applications in various areas, such as identifying digits on utility maps, postal mail zip codes, bank check amounts, and more. Offline handwritten digits possess distinct characteristics, including size, orientation, position, and thickness. Each person's handwriting is unique, making the classification process more difficult. Additionally, the high similarity between certain digits and the over fitting problem with high-dimensional data can increase computational time and cost. Consequently, numerous researchers have developed and implemented different machine learning algorithms to effectively address the issue of recognizing handwritten digits. This paper introduces a novel approach to improve the classification performance and accuracy of handwritten digit recognition. The proposed method utilizes an ensemble majority vote classifier, which combines three classifiers in each experiment by utilizing a range of different algorithms; Naïve Bayes (NB), Adaptive Boosting Algorithm (AdaBoost), K-Nearest Neighbors (K-NN), Multi-layer Perceptron Classifier (MLP), eXtreme Gradient Boosting algorithm (XGBoost), and Decision Tree Algorithm (DT) to form a single ensemble classifier. To validate the approach, we used the MINST free public handwritten dataset, to ascertain the classifier with the highest rate of accuracy in this study. The results obtained from the experiments demonstrated that the full performance of Ensemble model classifiers MLP, XGBOOST, and DT is superior to AdaBoost, K-NN, and NB. It achieved the highest accuracy level, reaching 98% in the experiment. The experimental results of this

Keywords: Image processing
Abstract: Computer vision for automated tracking is one of the topics of research and knowledge acquisition on migratory behavior. In this research, some elves were marked with four colors: red, orange, blue, and green, with single or double markings, and then introduced into an experimental medium and, a large database was collected by collecting many video observations. Elvers is segmented by background subtraction method and their motion information is extracted by connecting component classification. Elvers tracking is carried out by a Kalman filter, in order to improve the detection through local search focused on location prediction and path filtering to reduce estimation error.

Keywords: Image processing
Abstract: This research paper explores the utilization of CNNs for diagnosing Alzheimer’s disease, delving into techniques that enhance their interpretability and decision-making process. Feature visualization and class activation mapping provide insights into how CNNs detect the disease, emphasizing the importance of interpretability in medical imaging. The methodology involved skull stripping to isolate brain structures by removing non-brain tissues, followed by skull cropping to focus on the region of interest, improving the CNN model’s accuracy and efficiency. Through training on the preprocessed dataset, the research achieved an impressive 99.92% accuracy in Alzheimer’s disease detection, highlighting CNNs’ potential for diagnosis. The study underscores the significance of understanding neural network decision-making, offering valuable insights into deep-learning models for medical imaging tasks. These findings contribute to improved Alzheimer’s detection methods and enhance our understanding of applying deep-learning models in medical imaging, benefiting other diagnostic imaging tasks and advancing disease detection.

Keywords: Power systems, Control applications, Fuzzy and neural systems
Abstract: The automatic load frequency control for multi-area power systems has presented a significant challenge for power system engineers since It requires to design a robust controller for maintaining the balance between the generation and consumption of electricity within each area, as well as the entire power grid. This study focuses on the load-frequency control of an interconnected two-area power system, aiming to minimize the area control error through adaptive control method. Specifically, an adaptive linear neural network known as ADALINE is employed to dynamically adjust the parameters of a PID controller in real-time. This approach effectively governs both the frequency deviation and tie-line power flow deviation at the interconnection point. The outcomes of the study revealed that the employing of the adaptive LNN-PID controller results in reduced frequency deviations, small overshoots and minimal oscillations. Furthermore, the power system demonstrated more robustness against wide range of operational conditions, even when faced with significant load changes and system uncertainties. These findings underscore the effectiveness of intelligent control strategies in optimizing the performance of interconnected power systems.

Keywords: Power systems, Control applications, Renewable Energy
Abstract: Temperature control is a key factor for the proper and efficient operation of alkaline electrolysis systems. Alkaline electrolysis systems are considered as time-delay systems in the design of PID to control the temperature. In green hydrogen production, the electrolysis is coupled to an intermittent renewable energy source. This means that the current applied to the electrolysis is not fully controlled, leading to overshoot and undershoot of the cell temperature, impacting performance and even lifetime. Model-free control is applied to alkaline electrolysis to reduce the overshoot phenomena observed when the system is controlled by a PID. Also to achieve faster stabilization time of stack temperature. Simulation results show that model-free control ensures a stabilization time that is at least 15 min shorter than that of the controller used in comparison with a low level of overshoot.

Keywords: Power systems, Modeling and simulation, Control applications
Abstract: The incremental demand for electrical power increases power system complexity and the need to expand the power grid which is very expensive, also the changes in load demand during the day can affect the power quality parameters especially, the voltage amplitude. Flexible AC transmission systems FACTS can be presented as the solution to expand the power systems transmission lines capacity and reactive power compensation to regulate voltage in the power system. this paper tries to attract attention to the use of FACTS on Iraq’s power systems for power losses reduction and voltage regulation, by studying the device effects in two different ways, first is using Newton-Raphson power flow and loss calculations with and without FC-TCR, and the second is to simulate the device performance in MATLAB/Simulink to investigate its performance in voltage regulation. Finally, calculate the economic benefit of power loss reductions by using FC-TCR during base caseload, the device saves 192,428 per year by using it only 10 hours daily, and reduces voltage drop in point of common connection from 0.6% to 0.1%.

Keywords: Power systems, Modeling of complex systems, Optimal control
Abstract: In the context of the energy transition and the imperative to reduce greenhouse gas emissions, promoting electric mobility has become a priority. Electric vehicles (EVs), primarily reliant on the power grid for battery charging, pose potential consequences if not carefully managed. To meet this challenge, the integration of EVs charging stations powered by photovoltaic panels (PVCS) appears to be a promising solution. This work contributes to understanding and improving the power management for a PV-powered DC microgrid (MG) dedicated to recharge EVs, enabling more efficient use of energy resources and better integration of PV charging infrastructure into the existing power grid. The objective is to investigate a two layers power management algorithm. The first layer is dedicated to managing the power flow in the MG respecting the physical constraints of the system. The second layer focuses on modulating the charging profile for the EVs. The charging profile is determined based on the EVs characteristics communicated by the users, such as parking time, charging mode, initial and desired departure state of charge. For this, several scenarios are presented and analyzed. Analysis of these scenarios enables to determine the benefits of parking time knowledge for optimizing the use of PV energy, to reduce dependence on the power grid, and illustrating the impact of charging mode on the participation rate of PV production.

Keywords: Power systems, Optimization, Modeling and simulation
Abstract: This paper proposes a novel model for the Distribution Network Reconfiguration (DNR) Problem. The DNR consists of altering the state of the switches (opening or closing them) in the network to obtain a different topology, thus modifying the flow of power in the network. This optimization problem is often formulated to minimize the losses in the network and can be combined with other objectives as well. Our study introduces a novel Mixed Integer Linear Programming (MILP) model that includes the losses of the lines in the power balance constraints. The efficacy of the proposed model is assessed with the use of two variants of the IEEE-33 bus network, the standard one as well as a modified version where Distributed Generations (DGs) are added, and a simple three-bus network.

Keywords: Power systems, Renewable Energy, Modeling and simulation
Abstract: This work describes a comprehensive study on modeling and controlling a grid-connected Photovoltaic (PV) system .The proposed system consists of various components, including Photovoltaic generator to absorb and convert sunlight into electricity, an adaptation stage composed of a boost DC-DC converter and a DC-AC full-bridge inverter to converting alternating current (AC) electricity. The integrated control strategy is based of two main controllers: The technique of algorithm Maximum Power Point Tracking (MPPT) and Control of the active and reactive powers is performed using the direct power control (DPC) strategy. The objective of DPC approach is to supervise the instantaneous active and reactive powers with the purpose to regulate the DC bus voltage at the desired status, often employing the PI controller in the regulation loop. To experimentally test the proposed strategy a dSPACE 1104 was implemented. The simulation and experimental results obtained confirm the performances of the proposed technique.

Keywords: Power systems, Renewable Energy, Modeling and simulation
Abstract: The ultimate function of the electrical Grid is to ensure a balance between production and consumption. Unfortunately, in the case of excess production, the electrical grid is at risk of collapsing. Indeed, this paper presents a study on improving the electrical grid through the integration of renewable sources (photovoltaic solar and wind). These sources are known as decentralized productions. Renewable sources, particularly photovoltaic solar systems (PV), have been experiencing exponential development in recent years and offer numerous advantages. They are inexhaustible and environmentally non-polluting, also referred to as green energy. However, their integration into the electrical grid is not without consequences. It changes the profile of voltages, power flows, energy quality, etc. In this context, the PSAT tool in Matlab was used to calculate power flow and to study the stability of the electrical grid. To evaluate various electrical quantities of the IEEE 30 Bus system, both with and without decentralized sources, the "Newton-Raphson" algorithm and continuous power flow technique were employed. This allows us to observe the improvement of these sources on the electrical grid (voltages, power loss, …).

Keywords: Optimal control, Control applications, Linear and nonlinear systems
Abstract: In this paper we provide an Optimized feedback color control of multi-channel LEDs Lighting System. The proposed feedback color control enables the maintenance of desired lighting ambiances in rooms, accounting for external lighting sources and mitigating disturbances. The system incorporates a 18 channel spectral sensor, a MIMO PI controller, an antiwindup algorithm, an efficient optimization algorithm and an array of colored LEDs. Thanks to its excellent performance, the experimental results validate the algorithm.

Keywords: Optimal control, Control applications, Modeling and simulation
Abstract: We address a control-constrained optimal control problem pertaining to the transformation of quantum states. Our objective is to steer a quantum system from an initial state to a desired target state while adhering to the principles of the Liouville-von Neumann equation. To achieve this, we introduce a cost functional that balances the dual goals of fidelity maximization and energy consumption minimization. We derive optimality conditions in the form of the Pontryagin Maximum Principle (PMP) for the matrix-valued dynamics associated with this problem. Subsequently, we present a time-discretized computational scheme designed to solve the optimal control problem. This computational scheme is rooted in an indirect method grounded in the PMP, showcasing its versatility and efficacy. To illustrate the practicality and applicability of our methodology, we employ it to address the case of a spin 1/2 particle subjected to interaction with a magnetic field. Our findings shed light on the potential of this approach to tackle complex quantum control scenarios and contribute to the broader field of quantum state transformations.

Keywords: Optimal control, Control applications, Control algorithms implementation
Abstract: Battery management systems (BMSs) play a crucial role in the reliability of lithium-ion batteries in electric vehicles. A BMS aims to maximize the efficiency and lifespan of the battery cells in the managed battery pack while ensuring the electric vehicle safety. This paper proposes a novel methodology of the BMS based on safe sets. A safe set of the Li-Ion battery cell is the set of admissible states and corresponding admissible currents on a fixed prediction horizon. Thus, the methodology based on safe sets allows one or more current changes, i.e., constant or multidimensional current profiles on the prediction horizon that maintain the posed cell constraints. Furthermore, the paper offers a computational procedure for the safe set approximation for which the nonlinear dual-polarization equivalent circuit model (ECM) of the cell is used. Due to the non-convexity of the safe set of the battery cell, the safe set is constructed as a union of single polytopes for different ranges of state of charge. The proposed methodology is carried out throughout a case study by comparing safe sets with one-, two-, and five-dimensional current profiles.

Keywords: Optimal control, Intelligent and AI based control, Estimations and identification
Abstract: In this work, an optimal control policy that stabilizes an Euler-Lagrange system of unknown dynamics and satisfies predetermined response criteria is obtained. The proposed methodology consists of two stages. In the first phase of the algorithm, a neural network is trained online via an iterative process that captures the unknown system dynamics. In the second phase, based on the acquired dynamics from the previous stage, a successive approximation algorithm is applied to find a near optimal control law that takes into consideration prescribed performance measures, such as convergence speed and steady state error. In this way, we manage to induce optimality in the prescribed performance control technique and improve its operation with respect to a state and input related integral cost. Our claims are validated through a simulated paradigm that confirms the success of both the identification process and the minimization of the cost function.

Keywords: Optimal control, Motion control, Robotics
Abstract: This work is mainly focused on the optimal motion planning problem of a Wheeled Inverted Pendulum (WIP) system, which is one of the benchmark mechanical systems used for the study of nonholonomic systems. Designing motion planning control law while respecting the nonholonomic constraints possesses a challenging control problem. We address the problem where the given system has to reach any desired target configuration from an arbitrary initial configuration by finding an optimal path which takes into account the path curvature, control effort as well as obstacles in the environment. Throughout the course of motion, it is desirable to maintain the pendulum body at the upright position. To tackle this problem from a variational point of view, Pontryagin's Minimum Principle (PMP) is utilized for deriving the necessary optimality conditions which finds a collision free optimal path with a reduced control cost satisfying the path smoothness constraints defined via an objective function. Numerical simulations are carried out on the derived nonlinear WIP model to validate the proposed theoretical results.

Keywords: Optimal control
Abstract: This paper presents an optimal control approach for nonlinear switched systems. By employing a novel metaheuristic optimization method called Fire Hawk Optimizer (FHO). Thus, a hybrid optimal control strategy based on FHO algorithm is proposed. This approach allows reducing a performance criterion depending on the optimal switching instants. Lastly, we propose a hydraulic two-tank system to prove the advantages of the suggested method.

Keywords: Optimal control, Robust control and Hinfty control, Linear and nonlinear systems
Abstract: In this paper, the dynamical controller for time-invariant systems with multiplicative noises is designed. The external disturbance belongs to set of sequences of random vectors with given mean anisotropy. The solution of the problem is reduced to convex optimization one.

Keywords: Intelligent and AI based control, Control algorithms implementation, Networks optimization
Abstract: Recently, phishing attacks have risen to the top of the social engineering assaults that affect organizations, governments, and the general public. A phishing assault cost roughly 1.5 billion in 2012, according to an internet analysis. As the worldwide impact of phishing attempts grows, more effective phishing detection systems will be required to combat this threat. This paper has examined six classifiers to identify the top machine learning classifiers for phishing attack detection from a dataset of 48 features from 5000 legitimate websites retrieved from the Kaggle repository. This work is implemented using CatBoostClassifier (CB), light gradient

Keywords: Intelligent and AI based control, Embedded Systems, Estimations and identification
Abstract: Clinical Decision Support Systems (CDSS) systems emerged in utilizing patients’ data to make decisions about their care plan. We found that the literature shows discrepancy in the effectiveness of using such systems. In this study, we used multivariate meta-analysis to investigate the embedding of CDSS in health care, particularly in diabetic care. Based on our results, the usage of CDSSs is growing and gaining higher approval rates and embedding CDSS enhanced the quality of health care of diabetes patients.

Keywords: Intelligent and AI based control, Fault detection and Diagnostics, Autonomous Systems
Abstract: In this research, we extensively explored security challenges in Healthcare IoT (H-IoT) systems, focusing on prominent attacks such as Spoofing, Man-in-the-Middle (MitM), and Sybil attacks. We evaluated various security solutions, including Testing Resource Solutions, Cryptographic-Based Solutions,RSSI-Based Solutions, and Behavior Monitoring Solutions, considering the constraints of IoT devices. Our analysis involved efficiency scores, response times, and the impact of proposed solutions on patient data privacy and medical data integrity. Additionally, we introduced a context-aware security framework tailored for healthcare, addressing real-time response needs. The study provides a comprehensive understanding of H-IoT security, aiming to enhance patient-centric care and medical data protection.

Keywords: Intelligent and AI based control, Fuzzy and neural systems
Abstract: Fake news is a growing problem in today’s digital world with the increase in dependence on social media for information amongst the populace. Additionally, the proliferation of digital media platforms and the rise of user-generated content have facilitated the dissemination of false and misleading information at an unprecedented scale. This has serious consequences on individual and societal decision-making. Therefore, identifying fake news has become an important process to ensure the credibility and integrity of the consumed information. This work seeks to detect fake news using Natural Language Processing techniques and machine learning models.

In this work, several Natural Language Processing techniques were used for text processing and vectorization. A labeled dataset of news articles was used to train the models, which included various models with supervised machine learning algorithms. The performance of each model was evaluated using accuracy, precision, and confusion matrix metrics. Finally, the work includes a prediction pipeline that enables users to input news articles and obtain predictions from each of the trained models. Using these methods this work achieves an accuracy of up to 91%.

Keywords: Intelligent and AI based control, Image processing, Control applications
Abstract: These days, the world's population is helplessly witnessing the multiplication of severe weather phenomena such as floods. Floods are classified as unpredictable natural disasters that have a major negative impact on people's lives, material infrastructures, and economies. The scientific community agrees that climate change is at the root of the multiplication of these phenomena, and is relentlessly seeking innovative solutions for the detection, prevention, and optimal management of these phenomena and their consequences. In this article, we focus on the design and development of an intelligent electronic station capable of early detection of flood phenomena, to reduce the intervention time of protection services and limit the toll of material and, above all, human damage. To avoid false alarms, our electronic station will be equipped with a deep learning algorithm called Convolutional Neural Network (CNN), which plays an active role in the early flood detection process, recognizing and validating confirmed flood situations.

Keywords: Intelligent and AI based control, Modeling and simulation, Control applications
Abstract: Wildfires pose a significant threat to both the environment and human settlements, causing extensive damage and loss. Understanding and predicting fire susceptibility are paramount in crafting effective mitigation and response strategies. In this paper we will provide a comprehensive overview of data sources and methodologies employed in Fire Susceptibility Mapping (FSM), considering various Environmental Factors. It delves into the integration of Geographic Information Systems (GIS) and advanced Machine Learning (ML) techniques, including Statistical Models and Neural Networks, for precise classification of areas based on their susceptibility to wildfires. Through an exploration of various research works, this review aims to illustrate the practicality and effectiveness of the approaches used, contributing to a better understanding of wildfire risk assessment and mitigation strategies.

Keywords: Intelligent and AI based control, Modeling of complex systems, Renewable Energy
Abstract: Virtual synchronous machines (VSMs) have recently emerged as a practical solution for incorporating virtual mechanical inertia into power systems using power electronic converters to add frequency support to the grid to handle the increase in the rate of change of frequency (RoCoF) that may lead to stability problems. Analyzing this concept compared to frequency-droop-based control schemes provides valuable physics-based insights into the tuning and operation of VSM controllers. However, the design and control of VSMs necessitate accurate predictive models that can adapt to microgrids' dynamic and complex nature. Physics-informed neural networks (PINNs) is an emerging discipline that combines principles from physics and neural networks to enhance the accuracy and efficiency of predictive models. By incorporating the fundamental laws and equations that govern a system, PINNs guide the learning process of neural networks and improve the interpretability of the resulting models. This study proposes PINNs trained to predict the VSMs model while ensuring compliance with the fundamental principles of physics described by the conventional swing equation governing synchronous machines (SMs). The effectiveness of the predicted model has been verified via simulations carried out using MATLAB software, demonstrating evidence of its potential to improve microgrid stability and power quality.

Keywords: Observer design, Estimations and identification, Fault detection and Diagnostics
Abstract: This paper presents an adaptive observer design for simultaneous estimation of system parameters and state variables for a class of linear descriptor systems. Sufficient conditions for the existence of the observer, ensuring stability with respect to Linear Matrix Inequalities (LMIs) constraints through the utilization of Lyapunov stability theory, are provided. The methodology presented is a generalization of proportional and proportional-integral observers and It allows for some robustness with respect to uncertainties and modeling errors, as well as steady-state accuracy. A numerical example is given to illustrate our results.

Keywords: Observer design, Estimations and identification, Modeling and simulation
Abstract: This paper addresses the estimation problem of temperatures and reaction heat within an intensified heat exchanger/reactor (HEX reactor) from the available temperature measurements, namely the input and output temperature measurements. This estimation is performed by a high gain cascade observer incorporating a suitable saturation function to deal with the peaking phenomena which will referred to as a suitable high gain cascade observer (SHGCO). This observer has been developed from a recent research activity devoted to the high gain observer engineering features bearing in mind promising modelling results of the HEX reactor involving a cascade of sub-reactor models. The effectiveness of the proposed SHGCO has been illustrated throughout relevant simulation results with particular emphasis on the estimation accuracy and the remarkable peaking phenomenon reduction during transition periods.

Keywords: Observer design, Fault detection and Diagnostics, Control applications
Abstract: This paper proposes a new method for co-design of unknown input functional interval observer and fault tolerant control for LPV-switched systems with faults and unknown but bounded uncertainties. Accurate estimates are achieved by combining Input-to-State Stability (ISS) with multiple Lyapunov function. Unknown input functional interval observer reduces computation effort and eliminates the effects of disturbances while existence conditions don't need similarity transformation condition. The fault tolerant controller compensates the estimated fault and robustly tracks a chosen reference model. The method is successfully applied to vehicle lateral dynamics estimation and control. It shows its efficiency for estimating the lateral speed with a tight interval between the lower and the upper bounds. More generally, the proposed approach has a high potential for higher-order systems.

Keywords: Observer design, Linear and nonlinear systems, Time-delay systems
Abstract: This paper provides a high gain cascade observer for a class of nonlinear systems involving an unknown constant time-delay in the available output measurements. The underlying observer is obtained from a suitable high gain adaptive observer cascaded with a set of adequate predictors. The high gain adaptive observer is essentially used to perform an admissible estimation of the delayed system state variables together with the involved output delay using the available output measurements, while the underlying predictors provide an admissible prediction of the actual system state variables over the estimated output delay. The observer convergence is investigated using an appropriate Lyapunov approach and its effectiveness is illustrated by probing simulation results involving an academic example.

Keywords: Observer design, Modeling and simulation, Linear and nonlinear systems
Abstract: High gain observers have made great progress over the last three decades, in both theoretical and algorithmic domains. In this note, we have demonstrated how these observers can be used in mechanical systems, specifically through an application on a gantry crane. We address the problem of estimating the vertical cable angle and angular velocity of a gantry crane, based on the measurements of the gantry position and the cable length.

Keywords: Control applications, Linear and nonlinear systems, Robotics
Abstract: In this paper, a super twisting algorithm has been formulated for the tracking control of a two-wheeled inverted pendulum subject to external disturbances. Initially, a model of the two-wheeled inverted pendulum was established without resorting to linearization. Subsequently, a high gain observer was been designed for the two wheeled self balancing robot in order to estimate unmeasured states. Then, the designed High Gain Observer was combined with the Super Twisting algorithm to obtain a robust controller that is able to achieve the tracking objectives and to compensate the presence of disturbances. The stability of the closed-loop system is methodically substantiated through rigorous Lyapunov analysis. Numerical simulations meticulously underscore the performance of the implemented controller in the presence of disturbances.

Keywords: Intelligent and AI based control, Renewable Energy
Abstract: Today, it is essential to identify and classify transmission line ‎problems in order to maintain stability and a steady supply of ‎electricity. To ensure safety and reduce power loss owing to the ‎defect, a faulty section must be removed from a healthy section. ‎Nowadays, Machine learning (ML) is widely employed in all ‎aspects of daily life. In this paper, an ML-based system for fault ‎classification and fault detection has been proposed. For that, ‎in order to create normal and fault data (voltage/ current) of ‎four different types of faults, Matlab Simulink was used to ‎simulate the DFIG-WT linked to the grid via the transmission ‎line. The dataset obtained by Matlab is utilized to test and train ‎the models as well as to calculate and analyze the performance ‎of four algorithms by the use of recall, precision, and accuracy ‎metrics. The high-performing classifier is determined by ‎comparing the results of the confusion matrix for machine ‎learning methods.‎

Keywords: Intelligent and AI based control, Renewable Energy
Abstract: The goal of designing this project is to is to gather crucial ‎environmental data, such as temperature, lighting ‎intensity, gas leakage from batteries, or in the event of a ‎fire, and send it in two different ways, wired and ‎wireless. The practical design of the data collecting ‎system in both wired and wireless models is based on the ‎Arduino Uno. Several sensors, including a temperature ‎sensor, a gas sensor, and a light intensity sensor, were ‎used to retrieve the data. LabVIEW software was used to ‎mimic all of the data that was collected to monitor, ‎record, and track the data‏.‏‎ Data from simulations showed ‎that while the connected model provided more ‎bandwidth and increased security, the wireless model ‎achieved mobility and almost accurate findings.‎

Keywords: Intelligent and AI based control, Robotics, Linear and nonlinear systems
Abstract: The paper considers a linear discrete-time system operating in a repetitive mode to track a reference trajectory with a given accuracy. The system parameters are incompletely known and are described by the affine uncertainty model. A new iterative learning control design method based on information about the measured output signal is obtained; this method takes into account the saturation-type nonlinearity inherent in the actuators of robots and allows achieving the required accuracy. The problem statement is motivated by the development trends of high-precision smart and additive manufacturing as well as medical rehabilitation robots. An example illustrates the effectiveness of this method.

Keywords: Intelligent and AI based control
Abstract: Diabetes is a common disease that can lead to dangerous health complications, including heart disease, oral health, nerve damage, vision, hearing, chronic kidney disease, and other problems in feet, and mental health. There are many causes of diabetes such as obesity, age, lifestyle, lack of exercise, hereditary diabetes, high blood pressure, poor diet, etc. Over time, people with diabetes have a high risk of diseases such as heart disease, stroke, kidney failure, nerve damage, eye issues, etc. Early diagnosis of diabetes is a very important factor in reducing the incidence of these complications. Machine learning can be used to develop models that can help Early diagnosis, leading to faster and better patient outcomes. This paper uses machine learning algorithms to train models on patients’ data. The model was trained using two different datasets for detecting type-1 diabetes, type-2 diabetes, and whether the patient may have prediabetes. At the end of the paper, a proposal was made about developing the model to achieve cognitive artificial intelligence for the prediction of diabetes.

Keywords: Intelligent and AI based control
Abstract: Daily electrical power needs are dramatically rising around the world. To support the side of the Power Management System in residential distribution systems, Short-Term Load Forecasting is crucial. However, the resident’s behavior is somewhat improvised and unplanned in advance, causing extreme fluctuations and irregular data patterns in predicting residential loads, this led to creating a significant challenge for accurately predicting residential loads. Recent studies have focused on several variables to improve prediction, such as weather data and previous loads, the influence of the population consumption behavior factor on predicting residential loads has received less attention despite being a key component impacting energy consumption patterns. This paper proposes using Deep Learning (DL) techniques to predict a residential load a hybrid model of Convolutional Neural Networks (CNN) and Long Short-Term Memory to enhance the accuracy of prediction for private data of a Libyan family, Human Consumption Behavior will be added as an effective factor, will be extracted from a survey of consumption behavior of a family in the Benghazi city. The error rate is measured with Mean Squared Error, Mean Absolute Error, and Coefficient of Determination (R²) in this work. The experiments’ results showed the superiority of the model with noticeable improvement in the prediction results with using of the Consumption Behavior Factor (HCB) through decreasing the error metrics from (0.103 MSE, 0.222 MAE 0.898 R²) into (0.057 MSE, 0.155 MAE, 0.944 R²) comparing to the slight improvement of (LSTM) model with using of this factor from (0.549 MSE, 0.541 MAE, 0.347 R²) into (0.426 MSE, 0.441 MAE, 0.523 R²).

Keywords: Control applications, Embedded Systems, Mechatronics
Abstract: Lasertrackers are used for tracking moving targets. Usually, interferometric distance measurement is used to determine the 3D coordinates of the target. In this paper, a tracker network is presented that uses triangulation only. The approach is more robust e.g. in case of beam interruption. But, triangulation requires at least two cooperating lasertrackers. Hence, the triangulation method is extended to a sensor network, that also improves robustness by redundance. Moreover, multiple targets can be tracked by an alternating lasertracker. By alternating, the requirements of the triangulation method are satisfied temporarily, thus the determination of the 3D target coordinates is possible.

Keywords: Control applications, Linear and nonlinear systems, Optimization
Abstract: In this paper, we develop the Input-output finite time stability and stabilization for the class of Linear Time-Varying descriptors systems by considering the disturbance input bound. First, a sufficient condition for input-output finite time stability with non-zero initial condition are investigates. Then, the improved the feedback control laws to guarantee the input-output finite time stabilization are introduced. For this goal, several techniques are employed, which can be utilized to reduce the conservatisme of the system. Ultimately, the numerical example are given to illustrate the effectiveness of the proposed strategy.

Keywords: Control education, Control algorithms implementation, Control applications
Abstract: Information and Communication Technology (ICT) has become essential in education, with e-learning and cloud services providing solutions to the education sector and enabling it to compete more effectively. The Technology Acceptance Model (TAM) and its various versions have been recognized worldwide for measuring technology acceptance in learning and teaching. However, deficiencies in measuring technology acceptance for cloud computing among teachers in the Arab world and Libya have been revealed in previous studies. This study aims to develop the TAM model by including external factors to measure teachers' acceptance of this technology. The researchers added essential factors to the model and tested it with experts in the same field. The resulting model was used to measure the acceptance of cloud computing applications among secondary school teachers in AL-Marj City. The study found that the model's external factors enhanced its ability to measure technology acceptance in an educational context. The study provides a solid foundation for future research. New extensions and amendments to the model can be proposed to include various factors influencing the decision to adopt, accept, or reject a specific technology in teaching and learning processes.

Keywords: Control algorithms implementation, Control applications, Linear and nonlinear systems
Abstract: This paper investigates the synthesis of a Simple Adaptive Anti-Windup compensator in the context of simple adaptive control of dynamic systems with actuator saturations. For this purpose, we demonstrate the usefulness of combining the use of the properties of a dead zone function with a simple version of cousins of Barbalat’s Lemma to prove the bounded- ness of all gains and signals of the controlled system.

Keywords: Control algorithms implementation, Modeling and simulation, Transportation systems
Abstract: For a tilt rotor Vertical Take-off and Landing (VTOL) aircraft, the transition from hovering to forward flight has unique challenges, particularly in terms of aerodynamics, control systems, and vehicle design. This paper presents a modular flight simulation framework for VTOL aircraft comprising multiple tilt rotors. The control strategy using the combination of the Proportional-Integral-Derivative (PID) controller and Incremental Nonlinear Dynamic Inversion (INDI) controller is successfully implemented in the models. Each multirotor configuration is mathematically modeled with a focus on the transition phase. The simulations presented are successfully implemented for the vertical takeoff, transition phase and forward flight for each case of multirotor configuration. This research not only advances the state of the art in UAV technology but also opens new possibilities for different applications.

Keywords: Control algorithms implementation, Optimization, Power systems
Abstract: Voltage stability is crucial for power system operation. STATCOM, an adaptable FACTS controller, has a linear PI control with a limited range for nonlinear processes. To overcome this problem, optimization methods such as Genetic Algorithm (GA) and Particle Swarm Optimization (PSO) are used. In this paper, a hybrid GA-PSO was adopted to tune the PI controllers of the studied system. Simulation results are implemented in the MATLAB/Simulink architecture. These results show the effectiveness of the proposed optimization algorithm for minimizing the integral absolute error (IAE) of the grid voltage comparing to PI tuned by GA and PSO algorithms and the traditional PI.

Keywords: Image processing, Fuzzy and neural systems, Intelligent and AI based control
Abstract: Abstract—Digital image restoration has become important for many image applications. Therefore, Image Noise removal is an essential issue in an image processing fields. In this paper, we presented a hybrid system, based on Self Organizing Maps neural networks (SOM NN) and Median filter (MF), to eliminate Random Impulsive Noises (RIN) from grayscaled digital images. In our system we applied two main processes (features extraction process, detection process): in the features extraction process, feature vectors of two or three features, (central pixel value, standard deviation value) or (pixel value, standard deviation, Mean Difference value) respectively, were extracted. In the detection process, the self-organizing maps (SOM NN) are used as an impulse noise detector. This SOM NN module is trained using competitive learning algorithms. Then the detected corrupted pixel is modified using Median filter (MF) algorithm; otherwise, it is left unchanged. Finally, the results of this study are compared with the previous traditional and state-of-the-art methods results that applied on the same database. Our results are significantly outperforms other traditional methods and were comparable with the state-of-the-art methods results in terms of reconstruction quality and that are comparable to the FL using the MSE and PSNR measurements.

Keywords: Image processing, Intelligent and AI based control
Abstract: The present work develops a classification system based on the convolutional neural network (CNN) model. The datasets used in this work is brain Magnetic Resonance Imaging (MRI) images of the Kaggle website. Firstly, a data augmentation algorithm is performed in order to expand the kaggle dataset. Secondly, the brain images are cropped by the extreme point calculation method for removing the unwanted areas and justifying the size of images used. Finally, the convolutional neural network is used to generate the feature extraction in order to apply the Fully Connected Layers to classify the normal and tumor images. An accuracy of 94.51%, a sensitivity of 96.31% and a 92.51% specificity are provided by convolutional neural network classification.

Keywords: Image processing, Intelligent and AI based control
Abstract: While recognizing banknotes may seem simple to those with normal vision, it presents a significant challenge for individuals in the blind community, especially when dealing with paper currency. The importance of money in daily life necessitates real-time detection and recognition of banknotes for blind or visually impaired individuals to conduct their business transactions with confidence. In this paper, an object recognition system designed to assist visually impaired individuals in their daily business transactions is proposed. To aid the blind community in banknote detection and recognition, smart glasses equipped with an Espressif32 (ESP32) camera for currency detection have been implemented. The approach involves the utilization of Teachable Machine, a real-time classification algorithm trained on a custom dataset of Libyan banknotes. Once the algorithm identifies the banknote, the label is processed and converted into audio using Text-to-Speech (TTS) technology, providing the expected output. A dataset consisting of Libyan banknote images captured in various scenarios was assembled for the algorithm's training. Subsequently, the system's robustness was enhanced by applying various geometric changes to the images, allowing the construction of reliable training and validation sets. The effectiveness of the system was evaluated using a test dataset. Experimental results demonstrate the system's ability to rapidly and reliably detect and identify Libyan currency, achieving an impressive accuracy rate of 96%.

Keywords: Image processing, Intelligent and AI based control, Signal processing
Abstract: A new texture descriptor based on local neighborhood intensity variation is proposed in this paper, which enables contactless palmprint recognition. This proposed representation enables capturing the texture information of adjacent neighbours. It is therefore inspired by the principle that the neighbours of a given pixel contain a significant dataset of texture information, which can be exploited for an efficient texture representation in contactless palmprint recognition.The main benefit of using the mutual relationship between adjacent neighbors resides not only in considering the sign of the intensity difference between the central pixel and its neighbors, but also in taking into account the sign of the difference values between a given pixel and its adjacent neighbors, and between the central pixel and all neighboring pixels. As a result, the proposed model becomes much more robust when illumination is varied. Moreover, most local models, especially those of the LBP, are mainly focused on the sign information and therefore ignoring also its magnitude. It's important to emphasize that magnitude information has an auxiliary role, as it provides further information that complements the texture descriptor. It is therefore necessary to incorporate it into the proposed representation, in which the average absolute deviation of each pixel from its adjacent neighbors must be taken into account. This allows the development of a high-performance texture descriptor known as Patch Local Neighborhood Binary Pattern (P-LNBP). This descriptor generates sign and amplitude patterns based on the relative intensity difference between the pixel in question and the central pixel, where the corresponding adjacent neighbors are taken into account. Finally, these sign and amplitude pattern

Keywords: Image processing, Signal processing, Intelligent and AI based control
Abstract: When a higher level of exactness is needed, handwritten signature recognition is mainly employed to certify administrative and official papers. Despite extensive previous research, offline signature recognition often remains a challenge, especially when distinguishing authentic signatures from forgeries. Indeed, the difference in appearance between a true and a forged signature may be much smaller than that of authentic signatures. Hence, the present paper outlines a new approach for the offline signature representation using multiscale analysis. Indeed, this last presentation is intended to capture texture features over a wide range of resolutions. It is designed from binarized statistical features of the image, which are computed at different scales. The pre-learned filters, derived from natural images, are implemented on the signature images to reveal the signature structure and generate a discriminative image description. The reduced relevant data is assessed using the classifier for efficient offline signature recognition.

Keywords: Optimization, Linear and nonlinear systems, Mathematical systems theory
Abstract: This paper explores the synthesis of sufficient conditions for finite-time stability (FTS) of nonlinear quadratic systems.

An adequate choice of the form of the Lyapunov function makes it possible to group the conditions for quadratic and non quadrtic Lyapunov functions into a single theorem.

The design conditions are expressed in term of linear matrix inequalities (LMIs) by combining the notion of annihilator with a version of Finsler's lemma.

Keywords: Optimization
Abstract: The testing phase is one of the most important phases of the software life cycle, which ensures the efficiency and quality of the software product. Many tests are conducted on the software, not only to detect errors, but also to provide a complete picture of the quality of the outputs provided by the program and to increase confidence in all of the functions. Due to the difficulty in estimating most of the test data and the time spent in this process, many techniques are used in the automatic generation of test cases. The genetic algorithm is one of the most important used methods in this process, and it has achieved great success in generating limited and appropriate test cases to give a complete conceptualization of the function under test. The Junit testing framework is used in automated testing processes within test-based development methodologies, but it lacks of a mechanism to specify and generate a specific test data that put the program under test in the required scenario. In order for the programmer to be able to perform the test automatically in an integrated way within JUnit testing framework, the genetic algorithm has been improved in this research and emerged into the JUnit testing framework to fully automatic testing and generate various numeric test cases at the unit-test stage. Based on testing of the proposed algorithm within the Junit testing framework, it has proven its effectiveness and facilitation of generating perfect specific numeric test cases.

Keywords: Optimization, Power systems, Renewable Energy
Abstract: The temperature of photovoltaic (PV) cells plays a crucial role in determining the overall module performance, as it directly influences the efficiency and effectiveness of PV systems. The present study explores the influence of climatic conditions on the efficiency of photovoltaic (PV) systems, with a particular focus on the city of Benghazi, Libya. A numerical model, based on the principles of energy balance, was developed to forecast the PV cell efficiency under varying environmental parameters. Through meticulous modelling and examination, the performance of the photovoltaic system was thoroughly investigated. The numerical proposed model is validated with the published results where the comparison showed good agreement between the present and published results with a maximum deviation of 2.56 %. The findings revealed a significant correlation between PV cell efficiency and the temperature of the solar panel. As solar radiation levels increased to 1017 W/m2, the temperature of the PV panel soared up to 71.1°C, surpassing the ambient air temperature by 34.7°C. Such elevated temperatures were identified as a primary factor contributing to the deterioration of cell efficiency.

Keywords: Networks optimization, Modeling and simulation, Control of telecommunications systems
Abstract: The paper explores the impact of blockages on resource allocation in Mm-Wave cellular systems by applying a classical handover (HO) scheme. The scheme adjusts the number of reserved channels dedicated exclusively for the HO requests based on a predefined threshold of the probability of blocking HO calls to ensure efficient resource allocation and meet the Quality of Service (QoS) requirement. The study finds that an optimized and balanced scheme between the blocking probability of HO calls and the blocking probability of originating calls in the mm-wave BS is necessary. The results indicate that a higher number of reserved channels is required to satisfy the minimum threshold of the blocking probability of HO calls

Keywords: Networks optimization, Process control and instrumentation, Optimization
Abstract: With the rapid expansion of Internet of Things (IoT) technologies, low-power wide-area networks (LPWANs) have emerged as effective solutions for long-range wireless connectivity. Long Range Wide Area Network (LoRaWAN) is a promising LPWAN architecture that enables robust, energy-efficient communication for distributed IoT devices. This paper presents an experimental performance analysis of a LoRaWAN network deployed in urban and suburban environments across Benghazi, Libya. The results demonstrate LoRaWAN’s capabilities in terms of reliability, transmission range, and adaptation to varied conditions. Packets were transmitted between LoRa32 nodes using different spreading factor (SF) configurations. Key performance metrics, including packet error rate, received signal strength indicator (RSSI), and signal-to-noise ratio (SNR), were measured across outdoor and indoor scenarios. The findings showcase LoRaWAN’s resilience to channel noise and interference, achieving minimal packet loss over several kilometers. Furthermore, tuning parameters, such as the spreading factor, allow effective trade-offs between data rate, communication range, and power consumption. This research validates LoRaWAN as a versatile IoT networking solution and offers insights to inform optimal deployment in diverse real-world settings.

Keywords: Power systems, Transportation systems, Modeling and simulation
Abstract: Electric vehicles (EVs) have been presenting themselves as a modern load in the eclectic power system. EVs have a great potential to positively contribute to greenhouse gases emission reduction and to less reliance on fossil fuel combustion engine vehicles. The impact of electric vehicle charging on a typical electric distribution system in a specific geographical region in Libya is the focus of this paper. This research started by collecting the needed data and circuit models of an available distribution line in Al-Bayda city, power converter, and EV battery as a load, then implementing the overall model in Matlab/Simulink software to perform simulation studies. Single and three-phase chargers-based controlled rectifiers are built on the selected line at home and in public areas respectively. Two scenarios were considered, the first was to simulate the selected distribution line without chargers, while the second was to perform the simulation in the presence of chargers. The last task in the adopted methodology, in order to attain the objectives of the research, was to compare the two scenarios based on voltage and current waveforms at different stations. The main findings of this research were that the shape of voltages at the nearest busbars to the chargers were slightly impacted by these EV chargers, while the currents drawn from the grid by these EV chargers were almost affected, so injecting harmonics into the grid.

Keywords: Power systems
Abstract: The huge number of transformers in electrical networks ‎exposes them to the rise of ‎failures from the design stages to the ‎operation stages. Therefore, keeping the safety of transformers ‎from faults is ‎an urgent need to ensure continuity to provide ‎consumers with energy. Lately, the power system network of ‎Mellitah Complex, in the "north part of Libya", faced several ‎transformer ‎failures. One of these failures caused power ‎interruption on industrial loads in the seventh distribution ‎station. The reason for the interruption is a fault on a step-‎down transformer 20 KV /6.6 KV tag No: 85-910-ET-071B. ‎This paper aims to study and analyze the causes of frequent ‎transformer failures and to suggest practical solutions to ‎overcome these failures. The research is done based on the ‎recorded data on a program and a digital protection relay ‎device called “Sepam Protection”, reports of maintenance and ‎protection engineers, and the results of tests done on the ‎transformer under investigation. The investigation shows; the ‎main cause of the transformer failure was a misalignment ‎between the Voltage Transformer (VT) of the line (L3) and the ‎busbar.‎

Keywords: Power systems
Abstract: This work proposes an innovative hybrid method for accurate one-day electricity consumption forecasting. By combining artificial intelligence (AI) and statistical techniques, our hybrid model optimizes electricity production forecasting. Unlike methods used by RTE, our approach relies solely on consumption data, eliminating the need for additional variables. By preprocessing historical data and employing neural networks and statistical models, our hybrid method achieves exceptional accuracy, closely approaching RTE's performance, while significantly reducing data usage and computational demands. This promising approach streamlines electricity production planning, offering an efficient and environmentally friendly solution to environmental and economic challenges.

Keywords: Power systems
Abstract: SPSS, R, and Stata are widely recognized statistical software packages that encompass a range of analytical tools and techniques for examining statistical data. Within the realm of research, these software packages are progressively gaining traction as valuable tools for analyzing statistical data across diverse fields such as economics, biology, medicine, public health, and education. Among the various challenges encountered during statistical analysis, the issue of missing data has garnered significant attention from researchers in recent years. Missing data refers to the absence or incompleteness of certain data points within an analysis, which can arise due to factors like data entry errors, equipment malfunction, or participant attrition in a study. The presence of missing data can introduce biases, diminish statistical power, and compromise the accuracy of predictive models. In our paper, we employed these three statistical software packages to analyze datasets with missing values and subsequently compared the obtained results. While some studies have explored differences between statistical software packages, they have often overlooked the aspect of missing data and the relative efficiency of these packages in handling such data. Through our analysis of datasets containing missing values using these three distinct packages, we obtained results that will be expounded upon in the ensuing discussion section.

Keywords: Intelligent and AI based control, Networks optimization, Signal processing
Abstract: Like other languages, the Arabic dialect has both spoken and written forms. The first form is the true mother tongue of Arabic speakers explored on a daily basis. As for the second form, it represents the dialect of Modern Standard Arabic (MSA) which mainly constitutes the content of most Arabic databases due to its predominance in written form. However, the large size of Arabic text databases requires robust and accurate machine learning methods. Random Multimodal Deep Learning (RMDL) is one such deep learning approach that is explored in this article to address the issue of finding the best deep learning architecture and structure while simultaneously maintaining improvement accuracy and robustness for the classification and identification of Arabic dialects (MSA, Gulf (GLF), Egypt (EGY) and Levantine (Lev)) via a set of deep learning architectures. Experimental tests were carried out on the Arabic Online Commentary (AOC) database and showed good results in terms of performance evaluated using the RMDL approach compared to those obtained with other deep learning algorithms.

Keywords: Intelligent and AI based control, Optimization, Renewable Energy
Abstract: This paper studies the application of Artificial Intelligence (AI), precisely Reinforcement Learning (RL), to minimize energy costs of a grid-connected community Microgrid (MG). This MG is composed of photovoltaic (PV) panels, dynamic and static loads, and a battery storage system (BSS). RL algorithms learn through interacting with a MG simulator to make decisions in real time that minimizes energy costs without prior knowledge of uncertainties related to PV production, load consumption and electricity costs. In this article, two RL algorithms, Q-learning and State-Action-Reward-State-Action (SARSA), are studied. These algorithms were benchmarked with two baselines in order to test its effectiveness. The first baseline involves no storage, and the second baseline involves storage management provided by a linear programming (LP) algorithm. The main contribution of this work is the investigation of the RL-based energy management approaches on the MG voltage magnitude at the Point of Common Coupling (PCC). In this sense, an electrical model of the studied MG is adopted. Simulation results shows effective results of the data-driven methods (based on Q-learning and SARSA) where no physical constraints violating have been recorded.

Keywords: Intelligent and AI based control, Renewable Energy
Abstract: A crucial part of digital solutions, Machine Learning (ML) is a subset of artificial intelligence. The main goal of this work is to extract knowledge from data and known experiences in order to automate decision-making procedures. Thus, ML has the potential to change the way we manage and design cities’ power demand. Firstly, this paper presents a multiples homes classification into five categories depending mainly on their peak powers, off-peak powers, average powers, and enhanced powers. In this operation, two ML techniques are used to determines the best model: decision tree and K-Nearest Neighbors. Secondly, the paper proposes a Home Energy Management Approach (HEMA) which is divided into three layers. The goal of the approach is to smooth a home power profile that has a high power by integrating Wind Turbine (WT) and Plug-in Electric Vehicle (PEV) using particle swarm optimisation (PSO) algorithm.

Keywords: Intelligent and AI based control, Renewable Energy
Abstract: There is a global demand for clean, safe, reliable, and stable electrical energy. This energy can be efficiently managed in terms of load consumption, allowing consumers to effectively utilize their economic resources in a manner that aligns with the significance, responsibility, and sensitivity of their operations. This paper presents an optimal sizing of a photovoltaic generator (PVG), wing turbines (WT), and an energy storage sources (ESS) integrated in a company using Homer software and Machine learning. The case study focuses on the Libyan Investment Authority, which is one of the sovereign and investment funds with significant financial investment assets worldwide. The objective of the study is to cover 250 kW of power demand in the company by integrating renewable energy sources, ESS, and diesel generator. Consequently, the study examines the electric, financial, and environmental aspects of the project. The total system production amounts to 2,398,024 kWh per year, resulting in a total cost of 5.69 million.

Keywords: Intelligent and AI based control, Multivariable control, Estimations and identification
Abstract: This paper validates the research hypothesis placing the thematic categorization of records at the center of the discussion. It highlights the necessity of deepening the standardization of government actions record management processes. The paper seeks to contribute to the supportive role that machine learning (ML) technologies can play in archives and records management and to inform future practices and decision-making in the field. It demonstrates also the initial practical results of an ongoing research project on the thematic categorization of the University of West Attica records.