The proceedings contain 78 papers. The special focus in this conference is on Distributed Computing and optimization Techniques. The topics include: A Study on Different Types of Convolutions in Deep Learning in the A...
ISBN:
(纸本)9789811922800
The proceedings contain 78 papers. The special focus in this conference is on Distributed Computing and optimization Techniques. The topics include: A Study on Different Types of Convolutions in Deep Learning in the Area of Lane Detection;A Study on the Impact of DC Appliances and Direct DC Power System in India;preface;1 T-1D Single-Ended SRAM Cell Design for Low Power applications Using CMOS Technology;a Survey on Vehicle Detection and Classification for Electronic Toll Collection applications;a Systematic Study of Sign Language Recognition Systems Employing Machine Learning algorithms;ACS Fed Coplanar Monopole Antenna with Complementary Split Ring Resonator for WLAN and Satellite Communication applications;advance the Energy Usage in Cloud Centers Utilizing Hybrid Approach;Advanced Architecture of Analog to Digital Converter Derived from Half Flash ADC;an Assessment of Criss-Cross multilevel Inverter with Fault Tolerance for Electric Vehicle applications;an Energy-Efficient Load Balancing Approach for Fog Environment Using Scientific Workflow applications;an Ensemble Model to Extract Discriminative Features for Semantic Image Classification in Large Datasets;an Evaluation of Wireless Charging Technology for Electric Vehicle;automated Dam Data Acquisition and Analysis in Real-Time;a Local Descriptor and Histogram of Oriented Gradients for Makeup Invariant Face Recognition Under Uncontrolled Environment;Chaotic System Based Modified Hill Cipher Algorithm for Image Encryption Using HLS;Chronological-Squirrel Earth Worm optimization for Power Minimization Using Topology Management in MANET;Classification of Neurological Disorders with Facial Emotions and EEG;comparative Analysis of Machine Learning Approaches for the Early Diagnosis of Keratoconus;efficient Square Root Computation–An Analysis.
This work presents the basic concepts of energy harvesting (EH) techniques and circuits, with reference to maritime applications. The main EH sources and related sensors are recalled as well as the DC-DC conversion ci...
详细信息
ISBN:
(数字)9798331516390
ISBN:
(纸本)9798331516406
This work presents the basic concepts of energy harvesting (EH) techniques and circuits, with reference to maritime applications. The main EH sources and related sensors are recalled as well as the DC-DC conversion circuits. Multi-input energy sources, piezoelectric and thermoelectric, have been considered. algorithms for maximum power point tracking are also cited as well as energy storage options. Matlab-Simulink models are finally described.
multilevel inverters are becoming more popular in industrial applications for high power and voltage ranges [6], [7]. Harmonics is a vast issue in the multilevel Inverter. By applying the Selective Harmonic Eliminatio...
详细信息
multilevel inverters are becoming more popular in industrial applications for high power and voltage ranges [6], [7]. Harmonics is a vast issue in the multilevel Inverter. By applying the Selective Harmonic Elimination (SHE)-PWM technique and the best possible choice of switching angles, harmonics are eliminated, and the Total Harmonic Distortion is optimized and reduced (THD) [2]. Grey Wolf optimizer (GWO) algorithm is used to find the best switching angles for a Cascaded H-Bridge multilevel Inverter (CHMLI), which wm allow some lower-order harmonics to be eliminated while still retaining the necessary fundamental voltage [8]. THD determines the output voltage quality. This report builds the five-level CHB-MLI and 7-level CHBMLI by implementing different algorithms. Here Particle Swarm optimization (PSO) algorithm and GWO are used. A GWO is compared with the practical values of PSO. By comparing both, GWO has less THD. A five-level and 7 verify the proposed system-level CHMLI; the work is done in MATLAB/Simulink.
We introduce a self supervised framework for learning representations in the context of dictionary learning. We cast the problem as a kernel matching task between the input and the representation space, with constrain...
详细信息
ISBN:
(数字)9798350372250
ISBN:
(纸本)9798350372267
We introduce a self supervised framework for learning representations in the context of dictionary learning. We cast the problem as a kernel matching task between the input and the representation space, with constraints on the latent kernel. By adjusting these constraints, we demonstrate how the framework can adapt to different learning objectives. We then formulate a novel Alternate Direction Method of Multipli-ers (ADMM) based algorithm to solve the optimization problem and connect the dynamics to classical alternate minimization techniques. This approach offers a unique way of learning representations with kernel constraints, that enable us implicitly learn a generative map for the data from the learned representations which can have broad applications in representation learning tasks both in machine learning and neuro-science.
In the antenna design process, designers often modify antenna dimensions iteratively to meet specified performance criteria, using simulation software to guide adjustments based on predefined targets. In this paper, a...
详细信息
ISBN:
(数字)9798350366976
ISBN:
(纸本)9798350366983
In the antenna design process, designers often modify antenna dimensions iteratively to meet specified performance criteria, using simulation software to guide adjustments based on predefined targets. In this paper, a rectangular patch microstrip antenna is used as a research target, and it proposes the use of hyperparameter algorithms and machine learning to predict the optimal size of the antenna, focusing on the frequency range of 2 to 5 GHz. With simulated antenna widths ranging from 20 to 24 mm, lengths ranging from 17 to 21 mm, and slot intervals ranging from 0 to 5 mm. This research utilizes Decision Trees, Random Forests, and XGBoost as three distinct machine learning algorithms to evaluate their effectiveness. The results show that the random forest algorithm using the Grid Search CV achieves the best result with an MSE of 0.108, RMSE of 0.329, and R-squared of 0.94. Based on these results, the antenna design process can be accelerated by machine learning to meet the desired performance targets.
In this paper, the mantis search algorithm (MSA), is proposed to determine the switching angles for a three-phase (3-ph) eleven-level (11-lvl) cascaded H-bridge multilevel inverter (CHBMLI). MSA is applied to the sele...
详细信息
ISBN:
(数字)9798331522193
ISBN:
(纸本)9798331522209
In this paper, the mantis search algorithm (MSA), is proposed to determine the switching angles for a three-phase (3-ph) eleven-level (11-lvl) cascaded H-bridge multilevel inverter (CHBMLI). MSA is applied to the selective harmonic minimization pulse-width modulation (SHMPWM) technique to maintain the desired fundamental harmonic while minimizing undesired low-order harmonics. The performance of the MSA-SHMPWM technique is compared to that of a well-known optimization algorithm, the genetic algorithm (GA). For fair comparison, both GA- and MSA-SHMPWM techniques are implemented in MATLAB with the same population size, maximum number of iterations and maximum number of runs. Furthermore, a 3-ph ll-lvl CHBMLI is modeled using PSIM simulation software to validate the effectiveness of the MSA-SHMPWM technique. The MATLAB analysis results show that the MSA-SHMPWM technique achieves a minimum objective function (OF) values of 10– 19 , which is lower than the 10– 9 achieved by the GA-SHMPWM technique. This indicates that the MSA-SHMPWM technique determines switching angles with higher accuracy compared to the GA-SHMPWM technique. Additionally, the PSIM simulation results confirm that the switching angles determined by the MSA-SHMPWM technique, when applied to a 3-ph ll-lvl CHBMLI, generate sinusoidal staircase-like output phase and line-to-line voltage waveforms with the desired fundamental voltage and minimized undesired low-order harmonics. The total harmonic distortion (THD) of these voltage waveforms, when the CHBMLI is operated at modulation index of 0.60, are 36.61 % and 7.24 %, respectively.
With the development of 5G and IoT networks, Device-to-Device (D2D) communication has become a major paradigm in wireless communication. Most existing approaches for D2D resource allocation are usually time consuming ...
详细信息
ISBN:
(纸本)9781450399333
With the development of 5G and IoT networks, Device-to-Device (D2D) communication has become a major paradigm in wireless communication. Most existing approaches for D2D resource allocation are usually time consuming and demand a high computational budget, especially in heterogeneous deployments where the D2D links have different configurations (i.e., different number of transmit and receive antennas). Recently, Graph neural networks (GNNs) have been proposed to solve many problems in the networking domain and have significantly outperformed traditional algorithms, including throughput optimization problems in D2D networks. However, existing throughput optimization works either only apply to MISO or SISO D2D networks or require extremely long runtime on MIMO D2D networks, which makes it hard to apply them in real-world D2D applications. In this paper, we consider the throughput prediction problem across a fixed association of transmitters and receivers to maximize the total throughput in heterogeneous MIMO D2D networks. We model the interference between different link types as heterogeneous edges and learn the optimal beamforming policy using a heterogeneous GNN. Simulation results show that our proposed GNN-based approach achieves a significant speedup compared with the state-of-the-art algorithm, while providing robust performance on large-scale synthetic datasets.
This article delves into cloud computing technologies, addressing the challenges and concerns that arise in this domain, specifically focusing on resource allocation. We investigate and compare three optimization algo...
This article delves into cloud computing technologies, addressing the challenges and concerns that arise in this domain, specifically focusing on resource allocation. We investigate and compare three optimizationalgorithms—Particle Swarm optimization (PSO), Ant Colony optimization (ACO), and Genetic Algorithm (GA)—in the context of their applications, objectives, and mechanisms of action. Additionally, the article highlights the significance of each algorithm in optimizing cloud resources and discusses their advantages and drawbacks. For cloud computing with higher complexity, it becomes increasingly important to develop efficient strategies for managing and allocating resources, which in turn helps organizations control costs and improve performance. Prepared comparative analysis aims to provide a comprehensive understanding of the three algorithms, equipping decision-makers with the knowledge to select the most suitable optimization method for their specific cloud computing needs and challenges.
This work presents the development of a multilevel power inverter prototype with the use of optimized modulation which manages to significantly minimize the harmonic content of output voltages using Genetics Algorithm...
详细信息
ISBN:
(纸本)9783030867027;9783030867010
This work presents the development of a multilevel power inverter prototype with the use of optimized modulation which manages to significantly minimize the harmonic content of output voltages using Genetics algorithms. Additionally, it integrates a DC/DC converter that allows to regulate the RMS value of the inverter output voltage through the control of the DC bus voltage. A control loop is implemented that allows obtaining the optimal power quality by verifying compliance with the IEEE 1159 (1995) and IEEE 519 (1992) standards. Through this it is possible to avoid most of the related power quality phenomena such as sag, swell, flicker, undervoltage, etc. Finally, the prototype was successfully implemented and verified.
The performance of the three-phase multilevel inverter (MLI) is monitored and analyzed using genetic algorithm as an intelligent algorithm for solving mathematical equations accompanying the inverter operation. This p...
详细信息
ISBN:
(数字)9798350380040
ISBN:
(纸本)9798350380057
The performance of the three-phase multilevel inverter (MLI) is monitored and analyzed using genetic algorithm as an intelligent algorithm for solving mathematical equations accompanying the inverter operation. This performance is compared when employing the traditional space vector pulse width modulation (SVPWM) technique to drive the MLI. SVPWM is commonly used for applications in the power electronics field. The gating angles of the inverters are extracted using the selective harmonic elimination (SHE) concept for optimal switching using artificial algorithms and monitoring the performance of the inverter when loaded with a three-phase motor. Then monitoring the performance of the inverter when extracting these angles using SVPWM techniques. The inverter's voltages, currents, and load characteristics are analyzed, as well as the frequency analysis of the variables, while checking the active and reactive power levels injected to the load. A distinct advantage has been indicated for algorithmic technology compared to SVPWM technique for most variables. A detailed MATLAB model with reduced number of switches for three phase multilevel inverter is built and thoroughly analyzed for both techniques. Also, a practical 31 level single phase inverter is built for testing and measurement.
暂无评论