Computers & Electrical Engineering最新文献_第2页

Artificial neural network based decentralized current-sharing control for parallel connected DC-DC converters in DC microgrid application 直流微电网应用中基于人工神经网络的并联直流-直流转换器分散分流控制

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2024-10-14 DOI: 10.1016/j.compeleceng.2024.109731

To create a non-interrupted, economical and reliable direct current (DC) microgrid, multiple DC-DC converters can be connected in parallel. Current sharing among these multiple converters becomes essential for proper operation of the system. This study proposes an artificial neural network (ANN) based control technique for parallel connected DC-DC boost converters which ensures accurate current sharing according to the specified maximum limits. Levenberg-Marquardt algorithm-based ANN network is used to reduce the training time and, also to achieve near ≈ 100 % accuracy of the training data. ANN control provides better voltage regulation, accurate current sharing unlike the conventional proportional-integral (PI) based control which faces issues such as inaccurate current sharing, high transient, peak overshoots and steady state error during sudden change in the system. The efficient functioning of the proposed control method is verified by simulating two parallel connected DC-DC converters using MATLAB/Simulink. A hardware prototype of converters rating ≈ 250 W using TMS320F28379D Digital signal processor controller is also developed to verify the performance and effectiveness of the ANN based control in comparison to PI based technique. The ANN based technique is faster to achieve the reference voltage, and the peak overshoot is approximately 75 % lesser than the PI based control.

为了创建一个不间断、经济可靠的直流（DC）微电网，可以并联多个直流-直流转换器。这些多个转换器之间的电流共享对系统的正常运行至关重要。本研究针对并联的直流-直流升压转换器提出了一种基于人工神经网络（ANN）的控制技术，可确保按照指定的最大限制精确分流。基于 Levenberg-Marquardt 算法的 ANN 网络可缩短训练时间，并使训练数据的准确率接近 ≈ 100%。与传统的基于比例-积分（PI）的控制不同，ANN 控制可提供更好的电压调节和精确的电流分担，而传统的比例-积分（PI）控制则面临着电流分担不精确、瞬态高、峰值过冲以及系统突变时的稳态误差等问题。通过使用 MATLAB/Simulink 对两个并联的 DC-DC 转换器进行仿真，验证了所提出的控制方法的高效运作。此外，还使用 TMS320F28379D 数字信号处理器控制器开发了额定功率≈250 W 的转换器硬件原型，以验证基于 ANN 的控制与基于 PI 的技术相比的性能和有效性。基于 ANN 的技术实现参考电压的速度更快，峰值过冲比基于 PI 的控制少约 75%。

{"title":"Artificial neural network based decentralized current-sharing control for parallel connected DC-DC converters in DC microgrid application","authors":"","doi":"10.1016/j.compeleceng.2024.109731","DOIUrl":"10.1016/j.compeleceng.2024.109731","url":null,"abstract":"<div><div>To create a non-interrupted, economical and reliable direct current (DC) microgrid, multiple DC-DC converters can be connected in parallel. Current sharing among these multiple converters becomes essential for proper operation of the system. This study proposes an artificial neural network (ANN) based control technique for parallel connected DC-DC boost converters which ensures accurate current sharing according to the specified maximum limits. Levenberg-Marquardt algorithm-based ANN network is used to reduce the training time and, also to achieve near ≈ 100 % accuracy of the training data. ANN control provides better voltage regulation, accurate current sharing unlike the conventional proportional-integral (PI) based control which faces issues such as inaccurate current sharing, high transient, peak overshoots and steady state error during sudden change in the system. The efficient functioning of the proposed control method is verified by simulating two parallel connected DC-DC converters using MATLAB/Simulink. A hardware prototype of converters rating ≈ 250 W using TMS320F28379D Digital signal processor controller is also developed to verify the performance and effectiveness of the ANN based control in comparison to PI based technique. The ANN based technique is faster to achieve the reference voltage, and the peak overshoot is approximately 75 % lesser than the PI based control.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":4.0,"publicationDate":"2024-10-14","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142437854","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

PV-fed multi-output buck converter-based renewable energy storage system with extended current control for lifetime extension of Li-ion batteries 基于光伏馈电多输出降压转换器的可再生能源储能系统，具有延长锂离子电池寿命的扩展电流控制功能

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2024-10-12 DOI: 10.1016/j.compeleceng.2024.109757

Recently, there has been a visible intensification of research on increasing the cycle life of energy storage devices used in Photovoltaic (PV)-fed energy storage systems (ESS). Compared to existing battery technologies, Lithium-ion (Li-ion) batteries have advantages such as high energy density and high cycle life. However, many existing Hybrid energy storage system (HESS) suffers from the fact that Li-ion batteries have limited cycle life, calendar and cycle aging. Therefore, attention has shifted towards the charging techniques necessary to improve the service life of Li-ion batteries by increasing their cycle life and reducing capacity losses. One such approach is to charge the battery by gradually increasing it from a certain level, rather than starting the charging process directly with the maximum charging current. This study proposes Extended Current Control (ECC) to reduce battery capacity losses and extend service life in PV-fed HESSs. The maximum power point (MPP) of the PV module is provided by the Perturb and observe (P&O) algorithm via the supercapacitor (SC) converter, while ECC is performed via the battery converter. This approach requires less hardware, unlike optimization, rule, or filtering based techniques. Controlling the battery current over a large area protects the battery packs from high currents at the start of charge and reduces capacity losses by increasing cycle life. Compared to PV-fed ESS containing only battery packs, the proposed technique provides a 40 % improvement in battery charging current, an 8 % improvement in the converter duty ratio in reaching the MPP point, and a 31.69 % improvement in battery capacity fades.

最近，有关提高光伏（PV）储能系统（ESS）中使用的储能设备循环寿命的研究明显加强。与现有的电池技术相比，锂离子（Li-ion）电池具有高能量密度和高循环寿命等优点。然而，现有的许多混合能源存储系统（HESS）都存在锂离子电池循环寿命有限、日历和循环老化等问题。因此，人们开始关注必要的充电技术，以通过延长锂离子电池的循环寿命和减少容量损失来提高其使用寿命。其中一种方法是通过从某一水平逐渐增加电流的方式为电池充电，而不是直接以最大充电电流启动充电过程。本研究提出了扩展电流控制（ECC），以减少电池容量损耗，延长光伏供电 HESS 的使用寿命。光伏模块的最大功率点 (MPP) 由 Perturb and observe (P&O) 算法通过超级电容器 (SC) 转换器提供，而 ECC 则通过电池转换器执行。与基于优化、规则或滤波的技术不同，这种方法所需的硬件较少。在大面积范围内控制电池电流可保护电池组免受充电开始时的大电流影响，并通过延长循环寿命减少容量损失。与仅包含电池组的光伏供电 ESS 相比，所提出的技术可将电池充电电流提高 40%，将达到 MPP 点的转换器占空比提高 8%，将电池容量衰减提高 31.69%。

{"title":"PV-fed multi-output buck converter-based renewable energy storage system with extended current control for lifetime extension of Li-ion batteries","authors":"","doi":"10.1016/j.compeleceng.2024.109757","DOIUrl":"10.1016/j.compeleceng.2024.109757","url":null,"abstract":"<div><div>Recently, there has been a visible intensification of research on increasing the cycle life of energy storage devices used in Photovoltaic (PV)-fed energy storage systems (ESS). Compared to existing battery technologies, Lithium-ion (Li-ion) batteries have advantages such as high energy density and high cycle life. However, many existing Hybrid energy storage system (HESS) suffers from the fact that Li-ion batteries have limited cycle life, calendar and cycle aging. Therefore, attention has shifted towards the charging techniques necessary to improve the service life of Li-ion batteries by increasing their cycle life and reducing capacity losses. One such approach is to charge the battery by gradually increasing it from a certain level, rather than starting the charging process directly with the maximum charging current. This study proposes Extended Current Control (ECC) to reduce battery capacity losses and extend service life in PV-fed HESSs. The maximum power point (MPP) of the PV module is provided by the Perturb and observe (P&O) algorithm via the supercapacitor (SC) converter, while ECC is performed via the battery converter. This approach requires less hardware, unlike optimization, rule, or filtering based techniques. Controlling the battery current over a large area protects the battery packs from high currents at the start of charge and reduces capacity losses by increasing cycle life. Compared to PV-fed ESS containing only battery packs, the proposed technique provides a 40 % improvement in battery charging current, an 8 % improvement in the converter duty ratio in reaching the MPP point, and a 31.69 % improvement in battery capacity fades.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":4.0,"publicationDate":"2024-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142418988","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Revolutionizing protection dynamics in microgrids: Local validation environment and a novel global management control through multi-agent systems 革新微电网的保护动态：本地验证环境和通过多代理系统实现的新型全局管理控制

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2024-10-12 DOI: 10.1016/j.compeleceng.2024.109748

Amid rapid renewable sources expansion in distribution systems, novel operational and protection challenges emerge due to dynamic variations. To address these issues, an innovative hierarchical protection strategy based on an agent-based structure has been developed. This approach integrates a dual-tier system managing digital structure comparisons and modular agent fusion. At the upper tier, critical operational decisions are made while preserving protection coordination, while lower tiers oversee event analysis and relay updates. A key advantage of this system is its autonomous ability to sustain protection coordination during adjustments of topology and control operation scenarios. Integration of controlled harmonic insertion post-fault and control adaptations ensures stability. A validation agent divides the network through IDs, enabling precise communication and ensuring the effective functioning of supplementary protection across various zones, independent of performance failures in the master agent. Efficacy of this strategy has been validated through simulation studies within the University of Valle’s microgrid.

随着配电系统中可再生能源的快速扩张，动态变化带来了新的运行和保护挑战。为解决这些问题，我们开发了一种基于代理结构的创新分层保护策略。这种方法整合了一个双层系统，管理数字结构比较和模块化代理融合。在上层，在保持保护协调的同时做出关键的运行决策，而下层则负责事件分析和中继更新。该系统的一个关键优势在于，它能够在拓扑结构调整和控制操作方案期间自主维持保护协调。故障后受控谐波插入与控制适应的整合确保了稳定性。验证代理通过 ID 对网络进行划分，从而实现精确通信，并确保辅助保护在不同区域有效运行，不受主代理性能故障的影响。瓦莱大学微电网的模拟研究验证了这一策略的有效性。

{"title":"Revolutionizing protection dynamics in microgrids: Local validation environment and a novel global management control through multi-agent systems","authors":"","doi":"10.1016/j.compeleceng.2024.109748","DOIUrl":"10.1016/j.compeleceng.2024.109748","url":null,"abstract":"<div><div>Amid rapid renewable sources expansion in distribution systems, novel operational and protection challenges emerge due to dynamic variations. To address these issues, an innovative hierarchical protection strategy based on an agent-based structure has been developed. This approach integrates a dual-tier system managing digital structure comparisons and modular agent fusion. At the upper tier, critical operational decisions are made while preserving protection coordination, while lower tiers oversee event analysis and relay updates. A key advantage of this system is its autonomous ability to sustain protection coordination during adjustments of topology and control operation scenarios. Integration of controlled harmonic insertion post-fault and control adaptations ensures stability. A validation agent divides the network through IDs, enabling precise communication and ensuring the effective functioning of supplementary protection across various zones, independent of performance failures in the master agent. Efficacy of this strategy has been validated through simulation studies within the University of Valle’s microgrid.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":4.0,"publicationDate":"2024-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142418889","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

UTXOAnalysis: A distributed graph storage and analysis system for UTXO-based cryptocurrencies UTXOAnalysis：基于UTXO的加密货币的分布式图存储和分析系统

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2024-10-12 DOI: 10.1016/j.compeleceng.2024.109760

Relationship analysis of cryptocurrencies is crucial for understanding and regulating their ecosystems with graph structures. In particular, the relationships of the UTXO-based cryptocurrencies, which were developed earlier, form ecosystems with more integrated and larger graph structures. It is a challenge to efficiently store such large graphs and to efficiently query and analyze these graphs. In this paper, we propose UTXOAnalysis to solve these problems. UTXOAnalysis is a distributed graph storage and analysis system with a three-level structure. In the data collection framework, UTXOAnalysis adopts batch queries and block pruning. In the parsing and storage framework, UTXOAnalysis utilizes the parallel approaches of multi-graph parsing and storage. UTXOAnalysis also provides these methods for incremental data. In the analysis framework, UTXOAnalysis supplies address relationship analysis, transaction relationship analysis, and cryptocurrency flow tracing, which are the three basic analysis methods. In our experiments, we collected data from Bitcoin and Zcash to demonstrate the efficiency of data collection, parsing, storage, and basic analysis. UTXOAnalysis is more efficient in storing and analyzing UTXO-based cryptocurrencies than the baseline methods.

加密货币的关系分析对于理解和规范其具有图结构的生态系统至关重要。特别是，早期开发的基于 UTXO 的加密货币的关系形成了具有更综合、更大型图结构的生态系统。如何高效地存储这样的大型图以及如何高效地查询和分析这些图是一个挑战。在本文中，我们提出了UTXOAnalysis 来解决这些问题。UTXOAnalysis是一个具有三层结构的分布式图存储和分析系统。在数据收集框架中，UTXOAnalysis 采用批量查询和块剪枝技术。在解析和存储框架中，UTXOAnalysis 采用多图解析和存储的并行方法。UTXOAnalysis还为增量数据提供了这些方法。在分析框架中，UTXOAnalysis 提供了地址关系分析、交易关系分析和加密货币流追踪这三种基本分析方法。在实验中，我们收集了比特币和 Zcash 的数据，以展示数据收集、解析、存储和基本分析的效率。与基线方法相比，UTXOAnalysis 在存储和分析基于 UTXO 的加密货币方面效率更高。

{"title":"UTXOAnalysis: A distributed graph storage and analysis system for UTXO-based cryptocurrencies","authors":"","doi":"10.1016/j.compeleceng.2024.109760","DOIUrl":"10.1016/j.compeleceng.2024.109760","url":null,"abstract":"<div><div>Relationship analysis of cryptocurrencies is crucial for understanding and regulating their ecosystems with graph structures. In particular, the relationships of the UTXO-based cryptocurrencies, which were developed earlier, form ecosystems with more integrated and larger graph structures. It is a challenge to efficiently store such large graphs and to efficiently query and analyze these graphs. In this paper, we propose UTXOAnalysis to solve these problems. UTXOAnalysis is a distributed graph storage and analysis system with a three-level structure. In the data collection framework, UTXOAnalysis adopts batch queries and block pruning. In the parsing and storage framework, UTXOAnalysis utilizes the parallel approaches of multi-graph parsing and storage. UTXOAnalysis also provides these methods for incremental data. In the analysis framework, UTXOAnalysis supplies address relationship analysis, transaction relationship analysis, and cryptocurrency flow tracing, which are the three basic analysis methods. In our experiments, we collected data from Bitcoin and Zcash to demonstrate the efficiency of data collection, parsing, storage, and basic analysis. UTXOAnalysis is more efficient in storing and analyzing UTXO-based cryptocurrencies than the baseline methods.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":4.0,"publicationDate":"2024-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142418887","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

High precision DSRC and LiDAR data integration positioning method for autonomous vehicles based on CNN 基于 CNN 的自动驾驶汽车高精度 DSRC 和激光雷达数据集成定位方法

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2024-10-12 DOI: 10.1016/j.compeleceng.2024.109741

In order to improve the safe driving and automatic positioning capability of autonomous vehicles, a high-precision DSRC and LiDAR data integration positioning technology for autonomous vehicles based on CNN is proposed. Import the data of Dedicated Short Range Communications and Light Detection and Ranging for automatic driving vehicle positioning, carry out kinematic analysis of autonomous driving vehicles under multi-sensor fusion, and transform the data of DSRC and LiDAR sensors into tightly coupled coordinate systems; The CNN depth learning method is used to compensate the position and attitude tracking estimation error under the overall time stamp synchronization of the sensor through adaptive information tracking; The first and second order feedforward compensation is made for the positioning parameters of the autonomous driving vehicle using the PID model, and the point cloud feature matching model is fused to complete the estimation of the positioning attitude parameters of the autonomous driving vehicle. In order to eliminate the noise interference under the DSRC communication mechanism, the Kalman filter function is used to automatically optimize the constraint parameters in the point cloud feature detection model, and the positioning error parameters are dynamically filtered and adjusted; Kinematics analysis is carried out for the driving state of the vehicle, and the positioning error in the vehicle movement is controlled through the difference technology to achieve high-precision DSRC and LiDAR data integration positioning. The simulation results show that this method can integrate and locate the high-precision DSRC and LiDAR data of the autonomous vehicle, and the attitude estimation and positioning accuracy of the vehicle is good, while the error of the attitude parameter estimation of the autonomous vehicle is low.

为了提高自动驾驶汽车的安全驾驶和自动定位能力，提出了一种基于 CNN 的自动驾驶汽车高精度 DSRC 和激光雷达数据融合定位技术。导入专用短程通信和光探测与测距数据用于自动驾驶车辆定位，对多传感器融合下的自动驾驶车辆进行运动学分析，将DSRC和LiDAR传感器数据转化为紧密耦合的坐标系；采用 CNN 深度学习方法，通过自适应信息跟踪补偿传感器整体时戳同步下的位置和姿态跟踪估计误差；利用 PID 模型对自主驾驶车辆的定位参数进行一阶和二阶前馈补偿，融合点云特征匹配模型完成自主驾驶车辆的定位姿态参数估计。为了消除DSRC通信机制下的噪声干扰，利用卡尔曼滤波函数对点云特征检测模型中的约束参数进行自动优化，对定位误差参数进行动态滤波调整；对车辆的行驶状态进行运动学分析，通过差分技术控制车辆运动中的定位误差，实现高精度的DSRC和LiDAR数据融合定位。仿真结果表明，该方法能对自主车辆的高精度DSRC和LiDAR数据进行集成定位，车辆姿态估计和定位精度好，自主车辆姿态参数估计误差小。

{"title":"High precision DSRC and LiDAR data integration positioning method for autonomous vehicles based on CNN","authors":"","doi":"10.1016/j.compeleceng.2024.109741","DOIUrl":"10.1016/j.compeleceng.2024.109741","url":null,"abstract":"<div><div>In order to improve the safe driving and automatic positioning capability of autonomous vehicles, a high-precision DSRC and LiDAR data integration positioning technology for autonomous vehicles based on CNN is proposed. Import the data of Dedicated Short Range Communications and Light Detection and Ranging for automatic driving vehicle positioning, carry out kinematic analysis of autonomous driving vehicles under multi-sensor fusion, and transform the data of DSRC and LiDAR sensors into tightly coupled coordinate systems; The CNN depth learning method is used to compensate the position and attitude tracking estimation error under the overall time stamp synchronization of the sensor through adaptive information tracking; The first and second order feedforward compensation is made for the positioning parameters of the autonomous driving vehicle using the PID model, and the point cloud feature matching model is fused to complete the estimation of the positioning attitude parameters of the autonomous driving vehicle. In order to eliminate the noise interference under the DSRC communication mechanism, the Kalman filter function is used to automatically optimize the constraint parameters in the point cloud feature detection model, and the positioning error parameters are dynamically filtered and adjusted; Kinematics analysis is carried out for the driving state of the vehicle, and the positioning error in the vehicle movement is controlled through the difference technology to achieve high-precision DSRC and LiDAR data integration positioning. The simulation results show that this method can integrate and locate the high-precision DSRC and LiDAR data of the autonomous vehicle, and the attitude estimation and positioning accuracy of the vehicle is good, while the error of the attitude parameter estimation of the autonomous vehicle is low.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":4.0,"publicationDate":"2024-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142419090","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A distributed end-to-end fair bandwidth allocation algorithm for multi-path networks 多路径网络的分布式端到端公平带宽分配算法

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2024-10-11 DOI: 10.1016/j.compeleceng.2024.109635

Multi-path transmission significantly improves network performance, yet it complicates the problem of fair resource allocation. While traditional fair bandwidth allocation schemes thrive in single-path settings, they often falter when applied to multi-path environments, highlighting the challenge of achieving fair bandwidth sharing in such networks. To tackle this issue, the concept of "max-min similarity" of queuing delays has been introduced based on insight into the intrinsic interactions between queuing packets, delays, and bandwidth allocation, which leads to a formal definition of max-min fair bandwidth allocation for multi-path environments. Theoretical analysis shows that the max-min similarity of queuing delays is a sufficient condition for max-min fair bandwidth allocation in single bottleneck environments. A novel distributed end-to-end fair bandwidth allocation algorithm, named DMFBA, is then proposed, which separates the control into flow-level and transmission path-level. In achieving max-min similarity in queuing delays by dynamically adjusting the distribution of flows’ queuing packet quotas across paths it achieves the goal of max-min fair bandwidth allocation. Two sets of numerical simulation experiments were conducted and the results show that DMFBA has less overhead and faster convergence than the traditional utility fair algorithms.

多路径传输大大提高了网络性能，但也使公平资源分配问题变得更加复杂。传统的公平带宽分配方案在单路径环境中表现出色，但应用于多路径环境时往往会出现问题，这凸显了在此类网络中实现公平带宽共享所面临的挑战。为了解决这个问题，我们基于对队列数据包、延迟和带宽分配之间内在相互作用的洞察，引入了队列延迟的 "最大最小相似性 "概念，从而为多路径环境下的最大最小公平带宽分配下了一个正式定义。理论分析表明，队列延迟的最大最小相似性是单瓶颈环境下最大最小公平带宽分配的充分条件。随后提出了一种名为 DMFBA 的新型分布式端到端公平带宽分配算法，该算法将控制分为流量级和传输路径级。通过动态调整不同路径上流量队列包配额的分布，实现队列延迟的最大最小相似性，从而达到最大最小公平带宽分配的目标。我们进行了两组数值模拟实验，结果表明，与传统的实用公平算法相比，DMFBA 的开销更小，收敛速度更快。

{"title":"A distributed end-to-end fair bandwidth allocation algorithm for multi-path networks","authors":"","doi":"10.1016/j.compeleceng.2024.109635","DOIUrl":"10.1016/j.compeleceng.2024.109635","url":null,"abstract":"<div><div>Multi-path transmission significantly improves network performance, yet it complicates the problem of fair resource allocation. While traditional fair bandwidth allocation schemes thrive in single-path settings, they often falter when applied to multi-path environments, highlighting the challenge of achieving fair bandwidth sharing in such networks. To tackle this issue, the concept of \"max-min similarity\" of queuing delays has been introduced based on insight into the intrinsic interactions between queuing packets, delays, and bandwidth allocation, which leads to a formal definition of max-min fair bandwidth allocation for multi-path environments. Theoretical analysis shows that the max-min similarity of queuing delays is a sufficient condition for max-min fair bandwidth allocation in single bottleneck environments. A novel distributed end-to-end fair bandwidth allocation algorithm, named DMFBA, is then proposed, which separates the control into flow-level and transmission path-level. In achieving max-min similarity in queuing delays by dynamically adjusting the distribution of flows’ queuing packet quotas across paths it achieves the goal of max-min fair bandwidth allocation. Two sets of numerical simulation experiments were conducted and the results show that DMFBA has less overhead and faster convergence than the traditional utility fair algorithms.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":4.0,"publicationDate":"2024-10-11","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142421048","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Spatiotemporal Action Detection Using 2D CNN and 3D CNN 使用二维 CNN 和三维 CNN 进行时空动作检测

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2024-10-10 DOI: 10.1016/j.compeleceng.2024.109739

In order to address the low accuracy issue in human spatiotemporal action detection tasks, this study proposes a more effective CNN framework. Like YOWO model, we also use CNN for feature extraction, however, we only utilize the extracted spatiotemporal features for action recognition and the fused features of spatiotemporal and spatial information for action localization. Additionally, in the action localization branch, we make improvements to the original channel fusion and attention mechanism (CFAM). We introduce a combination of convolution and attention mechanisms to selectively replace the traditional convolutions, enabling more effective utilization of the fused features. Finally, in order to make the model more accurate for bounding box regression, we use CIoU loss instead of the offset loss. Results show that our proposed method achieves frame-mAP scores (@IoU 0.5) of 75.73 % and 83.13 % on JHMDB-21 and UCF101–24 datasets, respectively. For video-mAP, we obtain 88.96 %, 85.81 % and 68.59 % at IoU threshold of 0.2,0.5 and 0.75 on JHMDB-21 dataset and 75.05 %, 69.72 % and 48.95 % at IoU threshold of 0.1,0.2 and 0.5 on UCF101–24 dataset.

为了解决人类时空动作检测任务中的低准确率问题，本研究提出了一种更有效的 CNN 框架。与 YOWO 模型一样，我们也使用 CNN 进行特征提取，但我们只将提取的时空特征用于动作识别，而将时空信息的融合特征用于动作定位。此外，在动作定位分支中，我们对原有的通道融合和注意力机制（CFAM）进行了改进。我们引入了卷积和注意力机制的组合，选择性地取代了传统的卷积，从而更有效地利用了融合后的特征。最后，为了使边界框回归模型更加精确，我们使用 CIoU 损失代替偏移损失。结果表明，我们提出的方法在 JHMDB-21 和 UCF101-24 数据集上的帧-映射得分（@IoU 0.5）分别达到 75.73 % 和 83.13 %。在视频映射率方面，当 IoU 阈值为 0.2、0.5 和 0.75 时，我们在 JHMDB-21 数据集上分别获得了 88.96 %、85.81 % 和 68.59 %；当 IoU 阈值为 0.1、0.2 和 0.5 时，我们在 UCF101-24 数据集上分别获得了 75.05 %、69.72 % 和 48.95 %。

引用次数: 0

Hybrid leader corona virus herd optimizer with multilevel thresholding techniques for foreground and background image segmentation 混合领导者日冕病毒群优化器与多级阈值技术用于前景和背景图像分割

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2024-10-10 DOI: 10.1016/j.compeleceng.2024.109569

In computer vision and image processing, an important role is given to image segmentation. As it can accurately extract and identify specific regions within an image. One of the key challenges that remain in this is achieving precise recognition and extraction of segmented regions. This can be solved by multilevel optimization techniques and this offers an effective solution. Recently, multilevel thresholding has emerged as an important technique for image segmentation as the separation of image pixels into various classes by the selection of optimal threshold values. However, when the number of thresholds increases, the computational complexity of multilevel thresholding also increases. To address this issue, various optimization algorithms are employed by the researchers. Thus, the Hybrid optimization techniques are integrated to enhance the overall efficacy of image segmentation processes. Thus, this research devised a Multithreshold-Hybrid Leader Coronavirus Herd Optimizer+kernel based Bayesian fuzzy clustering (Multithreshold-HLCHO+kernel-BFC) for the segmentation of foreground and background images. Here, the noise from the input image is removed using the Non-Local Means (NLM) filter and also the region of interest (ROI) extraction is done. The foreground and background image segmentation is done using the multilevel thresholding techniques, wherein threshold values are optimally generated utilizing HLCHO with multiobjectives like Renyi, Masi and Tsallic. Moreover, the integration of Hybrid Leader Based Optimization (HLBO) with Coronavirus Herd Immunity Optimizer (CHIO) forms the HLCHO. Also, the kernel-based BFC is employed for the segmentation of background and foreground images. Finally, the segmented output is achieved by fusing these two outputs using a fusion process. Additionally, Multithreshold-HLCHO+kernel-BFC acquired a maximum value of 0.891 for dice coefficient, 37.174 dB for PSNR, 0.918 for uniformity measure, and 0.297 for Mean Squared error (MSE).

在计算机视觉和图像处理中，图像分割发挥着重要作用。因为它可以准确提取和识别图像中的特定区域。实现精确识别和提取分割区域是其中仍然存在的关键挑战之一。这可以通过多级优化技术来解决，它提供了一种有效的解决方案。最近，多级阈值技术作为一种重要的图像分割技术出现了，它通过选择最佳阈值将图像像素分成不同的类别。然而，当阈值数量增加时，多级阈值的计算复杂度也随之增加。为了解决这个问题，研究人员采用了各种优化算法。因此，混合优化技术被整合在一起，以提高图像分割过程的整体效率。因此，本研究设计了一种多阈值-混合领导冠状病毒群优化器+基于内核的贝叶斯模糊聚类（Multithreshold-HLCHO+kernel-BFC），用于前景和背景图像的分割。在此，使用非局部均值（NLM）滤波器去除输入图像中的噪声，并提取感兴趣区域（ROI）。使用多级阈值技术进行前景和背景图像分割，其中阈值是利用 HLCHO 与 Renyi、Masi 和 Tsallic 等多目标优化生成的。此外，基于混合领导优化（HLBO）与冠状病毒群免疫优化器（CHIO）的集成构成了 HLCHO。同时，基于核的 BFC 被用于分割背景和前景图像。最后，通过使用融合程序将这两个输出结果融合在一起，实现分割输出。此外，Multithreshold-HLCHO+kernel-BFC 的骰子系数最大值为 0.891，PSNR 为 37.174 dB，均匀性测量值为 0.918，平均平方误差 (MSE) 为 0.297。

{"title":"Hybrid leader corona virus herd optimizer with multilevel thresholding techniques for foreground and background image segmentation","authors":"","doi":"10.1016/j.compeleceng.2024.109569","DOIUrl":"10.1016/j.compeleceng.2024.109569","url":null,"abstract":"<div><div>In computer vision and image processing, an important role is given to image segmentation. As it can accurately extract and identify specific regions within an image. One of the key challenges that remain in this is achieving precise recognition and extraction of segmented regions. This can be solved by multilevel optimization techniques and this offers an effective solution. Recently, multilevel thresholding has emerged as an important technique for image segmentation as the separation of image pixels into various classes by the selection of optimal threshold values. However, when the number of thresholds increases, the computational complexity of multilevel thresholding also increases. To address this issue, various optimization algorithms are employed by the researchers. Thus, the Hybrid optimization techniques are integrated to enhance the overall efficacy of image segmentation processes. Thus, this research devised a Multithreshold-Hybrid Leader Coronavirus Herd Optimizer+kernel based Bayesian fuzzy clustering (Multithreshold-HLCHO+kernel-BFC) for the segmentation of foreground and background images. Here, the noise from the input image is removed using the Non-Local Means (NLM) filter and also the region of interest (ROI) extraction is done. The foreground and background image segmentation is done using the multilevel thresholding techniques, wherein threshold values are optimally generated utilizing HLCHO with multiobjectives like Renyi, Masi and Tsallic. Moreover, the integration of Hybrid Leader Based Optimization (HLBO) with Coronavirus Herd Immunity Optimizer (CHIO) forms the HLCHO. Also, the kernel-based BFC is employed for the segmentation of background and foreground images. Finally, the segmented output is achieved by fusing these two outputs using a fusion process. Additionally, Multithreshold-HLCHO+kernel-BFC acquired a maximum value of 0.891 for dice coefficient, 37.174 dB for PSNR, 0.918 for uniformity measure, and 0.297 for Mean Squared error (MSE).</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":4.0,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142421049","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Edge-cloud collaboration for low-latency, low-carbon, and cost-efficient operations 边缘-云协作，实现低延迟、低碳和经济高效的运营

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2024-10-10 DOI: 10.1016/j.compeleceng.2024.109758

The growing demand for low-latency services and the increasing impact of carbon emissions pose challenges to traditional cloud computing architectures. Hence, to address the high latency limitations of traditional cloud computing and leverage the advantages of abundant renewable energy sources (RESs) and low-priced electricity of remote clouds, we design an edge-cloud collaboration system to distribute mixed workloads, aiming at meeting delay requirements while reducing carbon emissions and improving operating profits. Specifically, delay-sensitive workloads are allocated to nearby edge clouds, while delay-tolerant workloads are assigned to remote core clouds. Additionally, a multi-level scheduling strategy is proposed to flexibly allocate delay-tolerant workloads. Beyond responding to RES generation and electricity price signals, this strategy prioritizes workloads and reduces the supply of high-priced electricity to low-priority workloads, further decreasing electricity costs. Finally, we use Alibaba workload traces to evaluate the proposed strategy. Simulation results demonstrate that the proposed edge-cloud collaboration system can reduce the average response delay of delay-sensitive workloads by 33.42 times compared to the traditional cloud system. Additionally, compared to the effective energy storage systems (ESSs)-based algorithm, the proposed strategy not only reduces carbon emissions by 3.14% but also increases operating profits by 18.78%. These results highlight its potential to enhance environmental sustainability, economic benefits, and Quality of Service (QoS).

对低延迟服务日益增长的需求和碳排放日益严重的影响给传统的云计算架构带来了挑战。因此，为了解决传统云计算的高延迟限制，并利用远程云丰富的可再生能源（RES）和低价电力的优势，我们设计了一种边缘云协作系统来分配混合工作负载，旨在满足延迟要求的同时减少碳排放并提高运营利润。具体来说，将对延迟敏感的工作负载分配给附近的边缘云，而将耐延迟的工作负载分配给远程核心云。此外，还提出了一种多级调度策略，以灵活分配延迟容忍工作负载。除了响应可再生能源发电和电价信号外，该策略还能确定工作负载的优先级，减少对低优先级工作负载的高价电力供应，从而进一步降低电力成本。最后，我们使用阿里巴巴工作负载跟踪来评估所提出的策略。仿真结果表明，与传统云系统相比，所提出的边缘云协作系统可将延迟敏感型工作负载的平均响应延迟降低 33.42 倍。此外，与基于有效能源存储系统（ESS）的算法相比，所提出的策略不仅减少了 3.14% 的碳排放，还增加了 18.78% 的运营利润。这些结果凸显了它在提高环境可持续性、经济效益和服务质量（QoS）方面的潜力。

{"title":"Edge-cloud collaboration for low-latency, low-carbon, and cost-efficient operations","authors":"","doi":"10.1016/j.compeleceng.2024.109758","DOIUrl":"10.1016/j.compeleceng.2024.109758","url":null,"abstract":"<div><div>The growing demand for low-latency services and the increasing impact of carbon emissions pose challenges to traditional cloud computing architectures. Hence, to address the high latency limitations of traditional cloud computing and leverage the advantages of abundant renewable energy sources (RESs) and low-priced electricity of remote clouds, we design an edge-cloud collaboration system to distribute mixed workloads, aiming at meeting delay requirements while reducing carbon emissions and improving operating profits. Specifically, delay-sensitive workloads are allocated to nearby edge clouds, while delay-tolerant workloads are assigned to remote core clouds. Additionally, a multi-level scheduling strategy is proposed to flexibly allocate delay-tolerant workloads. Beyond responding to RES generation and electricity price signals, this strategy prioritizes workloads and reduces the supply of high-priced electricity to low-priority workloads, further decreasing electricity costs. Finally, we use Alibaba workload traces to evaluate the proposed strategy. Simulation results demonstrate that the proposed edge-cloud collaboration system can reduce the average response delay of delay-sensitive workloads by 33.42 times compared to the traditional cloud system. Additionally, compared to the effective energy storage systems (ESSs)-based algorithm, the proposed strategy not only reduces carbon emissions by 3.14% but also increases operating profits by 18.78%. These results highlight its potential to enhance environmental sustainability, economic benefits, and Quality of Service (QoS).</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":4.0,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142419093","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

An intelligent ANFIS-based fault detection, classification, and location model for VSC-HVDC systems based on hybrid PCA-DWT signal processing technique 基于混合 PCA-DWT 信号处理技术的 VSC-HVDC 系统 ANFIS 智能故障检测、分类和定位模型

IF 4 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2024-10-10 DOI: 10.1016/j.compeleceng.2024.109763

Designing an accurate model for fault detection, classification, and location is vital from the protection viewpoint. The adaptive neuro-fuzzy inference system (ANFIS) is a commonly used learning-based fault location model that performs independently from the propagating wave characteristics, whose performance can be improved by optimizing the membership functions associated with the inputs. Accordingly, two metaheuristic algorithms with the quick-search capability of the population space, i.e., Harris Hawks optimization (HHO) and cuckoo search (CS) optimization algorithms, are used in this paper to optimize the ANFIS, and their performances are compared with the traditional ANFIS training algorithms. Moreover, principal component analysis (PCA) is employed to detect the fault, and the discrete wavelet transform (DWT) strategy is exploited to acquire the ANFIS training and testing dataset according to the statistics T² obtained by PCA. Three statistical indices, i.e., mean value, standard deviation, and norm entropy, are computed corresponding to the extracted wavelet coefficients from the current signal and applied to train and test the ANFIS. Optimized ANFIS conducts fault classification and location tasks, and the accuracy of the proposed model is compared with the commonly used traveling wave (TW) –based models and recently proposed fault location methods in the literature. Three fault types on the DC-link with three fault resistances are examined to confirm the fault classification superiority of the proposed model and its fault type/impedance-independent estimation capability. An accuracy rate of 99.995% is obtained for the fault locating task, while fault detecting and classifying are accomplished with an accuracy of 100%. Simulations and numerical studies are performed in MATLAB software.

从保护角度来看，设计一个准确的故障检测、分类和定位模型至关重要。自适应神经模糊推理系统（ANFIS）是一种常用的基于学习的故障定位模型，其性能不受传播波特性的影响，可通过优化与输入相关的成员函数来提高其性能。因此，本文采用了两种具有群体空间快速搜索能力的元启发式算法，即哈里斯-霍克斯优化算法（HHO）和布谷鸟搜索优化算法（CS）来优化 ANFIS，并将其性能与传统 ANFIS 训练算法进行了比较。此外，本文还采用主成分分析（PCA）来检测故障，并利用离散小波变换（DWT）策略，根据 PCA 得到的统计量 T2 获取 ANFIS 训练和测试数据集。根据从当前信号中提取的小波系数，计算出与之相对应的三个统计指标，即平均值、标准偏差和规范熵，并将其应用于 ANFIS 的训练和测试。优化后的 ANFIS 执行故障分类和定位任务，并将所提模型的准确性与常用的基于行波（TW）的模型和近期文献中提出的故障定位方法进行比较。研究了直流链路上的三种故障类型和三种故障电阻，以证实所提模型的故障分类优势及其与故障类型/阻抗无关的估计能力。故障定位任务的准确率为 99.995%，而故障检测和分类的准确率为 100%。仿真和数值研究在 MATLAB 软件中进行。

{"title":"An intelligent ANFIS-based fault detection, classification, and location model for VSC-HVDC systems based on hybrid PCA-DWT signal processing technique","authors":"","doi":"10.1016/j.compeleceng.2024.109763","DOIUrl":"10.1016/j.compeleceng.2024.109763","url":null,"abstract":"<div><div>Designing an accurate model for fault detection, classification, and location is vital from the protection viewpoint. The adaptive neuro-fuzzy inference system (ANFIS) is a commonly used learning-based fault location model that performs independently from the propagating wave characteristics, whose performance can be improved by optimizing the membership functions associated with the inputs. Accordingly, two metaheuristic algorithms with the quick-search capability of the population space, i.e., Harris Hawks optimization (HHO) and cuckoo search (CS) optimization algorithms, are used in this paper to optimize the ANFIS, and their performances are compared with the traditional ANFIS training algorithms. Moreover, principal component analysis (PCA) is employed to detect the fault, and the discrete wavelet transform (DWT) strategy is exploited to acquire the ANFIS training and testing dataset according to the statistics <em>T</em><sup>2</sup> obtained by PCA. Three statistical indices, i.e., mean value, standard deviation, and norm entropy, are computed corresponding to the extracted wavelet coefficients from the current signal and applied to train and test the ANFIS. Optimized ANFIS conducts fault classification and location tasks, and the accuracy of the proposed model is compared with the commonly used traveling wave (TW) –based models and recently proposed fault location methods in the literature. Three fault types on the DC-link with three fault resistances are examined to confirm the fault classification superiority of the proposed model and its fault type/impedance-independent estimation capability. An accuracy rate of 99.995% is obtained for the fault locating task, while fault detecting and classifying are accomplished with an accuracy of 100%. Simulations and numerical studies are performed in MATLAB software.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":null,"pages":null},"PeriodicalIF":4.0,"publicationDate":"2024-10-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"142418886","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0