Journal of Engineering Research最新文献_第6页

Experimental investigation of micro fracture characteristics and the influence of water drive in shallow tight sandstone reservoirs 浅层致密砂岩储层微裂缝特征及水驱影响实验研究

IF 2.2 4区工程技术 Q3 ENGINEERING, MULTIDISCIPLINARY

Journal of Engineering Research

Pub Date : 2025-12-01 DOI: 10.1016/j.jer.2025.02.001

Yanlong Wang , Mingwei Wang , Zhendong Gao , Wen Wu , Qi Ni , Tao Li , Yu Yang

The description of reservoir fractures constitutes the basis for studying fracture formation and distribution, as well as for evaluating the significance of fracture effects in different formations, conducting effectiveness analysis, and predicting the underground fluid seepage network system. Given that the distribution, development methods, and effectiveness of fractured oil and gas reservoirs are mainly restricted by fracture development, accurately and comprehensively describing fracture types, filling conditions, formations, orientations, densities, and scales is of crucial importance for exploration, well network deployment, development plan design, and subsequent production management. Taking the natural fractures in the Yanchang Formation reservoir of the southern exploration area in the Ordos Basin, China, as the research object, a macroscopic description of core fractures was initially conducted. Combined with thin section identification, scanning electron microscopy, and energy spectrum technology, the scale, occurrence, nature, and filling conditions of natural fractures were clarified. The research findings indicate that a considerable number of micro-fractures, accounting for over 60 % of the total, show widths ranging from 10 to 20 µm within the target reservoir. The manifestation of micro-fractures varies among different lithologies: conglomerates have granular micro-fractures with impermeable cements; micro-fractures within silty conglomerates present a diverse range of shapes, sometimes distributed in bands; and those in mudstone appear as intermittent or spider-web patterns. The results of the water flooding experiment imply that longer and wider fractures decrease water flooding efficiency. During low-pressure water flooding, pore-scale fractures display a relatively weak response and only show their characteristics when the applied pressure reaches their ultimate extension pressure. The characterization and evaluation of natural fractures provide crucial references for the efficient development of tight sandstone reservoirs in the study area and offer valuable guidance for optimizing water flooding techniques and enhancing recovery efficiency.

储层裂缝的描述是研究裂缝形成和分布的基础，也是评价不同地层裂缝作用意义、进行有效性分析、预测地下流体渗流网络系统的基础。裂缝性油气藏的分布、开发方式和效益主要受裂缝发育程度的制约，准确、全面地描述裂缝类型、充填条件、地层、取向、密度、规模对勘探、井网部署、开发方案设计及后续生产管理具有至关重要的意义。以鄂尔多斯盆地南部探区延长组储层天然裂缝为研究对象，初步对岩心裂缝进行宏观描述。结合薄片鉴定、扫描电镜、能谱等技术，明确了天然裂缝的规模、产状、性质及充填条件。研究结果表明，在目标储层内，相当数量的微裂缝宽度在10 ~ 20 µm之间，占总数的60%以上。微裂缝在不同岩性下表现不同：砾岩为颗粒状微裂缝，胶结物不透水；粉质砾岩内微裂缝形态多样，有时呈带状分布；而泥岩中的那些则呈间歇性或蛛网状。水驱试验结果表明，裂缝越长越宽，水驱效率越低。在低压水驱过程中，孔隙尺度裂缝的响应相对较弱，只有当施加压力达到其极限延伸压力时，孔隙尺度裂缝才会表现出其特征。天然裂缝的表征与评价为研究区致密砂岩储层的高效开发提供了重要参考，为优化水驱技术、提高采收率提供了有价值的指导。

{"title":"Experimental investigation of micro fracture characteristics and the influence of water drive in shallow tight sandstone reservoirs","authors":"Yanlong Wang , Mingwei Wang , Zhendong Gao , Wen Wu , Qi Ni , Tao Li , Yu Yang","doi":"10.1016/j.jer.2025.02.001","DOIUrl":"10.1016/j.jer.2025.02.001","url":null,"abstract":"<div><div>The description of reservoir fractures constitutes the basis for studying fracture formation and distribution, as well as for evaluating the significance of fracture effects in different formations, conducting effectiveness analysis, and predicting the underground fluid seepage network system. Given that the distribution, development methods, and effectiveness of fractured oil and gas reservoirs are mainly restricted by fracture development, accurately and comprehensively describing fracture types, filling conditions, formations, orientations, densities, and scales is of crucial importance for exploration, well network deployment, development plan design, and subsequent production management. Taking the natural fractures in the Yanchang Formation reservoir of the southern exploration area in the Ordos Basin, China, as the research object, a macroscopic description of core fractures was initially conducted. Combined with thin section identification, scanning electron microscopy, and energy spectrum technology, the scale, occurrence, nature, and filling conditions of natural fractures were clarified. The research findings indicate that a considerable number of micro-fractures, accounting for over 60 % of the total, show widths ranging from 10 to 20 µm within the target reservoir. The manifestation of micro-fractures varies among different lithologies: conglomerates have granular micro-fractures with impermeable cements; micro-fractures within silty conglomerates present a diverse range of shapes, sometimes distributed in bands; and those in mudstone appear as intermittent or spider-web patterns. The results of the water flooding experiment imply that longer and wider fractures decrease water flooding efficiency. During low-pressure water flooding, pore-scale fractures display a relatively weak response and only show their characteristics when the applied pressure reaches their ultimate extension pressure. The characterization and evaluation of natural fractures provide crucial references for the efficient development of tight sandstone reservoirs in the study area and offer valuable guidance for optimizing water flooding techniques and enhancing recovery efficiency.</div></div>","PeriodicalId":48803,"journal":{"name":"Journal of Engineering Research","volume":"13 4","pages":"Pages 3919-3928"},"PeriodicalIF":2.2,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145739036","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Synthesis of a second order band-stop filter 二阶带阻滤波器的合成

IF 2.2 4区工程技术 Q3 ENGINEERING, MULTIDISCIPLINARY

Journal of Engineering Research

Pub Date : 2025-12-01 DOI: 10.1016/j.jer.2025.02.007

Fatima Almutairi

This paper discusses an innovative strategy for implementing a band-stop filter. The proposed band-stop filter is designed using minimally invasive filter architecture with an active inductor. The transfer function of invasive filter is based on the impedance ratio; therefore, the filter type can be determined by choosing the grounding impedance at input terminal of the inverting amplifier. In order to design a band-stop filter, it is essential to integrate an inductor into the impedance framework. This inductor plays a crucial role in the overall functionality of the filter. Moreover, the use of an active inductor is particularly advantageous as it not only helps to conserve space within the circuit but also provides the flexibility for tunability, allowing for adjustments to the filter’s performance as needed. The proposed filter was validated through both simulation and experimental results. The results of the simulation indicate that the proposed filter consumes 2.86 mW from 1.8 V power supply, achieving 2.1 MHz center frequency with 430 KHz bandwidth. The integrated noise is 162.56 μV from 100 KHz to 10 MHz. It achieves an in-band IIP3 of 13.7 dBm. While, the discrete components were used to obtain the experimental results, achieving 2.5 KHz with 26.25 dB attenuation and 1.1 KHz bandwidth.

本文讨论了一种实现带阻滤波器的创新策略。所提出的带阻滤波器采用带有源电感的微创滤波器结构设计。侵入式滤波器的传递函数基于阻抗比；因此，可以通过选择反相放大器输入端的接地阻抗来确定滤波器的类型。为了设计带阻滤波器，必须将电感集成到阻抗框架中。该电感器在滤波器的整体功能中起着至关重要的作用。此外，使用有源电感是特别有利的，因为它不仅有助于节省电路内的空间，而且还提供了可调性的灵活性，允许根据需要调整滤波器的性能。仿真和实验结果验证了该滤波器的有效性。仿真结果表明，该滤波器从1.8 V电源中消耗2.86 mW，在430 KHz带宽下实现2.1 MHz的中心频率。在100 KHz ~ 10 MHz范围内的综合噪声为162.56 μV。实现了13.7 dBm的带内IIP3。而使用离散元件获得实验结果，达到2.5 KHz，衰减26.25 dB，带宽1.1 KHz。

{"title":"Synthesis of a second order band-stop filter","authors":"Fatima Almutairi","doi":"10.1016/j.jer.2025.02.007","DOIUrl":"10.1016/j.jer.2025.02.007","url":null,"abstract":"<div><div>This paper discusses an innovative strategy for implementing a band-stop filter. The proposed band-stop filter is designed using minimally invasive filter architecture with an active inductor. The transfer function of invasive filter is based on the impedance ratio; therefore, the filter type can be determined by choosing the grounding impedance at input terminal of the inverting amplifier. In order to design a band-stop filter, it is essential to integrate an inductor into the impedance framework. This inductor plays a crucial role in the overall functionality of the filter. Moreover, the use of an active inductor is particularly advantageous as it not only helps to conserve space within the circuit but also provides the flexibility for tunability, allowing for adjustments to the filter’s performance as needed. The proposed filter was validated through both simulation and experimental results. The results of the simulation indicate that the proposed filter consumes 2.86 mW from 1.8 V power supply, achieving 2.1 MHz center frequency with 430 KHz bandwidth. The integrated noise is 162.56 <em>μ</em>V from 100 KHz to 10 MHz. It achieves an in-band IIP3 of 13.7 dBm. While, the discrete components were used to obtain the experimental results, achieving 2.5 KHz with 26.25 dB attenuation and 1.1 KHz bandwidth.</div></div>","PeriodicalId":48803,"journal":{"name":"Journal of Engineering Research","volume":"13 4","pages":"Pages 3410-3417"},"PeriodicalIF":2.2,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145739249","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

SSD-based innovations for improved construction management 基于ssd的创新改进了施工管理

IF 2.2 4区工程技术 Q3 ENGINEERING, MULTIDISCIPLINARY

Journal of Engineering Research

Pub Date : 2025-12-01 DOI: 10.1016/j.jer.2025.02.004

Li-Wei Lung, Yu-Ren Wang

Amidst a global labor shortage, construction site workers face heightened pressure, with recruits needing swift adaptation, amplifying safety risks in complex environments. Yet, integrating image detection intelligence technology offers a solution, aiding managers in effectively tackling challenges and bolstering efficiency and safety in site management. This study addresses the challenge of accurately detecting objects in construction sites, where obstacles such as small objects, occlusion, and high object density hinder detection performance. Enhancing the Single Shot MultiBox Detector (SSD) model with tailored data preprocessing grounded in the DIKW hierarchy, known as Data Preprocessing's Fusion SSD (DPF-SSD), significantly improves detection accuracy. This novel DPF-SSD methodology can resolve incomplete or missing data, noise, and outliers. Firstly, it filters high-quality images that exhibit complexity, blurriness, and consistency with object characteristics. Secondly, it converts the image format and clarifies labeling rules for object characteristics. Thirdly, it loads corresponding label information and pixel coordinates for training alongside the object image. Finally, it integrates loss functions like L1 smooth loss and Softmax Loss to enhance feature clarity and improve identification accuracy. The effectiveness of the mentioned optimization was validated using a custom dataset in Python, implemented within the TensorFlow framework. Experimental results revealed that the enhanced SSD achieved an 81.97 % mean average precision (mAP), signifying a noteworthy 15.85 % enhancement compared to the original model. This advancement holds promise for enhancing productivity and safety within construction sites, thereby facilitating the automation of site management processes.

在全球劳动力短缺的情况下，建筑工地工人面临着更大的压力，新员工需要迅速适应，这加大了复杂环境中的安全风险。然而，集成图像检测智能技术提供了一个解决方案，帮助管理人员有效地应对挑战，提高现场管理的效率和安全性。该研究解决了在建筑工地准确检测物体的挑战，在建筑工地，小物体、遮挡和高物体密度等障碍物会阻碍检测性能。通过基于DIKW层次结构的定制数据预处理，即数据预处理的融合SSD (DPF-SSD)，增强了单次多盒探测器（SSD）模型，显著提高了检测精度。这种新颖的DPF-SSD方法可以解决不完整或缺失的数据、噪声和异常值。首先，它过滤出具有复杂性、模糊性和与目标特征一致性的高质量图像。其次，对图像格式进行转换，明确目标特征标注规则；第三步，在目标图像旁边加载相应的标签信息和像素坐标进行训练。最后，结合L1平滑损失、Softmax损失等损失函数，增强特征清晰度，提高识别精度。使用Python中的自定义数据集验证了上述优化的有效性，该数据集在TensorFlow框架中实现。实验结果表明，改进后的SSD实现了81.97 %的平均精度（mAP），与原始模型相比，显著提高了15.85 %。这一进步有望提高建筑工地的生产力和安全性，从而促进工地管理过程的自动化。

{"title":"SSD-based innovations for improved construction management","authors":"Li-Wei Lung, Yu-Ren Wang","doi":"10.1016/j.jer.2025.02.004","DOIUrl":"10.1016/j.jer.2025.02.004","url":null,"abstract":"<div><div>Amidst a global labor shortage, construction site workers face heightened pressure, with recruits needing swift adaptation, amplifying safety risks in complex environments. Yet, integrating image detection intelligence technology offers a solution, aiding managers in effectively tackling challenges and bolstering efficiency and safety in site management. This study addresses the challenge of accurately detecting objects in construction sites, where obstacles such as small objects, occlusion, and high object density hinder detection performance. Enhancing the Single Shot MultiBox Detector (SSD) model with tailored data preprocessing grounded in the DIKW hierarchy, known as Data Preprocessing's Fusion SSD (DPF-SSD), significantly improves detection accuracy. This novel DPF-SSD methodology can resolve incomplete or missing data, noise, and outliers. Firstly, it filters high-quality images that exhibit complexity, blurriness, and consistency with object characteristics. Secondly, it converts the image format and clarifies labeling rules for object characteristics. Thirdly, it loads corresponding label information and pixel coordinates for training alongside the object image. Finally, it integrates loss functions like L1 smooth loss and Softmax Loss to enhance feature clarity and improve identification accuracy. The effectiveness of the mentioned optimization was validated using a custom dataset in Python, implemented within the TensorFlow framework. Experimental results revealed that the enhanced SSD achieved an 81.97 % mean average precision (mAP), signifying a noteworthy 15.85 % enhancement compared to the original model. This advancement holds promise for enhancing productivity and safety within construction sites, thereby facilitating the automation of site management processes.</div></div>","PeriodicalId":48803,"journal":{"name":"Journal of Engineering Research","volume":"13 4","pages":"Pages 3070-3084"},"PeriodicalIF":2.2,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145739314","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Task scheduling and load balancing in SDN-based cloud computing: A review of relevant research 基于sdn的云计算任务调度与负载均衡研究综述

IF 2.2 4区工程技术 Q3 ENGINEERING, MULTIDISCIPLINARY

Journal of Engineering Research

Pub Date : 2025-12-01 DOI: 10.1016/j.jer.2024.11.002

Masoumeh Mahdizadeh, Ahmadreza Montazerolghaem, Kamal Jamshidi

This article presents a comprehensive exploration of the architecture and various approaches in the domain of cloud computing and software-defined networks. The salient points addressed in this article encompass: Foundational Concepts: An overview of the foundational concepts and technologies of cloud computing, including software-defined cloud computing. Algorithm Evaluation: An introduction and evaluation of various algorithms aimed at enhancing network performance. These algorithms include Intelligent Rule-Based Metaheuristic Task Scheduling (IRMTS), reinforcement learning algorithms, task scheduling algorithms, and Priority-aware Semi-Greedy (PSG). Each of these algorithms contributes uniquely to optimizing Quality of Service (QoS) and data center efficiency. Resource Optimization: An introduction and examination of cloud network resource optimization based on presented results and practical experiments, including a comparison of the performance of different algorithms and approaches. Future Challenges: An investigation and presentation of challenges and future scenarios in the realm of cloud computing and software-defined networks. In conclusion, by introducing and analyzing simulators like Mininet and CloudSim, the article guides the reader in choosing the most suitable simulation tool for their project. Through its comprehensive analysis of the architecture, methodologies, and prevalent algorithms in cloud computing and software-defined networking, this article aids the reader in achieving a deeper understanding of the domain. Additionally, by presenting the findings and results of conducted research, it facilitates the discovery of the most effective and practical solutions for optimizing cloud network resources.

本文全面探讨了云计算和软件定义网络领域的体系结构和各种方法。本文讨论的重点包括：基本概念：概述云计算的基本概念和技术，包括软件定义的云计算。算法评估：介绍和评估各种旨在提高网络性能的算法。这些算法包括基于规则的智能启发式任务调度（IRMTS）、强化学习算法、任务调度算法和优先级感知半贪婪（PSG）。每种算法都有助于优化服务质量（QoS）和数据中心效率。资源优化：基于提出的结果和实际实验，介绍和检查云网络资源优化，包括不同算法和方法的性能比较。未来挑战：调查和展示云计算和软件定义网络领域的挑战和未来场景。总之，本文通过对Mininet和CloudSim等模拟器的介绍和分析，指导读者选择最适合自己项目的仿真工具。通过对云计算和软件定义网络中的体系结构、方法和流行算法的全面分析，本文帮助读者更深入地了解该领域。此外，通过展示所进行研究的发现和结果，它有助于发现优化云网络资源的最有效和实用的解决方案。

{"title":"Task scheduling and load balancing in SDN-based cloud computing: A review of relevant research","authors":"Masoumeh Mahdizadeh, Ahmadreza Montazerolghaem, Kamal Jamshidi","doi":"10.1016/j.jer.2024.11.002","DOIUrl":"10.1016/j.jer.2024.11.002","url":null,"abstract":"<div><div>This article presents a comprehensive exploration of the architecture and various approaches in the domain of cloud computing and software-defined networks. The salient points addressed in this article encompass: <em>Foundational Concepts</em>: An overview of the foundational concepts and technologies of cloud computing, including software-defined cloud computing. <em>Algorithm Evaluation</em>: An introduction and evaluation of various algorithms aimed at enhancing network performance. These algorithms include Intelligent Rule-Based Metaheuristic Task Scheduling (IRMTS), reinforcement learning algorithms, task scheduling algorithms, and Priority-aware Semi-Greedy (PSG). Each of these algorithms contributes uniquely to optimizing Quality of Service (QoS) and data center efficiency. <em>Resource Optimization</em>: An introduction and examination of cloud network resource optimization based on presented results and practical experiments, including a comparison of the performance of different algorithms and approaches. <em>Future Challenges</em>: An investigation and presentation of challenges and future scenarios in the realm of cloud computing and software-defined networks. In conclusion, by introducing and analyzing simulators like Mininet and CloudSim, the article guides the reader in choosing the most suitable simulation tool for their project. Through its comprehensive analysis of the architecture, methodologies, and prevalent algorithms in cloud computing and software-defined networking, this article aids the reader in achieving a deeper understanding of the domain. Additionally, by presenting the findings and results of conducted research, it facilitates the discovery of the most effective and practical solutions for optimizing cloud network resources.</div></div>","PeriodicalId":48803,"journal":{"name":"Journal of Engineering Research","volume":"13 4","pages":"Pages 3132-3146"},"PeriodicalIF":2.2,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145739318","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimized deep learning model for medical image diagnosis 优化的医学图像诊断深度学习模型

IF 2.2 4区工程技术 Q3 ENGINEERING, MULTIDISCIPLINARY

Journal of Engineering Research

Pub Date : 2025-12-01 DOI: 10.1016/j.jer.2024.11.003

Hussein Samma , Ali Salem Bin Sama , Qusay Shihab Hamad

Deep learning models, particularly convolutional neural networks (CNNs), have excelled in pattern recognition tasks like face recognition, pedestrian detection, and medical diagnosis. CNNs, known for their end-to-end feature extraction and classification, are widely used in computer vision. Pre-trained CNN models such as AlexNet, VGG-19, ResNet, and Inception can serve as feature extractors for various problems, though they often produce a large number of features. To address this challenge and reduce the number of features, this research employs an efficient two-layer feature selection algorithm. The first layer, a global search optimizer, guides the swarm in the search space to identify the optimal feature subset. The second layer refines and fine-tunes the solution obtained from the global search. The main contributions of this work encompass several aspects: (i) the introduction of a lightweight classification model that utilizes a subset of features, (ii) the use of an efficient optimizer to perform simultaneous feature and instance selection, and (iii) the direct implications for the radiology field, where it could serve as a supportive tool for diagnosis. To evaluate the proposed approach, it was tested on skin lesion diagnosis using 1113 abnormal and 1099 normal lesions. The pre-trained VGG-19 model provided a 1-D feature vector of size 4096, with a linear SVM used for classification. Results showed the optimizer reduced the feature vector by up to 86 % while achieving over 70 % accuracy. The proposed optimizer outperformed the Reinforcement Learning-based Memetic Particle Swarm Optimization (RLMPSO).

深度学习模型，特别是卷积神经网络（cnn），在人脸识别、行人检测和医疗诊断等模式识别任务中表现出色。cnn以端到端特征提取和分类而闻名，在计算机视觉领域得到了广泛的应用。预训练的CNN模型，如AlexNet、VGG-19、ResNet和Inception，可以作为各种问题的特征提取器，尽管它们通常会产生大量的特征。为了解决这一问题并减少特征的数量，本研究采用了一种高效的两层特征选择算法。第一层是全局搜索优化器，引导群在搜索空间中识别最优的特征子集。第二层对全局搜索得到的解进行细化和微调。这项工作的主要贡献包括几个方面：(i)引入了一个利用特征子集的轻量级分类模型，（ii）使用一个高效的优化器来同时执行特征和实例选择，以及（iii）对放射学领域的直接影响，在那里它可以作为诊断的支持工具。为了评估所提出的方法，使用1113个异常和1099个正常的皮肤病变进行了皮肤病变诊断测试。预训练的VGG-19模型提供了一个大小为4096的一维特征向量，并使用线性支持向量机进行分类。结果表明，优化器减少了多达86 %的特征向量，同时实现了超过70 %的准确率。该优化算法优于基于强化学习的模因粒子群优化算法（RLMPSO）。

{"title":"Optimized deep learning model for medical image diagnosis","authors":"Hussein Samma , Ali Salem Bin Sama , Qusay Shihab Hamad","doi":"10.1016/j.jer.2024.11.003","DOIUrl":"10.1016/j.jer.2024.11.003","url":null,"abstract":"<div><div>Deep learning models, particularly convolutional neural networks (CNNs), have excelled in pattern recognition tasks like face recognition, pedestrian detection, and medical diagnosis. CNNs, known for their end-to-end feature extraction and classification, are widely used in computer vision. Pre-trained CNN models such as AlexNet, VGG-19, ResNet, and Inception can serve as feature extractors for various problems, though they often produce a large number of features. To address this challenge and reduce the number of features, this research employs an efficient two-layer feature selection algorithm. The first layer, a global search optimizer, guides the swarm in the search space to identify the optimal feature subset. The second layer refines and fine-tunes the solution obtained from the global search. The main contributions of this work encompass several aspects: (i) the introduction of a lightweight classification model that utilizes a subset of features, (ii) the use of an efficient optimizer to perform simultaneous feature and instance selection, and (iii) the direct implications for the radiology field, where it could serve as a supportive tool for diagnosis. To evaluate the proposed approach, it was tested on skin lesion diagnosis using 1113 abnormal and 1099 normal lesions. The pre-trained VGG-19 model provided a 1-D feature vector of size 4096, with a linear SVM used for classification. Results showed the optimizer reduced the feature vector by up to 86 % while achieving over 70 % accuracy. The proposed optimizer outperformed the Reinforcement Learning-based Memetic Particle Swarm Optimization (RLMPSO).</div></div>","PeriodicalId":48803,"journal":{"name":"Journal of Engineering Research","volume":"13 4","pages":"Pages 3147-3155"},"PeriodicalIF":2.2,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145739319","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

A trust-based global expert system for disease diagnosis using hierarchical federated learning 基于信任的疾病诊断全局专家系统

IF 2.2 4区工程技术 Q3 ENGINEERING, MULTIDISCIPLINARY

Journal of Engineering Research

Pub Date : 2025-12-01 DOI: 10.1016/j.jer.2025.03.001

Farah M. Almulla , Mohammed A. Almulla

Many healthcare institutions have leveraged expert systems to assist in disease diagnosis and treatment recommendations. However, regulations require protecting patient data, this hinders the ability of these institutions to collaborate on developing a robust and comprehensive expert system. To address this concern, we proposed integrating Federated Learning (FL) and the joint expert system development. Each institution trains its local model on its data, and only the updated model parameters are sent to a central server. The server aggregates these updates to enhance a global model, thus enabling the institutions to contribute to a collective knowledge base while maintaining compliance with privacy regulations. However, a significant challenge in the federated learning solution is the presence of free riders. These are clients who benefit from the shared global model without truly contributing to its development. They may withhold their expertise, opting instead to rely on the contributions of more engaged clients. To handle this challenge, we propose a novel trust-oriented, coalition-based client selection process, ensuring that only clients who actively contribute to the construction of the global expert system are allowed model updates. Our experimental results demonstrate the effectiveness of this approach. Our solution accelerates the global model's convergence and improves its accuracy, which makes federated learning more resilient and effective in sensitive domains such as healthcare, where privacy concerns and collaborative knowledge sharing must coexist.

许多医疗机构已经利用专家系统来协助疾病诊断和治疗建议。然而，法规要求保护患者数据，这阻碍了这些机构合作开发强大而全面的专家系统的能力。为了解决这一问题，我们提出将联邦学习（FL）与联合专家系统开发相结合。每个机构都根据自己的数据训练本地模型，只有更新后的模型参数才会被发送到中央服务器。服务器聚合这些更新以增强全球模型，从而使机构能够在遵守隐私法规的同时为集体知识库做出贡献。然而，联邦学习解决方案中的一个重大挑战是存在搭便车者。这些客户受益于共享的全球模式，但并未真正为其发展做出贡献。他们可能会保留自己的专业知识，转而选择依赖更积极的客户的贡献。为了应对这一挑战，我们提出了一种新的以信任为导向的、基于联盟的客户选择过程，确保只有积极参与构建全球专家系统的客户才能被允许更新模型。实验结果证明了该方法的有效性。我们的解决方案加速了全球模型的融合并提高了其准确性，这使得联邦学习在医疗保健等敏感领域更具弹性和有效性，在这些领域中，隐私问题和协作知识共享必须共存。

{"title":"A trust-based global expert system for disease diagnosis using hierarchical federated learning","authors":"Farah M. Almulla , Mohammed A. Almulla","doi":"10.1016/j.jer.2025.03.001","DOIUrl":"10.1016/j.jer.2025.03.001","url":null,"abstract":"<div><div>Many healthcare institutions have leveraged expert systems to assist in disease diagnosis and treatment recommendations. However, regulations require protecting patient data, this hinders the ability of these institutions to collaborate on developing a robust and comprehensive expert system. To address this concern, we proposed integrating Federated Learning (FL) and the joint expert system development. Each institution trains its local model on its data, and only the updated model parameters are sent to a central server. The server aggregates these updates to enhance a global model, thus enabling the institutions to contribute to a collective knowledge base while maintaining compliance with privacy regulations. However, a significant challenge in the federated learning solution is the presence of free riders. These are clients who benefit from the shared global model without truly contributing to its development. They may withhold their expertise, opting instead to rely on the contributions of more engaged clients. To handle this challenge, we propose a novel trust-oriented, coalition-based client selection process, ensuring that only clients who actively contribute to the construction of the global expert system are allowed model updates. Our experimental results demonstrate the effectiveness of this approach. Our solution accelerates the global model's convergence and improves its accuracy, which makes federated learning more resilient and effective in sensitive domains such as healthcare, where privacy concerns and collaborative knowledge sharing must coexist.</div></div>","PeriodicalId":48803,"journal":{"name":"Journal of Engineering Research","volume":"13 4","pages":"Pages 3191-3197"},"PeriodicalIF":2.2,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145739323","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Exploring urban growth drivers in heritage areas using machine learning: The case of Gharb Sohail, Aswan 利用机器学习探索遗产地区的城市增长动力：以阿斯旺的Gharb Sohail为例

IF 2.2 4区工程技术 Q3 ENGINEERING, MULTIDISCIPLINARY

Journal of Engineering Research

Pub Date : 2025-12-01 DOI: 10.1016/j.jer.2025.01.006

Hana Hasanoon , Mohamed E. Elattar , Muhammad Salem , Omar Hamdy

Heritage areas are increasingly threatened by rapid urban expansion, driven by population growth and insufficient planning measures. These areas are vital as they preserve cultural identity, social cohesion, and economic vitality, making their preservation essential. This study investigates the drivers of urban expansion in Gharb Sohail, Aswan, a culturally and historically significant heritage area. Employing Geographic Information Systems (GIS) and advanced machine learning models, including the Land Change Modeler (LCM) and Multi-Layer Perceptron (MLP) neural networks, the research identifies key factors shaping urban growth and simulates future expansion scenarios. The findings indicate that urban sprawl within the study area is projected to cover 753.65 feddan by 2062. Proximity to the Nile River, mosques, and tourism infrastructure emerge as the dominant factors influencing urbanization, with Cramer's V values of 0.66, 0.50, and 0.49, respectively. The study forecasts that by 2062, the urbanized area will expand from 20.15 % to 27.39 % of the total study area, resulting in considerable encroachment on non-urban lands. This significant growth poses a direct threat to the integrity of Gharb Sohail's heritage areas, highlighting the urgent need for comprehensive urban management strategies. In response, the research advocates for the implementation of targeted urban planning measures, including strict urban growth regulation, the promotion of architectural continuity, and the integration of sustainable development practices in adjacent urban areas. These strategic recommendations offer actionable insights for urban planners and policymakers, providing a robust framework for balancing heritage conservation with the socio-economic demands of urban growth.

在人口增长和规划措施不足的驱动下，城市快速扩张对遗产区构成了越来越大的威胁。这些地区至关重要，因为它们保存着文化认同、社会凝聚力和经济活力，因此保护它们至关重要。本研究调查了阿斯旺Gharb Sohail城市扩张的驱动因素，这是一个文化和历史上重要的遗产地区。该研究利用地理信息系统（GIS）和先进的机器学习模型，包括土地变化建模器（LCM）和多层感知器（MLP）神经网络，确定了影响城市增长的关键因素，并模拟了未来的扩张情景。研究结果表明，到2062年，研究区城市扩张面积预计将达到753.65平方英尺。靠近尼罗河、清真寺和旅游基础设施成为影响城市化的主导因素，其克莱默V值分别为0.66、0.50和0.49。研究预测，到2062年，城市化面积占研究总面积的比例将从20.15% %扩大到27.39% %，对非城市土地的侵占将相当大。这种显著的增长对Gharb Sohail遗产地区的完整性构成了直接威胁，突出了对综合城市管理战略的迫切需要。对此，本研究主张实施有针对性的城市规划措施，包括严格的城市增长调控，促进建筑的连续性，以及在邻近城市区域整合可持续发展实践。这些战略建议为城市规划者和政策制定者提供了可行的见解，为平衡遗产保护与城市发展的社会经济需求提供了强有力的框架。

{"title":"Exploring urban growth drivers in heritage areas using machine learning: The case of Gharb Sohail, Aswan","authors":"Hana Hasanoon , Mohamed E. Elattar , Muhammad Salem , Omar Hamdy","doi":"10.1016/j.jer.2025.01.006","DOIUrl":"10.1016/j.jer.2025.01.006","url":null,"abstract":"<div><div>Heritage areas are increasingly threatened by rapid urban expansion, driven by population growth and insufficient planning measures. These areas are vital as they preserve cultural identity, social cohesion, and economic vitality, making their preservation essential. This study investigates the drivers of urban expansion in Gharb Sohail, Aswan, a culturally and historically significant heritage area. Employing Geographic Information Systems (GIS) and advanced machine learning models, including the Land Change Modeler (LCM) and Multi-Layer Perceptron (MLP) neural networks, the research identifies key factors shaping urban growth and simulates future expansion scenarios. The findings indicate that urban sprawl within the study area is projected to cover 753.65 feddan by 2062. Proximity to the Nile River, mosques, and tourism infrastructure emerge as the dominant factors influencing urbanization, with Cramer's V values of 0.66, 0.50, and 0.49, respectively. The study forecasts that by 2062, the urbanized area will expand from 20.15 % to 27.39 % of the total study area, resulting in considerable encroachment on non-urban lands. This significant growth poses a direct threat to the integrity of Gharb Sohail's heritage areas, highlighting the urgent need for comprehensive urban management strategies. In response, the research advocates for the implementation of targeted urban planning measures, including strict urban growth regulation, the promotion of architectural continuity, and the integration of sustainable development practices in adjacent urban areas. These strategic recommendations offer actionable insights for urban planners and policymakers, providing a robust framework for balancing heritage conservation with the socio-economic demands of urban growth.</div></div>","PeriodicalId":48803,"journal":{"name":"Journal of Engineering Research","volume":"13 4","pages":"Pages 2851-2859"},"PeriodicalIF":2.2,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145739368","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Optimal resilience-based restoration plan for disrupted interdependent infrastructure networks under uncertainty 不确定性下相互依赖基础设施网络中断的最优弹性恢复方案

IF 2.2 4区工程技术 Q3 ENGINEERING, MULTIDISCIPLINARY

Journal of Engineering Research

Pub Date : 2025-12-01 DOI: 10.1016/j.jer.2024.10.005

Yasser Almoghathawi

The functioning of vital services in a society depends heavily on the proper and continuous operation of critical infrastructure networks, including water, transportation, power, natural gas, and telecommunication. Such networks are essential for maintaining the economy, security, and overall quality of life. However, since these critical networks are interdependent; hence, they are susceptible to various types of disruptions caused by natural disasters, failures, or malicious activities, leading to varying degrees of performance impacts that could directly affect people's daily lives. Thus, a disruption to one network can quickly spread to others, resulting in significant impacts on daily life. Therefore, restoring such disrupted networks in a timely manner is crucial to ensure a prompt recovery and minimize the impact on daily life. In this work, we consider the problem of restoring a set of interdependent infrastructure networks following a disruptive event. We study two essential factors that can accelerate the recovery process: (i) the restoration facilities location, which are considered as dispatching points for restoration crews, and (ii) the routes of the restoration crews within the interdependent networks. Accordingly, we formulate two optimization models using mixed integer programming. The first model, restoration facility location model, finds the optimal location for restoration facilities within interdependent infrastructure networks considering uncertainty in disruptions. The second model, restoration crew routing model, determine the optimal routes for restoration crews within the interdependent infrastructure networks considering: (i) travelled distance, (ii) restoration time required for disrupted network components, and (iii) different importance of network components. The two optimization models are solved through generated interdependent power and water networks to demonstrate their effectiveness.

一个社会中重要服务的运作在很大程度上取决于关键基础设施网络的正常和持续运行，包括水、交通、电力、天然气和电信。这样的网络对于维持经济、安全和整体生活质量至关重要。然而，由于这些关键的网络是相互依存的；因此，它们很容易受到自然灾害、故障或恶意活动造成的各种类型的中断，从而导致不同程度的性能影响，直接影响人们的日常生活。因此，一个网络的中断可以迅速蔓延到其他网络，对日常生活造成重大影响。因此，及时恢复这些中断的网络对于确保迅速恢复和尽量减少对日常生活的影响至关重要。在这项工作中，我们考虑了在破坏性事件发生后恢复一组相互依赖的基础设施网络的问题。我们研究了加速恢复过程的两个重要因素：(1)修复设施的位置，这被认为是修复人员的调度点；(2)修复人员在相互依赖的网络中的路线。据此，我们用混合整数规划建立了两个优化模型。第一个模型是修复设施选址模型，该模型考虑了中断的不确定性，在相互依赖的基础设施网络中寻找修复设施的最优选址。第二个模型，即修复人员路线模型，在相互依赖的基础设施网络中确定修复人员的最佳路线，考虑：(i)行进距离，（ii）中断的网络组件所需的恢复时间，以及（iii）网络组件的不同重要性。通过生成的相互依赖的水电网络来求解这两个优化模型，以验证其有效性。

{"title":"Optimal resilience-based restoration plan for disrupted interdependent infrastructure networks under uncertainty","authors":"Yasser Almoghathawi","doi":"10.1016/j.jer.2024.10.005","DOIUrl":"10.1016/j.jer.2024.10.005","url":null,"abstract":"<div><div>The functioning of vital services in a society depends heavily on the proper and continuous operation of critical infrastructure networks, including water, transportation, power, natural gas, and telecommunication. Such networks are essential for maintaining the economy, security, and overall quality of life. However, since these critical networks are interdependent; hence, they are susceptible to various types of disruptions caused by natural disasters, failures, or malicious activities, leading to varying degrees of performance impacts that could directly affect people's daily lives. Thus, a disruption to one network can quickly spread to others, resulting in significant impacts on daily life. Therefore, restoring such disrupted networks in a timely manner is crucial to ensure a prompt recovery and minimize the impact on daily life. In this work, we consider the problem of restoring a set of interdependent infrastructure networks following a disruptive event. We study two essential factors that can accelerate the recovery process: (i) the restoration facilities location, which are considered as dispatching points for restoration crews, and (ii) the routes of the restoration crews within the interdependent networks. Accordingly, we formulate two optimization models using mixed integer programming. The first model, restoration facility location model, finds the optimal location for restoration facilities within interdependent infrastructure networks considering uncertainty in disruptions. The second model, restoration crew routing model, determine the optimal routes for restoration crews within the interdependent infrastructure networks considering: (i) travelled distance, (ii) restoration time required for disrupted network components, and (iii) different importance of network components. The two optimization models are solved through generated interdependent power and water networks to demonstrate their effectiveness.</div></div>","PeriodicalId":48803,"journal":{"name":"Journal of Engineering Research","volume":"13 4","pages":"Pages 3441-3459"},"PeriodicalIF":2.2,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145739182","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Two-Stage Aggregation based Federated Learning (TSA-FL) for Industrial Internet of Things 基于两阶段聚合的工业物联网联邦学习（TSA-FL）

IF 2.2 4区工程技术 Q3 ENGINEERING, MULTIDISCIPLINARY

Journal of Engineering Research

Pub Date : 2025-12-01 DOI: 10.1016/j.jer.2025.03.003

Atallo Kassaw Takele, Balázs Villányi

Federated learning is a privacy-focused machine learning technique that promises to improve the Industrial Internet of Things (IIoT) systems. It can be used to enhance predictive maintenance, quality control, energy management, supply chain optimization, and anomaly detection. However, the implementation of federated learning in IIoT faces several challenges, including security while sharing parameters, communication overhead, data heterogeneity, and edge device resource limitation. This paper proposed a Two-Stage Aggregation Federated Learning (TSA-FL) technique, where parameter aggregation is carried out in two phases to detect malicious nodes and minimize delays. During a specific round, parameters intercepted by the attacker and poisoned by Byzantine nodes may arrive later than legitimate ones, or they might not be received at all. Hence, the local server sets two time slots namely time-slot-1 for faster arriving first group of edge devices and time-slot-2 for late arriving second group of edge devices. Any device arriving after the time-slot-2 won’t be accepted and considered malicious. The initial time slots are estimated manually, but they can be adjusted later based on the available bandwidth and resources of edge devices. In some cases, Byzantine nodes and attackers with better computational resource may still be able to submit their manipulated parameters within the designated time slots (either time-slot-1 or time-slot-2). This results in another vulnerability in the aggregation process. In this scenario, we propose an anomaly detection mechanism for distinguishing abnormal parameters. Experimental evaluation of the proposed approach shows significant performance improvement by maintaining processing time over the existing state-of-the-art.

联邦学习是一种以隐私为重点的机器学习技术，有望改善工业物联网（IIoT）系统。它可用于增强预测性维护、质量控制、能源管理、供应链优化和异常检测。然而，在工业物联网中实施联邦学习面临着一些挑战，包括共享参数时的安全性、通信开销、数据异构性和边缘设备资源限制。本文提出了一种两阶段聚合联邦学习（TSA-FL）技术，该技术分两个阶段进行参数聚合，以检测恶意节点并最小化延迟。在一个特定的回合中，被攻击者拦截并被拜占庭节点毒害的参数可能比合法的参数晚到，或者根本没有收到。因此，本地服务器设置两个时隙，即较早到达的第一组边缘设备的时隙-1和较晚到达的第二组边缘设备的时隙-2。任何在时隙2之后到达的设备将不被接受并被视为恶意设备。初始时隙是手动估计的，但可以根据边缘设备的可用带宽和资源进行调整。在某些情况下，拜占庭节点和拥有更好计算资源的攻击者可能仍然能够在指定的时隙（时隙1或时隙2）内提交他们操纵的参数。这导致了聚合过程中的另一个漏洞。在这种情况下，我们提出了一种异常检测机制来识别异常参数。实验评估表明，通过保持处理时间，该方法的性能显著提高。

{"title":"Two-Stage Aggregation based Federated Learning (TSA-FL) for Industrial Internet of Things","authors":"Atallo Kassaw Takele, Balázs Villányi","doi":"10.1016/j.jer.2025.03.003","DOIUrl":"10.1016/j.jer.2025.03.003","url":null,"abstract":"<div><div>Federated learning is a privacy-focused machine learning technique that promises to improve the Industrial Internet of Things (IIoT) systems. It can be used to enhance predictive maintenance, quality control, energy management, supply chain optimization, and anomaly detection. However, the implementation of federated learning in IIoT faces several challenges, including security while sharing parameters, communication overhead, data heterogeneity, and edge device resource limitation. This paper proposed a Two-Stage Aggregation Federated Learning (TSA-FL) technique, where parameter aggregation is carried out in two phases to detect malicious nodes and minimize delays. During a specific round, parameters intercepted by the attacker and poisoned by Byzantine nodes may arrive later than legitimate ones, or they might not be received at all. Hence, the local server sets two time slots namely time-slot-1 for faster arriving first group of edge devices and time-slot-2 for late arriving second group of edge devices. Any device arriving after the time-slot-2 won’t be accepted and considered malicious. The initial time slots are estimated manually, but they can be adjusted later based on the available bandwidth and resources of edge devices. In some cases, Byzantine nodes and attackers with better computational resource may still be able to submit their manipulated parameters within the designated time slots (either time-slot-1 or time-slot-2). This results in another vulnerability in the aggregation process. In this scenario, we propose an anomaly detection mechanism for distinguishing abnormal parameters. Experimental evaluation of the proposed approach shows significant performance improvement by maintaining processing time over the existing state-of-the-art.</div></div>","PeriodicalId":48803,"journal":{"name":"Journal of Engineering Research","volume":"13 4","pages":"Pages 3208-3215"},"PeriodicalIF":2.2,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145739232","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0

Versa net fusion with hybrid feature selection for intrusion detection in IoT 基于混合特征选择的Versa net融合物联网入侵检测

IF 2.2 4区工程技术 Q3 ENGINEERING, MULTIDISCIPLINARY

Journal of Engineering Research

Pub Date : 2025-12-01 DOI: 10.1016/j.jer.2025.03.004

Afnan M. Alhassan

Internet of Things (IoT) networks require intrusion detection to protect against illegal access, but existing techniques sometimes struggle to handle enormous amounts of data accurately. To tackle these issues, the suggested method implements a hybrid feature selection and Versa Net-based detection framework. The process starts with thorough data pre-processing, which involves removing inconsistent and superfluous entries and managing missing values by mean imputation. Then, Z-score normalization is used to standardize numerical features and guarantee dataset consistency. A multimodal approach is used in feature extraction: While flow-based, entropy-based, correlation-based, protocol-based, statistical, and flow-based features offer a comprehensive overview of the data's underlying structure, Independent Component Analysis (ICA) separates statistically independent components. A hybrid optimization method for feature selection finds and selects the most essential features by fusing the advantages of the Pelican and Chimp algorithms. Versa Net results from the detection framework's integration of neural network architectures, including Squeeze Net, Google Net, Alex Net, and Dense Net. These networks use dense blocks, inception modules, fire modules, and convolutional layers to extract a variety of characteristics. The outputs are flattened, concatenated, and sent to a fully connected output layer for anomaly detection after feature extraction. The goal of this comprehensive strategy is to improve intrusion detection in IoT environments by increasing its accuracy to 0.978005 and efficiency.

物联网（IoT）网络需要入侵检测来防止非法访问，但现有技术有时难以准确处理大量数据。为了解决这些问题，建议的方法实现了一个混合特征选择和基于Versa net的检测框架。该过程从彻底的数据预处理开始，包括删除不一致和多余的条目，并通过平均imputation管理缺失值。然后采用Z-score归一化对数值特征进行标准化，保证数据集的一致性。特征提取中使用了多模态方法：基于流的、基于熵的、基于关联的、基于协议的、统计的和基于流的特征提供了数据底层结构的全面概述，而独立组件分析（ICA）将统计独立的组件分离出来。结合Pelican算法和Chimp算法的优点，提出了一种混合优化的特征选择方法。Versa Net是检测框架集成神经网络架构的结果，包括Squeeze Net、谷歌Net、Alex Net和Dense Net。这些网络使用密集块、初始模块、火焰模块和卷积层来提取各种特征。在特征提取后，输出被平面化、串联并送到全连接的输出层进行异常检测。该综合策略的目标是通过将其准确性提高到0.978005和效率来改善物联网环境中的入侵检测。

{"title":"Versa net fusion with hybrid feature selection for intrusion detection in IoT","authors":"Afnan M. Alhassan","doi":"10.1016/j.jer.2025.03.004","DOIUrl":"10.1016/j.jer.2025.03.004","url":null,"abstract":"<div><div>Internet of Things (IoT) networks require intrusion detection to protect against illegal access, but existing techniques sometimes struggle to handle enormous amounts of data accurately. To tackle these issues, the suggested method implements a hybrid feature selection and Versa Net-based detection framework. The process starts with thorough data pre-processing, which involves removing inconsistent and superfluous entries and managing missing values by mean imputation. Then, Z-score normalization is used to standardize numerical features and guarantee dataset consistency. A multimodal approach is used in feature extraction: While flow-based, entropy-based, correlation-based, protocol-based, statistical, and flow-based features offer a comprehensive overview of the data's underlying structure, Independent Component Analysis (ICA) separates statistically independent components. A hybrid optimization method for feature selection finds and selects the most essential features by fusing the advantages of the Pelican and Chimp algorithms. Versa Net results from the detection framework's integration of neural network architectures, including Squeeze Net, Google Net, Alex Net, and Dense Net. These networks use dense blocks, inception modules, fire modules, and convolutional layers to extract a variety of characteristics. The outputs are flattened, concatenated, and sent to a fully connected output layer for anomaly detection after feature extraction. The goal of this comprehensive strategy is to improve intrusion detection in IoT environments by increasing its accuracy to 0.978005 and efficiency.</div></div>","PeriodicalId":48803,"journal":{"name":"Journal of Engineering Research","volume":"13 4","pages":"Pages 3216-3230"},"PeriodicalIF":2.2,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145739233","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":4,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 0