Indoor localization is a fundamental capability for autonomous mobile robots operating in complex indoor environments, where visual degradation, sensor noise, and rotational motion often lead to accumulated drift. This paper presents MFL-SLAM, a practical multi-sensor fusion localization system that extends the ORB-SLAM3 framework by explicitly integrating wheel odometry with visual–inertial SLAM. Unlike conventional visual–inertial approaches, MFL-SLAM employs an Extended Kalman Filter (EKF) to tightly fuse Inertial Measurement Unit (IMU) and wheel odometry, effectively compensating for vibration-induced inertial drift and wheel slippage during rotational motion. The EKF fusion output is then incorporated as a prior in a nonlinear optimization back-end together with RGB-D visual constraints, enabling accurate and globally consistent pose estimation. Extensive experiments demonstrate that MFL-SLAM achieves a 47.3 % reduction in relative pose error compared to ORB-SLAM3 and reduces the average localization error to 0.29 m, outperforming ORB-SLAM2 and LIO-SAM across small- and large-scale indoor environments. These results indicate that the proposed fusion strategy provides a robust and deployable solution for reliable indoor mobile robot localization.
{"title":"Indoor mobile robot localization system based on ORB-SLAM3 and multi-sensor fusion","authors":"Siyong Fu, Qinghua Zhao, Qiuxiang Tao, Hesheng Liu, Qing Wang, Danjuan Liu","doi":"10.1016/j.aej.2026.01.029","DOIUrl":"10.1016/j.aej.2026.01.029","url":null,"abstract":"<div><div>Indoor localization is a fundamental capability for autonomous mobile robots operating in complex indoor environments, where visual degradation, sensor noise, and rotational motion often lead to accumulated drift. This paper presents MFL-SLAM, a practical multi-sensor fusion localization system that extends the ORB-SLAM3 framework by explicitly integrating wheel odometry with visual–inertial SLAM. Unlike conventional visual–inertial approaches, MFL-SLAM employs an Extended Kalman Filter (EKF) to tightly fuse Inertial Measurement Unit (IMU) and wheel odometry, effectively compensating for vibration-induced inertial drift and wheel slippage during rotational motion. The EKF fusion output is then incorporated as a prior in a nonlinear optimization back-end together with RGB-D visual constraints, enabling accurate and globally consistent pose estimation. Extensive experiments demonstrate that MFL-SLAM achieves a 47.3 % reduction in relative pose error compared to ORB-SLAM3 and reduces the average localization error to 0.29 m, outperforming ORB-SLAM2 and LIO-SAM across small- and large-scale indoor environments. These results indicate that the proposed fusion strategy provides a robust and deployable solution for reliable indoor mobile robot localization.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"137 ","pages":"Pages 194-205"},"PeriodicalIF":6.8,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146036659","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-21DOI: 10.1016/j.aej.2026.01.020
S.A. Alblowi , Hala Alzumi , A.N. Al Qarni , M. El Sayed , M.A. El Safty
The concepts of nano beta are introduced in this article along with their relationships. We also analyze some of their properties. Based on this concept, we introduce the topological approach to decision-making and its application in the medical field, which uses topological tools like nano beta. These mathematical tools are used in medicine to improve diagnostic accuracy and aid in the development of treatment plans. In a study on the diagnosis of chronic kidney disease, the authors show that this strategy is beneficial. Based on the results, the topological application of nano beta offers a trustworthy and accurate way to make medical decisions. This paper employs the concept of attribute and basis elimination in nano beta topology to identify the principal factors contributing to chronic kidney disease. Diabetes and high blood pressure were found to be the main things that put people at risk for CKD. This risk can be prevented by taking healthy food and proper medical care. To help physicians determine whether a patient has chronic kidney disease, we developed an algorithm. Healthcare providers can use this approach to accurately diagnose illnesses, develop effective treatment strategies, and assist patients in recovering. These can also act as a starting point for the creation of sophisticated nano systems. Finally, we describe a medical method that helps people with chronic kidney disease to determine the underlying cause of their illness.
{"title":"Topological approach with decision making based on nano beta and its application","authors":"S.A. Alblowi , Hala Alzumi , A.N. Al Qarni , M. El Sayed , M.A. El Safty","doi":"10.1016/j.aej.2026.01.020","DOIUrl":"10.1016/j.aej.2026.01.020","url":null,"abstract":"<div><div>The concepts of nano beta are introduced in this article along with their relationships. We also analyze some of their properties. Based on this concept, we introduce the topological approach to decision-making and its application in the medical field, which uses topological tools like nano beta. These mathematical tools are used in medicine to improve diagnostic accuracy and aid in the development of treatment plans. In a study on the diagnosis of chronic kidney disease, the authors show that this strategy is beneficial. Based on the results, the topological application of nano beta offers a trustworthy and accurate way to make medical decisions. This paper employs the concept of attribute and basis elimination in nano beta topology to identify the principal factors contributing to chronic kidney disease. Diabetes and high blood pressure were found to be the main things that put people at risk for CKD. This risk can be prevented by taking healthy food and proper medical care. To help physicians determine whether a patient has chronic kidney disease, we developed an algorithm. Healthcare providers can use this approach to accurately diagnose illnesses, develop effective treatment strategies, and assist patients in recovering. These can also act as a starting point for the creation of sophisticated nano systems. Finally, we describe a medical method that helps people with chronic kidney disease to determine the underlying cause of their illness.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"137 ","pages":"Pages 218-224"},"PeriodicalIF":6.8,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146036719","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-20DOI: 10.1016/j.aej.2026.01.016
Sita Rani , Ramesh Karnati , Vivek Patel , M.K. Ranganathaswamy , Prakhar Tomar , Aman Kataria , Amrindra Pal
The integration of Artificial Intelligence (AI) driven optimization techniques is transforming smart manufacturing in the industry 5.0 landscape leading to sustainable industrial processes. This review comprehensively explores AI-driven optimization methods that enhance efficiency, resilience, and sustainability in modern manufacturing ecosystems. It highlights the role of various AI - based algorithms in optimizing production processes, energy consumption, and supply chains. Along with this, it also presents the significance of AI-driven manufacturing in improving secure production by facilitating real-time monitoring, anomaly detection, and predictive maintenance. In this work, the authors also examine how AI contributes to human-centric manufacturing, addressing challenges such as resource utilization, waste reduction, and adaptive decision-making. Key advancements, limitations, and future research directions are analyzed to provide a holistic view of AI’s transformative potential. The findings underscore the necessity of AI-driven optimization for achieving sustainable, efficient, and flexible manufacturing processes in Industry 5.0. This work serves as a significant reference for researchers, industry professionals, and policymakers seeking to leverage AI for sustainable industrial advancements. This paper presents the comprehensive synthesis of AI-driven optimization techniques represented for the emerging Industry 5.0 model, prioritizing smart sustainable manufacturing. Unlike prior reviews, it systematically compares traditional and AI-based approaches, highlights the transformative synergy of advanced technologies like AI, IoT, digital twins, and blockchain for real-time, human-centric manufacturing, and details hybrid optimization methods integrating AI algorithms. This review uniquely maps the integration of these innovations with sustainability, adaptability, and mass personalization, presenting a roadmap to help industries employ intelligent, data-driven, and eco-friendly optimization solutions for future-ready manufacturing.
{"title":"AI-driven optimization techniques for smart sustainable manufacturing in Industry 5.0 ecosystem: A comprehensive review","authors":"Sita Rani , Ramesh Karnati , Vivek Patel , M.K. Ranganathaswamy , Prakhar Tomar , Aman Kataria , Amrindra Pal","doi":"10.1016/j.aej.2026.01.016","DOIUrl":"10.1016/j.aej.2026.01.016","url":null,"abstract":"<div><div>The integration of Artificial Intelligence (AI) driven optimization techniques is transforming smart manufacturing in the industry 5.0 landscape leading to sustainable industrial processes. This review comprehensively explores AI-driven optimization methods that enhance efficiency, resilience, and sustainability in modern manufacturing ecosystems. It highlights the role of various AI - based algorithms in optimizing production processes, energy consumption, and supply chains. Along with this, it also presents the significance of AI-driven manufacturing in improving secure production by facilitating real-time monitoring, anomaly detection, and predictive maintenance. In this work, the authors also examine how AI contributes to human-centric manufacturing, addressing challenges such as resource utilization, waste reduction, and adaptive decision-making. Key advancements, limitations, and future research directions are analyzed to provide a holistic view of AI’s transformative potential. The findings underscore the necessity of AI-driven optimization for achieving sustainable, efficient, and flexible manufacturing processes in Industry 5.0. This work serves as a significant reference for researchers, industry professionals, and policymakers seeking to leverage AI for sustainable industrial advancements. This paper presents the comprehensive synthesis of AI-driven optimization techniques represented for the emerging Industry 5.0 model, prioritizing smart sustainable manufacturing. Unlike prior reviews, it systematically compares traditional and AI-based approaches, highlights the transformative synergy of advanced technologies like AI, IoT, digital twins, and blockchain for real-time, human-centric manufacturing, and details hybrid optimization methods integrating AI algorithms. This review uniquely maps the integration of these innovations with sustainability, adaptability, and mass personalization, presenting a roadmap to help industries employ intelligent, data-driven, and eco-friendly optimization solutions for future-ready manufacturing.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"137 ","pages":"Pages 133-158"},"PeriodicalIF":6.8,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146036730","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-20DOI: 10.1016/j.aej.2026.01.019
Jiaxing Lu , Lingrong Kong , Yu Wang , Jiong Li
In the field of drilling engineering, the rubber screw motor adopts the cycloidal principle for drilling, and the rubber bushing will carbonize and fail when the drilling at 180°C and above. As a positive displacement motor, the cycloidal motor has high temperature and corrosion resistance, low speed, and high torque. It is used in traditional mechanical engineering fields, such as the hydraulic cycloidal motor of the hoist, while the hydraulic cycloidal motor applied in the drilling fluid field has been blank. Therefore, the cycloidal motor in drilling fluid medium is proposed in this paper, which is different from the traditional mechanical cycloidal motor in oil medium and applied in the field of high-temperature drilling engineering. Based on the clearance flow theory and energy conservation law, the mathematical model of fluid leakage of the cycloidal motor was calculated. Through Matlab programming, the cloud diagram of fluid pressure drop, clearance, and viscosity on the leakage of the cycloidal motor was calculated numerically. The characteristic curve of motor numerical calculation was compared and verified by experimental data of the OMT 160 cycloidal motor produced by Danfoss Company. A theoretical basis for selecting clearance of machining and manufacturing all-metal cycloidal motor were provided.
{"title":"Analysis of cycloid motor output characteristics based on fluid simulation","authors":"Jiaxing Lu , Lingrong Kong , Yu Wang , Jiong Li","doi":"10.1016/j.aej.2026.01.019","DOIUrl":"10.1016/j.aej.2026.01.019","url":null,"abstract":"<div><div>In the field of drilling engineering, the rubber screw motor adopts the cycloidal principle for drilling, and the rubber bushing will carbonize and fail when the drilling at 180°C and above. As a positive displacement motor, the cycloidal motor has high temperature and corrosion resistance, low speed, and high torque. It is used in traditional mechanical engineering fields, such as the hydraulic cycloidal motor of the hoist, while the hydraulic cycloidal motor applied in the drilling fluid field has been blank. Therefore, the cycloidal motor in drilling fluid medium is proposed in this paper, which is different from the traditional mechanical cycloidal motor in oil medium and applied in the field of high-temperature drilling engineering. Based on the clearance flow theory and energy conservation law, the mathematical model of fluid leakage of the cycloidal motor was calculated. Through Matlab programming, the cloud diagram of fluid pressure drop, clearance, and viscosity on the leakage of the cycloidal motor was calculated numerically. The characteristic curve of motor numerical calculation was compared and verified by experimental data of the OMT 160 cycloidal motor produced by Danfoss Company. A theoretical basis for selecting clearance of machining and manufacturing all-metal cycloidal motor were provided.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"137 ","pages":"Pages 159-170"},"PeriodicalIF":6.8,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146036657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-20DOI: 10.1016/j.aej.2026.01.024
Dinh-Cuong Hoang , Phan Xuan Tan , Anh-Nhat Nguyen , Ta Huu Anh Duong , Tuan-Minh Huynh , Minh-Anh Nguyen , Duc-Manh Nguyen , Minh-Duc Cao , Duc-Huy Ngo , Minh-Quang Vu , Van-Duc Vu , Van-Thiep Nguyen , Thu-Uyen Nguyen , Khanh-Toan Phan , Van-Hiep Duong
Existing three-dimensional (3D) anomaly detection approaches typically rely on reconstruction, external memory banks, or fixed-radius clustering and often fail to generalize to noisy, irregularly sampled industrial scans or to capture the full diversity of real defects. We present Vote3D-AD, a single-pass framework that trains only on defect-free data and addresses these gaps with two principal contributions. First, we introduce Varied Defect Synthesis (VDS), a saliency-guided pseudo-anomaly generator that produces diverse, physically plausible defects (bulges, dents, holes, cracks, and surface roughness) together with sensor-level degradations to narrow the synthetic-to-real gap. Second, we develop a vote-and-cluster architecture in which local geometric representations predict learned, scale-aware votes that encode both spatial and boundary cues, and a differentiable soft-assignment clustering module aggregates these votes into coherent anomaly regions without relying on fixed-radius grouping or external memory structures. We evaluated our method on the synthetic Anomaly-ShapeNet benchmark and a new industrial dataset using three metrics: point-level Area Under the Receiver Operating Characteristic curve (AUROC), object-level Area Under the Precision-Recall curve (AUPR), and F1-Score. On average across both benchmarks, our method improves point-level AUROC by 6.7%, AUPR by 10.1% and F1 by 11.2%, and improves object-level AUROC by 5.3%, AUPR by 3.8% and F1 by 5.4% over the strongest baseline, while maintaining inference speeds above 9 frames per second (FPS).
{"title":"Vote3D-AD: Unsupervised point cloud anomaly localization via varied defect synthesis and differentiable vote-clustering","authors":"Dinh-Cuong Hoang , Phan Xuan Tan , Anh-Nhat Nguyen , Ta Huu Anh Duong , Tuan-Minh Huynh , Minh-Anh Nguyen , Duc-Manh Nguyen , Minh-Duc Cao , Duc-Huy Ngo , Minh-Quang Vu , Van-Duc Vu , Van-Thiep Nguyen , Thu-Uyen Nguyen , Khanh-Toan Phan , Van-Hiep Duong","doi":"10.1016/j.aej.2026.01.024","DOIUrl":"10.1016/j.aej.2026.01.024","url":null,"abstract":"<div><div>Existing three-dimensional (3D) anomaly detection approaches typically rely on reconstruction, external memory banks, or fixed-radius clustering and often fail to generalize to noisy, irregularly sampled industrial scans or to capture the full diversity of real defects. We present Vote3D-AD, a single-pass framework that trains only on defect-free data and addresses these gaps with two principal contributions. First, we introduce Varied Defect Synthesis (VDS), a saliency-guided pseudo-anomaly generator that produces diverse, physically plausible defects (bulges, dents, holes, cracks, and surface roughness) together with sensor-level degradations to narrow the synthetic-to-real gap. Second, we develop a vote-and-cluster architecture in which local geometric representations predict learned, scale-aware votes that encode both spatial and boundary cues, and a differentiable soft-assignment clustering module aggregates these votes into coherent anomaly regions without relying on fixed-radius grouping or external memory structures. We evaluated our method on the synthetic Anomaly-ShapeNet benchmark and a new industrial dataset using three metrics: point-level Area Under the Receiver Operating Characteristic curve (AUROC), object-level Area Under the Precision-Recall curve (AUPR), and F1-Score. On average across both benchmarks, our method improves point-level AUROC by 6.7%, AUPR by 10.1% and F1 by 11.2%, and improves object-level AUROC by 5.3%, AUPR by 3.8% and F1 by 5.4% over the strongest baseline, while maintaining inference speeds above 9 frames per second (FPS).</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"137 ","pages":"Pages 171-193"},"PeriodicalIF":6.8,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146036658","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-20DOI: 10.1016/j.aej.2026.01.027
Jie He
Semantic segmentation requires both global context and fine-grained details, yet CNNs struggle with long-range dependencies and Transformers can under-represent low-level structure and be computationally heavy. We propose DGGAT, a dual-branch gated graph attention Transformer: a global branch models long-range context, while a graph-embedded local branch groups semantically related pixels into nodes and applies gated graph attention to sharpen boundaries and small objects. An Attention-Based Feature Selection Fusion Module (ASFM) fuses global and local features to suppress redundancy and balance detail with context. On Cityscapes, DGGAT reaches 84.7% mIoU with a ResNet-101 backbone (82.8% with ResNet-50), and on ADE20K it attains 48.2% mIoU, validating both accuracy and efficiency. These results demonstrate that DGGAT effectively integrates global semantics with fine-detail representation.
{"title":"DGGAT: Dual-branch gated graph attention transformer for high-accuracy semantic segmentation","authors":"Jie He","doi":"10.1016/j.aej.2026.01.027","DOIUrl":"10.1016/j.aej.2026.01.027","url":null,"abstract":"<div><div>Semantic segmentation requires both global context and fine-grained details, yet CNNs struggle with long-range dependencies and Transformers can under-represent low-level structure and be computationally heavy. We propose DGGAT, a dual-branch gated graph attention Transformer: a global branch models long-range context, while a graph-embedded local branch groups semantically related pixels into nodes and applies gated graph attention to sharpen boundaries and small objects. An Attention-Based Feature Selection Fusion Module (ASFM) fuses global and local features to suppress redundancy and balance detail with context. On Cityscapes, DGGAT reaches 84.7% mIoU with a ResNet-101 backbone (82.8% with ResNet-50), and on ADE20K it attains 48.2% mIoU, validating both accuracy and efficiency. These results demonstrate that DGGAT effectively integrates global semantics with fine-detail representation.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"137 ","pages":"Pages 123-132"},"PeriodicalIF":6.8,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146036708","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
As large language models (LLMs) become increasingly central to Arabic NLP applications, their effectiveness in linguistically diverse settings, particularly regions with rich dialectal variation such as Saudi Arabia, remains underexplored. Existing evaluation paradigms tend to prioritize high-resource languages or Modern Standard Arabic (MSA), overlooking regional linguistic and cultural specificities. This leads to performance limitations and cultural biases in real-world deployments. To address this gap, we introduce Absher, the first comprehensive and fine-grained benchmark designed to assess the understanding of LLMs regarding Saudi dialects and their embedded cultural nuances. Absher consists of over 18,000 multiple choice questions derived from a curated dataset of dialectal words, phrases, and proverbs sourced from five major Saudi regions. The benchmark spans six task categories: Meaning, True/False, Fill-in-the-Blank, Contextual Usage, Cultural Interpretation, and Location Recognition, enabling multifaceted evaluation across both linguistic and cultural dimensions. We perform zero-shot evaluations on six state-of-the-art open LLMs: ALLaM, LLaMA, Jais, Mistral, Qwen, and AceGPT. Our results reveal substantial performance variability across dialects and question types. Qwen achieved the highest overall accuracy, excelling in word-level questions (63%), while ALLaM outperformed others in the interpretation of proverbs (48% accuracy). All models struggled with content from underrepresented dialects, particularly Southern and Eastern variants, and with context-free True/False questions, highlighting weaknesses in dialect grounding and binary reasoning. These findings demonstrate the need for dialect-aware training and culturally aligned evaluation. We position Absher as a critical step toward more equitable and effective LLMs development for real-world Arabic applications.
{"title":"From words to proverbs: Evaluating LLMs’ linguistic and cultural competence in Saudi dialects with Absher","authors":"Renad Al-Monef , Hassan Alhuzali , Nora Alturayeif , Ashwag Alasmari","doi":"10.1016/j.aej.2025.12.066","DOIUrl":"10.1016/j.aej.2025.12.066","url":null,"abstract":"<div><div>As large language models (LLMs) become increasingly central to Arabic NLP applications, their effectiveness in linguistically diverse settings, particularly regions with rich dialectal variation such as Saudi Arabia, remains underexplored. Existing evaluation paradigms tend to prioritize high-resource languages or Modern Standard Arabic (MSA), overlooking regional linguistic and cultural specificities. This leads to performance limitations and cultural biases in real-world deployments. To address this gap, we introduce <span>Absher</span>, the first comprehensive and fine-grained benchmark designed to assess the understanding of LLMs regarding Saudi dialects and their embedded cultural nuances. <span>Absher</span> consists of over 18,000 multiple choice questions derived from a curated dataset of dialectal words, phrases, and proverbs sourced from five major Saudi regions. The benchmark spans six task categories: Meaning, True/False, Fill-in-the-Blank, Contextual Usage, Cultural Interpretation, and Location Recognition, enabling multifaceted evaluation across both linguistic and cultural dimensions. We perform zero-shot evaluations on six state-of-the-art open LLMs: ALLaM, LLaMA, Jais, Mistral, Qwen, and AceGPT. Our results reveal substantial performance variability across dialects and question types. Qwen achieved the highest overall accuracy, excelling in word-level questions (63%), while ALLaM outperformed others in the interpretation of proverbs (48% accuracy). All models struggled with content from underrepresented dialects, particularly Southern and Eastern variants, and with context-free True/False questions, highlighting weaknesses in dialect grounding and binary reasoning. These findings demonstrate the need for dialect-aware training and culturally aligned evaluation. We position <span>Absher</span> as a critical step toward more equitable and effective LLMs development for real-world Arabic applications.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"137 ","pages":"Pages 25-41"},"PeriodicalIF":6.8,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145993580","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-19DOI: 10.1016/j.aej.2026.01.021
Abdullah Al Mahfazur Rahman , Mohamad A. Alawad , Md. Moniruzzaman , Yazeed Alkhrijah , Badariah Bais , Abdulmajeed M. Alenezi , Mohammad Tariqul Islam
This paper presents a rotationally symmetric metamaterial (MTM) designed for electromagnetic interference (EMI) shielding and blood dielectric sensing applications. The geometry of the MTM unit cell ( ) is optimized through CST simulation. The array of unit cells ensures the S21 resonance at 5.961 GHz, with a broader bandwidth of 4.28 GHz (71.80 %) spanning from 3.75 to 8.03 GHz for the optimized dimensions of various segments of the rotationally symmetric unit cell. Utilizing field distribution, surface current, and effective parameter responses, the resonance phenomena are analyzed. The array structure of the MTM achieves a peak shielding effectiveness of 39.78 dB within the C-band while maintaining angular stability. Additionally, it performs nonlinear sensing responses, establishing a high-frequency deviation ranging from 4.037 to 4.230 GHz and demonstrating a high sensitivity of 4.44 %, which enables it to detect variations in blood dielectric properties. For sensing analysis, samples are replicated in a laboratory to accurately imitate blood dielectric properties. The performance of the designed MTM is validated by prototype measurements, which align well with the simulations. The findings confirm the design's effectiveness for EMI shielding in microwave communication and its potential for blood dielectric sensing in biomedical applications.
{"title":"Rotationally symmetric resonator-based metamaterial for wideband EMI shielding and blood dielectric property sensing applications","authors":"Abdullah Al Mahfazur Rahman , Mohamad A. Alawad , Md. Moniruzzaman , Yazeed Alkhrijah , Badariah Bais , Abdulmajeed M. Alenezi , Mohammad Tariqul Islam","doi":"10.1016/j.aej.2026.01.021","DOIUrl":"10.1016/j.aej.2026.01.021","url":null,"abstract":"<div><div>This paper presents a rotationally symmetric metamaterial (MTM) designed for electromagnetic interference (EMI) shielding and blood dielectric sensing applications. The geometry of the MTM unit cell (<span><math><mrow><mn>9.6</mn><mi>mm</mi><mo>×</mo></mrow></math></span> <span><math><mrow><mn>9.6</mn><mi>mm</mi><mo>×</mo><mn>1.6</mn><mi>mm</mi></mrow></math></span>) is optimized through CST simulation. The array of unit cells ensures the S<sub>21</sub> resonance at 5.961 GHz, with a broader bandwidth of 4.28 GHz (71.80 %) spanning from 3.75 to 8.03 GHz for the optimized dimensions of various segments of the rotationally symmetric unit cell. Utilizing field distribution, surface current, and effective parameter responses, the resonance phenomena are analyzed. The array structure of the MTM achieves a peak shielding effectiveness of 39.78 dB within the C-band while maintaining angular stability. Additionally, it performs nonlinear sensing responses, establishing a high-frequency deviation ranging from 4.037 to 4.230 GHz and demonstrating a high sensitivity of 4.44 %, which enables it to detect variations in blood dielectric properties. For sensing analysis, samples are replicated in a laboratory to accurately imitate blood dielectric properties. The performance of the designed MTM is validated by prototype measurements, which align well with the simulations. The findings confirm the design's effectiveness for EMI shielding in microwave communication and its potential for blood dielectric sensing in biomedical applications.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"137 ","pages":"Pages 101-122"},"PeriodicalIF":6.8,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146036660","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-19DOI: 10.1016/j.aej.2026.01.006
Mingfei Zeng, Yuting Lian
Power grid fault alarms play a crucial role in minimizing system damage and service disruptions. However, existing deep learning approaches frequently neglect the topological structure and physical characteristics of power system data, leading to suboptimal fault identification performance. To address this limitation, this paper proposes PhysLSTM-Attn, a novel physics-informed deep learning method for power network anomaly alarm denoising. The model incorporates Kirchhoff’s Laws directly within the feature embedding layer, ensuring that learned representations adhere to fundamental circuit conservation principles. A topology-aware bidirectional LSTM encoder captures both temporal dependencies and spatial relationships by integrating graph-Laplacian-based positional encodings into its gating mechanism. In addition, an electrical-distance-enhanced multi-head attention mechanism computes attention weights based on electrical coupling strength rather than semantic similarity, providing a more accurate reflection of device interactions. A multi-hop graph convolutional network further models cascading fault propagation across multiple electrical distances, while a confidence calibration module supplies reliability estimates to support decision-making. Comprehensive experiments on the East China Power Grid Alarm Dataset and the IEEE 118-Node Extended Dataset demonstrate accuracy improvements of 8.32 % and 7.43 % over the LSTM-Attention baseline, respectively.
{"title":"A power network anomaly alarm denoising method based on a hybrid LSTM-attention model","authors":"Mingfei Zeng, Yuting Lian","doi":"10.1016/j.aej.2026.01.006","DOIUrl":"10.1016/j.aej.2026.01.006","url":null,"abstract":"<div><div>Power grid fault alarms play a crucial role in minimizing system damage and service disruptions. However, existing deep learning approaches frequently neglect the topological structure and physical characteristics of power system data, leading to suboptimal fault identification performance. To address this limitation, this paper proposes PhysLSTM-Attn, a novel physics-informed deep learning method for power network anomaly alarm denoising. The model incorporates Kirchhoff’s Laws directly within the feature embedding layer, ensuring that learned representations adhere to fundamental circuit conservation principles. A topology-aware bidirectional LSTM encoder captures both temporal dependencies and spatial relationships by integrating graph-Laplacian-based positional encodings into its gating mechanism. In addition, an electrical-distance-enhanced multi-head attention mechanism computes attention weights based on electrical coupling strength rather than semantic similarity, providing a more accurate reflection of device interactions. A multi-hop graph convolutional network further models cascading fault propagation across multiple electrical distances, while a confidence calibration module supplies reliability estimates to support decision-making. Comprehensive experiments on the East China Power Grid Alarm Dataset and the IEEE 118-Node Extended Dataset demonstrate accuracy improvements of 8.32 % and 7.43 % over the LSTM-Attention baseline, respectively.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"137 ","pages":"Pages 76-88"},"PeriodicalIF":6.8,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146036709","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2026-01-19DOI: 10.1016/j.aej.2026.01.025
Yong Lu , Chenxu Wang , Xuze Gu , Xiuqin Pan , Yijin Gang
In order to effectively prevent the global spread of malaria, classical deep learning models have been applied to malaria detection. However, these models generally suffer from low accuracy. In order to address the identified limitations, an Efficient Target-Oriented YOLO model (ET-YOLO) is proposed in this thesis. To address the limited discriminability of C3k2 in malaria microscopy images, we redesigned it into C3k2fECA, which integrates efficient channel attention and a refined fusion pathway to emphasize parasite-related regions. We further developed C3k2fTR, leveraging Transformer-based global context modeling to remedy the loss of contextual cues and improve robustness under complex backgrounds. In addition, a lightweight ConvNeXt variant, CNeB (ConvNeXt Block), was incorporated to effectively reduce model parameters while maintaining strong representational capacity. The experimental results of the improved model on two different datasets demonstrate the effectiveness of the improved model, specifically achieving [email protected] of 86.2% and 77.9% on two different datasets, both of which outperform other traditional YOLO models, while the number of parameters is reduced by about 7.2% compared to the reference model. A balance has been achieved between detection accuracy and computational resource utilization, providing a practical technical solution for malaria control in resource-constrained regions.
{"title":"ET-YOLO:A study on a malaria pathogen detection model based on YOLO11","authors":"Yong Lu , Chenxu Wang , Xuze Gu , Xiuqin Pan , Yijin Gang","doi":"10.1016/j.aej.2026.01.025","DOIUrl":"10.1016/j.aej.2026.01.025","url":null,"abstract":"<div><div>In order to effectively prevent the global spread of malaria, classical deep learning models have been applied to malaria detection. However, these models generally suffer from low accuracy. In order to address the identified limitations, an Efficient Target-Oriented YOLO model (ET-YOLO) is proposed in this thesis. To address the limited discriminability of C3k2 in malaria microscopy images, we redesigned it into C3k2fECA, which integrates efficient channel attention and a refined fusion pathway to emphasize parasite-related regions. We further developed C3k2fTR, leveraging Transformer-based global context modeling to remedy the loss of contextual cues and improve robustness under complex backgrounds. In addition, a lightweight ConvNeXt variant, CNeB (ConvNeXt Block), was incorporated to effectively reduce model parameters while maintaining strong representational capacity. The experimental results of the improved model on two different datasets demonstrate the effectiveness of the improved model, specifically achieving [email protected] of 86.2% and 77.9% on two different datasets, both of which outperform other traditional YOLO models, while the number of parameters is reduced by about 7.2% compared to the reference model. A balance has been achieved between detection accuracy and computational resource utilization, providing a practical technical solution for malaria control in resource-constrained regions.</div></div>","PeriodicalId":7484,"journal":{"name":"alexandria engineering journal","volume":"137 ","pages":"Pages 14-24"},"PeriodicalIF":6.8,"publicationDate":"2026-01-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145993579","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"工程技术","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号