{"title":"Adaptive virtual inertia emulation based on policy gradient clipping for low-inertia microgrids with phase-locked loop dynamics","authors":"Ming Chang , Mohamed Salem , Faisal A. Mohamed","doi":"10.1016/j.compeleceng.2025.110477","DOIUrl":"10.1016/j.compeleceng.2025.110477","url":null,"abstract":"<div><div>The high-penetration of sustainable energy resources in the hybrid microgrids necessitates deploying the power electronic interface systems (e.g., rectifiers, inverters, and converters) for conversion purposes. However, the utilization of such technologies reduces the inertia of microgrids which highly threaten their stability. The stability challenges of microgrids are heightened when the phase-locked loop devices are installed in the converter-based systems. In this work, a fractional order disturbance-observer-based control (FO-DOBC) is developed for advanced virtual inertia control (AVIC) of microgrids with sustainable resources, electric vehicles, and storage units. In particular, the effect of phase-locked loop’s dynamics on the stability of microgrid is investigated. To dynamically respond to the disturbances in the microgrid and phase-locked loop’s dynamics, the coefficients embedded in the FO-DOBC are adaptively adjusted by the stochastic policy gradient clipping. By training the neural networks of SPGC, the FO-DOBC controller is designed in such a way that maximizes a reward function defined based on the system requirements. The comprehensive examinations based on the Arduino testbed are carried out to appraise the feasibility of the suggested virtual-based controller in a real-time framework. The real-time outcomes of the microgrid reveal that the AVIC based on FO-DOBC controller (designed by the stochastic policy gradient clipping) provides better responses than conventional virtual inertia control. Moreover, the suggested AVIC controller provides a higher level of stability against the reduction of inertia (between 1 % to 10 %) from its nominal value.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"125 ","pages":"Article 110477"},"PeriodicalIF":4.0,"publicationDate":"2025-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144178423","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}
Sana Farooq , Ayesha Altaf , Muhammad Shoaib , Faiza Iqbal , Nagwan Abdel Samee , Manal Abdullah Alohali , Imran Ashraf
{"title":"Analyzing post-quantum cryptographic algorithms efficiency for transport security layer","authors":"Sana Farooq , Ayesha Altaf , Muhammad Shoaib , Faiza Iqbal , Nagwan Abdel Samee , Manal Abdullah Alohali , Imran Ashraf","doi":"10.1016/j.compeleceng.2025.110437","DOIUrl":"10.1016/j.compeleceng.2025.110437","url":null,"abstract":"<div><div>The development of post-quantum computing has increasingly put at risk the viability of classical public-key cryptographic schemes. The National Institute of Standards and Technology (NIST) is actively working on standardizing post-quantum cryptography (PQC) algorithms, and this research assesses the performance of two fourth-round finalists: BIKE and Classic McEliece, within the Transport Layer Security (TLS) context. With OQS’s OpenSSL implementation, we benchmark key generation, encapsulation, decapsulation, and TLS handshake latency across multiple algorithm variants and operating systems. Our results illustrate that BIKE-L1 demonstrates the lowest TLS handshake latency of 6 ms and performs well on Linux, positioning it as optimal for secure communication when minimal delay is needed. In contrast, the Classic McEliece variants provide robust encryption security but have large key sizes coupled with high computational burdens, especially on Windows systems. These results inform the selection of PQC algorithms based on the limitations of the implementation platform, outline the objectives for the operation efficiency of algorithms, and help guide future research focused on optimization and deployment within real-world cryptographic systems.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"125 ","pages":"Article 110437"},"PeriodicalIF":4.0,"publicationDate":"2025-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144178425","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}
{"title":"A hybrid model based on advanced optimization algorithm, and deep learning model for wind turbine sensor condition monitoring and fault identification","authors":"Anfeng Zhu , Qiancheng Zhao , Tianlong Yang , Ling Zhou , Bing Zeng","doi":"10.1016/j.compeleceng.2025.110465","DOIUrl":"10.1016/j.compeleceng.2025.110465","url":null,"abstract":"<div><div>Sensors are crucial components of wind turbines, and their stable and reliable operation directly affects the safety and economic benefits of wind turbines. To effectively monitor the status of the sensors, this paper proposes a technique for monitoring the status and fault identification of wind turbine sensors based on multi-strategy optimization Harris Hawks optimization (MHHO) and deep belief network (DBN). Firstly, the input and output parameters of wind speed sensor and temperature sensor are selected using the mixed correlation index. Second, the MHHO-DBN-based wind turbine sensor state monitoring and fault identification model is established, the time sliding window performance evaluation index is constructed, and the threshold of the wind turbine sensor abnormality index is determined according to the interval estimation theory of statistics. Then, a mathematical model is established to identify the faults of sensors with abnormal states. Finally, the MHHO-DBN model is established to monitor the actual sensor state, and the mathematical model is used to identify the fault. The calculation results reveal that this technique can effectively monitor the state of the wind turbine sensors and recognize the sensor fault categories in time, which is of good engineering practical meaning for improving the safety of wind turbine operation.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"125 ","pages":"Article 110465"},"PeriodicalIF":4.0,"publicationDate":"2025-05-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144178424","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}
{"title":"Weighted error support algorithm for accelerated convergence channel equalizer","authors":"Haider A. Mohamed-Kazim , Ikhlas Abdel-Qader","doi":"10.1016/j.compeleceng.2025.110440","DOIUrl":"10.1016/j.compeleceng.2025.110440","url":null,"abstract":"<div><div>To accelerate the convergence speed of adaptive equalizers and improve the robustness against inter-symbol interference and noise, a new weighted-error-based adjustable stepsize is proposed. The algorithm intends to achieve this acceleration while maintaining a comparable steady-state error to other related approaches. The key innovation is in leveraging two patterns of gradient, named forced-change gradients, to adjust the stepsize. These are the gradient of error and the gradient of coefficients, in which we consider the difference between their estimated values at the current samples and a scaled version of their estimated values from previous samples. Combining these gradients, we create a weighted-error parameter used for adjusting the stepsize. The motivation of the proposed approach is to exploit the combination of these two gradients to accelerate the convergence speed as well as to reduce the steady-state error. The algorithm delivers distinguished outcomes, especially in terms of bit-error-rate, with fewer training symbols compared with others. Furthermore, the algorithm shows stable performance under various levels of the signal-to-noise ratio. The superiority in attenuating the inter-symbol interference and accelerating convergence is demonstrated through the simulation results. Computer simulations are also provided to support theoretical findings.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110440"},"PeriodicalIF":4.0,"publicationDate":"2025-05-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144178822","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}
{"title":"Active disturbance rejection control strategy for PMSMs in electric tractors based on fuzzy cascaded extended observers","authors":"Qiang Yu, Xinjia Pan, Xionglin He, Longze Liu, Zihong Jiang","doi":"10.1016/j.compeleceng.2025.110466","DOIUrl":"10.1016/j.compeleceng.2025.110466","url":null,"abstract":"<div><div>Compared with traditional fuel tractors, electric tractors have significant advantages in environmental sustainability, energy efficiency, and intelligence, which has become an important way to realize the \"dual-carbon\" goal in China's agricultural field. However, in practical applications, electric tractors often face the problem of high-torque load disturbance under complex working conditions, which poses a serious challenge to the stability and operational efficiency of the tractor. To address this problem, an active disturbance rejection control (ADRC) strategy based on fuzzy cascaded extended state observer (FCESO) is proposed in this paper for the speed control of the permanent magnet synchronous motor (PMSM) of the electric tractor. The proposed scheme employs a two-stage cascaded extended state observer architecture, enhancing the traditional observer to improve estimation accuracy and compensation capability for complex disturbances. Additionally, the dynamic adjustment of bandwidth parameters by combining fuzzy control further enhances the robustness and adaptability of the control strategy. To validate the effectiveness of the proposed strategy, comparative tests were performed on a 2.5 kW PMSM platform. The results indicate that, compared with the traditional PI control, ADRC, and cascade active disturbance rejection control (CADRC), the strategy proposed in this paper has obvious advantages in reducing speed fluctuation, speed overshoot, and improving dynamic response speed. Experimental results demonstrate that the proposed FCADRC exhibits only 12 % of the overshoot observed in PI control, while reducing steady-state speed fluctuation by 18 % and speed overshoot by 3 % compared to ADRC. The controller achieves steady-state errors that are 58 % of ADRC's values, with effective disturbance rejection under actual tractor load frequencies, confirming its superior adaptability to varying agricultural inertia conditions.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110466"},"PeriodicalIF":4.0,"publicationDate":"2025-05-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144166088","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}
{"title":"Metapath-guided graph neural networks for financial fraud detection","authors":"Junjie Qian, Guoxiang Tong","doi":"10.1016/j.compeleceng.2025.110428","DOIUrl":"10.1016/j.compeleceng.2025.110428","url":null,"abstract":"<div><div>Financial fraud detection is an important task to ensure the security of financial system. Graph neural networks has shown good results in the field of financial fraud detection. However there are problems of insufficient data mining and category imbalance in heterogeneous graphs of financial transaction networks. Therefore, this paper proposes Metapath Graph neural networks(Metapath-GNN), a graph neural network model based on metapath subgraph, for detecting financial frauds in complex transaction networks and hidden pattern states. Firstly, the subgraph is generated based on predefined metapath patterns by the metapath subgraph generation module. And the node selection is adjusted using the attention mechanism to improve the adaptability to the category imbalance data; then, an aggregation module is utilized to combine the subgraph and full graph information to generate more representative node embeddings. The effective information is fully exploited to enhance the detection performance of the model. Metapath-GNN is extensively evaluated on public datasets YelpChi, Amazon and Elliptic. In addition, for Elliptic, a real-world financial transaction dataset, the data labeling cost is reduced by a semi-supervised learning approach that makes full use of unlabeled data for training. The optimal performance is also achieved in the comparison experiments with the advanced methods. Such as F1-macro, Area Under the Receiver Operating Characteristic Curve(AUC) and Geometric Mean(GMean), by 11.33%, 1.26%, and 7.00% on YelpChi, 1.75%, 1.31% and 1.22% on Amazon, respectively. In Elliptic key indicator F1 improved by 6.78%. In T-Finance key metrics F1 improved by 1.28% and AUC by 3.54%.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110428"},"PeriodicalIF":4.0,"publicationDate":"2025-05-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144147772","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}
{"title":"Optimized phase disposition sinusoidal pulse width modulation classical multi-level inverters using curve fitting techniques and genetic algorithm","authors":"Salahaldeen AlZoubi, Majd Ghazi Batarseh","doi":"10.1016/j.compeleceng.2025.110464","DOIUrl":"10.1016/j.compeleceng.2025.110464","url":null,"abstract":"<div><div>This paper presents an open-loop optimization methodology for three classical multi-level inverter types: cascaded H-bridge, neutral point clamped, and flying capacitors inverters controlled by Phase Disposition Sinusoidal Pulse Width Modulation (PD-SPWM) to minimize Total Harmonic Distortion (THD) in a single-phase output voltage while maximizing efficiency. The optimization process encompasses data collection through simulating the three inverter types from three to seventeen levels in MATLAB/Simulink, optimizing the sinusoidal reference waveform amplitude and carrier waveform frequency for each level of the three multi-level inverter types corresponding to the lowest THD and within a predefined desired output voltage range. Curve-fitting techniques are then employed to predict the optimal operating points for levels beyond seventeen by analyzing the data collected from the simulations of the three multi-level inverters for levels below seventeen, identifying underlying patterns and relations, and then forecasting the futuristic data. A genetic algorithm is subsequently used to select an appropriate filter, ensuring that the THD remains below 5 %, in compliance with IEEE Standard 519. The optimal points obtained in this study are compared with existing works, demonstrating consistently lower THD across all configurations.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"125 ","pages":"Article 110464"},"PeriodicalIF":4.0,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144155160","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}
Kawtar Tifidat, Noureddine Maouhoub, Fatima Ezzahra Ait Salah
{"title":"Hybrid methodology for accurate modeling of PV modules' electrical performance by extracting the two-diode model's electrical parameters","authors":"Kawtar Tifidat, Noureddine Maouhoub, Fatima Ezzahra Ait Salah","doi":"10.1016/j.compeleceng.2025.110436","DOIUrl":"10.1016/j.compeleceng.2025.110436","url":null,"abstract":"<div><div>This investigation aims to model the electrical behavior of photovoltaic modules by extracting the seven parameters (R<sub>p</sub>, R<sub>s</sub>, A<sub>1</sub>, I<sub>01</sub>, A<sub>2</sub>, I<sub>02</sub>, and I<sub>ph</sub>) of their two-diode circuit model. A novel numerical-analytical hybrid approach is proposed to simplify the complexities caused by the photovoltaic equation’s non-linearity, allowing smooth incorporation into photovoltaic systems for circuit-level modeling. The proposed technique relies solely on PV modules’ characteristic points, eliminating the need for experimental measurements, which enhances its applicability. In the first step of the proposed identification process, the parameters R<sub>s</sub>, I<sub>ph</sub>, A<sub>1</sub>, and A<sub>2</sub>, with readily determined initial values, are extracted numerically. Second, the remaining parameters, which are strongly influenced by the module’s characteristics and manufacturing processes, are determined analytically based on the values obtained for the first four parameters. The performance was tested on various modules and benchmarked against well-established modeling techniques. The proposed method demonstrated high effectiveness in PV performance modeling, achieving an RMSE below 0.045 A. for current-voltage characteristics prediction, with results generated in less than 0.12 s. Moreover, the suggested method was also evaluated for the one-diode model and demonstrated high accuracy. The outcomes highlight its suitability for PV designers seeking straightforward modeling techniques and applicability in dynamic scenarios.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"125 ","pages":"Article 110436"},"PeriodicalIF":4.0,"publicationDate":"2025-05-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144155161","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}
Fei Wang , Jia Wu , Rui Ma , Yisha Liu , Zengshuai Qiu
{"title":"A cross-modal Siamese representation learning network for point cloud understanding","authors":"Fei Wang , Jia Wu , Rui Ma , Yisha Liu , Zengshuai Qiu","doi":"10.1016/j.compeleceng.2025.110426","DOIUrl":"10.1016/j.compeleceng.2025.110426","url":null,"abstract":"<div><div>Learning effective representations from unannotated point cloud data is a challenging task in self-supervised learning. Recently, methods that use point clouds and images for cross-modal learning have achieved impressive performance. However, these methods still have some shortcomings in exploring the latent information between these two modalities. To address this issue, we propose a cross-modal Siamese representation learning network called CrossSiamese. This network uses point clouds and their rendered images for cross-modal contrastive learning. We introduce an intra-modal prediction mechanism in the network to capture the internal information in the point cloud and image modalities. In addition, we introduce a cross-modal cross-prediction mechanism to capture mutual information between the two modalities. Experimental results show that our method improves the accuracy of linear classification for 3D objects by 0.4% on ModelNet40 and 1.7% on ScanObjectNN compared to existing baseline methods. Additionally, experiments on few-shot object classification and 3D object part segmentation further validate the effectiveness of our method. These results indicate that the representations learned by our method have generalization ability and can be effectively transferred to these three downstream tasks.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"126 ","pages":"Article 110426"},"PeriodicalIF":4.0,"publicationDate":"2025-05-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144134075","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}
{"title":"Graphically Residual Attentive Network for tackling aerial image occlusion","authors":"Praveen Kumar Pradhan , Kunal Purkayastha , Aaditya Lochan Sharma , Udayan Baruah , Biswaraj Sen , Palash Ghosal","doi":"10.1016/j.compeleceng.2025.110429","DOIUrl":"10.1016/j.compeleceng.2025.110429","url":null,"abstract":"<div><div>Deep learning has rapidly advanced, enabling new applications such as object detection, text recognition, and occlusion handling. However, challenges remain in the detection of objects in complex environments such as aerial images where things like motion blur, low light, and significant occlusion occur. This paper addresses a similar challenge by introducing a novel supervised framework, the Graphically Residual Attentive Network (GRESIDAN). In the same model, GRESIDAN integrates three synergistic pipelines for object detection, occlusion detection, and occlusion removal. GRESIDAN uses a residually attentive block combining ResNet-18 and a multi-headed attention mechanism to improve feature extraction and detection accuracy in low-quality, occluded aerial images. A graphically attentive occlusion detection pipeline is implemented to handle occlusion, segment better, and mask out the occluder in the aerial image. The GRESIDAN model is validated on the COCO-2017 dataset and a custom private aerial object detection dataset, outperforming the state-of-the-art methods in handling occlusion and detecting objects. Our contributions provide a robust solution to the problem of detecting and handling occluded objects in aerial imagery, pushing the boundaries of automated visual recognition in challenging real-world scenarios. The code for public use in training and testing is available on <span><span>GitHub</span><svg><path></path></svg></span>.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"125 ","pages":"Article 110429"},"PeriodicalIF":4.0,"publicationDate":"2025-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"144131244","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}