Computers & Electrical Engineering最新文献

{"title":"PAKMamba: Enhancing electricity load forecasting with periodic aggregation and Koopman analysis","authors":"Tao Shen ,&nbsp;Wenbin Shi ,&nbsp;Jingsheng Lei ,&nbsp;Qiwei Li","doi":"10.1016/j.compeleceng.2025.110113","DOIUrl":"10.1016/j.compeleceng.2025.110113","url":null,"abstract":"<div><div>In recent years, with the increasing proportion of renewable energy in power systems, the difficulty of system dispatching has also increased. Accurate power load forecasting is an important prerequisite for achieving flexible dispatching. Power load exhibits significant daily and weekly periodicity and non-stationary characteristics. Current deep learning models cannot fully capture the periodicity and non-stationary characteristics of power load, leading to insufficient prediction accuracy and scalability. To address this issue, this paper proposes a new prediction model, PAKMamba, which consists of a dual-layer Mamba, a Periodic Aggregation module, and a Koopman Temporal Detector. The dual-layer Mamba handles the forward and backward dependencies of the sequence in parallel. The Periodic Aggregation module is used to extract the periodic properties of the sequence to capture its local features. The Koopman Temporal Detector, combined with Koopman dynamics theory, more effectively handles the non-stationarity between sequences. Validation on five datasets demonstrates that PAKMamba achieves more accurate predictions compared to other benchmark models, with an average MSE of 0.1324 and an average MAE of 0.252 for four-step predictions on the AEP(American Electric Power) dataset.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110113"},"PeriodicalIF":4.0,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143208981","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":"Dynamic malware detection based on supervised contrastive learning","authors":"Shumian Yang,&nbsp;Yongqi Yang,&nbsp;Dawei Zhao,&nbsp;Lijuan Xu,&nbsp;Xin Li,&nbsp;Fuqiang Yu,&nbsp;Jiarui Hu","doi":"10.1016/j.compeleceng.2025.110108","DOIUrl":"10.1016/j.compeleceng.2025.110108","url":null,"abstract":"<div><div>Application Programming Interface (API) calls record interactions between a program and the operating system or other programs during runtime. Due to this precise tracking capability, API call information is extensively utilized in malware detection. However, most approaches only focus on the API names or simply combine them with API parameters, leading to insufficient semantic information extraction. Additionally, with the continuous development of network technology, the behavioral feature differences between malware and benign software are gradually blurring, which makes it difficult to detect hard samples (e.g., well-disguised or atypical malware) in static analysis or simple behavioral patterns. Therefore, in this paper, we propose DMASCL, a framework that utilizes <u>D</u>ynamic <u>M</u>alware Analysis based on <u>A</u>PI calls and <u>S</u>upervised <u>C</u>ontrastive <u>L</u>earning techniques, which encodes semantic as well as statistical features in each sample, dynamically compares samples from different categories, learns inter-sample differences and performs classification. In particular, we combine API names with API parameters to construct API sentences containing rich semantic information, and propose a hybrid feature encoder for obtaining the semantics and statistical features of API parameters. We then apply supervised contrastive learning techniques for further feature learning, utilizing Gaussian noise to construct contrast tasks. The model is optimized by combining cross-entropy loss for classification with supervised contrastive loss to reinforce relationships between samples, thus enhancing the model’s ability to recognize malicious behavior. Our method achieves a 98.42% F1-score and a 98.03% recall in Dataset 1. It achieves a 99.59% F1-score and a 98.86% recall in Dataset 2. The experimental results show an increase in accuracy of 1.05% and 2.27%, respectively, compared to the state-of-the-art method.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110108"},"PeriodicalIF":4.0,"publicationDate":"2025-02-05","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143208969","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":"Reinforced learning for demand side management of smart microgrid based forecasted hybrid renewable energy scenarios","authors":"Khwairakpam Chaoba Singh,&nbsp;Shakila Baskaran,&nbsp;Prakash Marimuthu","doi":"10.1016/j.compeleceng.2025.110127","DOIUrl":"10.1016/j.compeleceng.2025.110127","url":null,"abstract":"<div><div>Energy management on residential loads is crucial since the loads vary and costs are also high. Hence, to deal with that, this paper proposes a novel demand management strategy using an energy retailing procedure. Initially, the power of PV and wind systems are forecasted using a recurrent neural network, and then the forecasted power is used to feed a household load of six devices that are non-linear. To manage the power, the loads are regularly updated in the Q-table; if any loads get shut, then the power retailing is performed, from which the average cost of the power consumed is reduced by African vulture optimization. Further, demand management is tested by varying the hybrid power sources. Under PV, wind and battery scenarios, the net present value and levelized cost of energy are 5115.31$ and 8.7$/kWh, respectively.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110127"},"PeriodicalIF":4.0,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097549","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":"Adaptive EKF algorithm with novel fading memory: Design and validation","authors":"Emrah Zerdali ,&nbsp;Recep Yildiz ,&nbsp;Levent Ozbek ,&nbsp;Murat Barut","doi":"10.1016/j.compeleceng.2025.110130","DOIUrl":"10.1016/j.compeleceng.2025.110130","url":null,"abstract":"<div><div>In this paper, an adaptive extended Kalman filter (EKF) with a novel fading memory is proposed and validated through its application to the state estimation of an induction motor (IM). A standard EKF (SEKF) observer requires a stochastic system with complete dynamic or measurement equations to achieve estimations. However, in practice, these equations are often partially known or entirely unknown, which degrades the performance of the SEKF. To address this issue, an EKF observer with a novel fading memory is proposed and applied to the state estimation problem of an IM. To validate its effectiveness, the proposed adaptive fading EKF (AFEKF) is experimentally compared to the SEKF and an existing AFEKF from different perspectives. The results confirm that the proposed AFEKF achieves better estimation performance with reduced computational complexity compared to the existing AFEKF approach.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110130"},"PeriodicalIF":4.0,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097546","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":"Integrating individualised and similar group in knowledge tracing","authors":"Xin Liu ,&nbsp;Pan Hu ,&nbsp;You Peng Su","doi":"10.1016/j.compeleceng.2025.110105","DOIUrl":"10.1016/j.compeleceng.2025.110105","url":null,"abstract":"<div><div>In light of the accelerated growth of online education platforms, Knowledge Tracing (KT), paramount for anticipating learners’ academic performance, assumes an increasingly significant function in real-time content adaptation and forecasting learner outcomes in intelligent education systems. Despite using diverse, complex neural networks to extract features from users’ historical interaction data, researchers often restrict the scope of personalisation and fail to consider the interconnection between individualisation and group similarity. Furthermore, the specific treatment of distinctive characteristics within personalised information is often disregarded in discussions focusing on personalisation. We propose a novel Individualised Group Knowledge Tracing (IGKT) model to address this research gap. The model is designed to focus on learners’ individualised behaviours, employing an attention mechanism to facilitate the processing of significant actions contained within these behaviours. Furthermore, we investigate the problem-skill dimension in conjunction with extracting latent features of learning resources through learners’ study behaviours. In our investigation of group characteristics, we move beyond the conventional aggregation of analogous knowledge states, integrating a more nuanced and detailed set of learning behaviour traits among learners while examining the influence of learning resources on group similarities. The Q-matrix is employed to update learners’ skill mastery levels and learner similarities, thereby integrating personalised and group features in subsequent modules. Furthermore, we conduct a detailed examination of learners’ knowledge state, obtaining a more objective representation of their knowledge state from the perspective of learning resources. We have also designed a forgetting gate incorporating filtered personalised features to achieve an individualised forgetting mechanism. Extensive experiments on three public datasets demonstrate that our model achieves higher prediction accuracy and more precisely captures learners’ knowledge state. Our research findings not only showcase the superiority of our model but also provide valuable insights for future joint studies on personalisation and group characteristics in Knowledge Tracing (KT) models.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110105"},"PeriodicalIF":4.0,"publicationDate":"2025-02-04","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097977","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":"Scalable COVID-19 classification using map reduce framework and deep learning enabled hunter Jaya African vultures optimization","authors":"Bhagyashree R. Patle ,&nbsp;Vijayarajan V","doi":"10.1016/j.compeleceng.2024.109943","DOIUrl":"10.1016/j.compeleceng.2024.109943","url":null,"abstract":"<div><div>Coronavirus disease 2019 (COVID-19) is the infection caused by severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) which was first identified in patients with mild respiratory illness. Some patients having COVID-19 recover without special treatment, and some require severe medical attention. This COVID-19 medical data is of big data and it is significant to classify for intelligentization of medical information. In this paper, Hunter Jaya African Vultures Optimization-based Deep Belief Network (HJAVO_DBN) is proposed for big data classification. Initially, input data is subjected to Deep Fuzzy Clustering (DFC) for data partitioning, followed by a map-reduce framework that has a Mapper phase and a Reducer phase. In each mapper, pre-processing is done by Z-score normalization and feature fusion is done using the Kumar-Hassebrook measure and Deep Residual Network (DRN). Afterwards, the output from the mapper phase is subjected to the reducer phase, where all fused data are initially merged and then fed to data augmentation for performing oversampling. Finally, big data classification is done by DBN that is structurally optimized using HJAVO, formed by the integration of Hunter Prey-Optimizer (HPO) and Jaya African Vultures Optimization (JAVO). Furthermore, JAVO is formed by combining the Jaya Algorithm and African Vultures Optimization (AVO). Moreover, the proposed HJAVO_DBN is analyzed for its effectiveness depending on various measures, like accuracy, sensitivity, and specificity, with greater performance of 92.28 %, 91.60 %, and 92.99 %.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 109943"},"PeriodicalIF":4.0,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143148974","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":"AI-enabled Computational Intelligence Approach to Neurodevelopmental Disorders Detection Using rs-fMRI Data","authors":"Soham Bandyopadhyay ,&nbsp;Monalisa Sarma ,&nbsp;Debasis Samanta","doi":"10.1016/j.compeleceng.2025.110117","DOIUrl":"10.1016/j.compeleceng.2025.110117","url":null,"abstract":"<div><div>Neurodevelopmental disorders (NDDs), including ADHD and ASD, profoundly impact children and adolescents. Leveraging Machine Learning (ML), Deep Learning (DeepL) on Functional magnetic resonance imaging (fMRI) data offers enhanced insights, advancing the understanding and diagnostic capabilities of NDDs. Traditionally, researchers extract time series data from predefined brain regions (ROIs) using atlas-based methods and focus on generating brain functional connectivity using Pearson correlation by analyzing changes in signal amplitude over time. This conventional approach assumes that the brain’s structure can be modeled in a simple Euclidean space and predicted with conventional ML/DeepL techniques. However, these traditional methods have several drawbacks. Predefined ROI extraction fails to capture the inherent variability in brain connectivity patterns across individuals, potentially missing crucial information, while relying on Pearson correlation to analyze functional brain connectivity is sensitive to amplitude fluctuations caused by high neural oscillations, leading to inaccurate representations of true neural relationships. Modeling brain functional structure in Euclidean space does not account for the brain’s complex, non-linear neural dynamics, limiting the effectiveness of ML/DeepL models. To address these issues, we propose: 1) An approach that adapts ROIs for each subject using combined grouped Independent Component Analysis (ICA) and Dictionary Learning (DL), better representing individual brain topologies; 2) The application of Phase Locking Value (PLV) to estimate functional connectivity in the frequency domain, reducing sensitivity to amplitude variations while effectively capturing both linear and non-linear signal relationships; 3) The implementation of a Graph Convolutional Network (GCN) to address the brain’s non-Euclidean topological structure with graph architecture, enhancing the classification and diagnosis of neural disorders. This method was tested on the ADHD-200 dataset for ADHD and the ABIDE-I dataset for ASD, achieving high accuracy (94% ±1.3% for ADHD and 89.3% ±2.3% for ASD) through 10-fold cross-validation. The integration of data-driven ROI extraction, frequency-domain connectivity analysis, and non-Euclidean graph-based brain architecture representation collectively represents a novel approach to improving the understanding and prediction of NDDs.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110117"},"PeriodicalIF":4.0,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097550","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 new nonlinear-doped beta controller for robust and resilient photovoltaic systems","authors":"Njimboh Henry Alombah ,&nbsp;Ambe Harrison ,&nbsp;Jean de Dieu Nguimfack Ndongmo ,&nbsp;Hilaire Bertrand Fotsin ,&nbsp;Ateekh Ur Rehman","doi":"10.1016/j.compeleceng.2025.110128","DOIUrl":"10.1016/j.compeleceng.2025.110128","url":null,"abstract":"<div><div>This paper presents the development of a nonlinear controller for robust and rapid maximum power point tracking (MPPT) in photovoltaic (PV) systems. Utilizing a two-stage approach, the proposed method integrates a variable step-size beta-integral backstepping control (VSS-IBSC) algorithm for initial approximation of the MPP, followed by a voltage-based incremental conductance-integral backstepping control (VINC-IBSC) algorithm for precise MPP tracking. The controller is based on the beta algorithm, doped with nonlinear features to enhance performance under variable load and irradiance conditions. The controller's stability is validated through Lyapunov's law, ensuring robustness. Simulation results demonstrate the controller's ability to achieve an MPP tracking efficiency of 99.969 % with a rapid settling time of 0.3 ms. Compared to conventional incremental conductance (INC) and sliding mode controller (SMC), the proposed controller reduces power tracking overshoot to 1.91 % (from 5.84 % for SMC) and improves tracking accuracy, with mean absolute error (MAE) and mean square error (MSE) values of 0.07 and 0.03, respectively. Further robustness is demonstrated under abrupt load variations, where the proposed controller maintains MPP power with minimal deviation. During parametric uncertainty scenarios, the controller exhibits superior response times, settling at 0.84 ms (Scenario-A) and 1.26 ms (Scenario-B), significantly outperforming INC (11.85 ms, 9.8 ms) and SMC (5.57 ms, 3.51 ms). Experimental validation under real environmental conditions corroborates the simulation results, with the proposed controller achieving an energy yield of 23.4938 W·s, compared to 22.7795 W·s (INC) and 22.5988 W·s (SMC). This novel nonlinear doped beta controller demonstrates exceptional performance, providing a robust, sensor-free solution for PV systems under dynamic and challenging conditions.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110128"},"PeriodicalIF":4.0,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097543","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 taxonomical review: recent advancements in FACTS controllers on power systems with modern optimization techniques","authors":"Chandu Valuva,&nbsp;Subramani Chinnamuthu","doi":"10.1016/j.compeleceng.2025.110120","DOIUrl":"10.1016/j.compeleceng.2025.110120","url":null,"abstract":"<div><div>Electrical power systems extensively sustain successive reformations and advances, as well as mounting load demand and integration of renewable energy sources. Eventually, the system should adapt rapidly and be reliable for extended generations and varied load demands. The necessity of more significant, effective operation of an electrical network has led to innovational advancements in power generation and transmission technology. A power electronics-based controller, a Flexible Alternating Current Transmission System (FACTS), plays a significant role in boosting power quality, enhancing voltage stability, minimizing losses, enhancing system stability, and enabling efficient integration of renewable energy. Researchers have devised recent advancements in modern metaheuristic methods that further improve the deployment of FACTS, empowering their best size and location within the power grid. This review article provides a detailed explanation of the mathematical modeling of different FACTS controllers with their control parameters and the technical role of FACTS in modern power systems.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110120"},"PeriodicalIF":4.0,"publicationDate":"2025-02-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097542","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":"An efficient tangent search based power harnessing algorithm for photovoltaic energy generation system","authors":"Injila Sajid ,&nbsp;Adil Sarwar ,&nbsp;Mohd Tariq ,&nbsp;Farhad Ilahi Bakhsh ,&nbsp;Shafiq Ahmad ,&nbsp;Adamali Shah Noor Mohamed","doi":"10.1016/j.compeleceng.2025.110118","DOIUrl":"10.1016/j.compeleceng.2025.110118","url":null,"abstract":"<div><div>Solar energy is a key player in renewable energy, essential for meeting global demand and combating climate change. However, the efficiency of Photovoltaic (PV) systems depends on atmospheric conditions, with partial shading causing significant power losses. Managing the nonlinear I-V curve remains a challenge, and research on nature-inspired control methods emphasizes the need for optimization. The present study presents an innovative optimization algorithm called the Tangent Search Algorithm (TSA), which leverages mathematical concepts, particularly the tangent function, for Maximum Power Point Tracking (MPPT) control in PV systems. The proposed approach aims to overcome the limitations of existing methods, enabling rapid and efficient tracking of the GMPP even under partial shading. A comprehensive comparison between the TSA approach, Particle Swarm Optimization (PSO), and Cuckoo Search Algorithm (CSA) is conducted, involving in-depth qualitative, quantitative, and statistical analyses using specific case examples. The real-world applicability of the TSA-based MPPT controller is validated through hardware-in-the-loop verification. The proposed TSA method achieves an average convergence speed enhancement of up to 161.5 %. Key attributes of the proposed TSA control strategy include its simplicity in implementation, robustness, and a high power tracking efficiency of nearly 99.90 % under steady-state conditions.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110118"},"PeriodicalIF":4.0,"publicationDate":"2025-02-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143097978","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}