IEEE Canadian Journal of Electrical and Computer Engineering最新文献

{"title":"An Adaptive Pilot Protection for Wind Farm Transmission Line Employing Multiscale Morphological Gradient","authors":"An Li;Yingjun Zhuo;Siyu Lu;Luliang Zhang;Yujie Zhang;Lingteng Zeng;Baorong Zhou","doi":"10.1109/ICJECE.2026.3716247","DOIUrl":"https://doi.org/10.1109/ICJECE.2026.3716247","url":null,"abstract":"The complex fault-currents and prominent harmonic components of doubly-fed induction generators (DFIGs) pose significant sensitivity and mal-operation risks for traditional pilot protection on long-distance transmission lines. To address these issues, this article proposes a novel adaptive time-domain pilot protection scheme based on multiscale morphological gradients. This method uses morphological gradient with multiscale structural elements for robust fault-current characterization, extracting transient fault current features at both ends of the line, balancing computational efficiency with detailed waveform preservation. Fault detection is achieved through kurtosis of morphological gradients within a sliding window, complemented by an adaptive weighted sum of energy differences criterion to distinguish internal from external faults. The proposed method was validated using a wind farm transmission line model in power systems computer aided design (PSCAD)/electromagnetic transients, including DC (EMTDC). Extensive PSCAD/EMTDC simulations demonstrate that the proposed scheme achieves ultrafast fault detection within 1.5 ms and requires a minimal computational load of only 11 core operations (COps) per data point. Furthermore, by extracting and transmitting only scalar feature values via an adaptive alignment mechanism, it completely eliminates the reliance on strict time synchronization. The energy index maintains a distinct selective margin even under extreme boundary conditions, proving its superior practical reliability for wind farm export lines.","PeriodicalId":100619,"journal":{"name":"IEEE Canadian Journal of Electrical and Computer Engineering","volume":"49 4","pages":"605-616"},"PeriodicalIF":1.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148877939","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Federated Learning-Based Distributed Frequency Control of Networked Microgrids Under PMU Failures Using Fuzzy Neural Network FOPID Controller","authors":"Jeevitha Kandasamy;Sheila Mahapatra;Fahima Hajjej","doi":"10.1109/ICJECE.2026.3703402","DOIUrl":"https://doi.org/10.1109/ICJECE.2026.3703402","url":null,"abstract":"Modern power systems, with the large-scale integration of renewable energy (RE) and distributed energy resources (DERs), have evolved into networked microgrid systems (NMGSs). While this transition aligns with sustainable development goals, it also introduces significant reliability challenges, including frequency instability caused by low system inertia, slow stochastic variations in load demand, communication time delays, and cyber–physical disturbances. Such scenarios make traditional centralized control strategies ambiguous, as system performance can drop significantly even in a simple case of communication disruption. To overcome these issues, a new distributed control framework for NMGS is developed based on federated learning fuzzy neural network (NN) optimized fractional-order PID (FLFNN FOPID). It is a federated architecture in which agents exchange only parameter updates, not raw operational data. This strategy preserves data confidentiality, reduces communication overhead, and enables control of synchronization frequency among DER. The proposed controller is then rigorously validated across normal load conditions, random disturbances, and phasor measurement unit (PMU) failure cases, while its practical implementation feasibility is additionally ascertained via hardware-in-the-loop (HIL) experiments on an OPAL-RT real-time simulator. Compared to the NN FOPID baseline, the proposed FLFNN FOPID reduces integral absolute error (IAE) by more than 69% for <inline-formula> <tex-math>$Delta $ </tex-math></inline-formula>f1, with its value (<inline-formula> <tex-math>$2.4235times 10^{-3}$ </tex-math></inline-formula>) compared with that of NN FOPID’s (<inline-formula> <tex-math>$0.7313times 10^{-3}$ </tex-math></inline-formula>), while keeping track of improvement in almost similar ratios for each <inline-formula> <tex-math>$Delta $ </tex-math></inline-formula>f2 and <inline-formula> <tex-math>$Delta $ </tex-math></inline-formula>f3 independently too. It is important to note that the integral time absolute error (ITAE) in <inline-formula> <tex-math>$Delta $ </tex-math></inline-formula>f2 decreased by almost 93% from <inline-formula> <tex-math>$0.1513times 10^{-5}$ </tex-math></inline-formula> to <inline-formula> <tex-math>$0.0099times 10^{-5}$ </tex-math></inline-formula>, which demonstrates a much-improved transient behavior. In all three MGs, integral squared error (ISE) and integral time-weighted squared error (ITSE) are also minimized, indicating reduced oscillatory behavior and improved system stability. Settling time reduced from 6.2 to 6.6 s under NN FOPID to 3.5–3.9 s under the proposed controller, corresponding to faster stabilization of about ~44%. Additionally, reductions in peak magnitude, rise time, peak time, and absolute error collectively indicate improved steady-state accuracy. These results have made the proposed FLFNN FOPID framework a powerful, privacy-preserving, and communication-friendly solution for frequency regulation ","PeriodicalId":100619,"journal":{"name":"IEEE Canadian Journal of Electrical and Computer Engineering","volume":"49 4","pages":"634-652"},"PeriodicalIF":1.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148928256","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Impact of Wind and Sea Loadings on the Voyage of All-Electric Ship","authors":"Puspendu Ghosh;Subho Paul;Kanendra Naidu;Lilik Jamilatul Awalin","doi":"10.1109/ICJECE.2026.3718016","DOIUrl":"https://doi.org/10.1109/ICJECE.2026.3718016","url":null,"abstract":"An all-electric ship (AES) generally employs integrated electric propulsion and coordinated onboard power supply to achieve economically efficient and environmentally sustainable maritime transportation. This article aims to evaluate the impact of wind and sea loadings on AES voyage performance and economic efficiency using a deterministic single-stage optimization model. The proposed multiobjective model optimizes total operating cost (TOC) and voyage unmet distance while satisfying navigation, propulsion, and electrical operating constraints. Wind and wave loads are modeled using analytical and semi-empirical wind-based resistance models to assess the resulting changes in propulsion power demand and attainable sailing speed. The optimization method is formulated as a mixed-integer linear programming (MILP) problem and solved using the MATLAB built-in intlinprog solver. Tradeoffs between conflicting objectives are analyzed through the Pareto front of the obtained nondominated solutions. The simulation results demonstrate that wind and wave loads significantly influence AES operation by increasing ship resistance, which in turn affects propulsion power demand.","PeriodicalId":100619,"journal":{"name":"IEEE Canadian Journal of Electrical and Computer Engineering","volume":"49 4","pages":"653-664"},"PeriodicalIF":1.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148928297","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Efficient Solar-Powered EV Charging System With High-Gain DC–DC Converter","authors":"Shantanu Saxena;Rajan Kumar;Nalin Kumar Sharma;Sudhakar Modem","doi":"10.1109/ICJECE.2026.3722929","DOIUrl":"https://doi.org/10.1109/ICJECE.2026.3722929","url":null,"abstract":"This article investigates an efficient solar-powered electric vehicle (EV) charging system using a high-gain DC–DC converter to address the limited charging infrastructure in EV adoption. The proposed architecture addresses the limitations of conventional multistage solar EV charging systems by eliminating the need for separate power conversion stages for photovoltaic (PV) power extraction and battery current regulation. This topology consolidates the functions of two separate converters, providing a cost-effective solution for EV charging. Employing maximum power point tracking (MPPT), the system optimizes power extraction from solar panels, enhancing the charging process. The incorporated charge controller ensures precise regulation of maximum output current, mitigating undershoot and overshoot, thereby ensuring safe and stable charging even amidst input power fluctuations. The system’s robustness under variable input power conditions underscores its suitability for diverse scenarios. Moreover, this approach can resolve the issue of limited charging systems in remote areas with efficient usage of renewable solar energy. The simulation and experimental results are presented to validate the system’s performance and feasibility.","PeriodicalId":100619,"journal":{"name":"IEEE Canadian Journal of Electrical and Computer Engineering","volume":"49 4","pages":"665-674"},"PeriodicalIF":1.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149133722","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluating the Robustness of Counterfactual Generating Methods for Tree-Based Ensembles","authors":"Rashinda Wijethunga;Vinura Galwaduge;Ayan Sadhu;Jagath Samarabandu","doi":"10.1109/ICJECE.2026.3685607","DOIUrl":"https://doi.org/10.1109/ICJECE.2026.3685607","url":null,"abstract":"Counterfactuals (CFs) are a widely studied approach for explaining machine learning models. They aim to identify a set of minimal and meaningful perturbations in input features required to alter a prediction of a model, thereby offering insights into decision boundaries. For instance, a CF may describe the changes needed for a rejected loan application to be accepted. Despite their usefulness, CFs are sensitive to model dynamics. Updates to training data, modifications of model parameters, or shifts in the underlying distribution can alter decision surfaces, potentially invalidating previously generated explanations. This instability raises critical concerns about the robustness and reliability of CFs in practical applications. Although tree-based ensemble models built using bagging and boosting are extensively used in real-world contexts, understanding the robustness of CFs for these models is still in its infancy. To address this gap, this work conducts an empirical investigation into the robustness of CF explanations for tree-based classifiers. In this work, several publicly available CF generation methods are evaluated across a range of experimental settings involving both bagging and boosting tree-based classifier models. Moreover, a novel robustness-oriented CF generation method is proposed that exploits the ensemble of trees in the classifier models. The results reveal substantial variation in robustness across methods and model types, raising concerns about the reliability of CFs in dynamic environments. This study advances understanding of the limitations and potential of existing techniques, and provides practical guidance for practitioners applying CFs to tree-based ensemble models. The code used for our experiments is available on <uri>https://github.com/VinuraD/CFRobustness-of-Trees</uri>.","PeriodicalId":100619,"journal":{"name":"IEEE Canadian Journal of Electrical and Computer Engineering","volume":"49 4","pages":"617-633"},"PeriodicalIF":1.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148928257","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Lightweight Autoencoder for Event-Based Vision on Low-Latency Energy-Constrained Edge Devices","authors":"Riadul Islam;Joey Mulé;Dhandeep Challagundla;Shahmir Rizvi;Sean Carson;Rachit Saini","doi":"10.1109/ICJECE.2026.3727217","DOIUrl":"https://doi.org/10.1109/ICJECE.2026.3727217","url":null,"abstract":"Event-based vision has emerged as a promising paradigm for energy-aware artificial intelligence (AI), offering sparse, low-latency visual signals that reduce redundant data processing and support sustainable edge computing. However, the asynchronous and noise-prone nature of event streams creates challenges for conventional deep learning models, which are often too computationally intensive for low-power embedded platforms. This work presents a compact and configurable event-driven autoencoder (AE) that efficiently compresses neuromorphic data while preserving essential spatiotemporal structure for downstream inference. The architecture integrates lightweight convolutional encoding with robust performance under adaptive event thresholding and a minimal classifier head, enabling substantial reductions in computational cost without degrading recognition fidelity. Extensive evaluations on the smart event face dataset (SEFD) and event-based crossing dataset (EBCD) show that the proposed framework achieves competitive or superior accuracy compared to YOLOv9 while requiring up to <inline-formula> <tex-math>$35.6times $ </tex-math></inline-formula> fewer parameters. To assess real-world sustainability, the model is deployed on resource-constrained hardware: a Raspberry Pi 4B and an NVIDIA Jetson Nano. On NVIDIA Jetson Nano, it delivers real-time throughput of 44.8 frames/s. On a Raspberry Pi 4B CPU, the 50% AE classifier consumes 16.19 J for the evaluated inference workload, corresponding to approximately <inline-formula> <tex-math>$726.3times $ </tex-math></inline-formula> lower energy consumption than YOLOv9 under the same evaluation protocol. These results demonstrate the potential of compact event-driven models to advance environmentally conscious, low-power AI systems for high-speed perception in autonomous, mobile, and embedded computing environments.","PeriodicalId":100619,"journal":{"name":"IEEE Canadian Journal of Electrical and Computer Engineering","volume":"49 4","pages":"675-685"},"PeriodicalIF":1.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149131288","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Energy-Efficient PM2.5-Based Air-Quality Forecasting Using Bayesian-Optimized CNNs on Embedded Devices","authors":"Ch Madhu Bhushan;Firoj Gazi;Md Muzakkir Hussain","doi":"10.1109/ICJECE.2026.3709151","DOIUrl":"https://doi.org/10.1109/ICJECE.2026.3709151","url":null,"abstract":"Real-time PM_2.5 forecasting on microcontroller-class devices is challenging because conventional learning models often require high memory, computation, and power. This article proposes a Bayesian-optimized convolutional neural network (BOCNN) for lightweight PM_2.5-based air-quality forecasting on embedded devices. The proposed approach uses Bayesian optimization to tune key CNN hyperparameters, including convolutional filters, kernel sizes, dense units, dropout rates, and learning rates, and then converts the optimized model to TensorFlow Lite for TinyML deployment. The model was evaluated on the Delhi 2024 CPCB air-quality dataset and compared with MLP, FFNN, CNN, LSTM, GRU, BiLSTM, Random forest (RF), Support Vector Regression (SVR), and XGBoost baselines. The proposed BOCNN achieved MAE, MSE, RMSE, and <inline-formula> <tex-math>$R^{2}$ </tex-math></inline-formula> values of 0.011093, 0.000317, 0.017815, and 0.985727, respectively. On the original PM_2.5 scale, it obtained an MAE of 9.012691 and an RMSE of 14.473820. Compared with the standard CNN baseline, the BOCNN reduced RMSE by approximately 40.73% and improved <inline-formula> <tex-math>$R^{2}$ </tex-math></inline-formula> from 0.959377 to 0.985727. For embedded deployment, the dynamically quantized TensorFlow Lite model achieved a compact size of 12.19 kB, 4344 bytes of runtime memory, 11.40 ms MCU inference latency, and a 40.45% lower power consumption than the CNN baseline. These results show that BOCNN provides an effective accuracy-efficiency tradeoff for low-power PM_2.5 forecasting in IoT-enabled smart-city environments.","PeriodicalId":100619,"journal":{"name":"IEEE Canadian Journal of Electrical and Computer Engineering","volume":"49 4","pages":"686-695"},"PeriodicalIF":1.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"149188423","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A Novel Fusion Pipeline of Foundational and Deep Learning Models for Footwear Slip Resistance Evaluation","authors":"Shaghayegh Chavoshian;Michael Brudno;Atena Roshan Fekr","doi":"10.1109/ICJECE.2026.3701534","DOIUrl":"https://doi.org/10.1109/ICJECE.2026.3701534","url":null,"abstract":"Slips and falls remain a major public health concern, especially in winter conditions and among older adults. Using slip resistance footwear is a major factor in reducing slip risk, but this feature is typically assessed in laboratories and is not accessible to the public. In this article, we presented a novel AI-driven pipeline that estimates footwear slip resistance scores directly from the images of outsole. Our approach fused foundational models with domain-adapted deep learning techniques, comprising four core stages: 1) fine-tuned segmentation using segment anything model (SAM) to detect contact regions of the outsole; 2) tile-based material classification using convolutional neural networks (CNNs) and vision transformers (ViTs) trained on a curated dataset of footwear materials; 3) visual overlay construction for interpretability; and 4) multioutput CNN regression to estimate slip resistance scores. We evaluated our system on 150 footwear images with experimentally validated slip resistance scores and achieved a mean absolute percentage error (MAPE) between 18.56% and 31.04%. This work enables accessible, image-based slip risk evaluation and represents a significant step toward AI-powered injury prevention in real-world settings.","PeriodicalId":100619,"journal":{"name":"IEEE Canadian Journal of Electrical and Computer Engineering","volume":"49 4","pages":"592-604"},"PeriodicalIF":1.6,"publicationDate":"2026-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148877946","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Permanent Magnet Synchronous Motor Drive Controller With Optimum Regenerative Braking for Electric Vehicles","authors":"P. K. Prathibha;Elizabeth Rita Samuel","doi":"10.1109/ICJECE.2026.3697845","DOIUrl":"https://doi.org/10.1109/ICJECE.2026.3697845","url":null,"abstract":"The proposed work presents the design and implementation of a permanent magnet synchronous motor (PMSM) drive using field-oriented control (FOC) for electric vehicles (EVs), with a detailed evaluation of its regenerative braking capability. Experimental investigations conducted on a 2-kW PMSM test bench demonstrate regenerative power peaks approaching 30 kW, charging currents up to 20 A, and an 18%–22% improvement in recovered braking energy under optimized 60% braking conditions, thereby validating the proposed short-circuit switching strategy (SCSS). The adaptive switching strategy dynamically regulates the braking torque based on the vehicle speed and braking demand, enabling enhanced energy recovery during regenerative braking without the use of additional power converters. The regenerated power is experimentally evaluated across a range of operating speeds; under optimal braking scenarios, simulation models are also employed to analyze the system behavior. The MATLAB/Simulink environment is employed to simulate and analyze the proposed system. This work presents an energy-efficient strategy for modern EV propulsion systems, contributing to extended driving range and improved system longevity.","PeriodicalId":100619,"journal":{"name":"IEEE Canadian Journal of Electrical and Computer Engineering","volume":"49 3","pages":"485-494"},"PeriodicalIF":1.6,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148512592","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Analyzing Patterns of Consumption on Distributed Units Location and Costs in Electrical Systems","authors":"Meisam Mahdavi;S. M. Muyeen;Abdullah G. Alharbi;Francisco Jurado","doi":"10.1109/ICJECE.2026.3710423","DOIUrl":"https://doi.org/10.1109/ICJECE.2026.3710423","url":null,"abstract":"The growing adoption of distributed generation (DG) has made it important to improve the performance of distribution networks. While numerous studies have focused on optimal DG siting and sizing for loss minimization, most approaches rely on peak load or static demand assumptions, overlooking the temporal variability of electricity consumption. However, load patterns directly influence power losses, energy procurement from the main grid, and the operational and investment costs associated with DG deployment. Ignoring these variations may lead to inaccurate cost estimation and suboptimal allocation decisions. This article investigates the impact of hourly certain and uncertain load profiles on the renewable DG allocation in radial systems through a mixed-integer conic linear programming (MICLP) model that minimizes the total annualized system cost, including energy loss costs, DG investment and operating costs, and energy purchase expenses. Four representative load profiles are incorporated into the optimization framework to evaluate their influence on DG placement decisions. The model is implemented in AMPL using CPLEX and validated on benchmark distribution systems ranging from 7 to 69 buses. Simulation results confirm that incorporating load variability and uncertainty significantly influences the DG placement and improves the accuracy of cost and loss evaluation compared with constant-load planning approaches. The findings demonstrate the importance of considering realistic consumption patterns in achieving more reliable and economically efficient DG allocation solutions.","PeriodicalId":100619,"journal":{"name":"IEEE Canadian Journal of Electrical and Computer Engineering","volume":"49 3","pages":"558-568"},"PeriodicalIF":1.6,"publicationDate":"2026-06-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"148780987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}