Pub Date : 2026-01-24DOI: 10.1016/j.compeleceng.2026.110947
Jihao Zhang, Guangwei Zhang, Ping Li, Chang Liu, Peng Gong
Robust classification of radar signals under low signal-to-noise ratio (SNR) conditions is critical for target recognition, electronic warfare, and radar emitter identification. However, the performance of conventional methods deteriorates severely in noisy environments due to interference and clutter. This paper proposes an effective classification framework based on a Support Vector Machine (SVM) that exploits the joint discriminative power of wavelet entropy and empirical mode decomposition (EMD) entropy. These two entropy measures characterize the intrinsic complexity and time–frequency structure of radar signals corrupted by noise and are combined into a compact two-dimensional feature vector. Extensive experiments on three representative radar modulation types—pulse Doppler (PD), linear frequency modulation (LFM), and pseudo-code phase modulation (PCPM)—demonstrate the robustness of the proposed method over a wide SNR range from −10 dB to 10 dB. The proposed classifier achieves 100% accuracy when the SNR is above 0 dB, maintains 95% accuracy at −5 dB, and still attains 83% accuracy at −10 dB. In comparative tests, it further achieves 56.7% accuracy at −15 dB, outperforming or matching several state-of-the-art SVM-based and deep-learning-based approaches.
{"title":"Classification of radar signals modulation based on SVM using wavelet entropy and empirical mode decomposition entropy","authors":"Jihao Zhang, Guangwei Zhang, Ping Li, Chang Liu, Peng Gong","doi":"10.1016/j.compeleceng.2026.110947","DOIUrl":"10.1016/j.compeleceng.2026.110947","url":null,"abstract":"<div><div>Robust classification of radar signals under low signal-to-noise ratio (SNR) conditions is critical for target recognition, electronic warfare, and radar emitter identification. However, the performance of conventional methods deteriorates severely in noisy environments due to interference and clutter. This paper proposes an effective classification framework based on a Support Vector Machine (SVM) that exploits the joint discriminative power of wavelet entropy and empirical mode decomposition (EMD) entropy. These two entropy measures characterize the intrinsic complexity and time–frequency structure of radar signals corrupted by noise and are combined into a compact two-dimensional feature vector. Extensive experiments on three representative radar modulation types—pulse Doppler (PD), linear frequency modulation (LFM), and pseudo-code phase modulation (PCPM)—demonstrate the robustness of the proposed method over a wide SNR range from −10 dB to 10 dB. The proposed classifier achieves 100% accuracy when the SNR is above 0 dB, maintains 95% accuracy at −5 dB, and still attains 83% accuracy at −10 dB. In comparative tests, it further achieves 56.7% accuracy at −15 dB, outperforming or matching several state-of-the-art SVM-based and deep-learning-based approaches.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"132 ","pages":"Article 110947"},"PeriodicalIF":4.9,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146080168","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}
Pub Date : 2026-01-24DOI: 10.1016/j.compeleceng.2026.110991
Guoping Zou , Peiliang Ma , Zhenguo Wang , Yuhang Li , Yongkang Peng
Due to the extremely small proportion of bolts and pins in inspection images, traditional methods are difficult to detect important defects such as bolt damage and pin missing. To overcome this limitation, this paper presents a novel approach for detecting missing pins in transmission lines, utilizing super-resolution reconstruction and cascade model. Firstly, in the first-stage, a standard You Only Look Once Version 8(YOLOv8) network is used to identify connection fittings containing pins in the image, eliminating the interference of common bolts without pins. Subsequently, the images of the connecting fittings are cropped and forwarded to the improved YOLOv8 network in the second-stage, where the normal and missing pins are distinguished. To enhance image clarity and resolve the low-resolution issues of small-sized targets, this study employs Real-ESRGAN (Enhanced Super-Resolution Generative Adversarial Network) for processing cropped connection fittings images. Additionally, in the second-stage network, the network structure was improved by replacing the standard convolution and pooling layers of YOLOv8 with Space-to-Depth (SPD) convolution modules, significantly enhancing the model's ability to extract features from low-resolution images and small objects. The experimental results indicate that compared to original YOLOv8 single-stage model and cascade model, the improved model proposed in this paper has improved the mean average precision (mAP) by 39.2 and 6.8 percentage points, respectively.
由于螺栓和销钉在检测图像中所占比例极小，传统方法难以检测出螺栓损坏、销钉缺失等重要缺陷。为了克服这一限制，本文提出了一种利用超分辨率重建和级联模型检测传输线中缺失引脚的新方法。首先，在第一阶段，使用标准的You Only Look Once Version 8（YOLOv8）网络来识别图像中包含销钉的连接配件，消除了没有销钉的普通螺栓的干扰。随后，在第二阶段，连接接头的图像被裁剪并转发到改进的YOLOv8网络，在那里区分正常和缺失的引脚。为了提高图像清晰度和解决小尺寸目标的低分辨率问题，本研究采用Real-ESRGAN（增强型超分辨率生成对抗网络）对裁剪的连接配件图像进行处理。此外，在第二阶段网络中，改进了网络结构，将YOLOv8的标准卷积层和池化层替换为Space-to-Depth （SPD）卷积模块，显著增强了模型从低分辨率图像和小物体中提取特征的能力。实验结果表明，与原始的YOLOv8单级模型和串级模型相比，本文提出的改进模型的平均精度（mAP）分别提高了39.2和6.8个百分点。
{"title":"A detection method for pin defects in transmission lines based on super-resolution reconstruction and cascade design network","authors":"Guoping Zou , Peiliang Ma , Zhenguo Wang , Yuhang Li , Yongkang Peng","doi":"10.1016/j.compeleceng.2026.110991","DOIUrl":"10.1016/j.compeleceng.2026.110991","url":null,"abstract":"<div><div>Due to the extremely small proportion of bolts and pins in inspection images, traditional methods are difficult to detect important defects such as bolt damage and pin missing. To overcome this limitation, this paper presents a novel approach for detecting missing pins in transmission lines, utilizing super-resolution reconstruction and cascade model. Firstly, in the first-stage, a standard You Only Look Once Version 8(YOLOv8) network is used to identify connection fittings containing pins in the image, eliminating the interference of common bolts without pins. Subsequently, the images of the connecting fittings are cropped and forwarded to the improved YOLOv8 network in the second-stage, where the normal and missing pins are distinguished. To enhance image clarity and resolve the low-resolution issues of small-sized targets, this study employs Real-ESRGAN (Enhanced Super-Resolution Generative Adversarial Network) for processing cropped connection fittings images. Additionally, in the second-stage network, the network structure was improved by replacing the standard convolution and pooling layers of YOLOv8 with Space-to-Depth (SPD) convolution modules, significantly enhancing the model's ability to extract features from low-resolution images and small objects. The experimental results indicate that compared to original YOLOv8 single-stage model and cascade model, the improved model proposed in this paper has improved the mean average precision (mAP) by 39.2 and 6.8 percentage points, respectively.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"132 ","pages":"Article 110991"},"PeriodicalIF":4.9,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146080163","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}
Pub Date : 2026-01-24DOI: 10.1016/j.compeleceng.2026.110978
Muhammad Bilal Zia, Xujuan Zhou, Raj Gururajan, Ka Ching Chan
The accurate detection of parasite eggs is a critical challenge in medical and veterinary diagnostics, as parasites can rapidly infect animals, humans, and even plants, causing serious health concerns. Traditional egg identification methods are highly dependent on manual microscopy, which is time consuming, skill intensive, and prone to human error, particularly in the recognition of subtle or overlapping egg features. To address these limitations, we introduce ParaVisionNet, a multitask deep learning framework that integrates Vision Transformers (ViT), Feature Pyramid Networks (FPN), and Mask Region-based Convolutional Neural Network (Mask R-CNN). This architecture is designed to detect, segment, and classify parasite eggs simultaneously in microscopic images. ViT serves as the backbone, extracting rich, high-dimensional feature maps. These are then organized into a multi-scale representation using FPN, enhancing feature clarity across different resolutions. The Region Proposal Network (RPN) proposes candidate egg regions, which are then refined by Mask Region-based Convolutional Neural Network (Mask R-CNN) with Region of Interest (ROI) align to produce precise masks and class predictions. Unlike previous ViT FPN or Swin Mask R-CNN hybrids that optimize prediction tasks independently or in sequential stages, ParaVisionNet does unified multitask inference in one pass by sharing RoI aligned features for detection, instance segmentation, and parasite type classification. Furthermore, Monte Carlo Dropout has also been incorporated within both the transformer encoder and FPN branches so that the uncertainty can be propagated throughout the prediction heads and result in the production of spatial entropy maps that indicate where uncertainty is concentrated. To the best of our knowledge, this is the first parasite microscopy framework capable of producing bounding boxes, instance masks, species classification, and uncertainty estimates from a single end-to-end training process. The model was extensively trained for over 50 epochs and tested on three datasets: the Sheep Egg dataset, Chula-Parasite Egg-11, and a custom Human Hookworm Egg dataset. It achieved remarkable results with 98.87% detection accuracy, 97.99% classification accuracy, and 98.98% multitasking accuracy, outperforming current state-of-the-art approaches. In practice, a single multitask pass reduces workflow steps and compute compared to running separate models, and the uncertainty maps help technicians triage ambiguous cases for review. These results show that ParaVisionNet is not only accurate, but is also a practical diagnostic tool in resource-limited settings.
{"title":"ParaVisionNet: A multitask vision transformer framework for accurate detection and classification of parasitic eggs in microscopy images","authors":"Muhammad Bilal Zia, Xujuan Zhou, Raj Gururajan, Ka Ching Chan","doi":"10.1016/j.compeleceng.2026.110978","DOIUrl":"10.1016/j.compeleceng.2026.110978","url":null,"abstract":"<div><div>The accurate detection of parasite eggs is a critical challenge in medical and veterinary diagnostics, as parasites can rapidly infect animals, humans, and even plants, causing serious health concerns. Traditional egg identification methods are highly dependent on manual microscopy, which is time consuming, skill intensive, and prone to human error, particularly in the recognition of subtle or overlapping egg features. To address these limitations, we introduce ParaVisionNet, a multitask deep learning framework that integrates Vision Transformers (ViT), Feature Pyramid Networks (FPN), and Mask Region-based Convolutional Neural Network (Mask R-CNN). This architecture is designed to detect, segment, and classify parasite eggs simultaneously in microscopic images. ViT serves as the backbone, extracting rich, high-dimensional feature maps. These are then organized into a multi-scale representation using FPN, enhancing feature clarity across different resolutions. The Region Proposal Network (RPN) proposes candidate egg regions, which are then refined by Mask Region-based Convolutional Neural Network (Mask R-CNN) with Region of Interest (ROI) align to produce precise masks and class predictions. Unlike previous ViT FPN or Swin Mask R-CNN hybrids that optimize prediction tasks independently or in sequential stages, ParaVisionNet does unified multitask inference in one pass by sharing RoI aligned features for detection, instance segmentation, and parasite type classification. Furthermore, Monte Carlo Dropout has also been incorporated within both the transformer encoder and FPN branches so that the uncertainty can be propagated throughout the prediction heads and result in the production of spatial entropy maps that indicate where uncertainty is concentrated. To the best of our knowledge, this is the first parasite microscopy framework capable of producing bounding boxes, instance masks, species classification, and uncertainty estimates from a single end-to-end training process. The model was extensively trained for over 50 epochs and tested on three datasets: the Sheep Egg dataset, Chula-Parasite Egg-11, and a custom Human Hookworm Egg dataset. It achieved remarkable results with 98.87% detection accuracy, 97.99% classification accuracy, and 98.98% multitasking accuracy, outperforming current state-of-the-art approaches. In practice, a single multitask pass reduces workflow steps and compute compared to running separate models, and the uncertainty maps help technicians triage ambiguous cases for review. These results show that ParaVisionNet is not only accurate, but is also a practical diagnostic tool in resource-limited settings.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"132 ","pages":"Article 110978"},"PeriodicalIF":4.9,"publicationDate":"2026-01-24","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026042","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}
Pub Date : 2026-01-22DOI: 10.1016/j.compeleceng.2026.110942
G.V. Sriramakrishnan , Ashapu Bhavani , V. Srilakshmi , B. Kiran Kumar
Developmental Dysplasia of the Hip (DDH) is a disease which affects newborn babies and young children. In DDH, the acetabulum may be shallow, or the femoral head may not fit correctly, which causes dislocation or instability of the hip joint. The early detection of DDH failed while the symptoms were mild, which led to delayed treatment and caused severe complications. Thus, the Hybrid Pyramid ResNeXt (HP-ResNeXt) is developed for detecting DDH using hip X-radiation (X-ray) images. Hip X-ray images are sourced from a database, and unwanted noise is removed through a Gaussian Adaptive Bilateral Filter (GABF). Then, a noise-free image is passed to the misshapen pelvis landmark detection phase, where Pyramid Non-local UNet (PN-UNet) is used to identify the affected pelvis region. Entropy-based Local Neighborhood Difference Pattern (LNDP) features, and Gray Level Co-Occurrence Matrix (GLCM) are extracted. Finally, the HP-ResNeXt method is applied for DDH detection, which integrates the advantages of Pyramid Network (PyramidNet) and ResNeXt. The newly introduced HP-ResNeXt approach achieved a True Positive Rate (TPR) of 93.272%, a True Negative Rate (TNR) of 92.567%, and an accuracy of 92.588% with a K-value of 8.
{"title":"HP-ResNeXt: Hybrid Pyramid ResNeXt for Detection of Developmental Dysplasia of the Hip in X-ray Image","authors":"G.V. Sriramakrishnan , Ashapu Bhavani , V. Srilakshmi , B. Kiran Kumar","doi":"10.1016/j.compeleceng.2026.110942","DOIUrl":"10.1016/j.compeleceng.2026.110942","url":null,"abstract":"<div><div>Developmental Dysplasia of the Hip (DDH) is a disease which affects newborn babies and young children. In DDH, the acetabulum may be shallow, or the femoral head may not fit correctly, which causes dislocation or instability of the hip joint. The early detection of DDH failed while the symptoms were mild, which led to delayed treatment and caused severe complications. Thus, the Hybrid Pyramid ResNeXt (HP-ResNeXt) is developed for detecting DDH using hip X-radiation (X-ray) images. Hip X-ray images are sourced from a database, and unwanted noise is removed through a Gaussian Adaptive Bilateral Filter (GABF). Then, a noise-free image is passed to the misshapen pelvis landmark detection phase, where Pyramid Non-local UNet (PN-UNet) is used to identify the affected pelvis region. Entropy-based Local Neighborhood Difference Pattern (LNDP) features, and Gray Level Co-Occurrence Matrix (GLCM) are extracted. Finally, the HP-ResNeXt method is applied for DDH detection, which integrates the advantages of Pyramid Network (PyramidNet) and ResNeXt. The newly introduced HP-ResNeXt approach achieved a True Positive Rate (TPR) of 93.272%, a True Negative Rate (TNR) of 92.567%, and an accuracy of 92.588% with a K-value of 8.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"132 ","pages":"Article 110942"},"PeriodicalIF":4.9,"publicationDate":"2026-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026041","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}
Pub Date : 2026-01-22DOI: 10.1016/j.compeleceng.2026.110967
Anil Saini, Kewal Krishan, Manoj Singh Gaur
The increasing convergence of Operational Technology (OT) and Information Technology (IT) has fundamentally transformed modern industrial environments. This shift has driven advancements in automation, data-driven decision-making, and system integration. This paper presents a systematic and comprehensive survey of ICS security, integrating design principles, risk models, threat taxonomies, and mitigation strategies. Using a Structured Literature Review (SLR) process inspired by the PRISMA guidelines, more than 180 studies (published between 2010 and 2025) were screened from reputable indexes and databases. The selected literature was synthesized to construct a multi-layered taxonomy linking architecture-level vulnerabilities, attack methodologies, and defensive frameworks.
Risk assessment frameworks were analyzed through standardized models such as NIST 800-82, IEC 62443, and the Cyber PHA methodology to ensure methodological rigor and comparability. Advanced paradigms such as Zero Trust Architecture (ZTA) and anomaly-based Intrusion Detection Systems (IDS) are discussed through a comparative synthesis of reported results in ICS/OT deployments, highlighting observed detection performance and operational trade-offs. This transparent, structured, and evidence-based review provides a coherent framework for enhancing ICS resilience in converged IT/OT environments. The findings provide researchers with a structured roadmap for innovation, practitioners with validated guidance for securing deployments, and policymakers with an evidence base for developing resilient critical-infrastructure standards.
{"title":"Analyzing ICS security: A survey of design principles, risks, threats, and mitigation methods","authors":"Anil Saini, Kewal Krishan, Manoj Singh Gaur","doi":"10.1016/j.compeleceng.2026.110967","DOIUrl":"10.1016/j.compeleceng.2026.110967","url":null,"abstract":"<div><div>The increasing convergence of Operational Technology (OT) and Information Technology (IT) has fundamentally transformed modern industrial environments. This shift has driven advancements in automation, data-driven decision-making, and system integration. This paper presents a systematic and comprehensive survey of ICS security, integrating design principles, risk models, threat taxonomies, and mitigation strategies. Using a Structured Literature Review (SLR) process inspired by the PRISMA guidelines, more than 180 studies (published between 2010 and 2025) were screened from reputable indexes and databases. The selected literature was synthesized to construct a multi-layered taxonomy linking architecture-level vulnerabilities, attack methodologies, and defensive frameworks.</div><div>Risk assessment frameworks were analyzed through standardized models such as NIST 800-82, IEC 62443, and the Cyber PHA methodology to ensure methodological rigor and comparability. Advanced paradigms such as Zero Trust Architecture (ZTA) and anomaly-based Intrusion Detection Systems (IDS) are discussed through a comparative synthesis of reported results in ICS/OT deployments, highlighting observed detection performance and operational trade-offs. This transparent, structured, and evidence-based review provides a coherent framework for enhancing ICS resilience in converged IT/OT environments. The findings provide researchers with a structured roadmap for innovation, practitioners with validated guidance for securing deployments, and policymakers with an evidence base for developing resilient critical-infrastructure standards.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"132 ","pages":"Article 110967"},"PeriodicalIF":4.9,"publicationDate":"2026-01-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146026040","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}
Pub Date : 2026-01-21DOI: 10.1016/j.compeleceng.2026.110934
Pascal A. Schirmer, Daniel Glose
Power electronic converters play a fundamental role in society’s electrification efforts, as they enable enhanced energy efficiency and contribute significantly to reducing the global carbon footprint. These converters are essential components across various sectors, including transportation, industrial automation, renewable energy generation, and power distribution systems. The advancement and optimization of highly efficient power converters directly impact the performance, reliability, and sustainability of these applications. To achieve optimal designs, it is critical to evaluate multiple factors early in the development process, such as waveform quality, electrical behavior, and thermal management. This article introduces PyPowerSim, an open-source Python library designed to streamline the early-phase evaluation of power electronic converter designs. PyPowerSim provides tools for the efficient assessment of modulator performance as well as both steady-state and transient load conditions, thereby facilitating the cost-effective selection of components and design parameters. Moreover, the library includes an interface for configuring switching devices using detailed manufacturer datasheet parameters, enabling accurate modeling of device behavior under various operating conditions. Extensive validation against commercial solvers, such as PLECS, demonstrates that PyPowerSim achieves a relative error margin ranging from 0.1% to 6.8%, confirming its reliability and suitability for early design stages.
{"title":"PyPowerSim: A Python toolkit for analysis of waveform distortions, power losses, and self-heating of standard converter topologies","authors":"Pascal A. Schirmer, Daniel Glose","doi":"10.1016/j.compeleceng.2026.110934","DOIUrl":"10.1016/j.compeleceng.2026.110934","url":null,"abstract":"<div><div>Power electronic converters play a fundamental role in society’s electrification efforts, as they enable enhanced energy efficiency and contribute significantly to reducing the global carbon footprint. These converters are essential components across various sectors, including transportation, industrial automation, renewable energy generation, and power distribution systems. The advancement and optimization of highly efficient power converters directly impact the performance, reliability, and sustainability of these applications. To achieve optimal designs, it is critical to evaluate multiple factors early in the development process, such as waveform quality, electrical behavior, and thermal management. This article introduces <span>PyPowerSim</span>, an open-source Python library designed to streamline the early-phase evaluation of power electronic converter designs. <span>PyPowerSim</span> provides tools for the efficient assessment of modulator performance as well as both steady-state and transient load conditions, thereby facilitating the cost-effective selection of components and design parameters. Moreover, the library includes an interface for configuring switching devices using detailed manufacturer datasheet parameters, enabling accurate modeling of device behavior under various operating conditions. Extensive validation against commercial solvers, such as PLECS, demonstrates that <span>PyPowerSim</span> achieves a relative error margin ranging from 0.1% to 6.8%, confirming its reliability and suitability for early design stages.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"132 ","pages":"Article 110934"},"PeriodicalIF":4.9,"publicationDate":"2026-01-21","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146001854","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}
Pub Date : 2026-01-20DOI: 10.1016/j.compeleceng.2026.110988
Jingde Jia , Gang Liu , Yifan Li , Rujian Chen , Yisheng Cao , Gang Xiao , Jianchao Tang
The stochastic and intermittent nature of solar energy poses major challenges for photovoltaic (PV) power forecasting. To address this, we propose a Dynamic Weather-Based Forecasting framework (DWBF) that integrates feature principal component analysis (FPCA) with an adaptive encoder–decoder structure. FPCA is employed to reduce dimensionality while preserving key meteorological information. A convolutional neural network (CNN) with a multi-attention mechanism serves as a shared encoder, capturing global dependencies across weather conditions. Based on solar radiation thresholds, input data is classified into sunny, cloudy, and rainy categories, and the model dynamically selects appropriate decoders: a long short-term memory (LSTM) decoder for sunny days to model stable temporal patterns; a transformer decoder for cloudy days to handle nonlinear variations; and a temporal convolutional network (TCN) decoder for rainy days to process sparse and noisy data. Additionally, Gaussian noise smoothing and adaptive interpolation enhance robustness under data-sparse conditions. Experimental results show that the proposed DWBF consistently outperforms traditional single architecture models across multiple metrics, including root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and coefficient of determination (). Overall, DWBF offers a flexible, accurate, and efficient solution for PV power forecasting by combining feature selection, weather-adaptive decoding, and targeted optimization.
{"title":"Photovoltaic power forecasting under dynamic weather conditions: An adaptive encoder–decoder framework with feature dimensionality optimization","authors":"Jingde Jia , Gang Liu , Yifan Li , Rujian Chen , Yisheng Cao , Gang Xiao , Jianchao Tang","doi":"10.1016/j.compeleceng.2026.110988","DOIUrl":"10.1016/j.compeleceng.2026.110988","url":null,"abstract":"<div><div>The stochastic and intermittent nature of solar energy poses major challenges for photovoltaic (PV) power forecasting. To address this, we propose a Dynamic Weather-Based Forecasting framework (DWBF) that integrates feature principal component analysis (FPCA) with an adaptive encoder–decoder structure. FPCA is employed to reduce dimensionality while preserving key meteorological information. A convolutional neural network (CNN) with a multi-attention mechanism serves as a shared encoder, capturing global dependencies across weather conditions. Based on solar radiation thresholds, input data is classified into sunny, cloudy, and rainy categories, and the model dynamically selects appropriate decoders: a long short-term memory (LSTM) decoder for sunny days to model stable temporal patterns; a transformer decoder for cloudy days to handle nonlinear variations; and a temporal convolutional network (TCN) decoder for rainy days to process sparse and noisy data. Additionally, Gaussian noise smoothing and adaptive interpolation enhance robustness under data-sparse conditions. Experimental results show that the proposed DWBF consistently outperforms traditional single architecture models across multiple metrics, including root mean square error (RMSE), mean absolute error (MAE), mean absolute percentage error (MAPE), and coefficient of determination (<span><math><msup><mrow><mi>R</mi></mrow><mrow><mn>2</mn></mrow></msup></math></span>). Overall, DWBF offers a flexible, accurate, and efficient solution for PV power forecasting by combining feature selection, weather-adaptive decoding, and targeted optimization.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"132 ","pages":"Article 110988"},"PeriodicalIF":4.9,"publicationDate":"2026-01-20","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"146001853","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}
Pub Date : 2026-01-17DOI: 10.1016/j.compeleceng.2026.110943
Ming-An Chung , Ting-Lan Lin , Ding-Yuan Chen , Bang-Hao Liu , Kun-Hu Jiang , Yangming Wen , Mohammad Shahid
The image sensors capture image signals in a color filter array (CFA) format. After demosaicking and RGB-to-YUV conversion, YUV 420 subsampling is performed for image/video compression. In recent work, YUV 420 subsampling is considered in either of two schemes: subsampling the chrominance while keeping the luminance values the same, or finding optimal luminance values given subsampled chrominance values. In this paper, we extended prior work by reducing the search space to a few Y candidates by observing multiple intervals in the pixel distortion curve, and by developing more flexible, structured cost functions to enable further optimization of the recovered pixels. The closed-form solution still requires a parameter set for each pixel location. Therefore, several methods for reducing complexity are proposed. In comparison to previous methods evaluated on two benchmark datasets, IMAX and SCI, our approach consistently improves image quality (measured in dB) while incurring only minimal increases in computation time (in seconds). Specifically, for the SCI dataset, relative to the Unoptimized Luminance method, we achieve an average CPSNR increase of 3.69 to 7.15 dB, accompanied by an increase in computation time of 12.35 to 13.63 s. In contrast, the Optimized Luminance method yields an average CPSNR improvement of 2.84 to 5.67 dB, with a lower computation time of 0.24 to 3.94 s. For the IMAX dataset, when compared to the unoptimized Luminance method, we note an average CPSNR enhancement of 1.66 to 4.58 dB, with a corresponding rise in computation time of 7.00 to 8.71 s. Meanwhile, the Optimized Luminance method results in an average CPSNR increase of 0.4 to 3.73 dB, with a modest computation time increase of 2.07 to 2.86 s.
{"title":"Optimization of subsampled chrominance and luminance for color image signals","authors":"Ming-An Chung , Ting-Lan Lin , Ding-Yuan Chen , Bang-Hao Liu , Kun-Hu Jiang , Yangming Wen , Mohammad Shahid","doi":"10.1016/j.compeleceng.2026.110943","DOIUrl":"10.1016/j.compeleceng.2026.110943","url":null,"abstract":"<div><div>The image sensors capture image signals in a color filter array (CFA) format. After demosaicking and RGB-to-YUV conversion, YUV 420 subsampling is performed for image/video compression. In recent work, YUV 420 subsampling is considered in either of two schemes: subsampling the chrominance while keeping the luminance values the same, or finding optimal luminance values given subsampled chrominance values. In this paper, we extended prior work by reducing the search space to a few Y candidates by observing multiple intervals in the pixel distortion curve, and by developing more flexible, structured cost functions to enable further optimization of the recovered pixels. The closed-form solution still requires a parameter set for each pixel location. Therefore, several methods for reducing complexity are proposed. In comparison to previous methods evaluated on two benchmark datasets, IMAX and SCI, our approach consistently improves image quality (measured in dB) while incurring only minimal increases in computation time (in seconds). Specifically, for the SCI dataset, relative to the Unoptimized Luminance method, we achieve an average CPSNR increase of 3.69 to 7.15 dB, accompanied by an increase in computation time of 12.35 to 13.63 s. In contrast, the Optimized Luminance method yields an average CPSNR improvement of 2.84 to 5.67 dB, with a lower computation time of 0.24 to 3.94 s. For the IMAX dataset, when compared to the unoptimized Luminance method, we note an average CPSNR enhancement of 1.66 to 4.58 dB, with a corresponding rise in computation time of 7.00 to 8.71 s. Meanwhile, the Optimized Luminance method results in an average CPSNR increase of 0.4 to 3.73 dB, with a modest computation time increase of 2.07 to 2.86 s.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"131 ","pages":"Article 110943"},"PeriodicalIF":4.9,"publicationDate":"2026-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978230","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}
Pub Date : 2026-01-17DOI: 10.1016/j.compeleceng.2026.110977
Renjin , Liyunhe , Gongshenggao , Biantao
A microgrid is an advanced infrastructure that offers increased sustainability, dependability, and local energy autonomy by incorporating renewable and hybrid energy sources into the utility system. However, uncertainties arising from the intermittent nature of renewable sources, fluctuating loads, and dynamic electricity market prices present significant challenges for efficient operation. Traditional heuristic-based energy management systems (EMS) rely on forecasted data but often lack precision and adaptability under real-world variability. To address these limitations, this research proposes a novel Fuzzy Logic Controller-based EMS (FLC-EMS) for optimizing microgrid performance. Unlike rigid rule-based or computationally intensive linear programming (LP) methods, the proposed FLC-EMS combines intelligent decision-making with responsiveness and cost-effectiveness. Simulation results demonstrate that the FLC-EMS outperforms both heuristic and LP-based EMS strategies. Specifically, it achieves cost savings of approximately 8.1% on clear days and 16.6% on cloudy days compared to heuristic methods, while offering additional savings of 1.6–5.5% over LP-based optimization. Furthermore, FLC-EMS reduces grid energy usage and effectively manages state-of-charge (SoC) variations, resulting in enhanced utilization of renewable resources and lower reliance on grid power. The integrated microgrid model and EMS framework developed in this study serve as a robust platform for smart grid applications, offering scalability, real-time adaptability, and improved consumer economics. This work positions the FLC-EMS as a promising candidate for advanced microgrid control, paving the way for resilient and intelligent next-generation power systems.
{"title":"A hybrid fuzzy logic-based energy management strategy for grid-connected photovoltaic microgrids with energy storage optimization","authors":"Renjin , Liyunhe , Gongshenggao , Biantao","doi":"10.1016/j.compeleceng.2026.110977","DOIUrl":"10.1016/j.compeleceng.2026.110977","url":null,"abstract":"<div><div>A microgrid is an advanced infrastructure that offers increased sustainability, dependability, and local energy autonomy by incorporating renewable and hybrid energy sources into the utility system. However, uncertainties arising from the intermittent nature of renewable sources, fluctuating loads, and dynamic electricity market prices present significant challenges for efficient operation. Traditional heuristic-based energy management systems (EMS) rely on forecasted data but often lack precision and adaptability under real-world variability. To address these limitations, this research proposes a novel Fuzzy Logic Controller-based EMS (FLC-EMS) for optimizing microgrid performance. Unlike rigid rule-based or computationally intensive linear programming (LP) methods, the proposed FLC-EMS combines intelligent decision-making with responsiveness and cost-effectiveness. Simulation results demonstrate that the FLC-EMS outperforms both heuristic and LP-based EMS strategies. Specifically, it achieves cost savings of approximately 8.1% on clear days and 16.6% on cloudy days compared to heuristic methods, while offering additional savings of 1.6–5.5% over LP-based optimization. Furthermore, FLC-EMS reduces grid energy usage and effectively manages state-of-charge (SoC) variations, resulting in enhanced utilization of renewable resources and lower reliance on grid power. The integrated microgrid model and EMS framework developed in this study serve as a robust platform for smart grid applications, offering scalability, real-time adaptability, and improved consumer economics. This work positions the FLC-EMS as a promising candidate for advanced microgrid control, paving the way for resilient and intelligent next-generation power systems.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"131 ","pages":"Article 110977"},"PeriodicalIF":4.9,"publicationDate":"2026-01-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145978231","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}
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