Computers & Electrical Engineering最新文献_第9页

Enhanced fetal electrocardiogram extraction via optimized template subtraction: A simplified approach for improved prenatal monitoring 通过优化模板减法增强胎儿心电图提取：一种改进产前监测的简化方法

IF 4.9 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-12-02 DOI: 10.1016/j.compeleceng.2025.110880

Mehdi Hosseinzadeh , Saqib Ali , Amir Masoud Rahmani , Mohammad Adeli , Aso Darwesh , Thantrira Porntaveetus , Sang-Woong Lee

Fetal electrocardiogram (fECG) estimation is critical for assessing prenatal cardiovascular health; yet existing template-subtraction methods suffer from performance limitations and unnecessary complexity when hybridized with auxiliary techniques. Here, we present a novel, stand-alone template-subtraction system that significantly improves fECG signal extraction without the need for combinatorial approaches. Our method innovatively employs (a) ensemble averaging of low-pass-filtered maternal abdominal signals and their envelope for precise maternal R-peak localization, (b) dynamic R–R interval analysis to mitigate false positives/negatives in both maternal and fetal peak detection, and (c) fECG ensemble averaging to enhance fetal R-peak identification.

Validated on two public datasets, the system achieved near-perfect maternal R-peak detection and substantially improved fetal R-peak detection, matching the performance of complex hybrid systems while reducing computational burden. Additional analysis revealed a strong correlation between the estimated fECG and the direct fetal scalp electrocardiogram, with both detection accuracy and waveform fidelity positively associated with signal quality. Furthermore, the system demonstrated robust fetal heart rate (fHR) estimation on a third dataset. These results underscore the potential of optimized template subtraction as a clinically viable, low-complexity framework for non-invasive prenatal monitoring and fetal cardiac electrophysiology research.

胎儿心电图（fECG）估计是评估产前心血管健康的关键；然而，现有的模板减法方法在与辅助技术混合使用时存在性能限制和不必要的复杂性。在这里，我们提出了一种新的，独立的模板减法系统，该系统显着提高了feg信号的提取，而不需要组合方法。我们的方法创新地采用了(a)低通滤波母体腹部信号及其包络的集合平均来精确定位母体r -峰，(b)动态R-R区间分析来减少母体和胎儿峰检测中的假阳性/阴性，以及(c) fECG集合平均来增强胎儿r -峰识别。在两个公共数据集上验证，该系统实现了近乎完美的母体r-峰检测，并大大提高了胎儿r-峰检测，在减少计算负担的同时达到了复杂混合系统的性能。进一步的分析显示，估计的fECG和直接胎儿头皮心电图之间有很强的相关性，检测精度和波形保真度都与信号质量呈正相关。此外，该系统在第三个数据集上展示了稳健的胎儿心率（fHR）估计。这些结果强调了优化的模板减法作为一种临床可行的、低复杂性的框架，用于无创产前监测和胎儿心脏电生理研究的潜力。

{"title":"Enhanced fetal electrocardiogram extraction via optimized template subtraction: A simplified approach for improved prenatal monitoring","authors":"Mehdi Hosseinzadeh , Saqib Ali , Amir Masoud Rahmani , Mohammad Adeli , Aso Darwesh , Thantrira Porntaveetus , Sang-Woong Lee","doi":"10.1016/j.compeleceng.2025.110880","DOIUrl":"10.1016/j.compeleceng.2025.110880","url":null,"abstract":"<div><div>Fetal electrocardiogram (fECG) estimation is critical for assessing prenatal cardiovascular health; yet existing template-subtraction methods suffer from performance limitations and unnecessary complexity when hybridized with auxiliary techniques. Here, we present a novel, stand-alone template-subtraction system that significantly improves fECG signal extraction without the need for combinatorial approaches. Our method innovatively employs (a) ensemble averaging of low-pass-filtered maternal abdominal signals and their envelope for precise maternal R-peak localization, (b) dynamic R–R interval analysis to mitigate false positives/negatives in both maternal and fetal peak detection, and (c) fECG ensemble averaging to enhance fetal R-peak identification.</div><div>Validated on two public datasets, the system achieved near-perfect maternal R-peak detection and substantially improved fetal R-peak detection, matching the performance of complex hybrid systems while reducing computational burden. Additional analysis revealed a strong correlation between the estimated fECG and the direct fetal scalp electrocardiogram, with both detection accuracy and waveform fidelity positively associated with signal quality. Furthermore, the system demonstrated robust fetal heart rate (fHR) estimation on a third dataset. These results underscore the potential of optimized template subtraction as a clinically viable, low-complexity framework for non-invasive prenatal monitoring and fetal cardiac electrophysiology research.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"130 ","pages":"Article 110880"},"PeriodicalIF":4.9,"publicationDate":"2025-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685498","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}

引用次数: 0

GCN-PF: Graph convolutional network integrated particle filter for vocal melody extraction GCN-PF：用于人声旋律提取的图卷积网络集成粒子滤波

IF 4.9 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-12-02 DOI: 10.1016/j.compeleceng.2025.110883

Xiaozhen Peng , Weiwei Zhang , De Hu , Hao Cheng , Xiaokai Liu

Extracting vocal melody from polyphony music, which is correlated with various music related consumer products, is a challenging task. Existing traditional methods rely on manually engineered features that may not generalize to real-world variations, while deep learning-based methods suffer from weak theoretical foundations and poor interpretability. To address these limitations, a method that integrates Graph Convolutional Network (GCN) into Particle Filter (PF) for vocal melody extraction from complex polyphony music is proposed in this paper. Specifically, the logistic distribution is employed to model pitch transition probability, and a GCN is utilized to model the likelihood function for particle weight allocation. Additionally, a zero-frequency allocation strategy is proposed at the prediction stage to detect the unvoiced frames, and a zero-frequency particle revalidation strategy is introduced at the update stage to reduce the misjudgment of unvoiced frames. The proposed method can infer vocal melody frame by frame based on the GCN integrated particle filter, combining the strong learning ability of neural networks with the theoretical foundation and interpretability of Bayesian theory. Moreover, the GCN-based likelihood function modeling can efficiently derive pitch likelihood with low computational complexity. Experimental results demonstrate that the proposed method obtains average overall accuracy of 89.05% on all test recordings, which is comparable to state-of-the-art (SOTA) method, but its parameter amount is about 24 times fewer, making it suitable for scenarios with limited computation and memory resources.

从与各种音乐相关的消费产品相关的复调音乐中提取人声旋律是一项具有挑战性的任务。现有的传统方法依赖于人工设计的特征，这些特征可能无法推广到现实世界的变化，而基于深度学习的方法则存在理论基础薄弱和可解释性差的问题。为了解决这些局限性，本文提出了一种将图卷积网络（GCN）与粒子滤波（PF）相结合的方法，用于从复杂复调音乐中提取人声旋律。具体而言，采用logistic分布建模pitch transition probability，采用GCN建模粒子权重分配的似然函数。此外，在预测阶段提出了零频率分配策略来检测未发音帧，在更新阶段引入了零频率粒子重验证策略来减少未发音帧的误判。该方法基于GCN集成粒子滤波，将神经网络强大的学习能力与贝叶斯理论的理论基础和可解释性相结合，可以逐帧推断人声旋律。此外，基于gcn的似然函数建模可以有效地推导出基音似然，且计算复杂度较低。实验结果表明，该方法在所有测试记录上的平均总体准确率为89.05%，与最先进的SOTA方法相当，但其参数数量减少了约24倍，适用于计算和内存资源有限的场景。

{"title":"GCN-PF: Graph convolutional network integrated particle filter for vocal melody extraction","authors":"Xiaozhen Peng , Weiwei Zhang , De Hu , Hao Cheng , Xiaokai Liu","doi":"10.1016/j.compeleceng.2025.110883","DOIUrl":"10.1016/j.compeleceng.2025.110883","url":null,"abstract":"<div><div>Extracting vocal melody from polyphony music, which is correlated with various music related consumer products, is a challenging task. Existing traditional methods rely on manually engineered features that may not generalize to real-world variations, while deep learning-based methods suffer from weak theoretical foundations and poor interpretability. To address these limitations, a method that integrates Graph Convolutional Network (GCN) into Particle Filter (PF) for vocal melody extraction from complex polyphony music is proposed in this paper. Specifically, the logistic distribution is employed to model pitch transition probability, and a GCN is utilized to model the likelihood function for particle weight allocation. Additionally, a zero-frequency allocation strategy is proposed at the prediction stage to detect the unvoiced frames, and a zero-frequency particle revalidation strategy is introduced at the update stage to reduce the misjudgment of unvoiced frames. The proposed method can infer vocal melody frame by frame based on the GCN integrated particle filter, combining the strong learning ability of neural networks with the theoretical foundation and interpretability of Bayesian theory. Moreover, the GCN-based likelihood function modeling can efficiently derive pitch likelihood with low computational complexity. Experimental results demonstrate that the proposed method obtains average overall accuracy of 89.05% on all test recordings, which is comparable to state-of-the-art (SOTA) method, but its parameter amount is about 24 times fewer, making it suitable for scenarios with limited computation and memory resources.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"130 ","pages":"Article 110883"},"PeriodicalIF":4.9,"publicationDate":"2025-12-02","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145684732","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}

引用次数: 0

A 6X–3X voltage gain thirteen level switched capacitor based inverter with reduced switches count for low voltage sources 一个6X-3X电压增益十三电平开关电容为基础的逆变器与减少开关计数的低压源

IF 4.9 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-12-01 DOI: 10.1016/j.compeleceng.2025.110864

Ahmed R. Hasouna, Sabry A. Mahmoud, Awad E. El-Sabbe, Dina S.M. Osheba

This article presents single source Switched Capacitor Multilevel Inverter (SC-MLI) capable of generating 13 voltage levels with either sixfold or threefold voltage gain. The voltage gain is selected based on the adopted switching pattern without any need to change the MLI connections. It utilizes 9 unidirectional switches, one bidirectional switch, three diodes, and three capacitors charged using a charging inductor. Owing to its inherent feature of voltage boosting, the MLI is suitable for low voltage DC sources such as fuel cell and Photovoltaic (PV) applications. The simple phase disposition sinusoidal pulse width modulation (PD-SPWM) strategy is adopted to achieve voltage balance of the utilized capacitors. The topology power losses and efficiency are investigated at different power levels using PSIM. The topology achieves 92.5% efficiency at

450 W

output power. The simulation is conducted using PLECS and a laboratory prototype is built to validate the simulation results. The proposed inverter performance is assessed under multiple loading conditions, modulation indices, and step-change situations. The proposed topology shows an overall advantage in terms of the cost function and hardware requirements through a comparative study implemented with similar topologies in literature.

本文介绍了单源开关电容多电平逆变器（SC-MLI），能够产生13个电压电平，具有六倍或三倍的电压增益。在不改变MLI连接的情况下，根据所采用的开关模式选择电压增益。它使用9个单向开关，一个双向开关，3个二极管和3个使用充电电感充电的电容器。由于其固有的升压特性，MLI适用于低压直流电源，如燃料电池和光伏（PV）应用。采用简单相位配置正弦脉宽调制（PD-SPWM）策略实现所利用电容器的电压平衡。利用PSIM研究了不同功率水平下的拓扑损耗和效率。该拓扑在450W输出功率下效率可达92.5%。利用PLECS进行了仿真，并建立了实验室样机对仿真结果进行了验证。在多种负载条件、调制指标和阶跃变化情况下对逆变器的性能进行了评估。通过与文献中实现的类似拓扑的比较研究，所提出的拓扑在成本函数和硬件要求方面显示出总体优势。

{"title":"A 6X–3X voltage gain thirteen level switched capacitor based inverter with reduced switches count for low voltage sources","authors":"Ahmed R. Hasouna, Sabry A. Mahmoud, Awad E. El-Sabbe, Dina S.M. Osheba","doi":"10.1016/j.compeleceng.2025.110864","DOIUrl":"10.1016/j.compeleceng.2025.110864","url":null,"abstract":"<div><div>This article presents single source Switched Capacitor Multilevel Inverter (SC-MLI) capable of generating 13 voltage levels with either sixfold or threefold voltage gain. The voltage gain is selected based on the adopted switching pattern without any need to change the MLI connections. It utilizes 9 unidirectional switches, one bidirectional switch, three diodes, and three capacitors charged using a charging inductor. Owing to its inherent feature of voltage boosting, the MLI is suitable for low voltage DC sources such as fuel cell and Photovoltaic (PV) applications. The simple phase disposition sinusoidal pulse width modulation (PD-SPWM) strategy is adopted to achieve voltage balance of the utilized capacitors. The topology power losses and efficiency are investigated at different power levels using PSIM. The topology achieves 92.5% efficiency at <span><math><mrow><mn>450</mn><mspace></mspace><mi>W</mi></mrow></math></span> output power. The simulation is conducted using PLECS and a laboratory prototype is built to validate the simulation results. The proposed inverter performance is assessed under multiple loading conditions, modulation indices, and step-change situations. The proposed topology shows an overall advantage in terms of the cost function and hardware requirements through a comparative study implemented with similar topologies in literature.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"130 ","pages":"Article 110864"},"PeriodicalIF":4.9,"publicationDate":"2025-12-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685501","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}

引用次数: 0

A hybrid lightweight feature extraction assisted ensemble approach for intrusion detection with ESMOTE-based class imbalance handling in IoT networks 物联网网络中基于esmote类不平衡处理的混合轻量级特征提取辅助集成入侵检测方法

IF 4.9 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-11-30 DOI: 10.1016/j.compeleceng.2025.110846

T. Ammannamma, ASN Chakravarthy

Modern network environments need sophisticated techniques for intrusion detection in order to reliably identify and neutralize threats in the face of increasing volumes and complexity of data. The novel intrusion detection framework proposed in this research combines robust machine learning classification, deep learning-based feature extraction, and advanced normalization techniques. The network traffic data is first normalized to reduce the impact of extreme values and ensure consistent data properties. After the pre-processing, the class imbalance issue is solved by using the Enhanced Synthetic Minority Oversampling Technique (ESMOTE) to balance the classes and reduce the overfitting issue. The balanced data is then processed through a hybrid model, the Semi-skipping Layered Gated Recurrent Autoencoder combined with Efficient Network (SLGRAE-ENet) for deep feature extraction. SLGRAE-ENet incorporates a semi-skipping layer within the Gated Recurrent Autoencoder (GRAE) for improving accuracy. The optimal feature set is selected by using Iterative Minimum Redundancy Maximum Relevance (ImRMR) to minimize the training time and reduce the feature dimensionality issue. An ensemble of machine learning classifiers, such as XGBoost, AdaBoost, and Random Forest, is used for the classification stage to enhance detection performance. Experimental results show that the proposed approach attains high accuracies of 99.7 %, 99.94 %, 99.96 %, 99.76 %, 99.75 % and 98.99 % on the Bot-IoT, CIC-DDoS2019, CSE-CIC-IDS2018, NSL-KDD, CIC-IoT2023, and CIC-IoMT 2024 datasets, respectively. Thus, the proposed integrated approach enhances the accuracy, precision, recall, F-measure, and overall efficiency of intrusion detection systems while minimizing the FAR and MSE.

现代网络环境需要复杂的入侵检测技术，以便在面对日益增长的数据量和复杂性时可靠地识别和消除威胁。本研究提出的新型入侵检测框架结合了鲁棒的机器学习分类、基于深度学习的特征提取和先进的归一化技术。首先对网络流量数据进行归一化处理，减少极值的影响，保证数据属性的一致性。预处理后，采用增强合成少数派过采样技术（Enhanced Synthetic Minority Oversampling Technique， ESMOTE）平衡类，减少过拟合问题，解决类不平衡问题。然后通过混合模型处理平衡数据，半跳过分层门控循环自编码器结合高效网络（SLGRAE-ENet）进行深度特征提取。SLGRAE-ENet在门控循环自动编码器（GRAE）内集成了一个半跳频层，以提高精度。采用迭代最小冗余最大相关性（ImRMR）方法选择最优特征集，最大限度地减少训练时间和特征维数问题。机器学习分类器的集成，如XGBoost， AdaBoost和Random Forest，用于分类阶段以提高检测性能。实验结果表明，该方法在Bot-IoT、CIC-DDoS2019、CSE-CIC-IDS2018、NSL-KDD、CIC-IoT2023和CIC-IoMT 2024数据集上的准确率分别达到99.7%、99.94%、99.96%、99.76%、99.75%和98.99%。因此，所提出的集成方法在最小化FAR和MSE的同时，提高了入侵检测系统的准确性、精密度、召回率、f -测度和整体效率。

{"title":"A hybrid lightweight feature extraction assisted ensemble approach for intrusion detection with ESMOTE-based class imbalance handling in IoT networks","authors":"T. Ammannamma, ASN Chakravarthy","doi":"10.1016/j.compeleceng.2025.110846","DOIUrl":"10.1016/j.compeleceng.2025.110846","url":null,"abstract":"<div><div>Modern network environments need sophisticated techniques for intrusion detection in order to reliably identify and neutralize threats in the face of increasing volumes and complexity of data. The novel intrusion detection framework proposed in this research combines robust machine learning classification, deep learning-based feature extraction, and advanced normalization techniques. The network traffic data is first normalized to reduce the impact of extreme values and ensure consistent data properties. After the pre-processing, the class imbalance issue is solved by using the Enhanced Synthetic Minority Oversampling Technique (ESMOTE) to balance the classes and reduce the overfitting issue. The balanced data is then processed through a hybrid model, the Semi-skipping Layered Gated Recurrent Autoencoder combined with Efficient Network (SLGRAE-ENet) for deep feature extraction. SLGRAE-ENet incorporates a semi-skipping layer within the Gated Recurrent Autoencoder (GRAE) for improving accuracy. The optimal feature set is selected by using Iterative Minimum Redundancy Maximum Relevance (ImRMR) to minimize the training time and reduce the feature dimensionality issue. An ensemble of machine learning classifiers, such as XGBoost, AdaBoost, and Random Forest, is used for the classification stage to enhance detection performance. Experimental results show that the proposed approach attains high accuracies of 99.7 %, 99.94 %, 99.96 %, 99.76 %, 99.75 % and 98.99 % on the Bot-IoT, CIC-DDoS2019, CSE-CIC-IDS2018, NSL-KDD, CIC-IoT2023, and CIC-IoMT 2024 datasets, respectively. Thus, the proposed integrated approach enhances the accuracy, precision, recall, F-measure, and overall efficiency of intrusion detection systems while minimizing the FAR and MSE.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"130 ","pages":"Article 110846"},"PeriodicalIF":4.9,"publicationDate":"2025-11-30","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145685500","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}

引用次数: 0

A quadratic voltage boost switched capacitor inverter (QVBSCI) with single voltage source 二次电压升压开关电容逆变器（QVBSCI）与单一电压源

IF 4.9 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-11-29 DOI: 10.1016/j.compeleceng.2025.110856

Vijayakumar Arun , Albert Alexander Stonier , Geno Peter , Samat Iderus

The Switched Capacitor (SC) method is utilized to provide higher voltage levels with stored capacitor energy thus reducing the requirement for a large amount of DC power. With traditional designs, designing large-level inverters is challenging, as it requires a greater number of sources and switches, which increases the system's complexity. In this paper, a Quadratic Voltage Boost Switched Capacitor Inverter (QVBSCI) is proposed which has a single DC source and ten switches and two capacitors to provide a nine-level output. A level shifted phase disposition pulse width modulation (LS-PDPWM) strategy is used to control the proposed topology. The proposed topology provides four times voltage boosting, self-voltage balancing of capacitor and capable of efficiently handling inductive loads. An elegant series-and-parallel connection of SC’s configuration is proposed to obtain the Quadratic Voltage Boost operations. The present research essentially discusses operational concepts and notably the mechanism of conserving capacitor voltage balance. Complete assessments of the losses incurred are appropriated in order to highlight its exceptional performance, through a meticulous comparison with the recent similar topologies. The effectiveness of the proposed QVBSCI is further supported by extensive simulations using MATLAB/Simulink in different test states as well as a laboratory setup.

开关电容（SC）方法被用来提供更高的电压水平与存储电容能量，从而减少了对大量直流电源的需求。对于传统的设计，设计大电平逆变器是具有挑战性的，因为它需要更多的源和开关，这增加了系统的复杂性。本文提出了一种二次电压升压开关电容逆变器（QVBSCI），该逆变器由一个直流电源、十个开关和两个电容提供九电平输出。采用电平移相配置脉宽调制（LS-PDPWM）策略控制所提出的拓扑结构。所提出的拓扑结构提供四倍升压，电容器的自电压平衡，能够有效地处理感性负载。提出了一种优雅的串联和并联SC结构，以获得二次电压升压操作。本研究主要讨论了操作概念，特别是保持电容器电压平衡的机制。通过与最近的类似拓扑进行细致的比较，对所造成的损失进行全面评估，以突出其出色的表现。利用MATLAB/Simulink在不同测试状态和实验室设置下的大量仿真进一步支持了所提出的QVBSCI的有效性。

{"title":"A quadratic voltage boost switched capacitor inverter (QVBSCI) with single voltage source","authors":"Vijayakumar Arun , Albert Alexander Stonier , Geno Peter , Samat Iderus","doi":"10.1016/j.compeleceng.2025.110856","DOIUrl":"10.1016/j.compeleceng.2025.110856","url":null,"abstract":"<div><div>The Switched Capacitor (SC) method is utilized to provide higher voltage levels with stored capacitor energy thus reducing the requirement for a large amount of DC power. With traditional designs, designing large-level inverters is challenging, as it requires a greater number of sources and switches, which increases the system's complexity. In this paper, a Quadratic Voltage Boost Switched Capacitor Inverter (QVBSCI) is proposed which has a single DC source and ten switches and two capacitors to provide a nine-level output. A level shifted phase disposition pulse width modulation (LS-PDPWM) strategy is used to control the proposed topology. The proposed topology provides four times voltage boosting, self-voltage balancing of capacitor and capable of efficiently handling inductive loads. An elegant series-and-parallel connection of SC’s configuration is proposed to obtain the Quadratic Voltage Boost operations. The present research essentially discusses operational concepts and notably the mechanism of conserving capacitor voltage balance. Complete assessments of the losses incurred are appropriated in order to highlight its exceptional performance, through a meticulous comparison with the recent similar topologies. The effectiveness of the proposed QVBSCI is further supported by extensive simulations using MATLAB/Simulink in different test states as well as a laboratory setup.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"130 ","pages":"Article 110856"},"PeriodicalIF":4.9,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145618609","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}

引用次数: 0

Improved link net-based energy prediction: Clustering-based routing and improved kernel-LMS algorithm for data aggregation in WSN 改进的基于链路网络的能量预测：基于聚类的路由和改进的核- lms算法用于WSN中的数据聚合

IF 4.9 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-11-29 DOI: 10.1016/j.compeleceng.2025.110855

Vikram Sadashiv Gawali, Milind Pande, Munir Sayyad, Raghunath S. Bhadade

Advancements in Wireless Sensor Networks (WSNs) are critical to enabling smart computing applications. These networks are composed of compact, battery-powered sensor nodes that operate under strict energy and resource constraints. Such limitations often lead to uneven energy consumption, adversely affecting network lifespan. Although clustering techniques help reduce energy usage, conventional methods frequently fail to ensure balanced energy distribution among nodes. Hence, this research presents the Secretary Bird Merged Walrus Optimization (SBMWO) method for effective cluster head (CH) selection, taking into account a number of constraints such link lifetime, risk, delay, energy prediction, and distance in order to handle these issues. An Improved LinkNet-based Transmission Model (ILNTM) is employed to predict the residual energy of nodes, aiding in optimal CH selection and enhancing network longevity. Following CH selection, an Improved Kernel-based Least Mean Square Algorithm (IK-LMSA) is used for data aggregation, effectively eliminating redundant data and improving both energy efficiency and network lifetime. Experimental results show that the proposed SBMWO method achieves a competitive mean performance score of 0.962, outperforming several existing methods. These outcomes confirm the effectiveness of the proposed framework in enhancing energy prediction accuracy, data aggregation efficiency, and network stability in WSNs.

无线传感器网络（wsn）的进步对于实现智能计算应用至关重要。这些网络由紧凑的、电池供电的传感器节点组成，在严格的能源和资源限制下运行。这种限制通常会导致能源消耗不均匀，从而对网络寿命产生不利影响。虽然聚类技术有助于减少能量的使用，但传统的方法往往不能确保节点之间的能量均衡分布。因此，本研究提出了秘书鸟合并海象优化（SBMWO）方法来有效地选择簇头（CH），该方法考虑了链路寿命、风险、延迟、能量预测和距离等约束条件来处理这些问题。采用一种改进的基于链路网络的传输模型（ILNTM）来预测节点的剩余能量，有助于优化CH选择并提高网络寿命。在选择CH之后，使用改进的基于核的最小均方算法（IK-LMSA）进行数据聚合，有效地消除了冗余数据，提高了能源效率和网络寿命。实验结果表明，所提出的SBMWO方法的竞争平均性能得分为0.962，优于现有的几种方法。这些结果证实了该框架在提高WSNs能量预测精度、数据聚合效率和网络稳定性方面的有效性。

{"title":"Improved link net-based energy prediction: Clustering-based routing and improved kernel-LMS algorithm for data aggregation in WSN","authors":"Vikram Sadashiv Gawali, Milind Pande, Munir Sayyad, Raghunath S. Bhadade","doi":"10.1016/j.compeleceng.2025.110855","DOIUrl":"10.1016/j.compeleceng.2025.110855","url":null,"abstract":"<div><div>Advancements in Wireless Sensor Networks (WSNs) are critical to enabling smart computing applications. These networks are composed of compact, battery-powered sensor nodes that operate under strict energy and resource constraints. Such limitations often lead to uneven energy consumption, adversely affecting network lifespan. Although clustering techniques help reduce energy usage, conventional methods frequently fail to ensure balanced energy distribution among nodes. Hence, this research presents the Secretary Bird Merged Walrus Optimization (SBMWO) method for effective cluster head (CH) selection, taking into account a number of constraints such link lifetime, risk, delay, energy prediction, and distance in order to handle these issues. An Improved LinkNet-based Transmission Model (ILNTM) is employed to predict the residual energy of nodes, aiding in optimal CH selection and enhancing network longevity. Following CH selection, an Improved Kernel-based Least Mean Square Algorithm (IK-LMSA) is used for data aggregation, effectively eliminating redundant data and improving both energy efficiency and network lifetime. Experimental results show that the proposed SBMWO method achieves a competitive mean performance score of 0.962, outperforming several existing methods. These outcomes confirm the effectiveness of the proposed framework in enhancing energy prediction accuracy, data aggregation efficiency, and network stability in WSNs.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"130 ","pages":"Article 110855"},"PeriodicalIF":4.9,"publicationDate":"2025-11-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145618608","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}

引用次数: 0

A hybrid deep learning model for adversarially resilient Internet traffic prediction 对抗弹性互联网流量预测的混合深度学习模型

IF 4.9 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-11-28 DOI: 10.1016/j.compeleceng.2025.110832

Sajal Saha , Saikat Das , Glaucio H.S. Carvalho

Accurate time series forecasting is crucial for Internet traffic telemetry, particularly in security-sensitive applications where resilience against adversarial perturbations is critical. This paper introduces ConvLSTMTransNet, a novel hybrid deep learning model that integrates Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and Transformer encoders to effectively capture spatial–temporal dependencies in network traffic data. Beyond its architectural novelty, a key contribution of this work is its emphasis on adversarial robustness—a dimension often overlooked in existing forecasting models. ConvLSTMTransNet is evaluated under both benign and adversarial conditions using the Fast Gradient Sign Method (FGSM) to simulate adversarial attacks. Results on real-world Internet traffic data demonstrate that our model not only outperforms baseline models (RNN, LSTM, GRU) in terms of Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Weighted Absolute Percentage Error (WAPE), but also exhibits superior resilience to adversarial perturbations. These findings underscore the model’s practicality for deployment in critical network environments where both forecasting accuracy and robustness are imperative.

准确的时间序列预测对于互联网流量遥测至关重要，特别是在安全敏感的应用中，对抗对抗性扰动的弹性至关重要。本文介绍了一种新型的混合深度学习模型ConvLSTMTransNet，该模型集成了卷积神经网络（cnn）、长短期记忆（LSTM）网络和Transformer编码器，以有效捕获网络流量数据中的时空依赖性。除了其架构的新颖性之外，这项工作的一个关键贡献是它对对抗鲁棒性的强调——一个在现有预测模型中经常被忽视的维度。利用快速梯度符号法（FGSM）模拟对抗性攻击，对良性和敌对条件下的ConvLSTMTransNet进行了评估。真实互联网流量数据的结果表明，我们的模型不仅在平均绝对误差（MAE）、均方根误差（RMSE）和加权绝对百分比误差（WAPE）方面优于基准模型（RNN、LSTM、GRU），而且对对抗性扰动也表现出优越的弹性。这些发现强调了该模型在关键网络环境中部署的实用性，在这些环境中，预测准确性和稳健性都是必不可少的。

{"title":"A hybrid deep learning model for adversarially resilient Internet traffic prediction","authors":"Sajal Saha , Saikat Das , Glaucio H.S. Carvalho","doi":"10.1016/j.compeleceng.2025.110832","DOIUrl":"10.1016/j.compeleceng.2025.110832","url":null,"abstract":"<div><div>Accurate time series forecasting is crucial for Internet traffic telemetry, particularly in security-sensitive applications where resilience against adversarial perturbations is critical. This paper introduces ConvLSTMTransNet, a novel hybrid deep learning model that integrates Convolutional Neural Networks (CNNs), Long Short-Term Memory (LSTM) networks, and Transformer encoders to effectively capture spatial–temporal dependencies in network traffic data. Beyond its architectural novelty, a key contribution of this work is its emphasis on adversarial robustness—a dimension often overlooked in existing forecasting models. ConvLSTMTransNet is evaluated under both benign and adversarial conditions using the Fast Gradient Sign Method (FGSM) to simulate adversarial attacks. Results on real-world Internet traffic data demonstrate that our model not only outperforms baseline models (RNN, LSTM, GRU) in terms of Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Weighted Absolute Percentage Error (WAPE), but also exhibits superior resilience to adversarial perturbations. These findings underscore the model’s practicality for deployment in critical network environments where both forecasting accuracy and robustness are imperative.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"130 ","pages":"Article 110832"},"PeriodicalIF":4.9,"publicationDate":"2025-11-28","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145618606","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}

引用次数: 0

A lightweight hyperchaotic memristor-based encryption for secure power and data synchronous transmission in DC microgrids 基于轻量级超混沌忆阻器的直流微电网安全电力和数据同步传输加密

IF 4.9 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-11-25 DOI: 10.1016/j.compeleceng.2025.110857

Chiemeka L. Maxwell , Dongsheng Yu , Yang Leng

This paper presents a lightweight encryption strategy for secure communication in power and data synchronous transmission (PDST) systems, leveraging a stochastic memristor-based two-dimensional (2D) chaotic map. While PDST frameworks have advanced in modulation design and converter integration, the security of the transmitted data remains an underexplored challenge - particularly for embedded, real-time applications. To address this, we propose a coupled memristor-tent map system that generates high-entropy pseudorandom sequences through bidirectional nonlinear interactions, forming the basis of a robust key generation scheme. The resulting hyperchaotic system exhibits strong sensitivity to initial conditions and supports cryptographic operations on the DC bus of switched-mode power supplies (SMPSs). A differential quadrature phase shift keying (DQPSK) scheme is adopted for communication. We implement the encryption framework in MATLAB/Simulink and validate its performance through entropy analysis, bit error rate (BER) analysis, and full NIST randomness evaluation. A hardware-in-the-loop (HIL) setup is used to demonstrate the PDST system in real-time. Results confirm that the proposed method achieves high randomness, and strong resistance to algebraic and differential attacks, making it well-suited for secure and scalable deployment in industrial PDST applications.

本文提出了一种用于电力和数据同步传输（PDST）系统中安全通信的轻量级加密策略，利用基于随机忆阻器的二维（2D）混沌映射。虽然PDST框架在调制设计和转换器集成方面取得了进步，但传输数据的安全性仍然是一个未被充分探索的挑战，特别是对于嵌入式实时应用。为了解决这个问题，我们提出了一个耦合的忆阻器-tent映射系统，该系统通过双向非线性相互作用生成高熵伪随机序列，形成了鲁棒密钥生成方案的基础。所得到的超混沌系统对初始条件具有很强的敏感性，并支持开关电源（smps）直流总线上的加密操作。通信采用差分正交相移键控（DQPSK）方案。我们在MATLAB/Simulink中实现了加密框架，并通过熵分析、误码率（BER）分析和全NIST随机性评估验证了其性能。硬件在环（HIL）设置用于实时演示PDST系统。结果表明，该方法具有较高的随机性，并且具有较强的抗代数和差分攻击能力，非常适合工业PDST应用的安全可扩展部署。

{"title":"A lightweight hyperchaotic memristor-based encryption for secure power and data synchronous transmission in DC microgrids","authors":"Chiemeka L. Maxwell , Dongsheng Yu , Yang Leng","doi":"10.1016/j.compeleceng.2025.110857","DOIUrl":"10.1016/j.compeleceng.2025.110857","url":null,"abstract":"<div><div>This paper presents a lightweight encryption strategy for secure communication in power and data synchronous transmission (PDST) systems, leveraging a stochastic memristor-based two-dimensional (2D) chaotic map. While PDST frameworks have advanced in modulation design and converter integration, the security of the transmitted data remains an underexplored challenge - particularly for embedded, real-time applications. To address this, we propose a coupled memristor-tent map system that generates high-entropy pseudorandom sequences through bidirectional nonlinear interactions, forming the basis of a robust key generation scheme. The resulting hyperchaotic system exhibits strong sensitivity to initial conditions and supports cryptographic operations on the DC bus of switched-mode power supplies (SMPSs). A differential quadrature phase shift keying (DQPSK) scheme is adopted for communication. We implement the encryption framework in MATLAB/Simulink and validate its performance through entropy analysis, bit error rate (BER) analysis, and full NIST randomness evaluation. A hardware-in-the-loop (HIL) setup is used to demonstrate the PDST system in real-time. Results confirm that the proposed method achieves high randomness, and strong resistance to algebraic and differential attacks, making it well-suited for secure and scalable deployment in industrial PDST applications.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"130 ","pages":"Article 110857"},"PeriodicalIF":4.9,"publicationDate":"2025-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145618607","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}

引用次数: 0

Photovoltaic power prediction via APSformer combined with secondary decomposition and optimization algorithms 结合二次分解和优化算法的aptransformer光伏功率预测

IF 4.9 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-11-25 DOI: 10.1016/j.compeleceng.2025.110862

Lingfeng Li , Haipeng Liu , Xiongfeng Zhao , Weihao Ren , Yuanyuan Guo , Jiwen Yu

Accurate Photovoltaic (PV) power forecasting is essential for grid stability and optimal dispatching, yet remains challenging due to the inherent intermittency of solar irradiance. To address this, the study first constructs a secondary decomposition strategy called CFVMD, which integrates Complementary Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Fuzzy Entropy (FE), and Variational Mode Decomposition (VMD). By integrating CFVMD with the rime optimization algorithm (RIME) and an improved APSformer, a novel hybrid model designated as CFVMD-RIME-APSformer is proposed. Specifically, the APSformer component employs an Adaptive ProbSparse Self-Attention mechanism to address the limitations of the standard Informer by dynamically allocating query vectors according to sparsity metric distributions, thereby reducing computational redundancy. The model was validated using a four-month dataset from Alice Springs, Australia. Experimental results demonstrate significant improvements over state-of-the-art benchmarks. Specifically, averaged across all seasons, the proposed model reduced the Root Mean Square Error (RMSE) by 27.92%, Mean Absolute Error (MAE) by 14.44%, and the symmetric Mean Absolute Percentage Error (SMAPE) by 19.50% compared to the original Informer. These findings substantiate the model’s superior accuracy and robustness across diverse seasonal conditions.

准确的光伏发电功率预测对电网稳定和优化调度至关重要，但由于太阳辐照度固有的间歇性，这一预测仍然具有挑战性。为了解决这个问题，本研究首先构建了一种称为CFVMD的二次分解策略，该策略集成了自适应噪声（CEEMDAN）、模糊熵（FE）和变分模态分解（VMD）的互补集成经验模态分解（CFVMD）。通过将CFVMD与时间优化算法（rime）和改进的APSformer相结合，提出了一种新的混合模型CFVMD- rime -APSformer。具体来说，APSformer组件采用自适应ProbSparse自关注机制，通过根据稀疏度度量分布动态分配查询向量来解决标准Informer的局限性，从而减少了计算冗余。该模型使用澳大利亚爱丽斯泉四个月的数据集进行了验证。实验结果表明，与最先进的基准相比，有了显著的改进。具体而言，与原始Informer模型相比，该模型在所有季节的平均误差（RMSE）降低了27.92%，平均绝对误差（MAE）降低了14.44%，对称平均绝对百分比误差（SMAPE）降低了19.50%。这些发现证实了该模型在不同季节条件下的优越准确性和稳健性。

{"title":"Photovoltaic power prediction via APSformer combined with secondary decomposition and optimization algorithms","authors":"Lingfeng Li , Haipeng Liu , Xiongfeng Zhao , Weihao Ren , Yuanyuan Guo , Jiwen Yu","doi":"10.1016/j.compeleceng.2025.110862","DOIUrl":"10.1016/j.compeleceng.2025.110862","url":null,"abstract":"<div><div>Accurate Photovoltaic (PV) power forecasting is essential for grid stability and optimal dispatching, yet remains challenging due to the inherent intermittency of solar irradiance. To address this, the study first constructs a secondary decomposition strategy called CFVMD, which integrates Complementary Ensemble Empirical Mode Decomposition with Adaptive Noise (CEEMDAN), Fuzzy Entropy (FE), and Variational Mode Decomposition (VMD). By integrating CFVMD with the rime optimization algorithm (RIME) and an improved APSformer, a novel hybrid model designated as CFVMD-RIME-APSformer is proposed. Specifically, the APSformer component employs an Adaptive ProbSparse Self-Attention mechanism to address the limitations of the standard Informer by dynamically allocating query vectors according to sparsity metric distributions, thereby reducing computational redundancy. The model was validated using a four-month dataset from Alice Springs, Australia. Experimental results demonstrate significant improvements over state-of-the-art benchmarks. Specifically, averaged across all seasons, the proposed model reduced the Root Mean Square Error (RMSE) by 27.92%, Mean Absolute Error (MAE) by 14.44%, and the symmetric Mean Absolute Percentage Error (SMAPE) by 19.50% compared to the original Informer. These findings substantiate the model’s superior accuracy and robustness across diverse seasonal conditions.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"130 ","pages":"Article 110862"},"PeriodicalIF":4.9,"publicationDate":"2025-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145618603","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}

引用次数: 0

Enhancement Of power quality using UPQC with adaptive fuzzy neural network-bidirectional long short-term memory-based bubble-net motion interactive optimization 基于UPQC的自适应模糊神经网络双向长短期记忆气泡网运动交互优化提高电能质量

IF 4.9 3区计算机科学 Q1 COMPUTER SCIENCE, HARDWARE & ARCHITECTURE

Computers & Electrical Engineering

Pub Date : 2025-11-25 DOI: 10.1016/j.compeleceng.2025.110843

Harshal Vitthalrao Takpire , Mukesh Kumar , Saurabh Sureshrao Jadhao

In the power electronic equipments, dynamic and adjustable solutions are considered to mitigate power quality (PQ) issues, which contribute to reduce voltage fluctuations and improve the system stability performance. However, non-linear loads and unpredictable fault patterns increased load current distortions, which lead the flickers and fluctuations in the power system. Hence, the research work adopts the control strategy of Bubble-net Motion interactive optimization based Adaptive Fuzzy Neural Network-Bidirectional Long Short-Term Memory (BMIO-AFBTM) for the Unified Power Quality Conditioner (UPQC). Moreover, a 5-bus distribution network (DN) is considered to reduce system cost and loss effects. AFBTM model controls the series and shunt converter and provides the compensating signals to contribute to PQ issue mitigation. Furthermore, the BMIO algorithm increases the convergence, sensitivity, and parameter selection ability. In the series converter of the UPQC device, the reference signal generation and series pulse generation regulate the rated load and voltage. Moreover, the source current and DC link voltage are regulated by the shunt control technique in the UPQC. The developed model of BMIO-AFBTM achieves the efficient performance metrics of injected voltage, injected reactive power, load voltage, reactive power, and power values are 0.9757pu, 103.77kvar, 0.9751pu, 279.06kvar, and 204.3 kW, respectively.

在电力电子设备中，采用动态和可调的解决方案来缓解电能质量问题，有助于降低电压波动，提高系统的稳定性。然而，非线性负荷和不可预测的故障模式增加了负荷电流的畸变，从而导致电力系统的闪烁和波动。因此，研究工作采用基于自适应模糊神经网络双向长短期记忆（BMIO-AFBTM）的气泡网运动交互优化控制策略对统一电能质量调节器（UPQC）进行控制。此外，还考虑了5总线配电网（DN），以降低系统成本和损耗效应。AFBTM模型控制串联和并联转换器，并提供补偿信号，有助于缓解PQ问题。此外，BMIO算法提高了收敛性、灵敏度和参数选择能力。在UPQC装置的串联变换器中，参考信号产生和串联脉冲产生调节额定负载和电压。UPQC采用分流控制技术对源电流和直流环节电压进行调节。所开发的BMIO-AFBTM模型实现了注入电压、注入无功功率、负载电压、无功功率和功率值分别为0.9757pu、103.77kvar、0.9751pu、279.06kvar和204.3 kW的高效性能指标。

{"title":"Enhancement Of power quality using UPQC with adaptive fuzzy neural network-bidirectional long short-term memory-based bubble-net motion interactive optimization","authors":"Harshal Vitthalrao Takpire , Mukesh Kumar , Saurabh Sureshrao Jadhao","doi":"10.1016/j.compeleceng.2025.110843","DOIUrl":"10.1016/j.compeleceng.2025.110843","url":null,"abstract":"<div><div>In the power electronic equipments, dynamic and adjustable solutions are considered to mitigate power quality (PQ) issues, which contribute to reduce voltage fluctuations and improve the system stability performance. However, non-linear loads and unpredictable fault patterns increased load current distortions, which lead the flickers and fluctuations in the power system. Hence, the research work adopts the control strategy of Bubble-net Motion interactive optimization based Adaptive Fuzzy Neural Network-Bidirectional Long Short-Term Memory (BMIO-AFBTM) for the Unified Power Quality Conditioner (UPQC). Moreover, a 5-bus distribution network (DN) is considered to reduce system cost and loss effects. AFBTM model controls the series and shunt converter and provides the compensating signals to contribute to PQ issue mitigation. Furthermore, the BMIO algorithm increases the convergence, sensitivity, and parameter selection ability. In the series converter of the UPQC device, the reference signal generation and series pulse generation regulate the rated load and voltage. Moreover, the source current and DC link voltage are regulated by the shunt control technique in the UPQC. The developed model of BMIO-AFBTM achieves the efficient performance metrics of injected voltage, injected reactive power, load voltage, reactive power, and power values are 0.9757pu, 103.77kvar, 0.9751pu, 279.06kvar, and 204.3 kW, respectively.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"130 ","pages":"Article 110843"},"PeriodicalIF":4.9,"publicationDate":"2025-11-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145618604","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}

引用次数: 0