Pub Date : 2025-02-09DOI: 10.1016/j.compeleceng.2025.110083
Ning Wang , Ya Li , Yuanbang Li
In the era of big data and the Energy Internet of Things (eIoT), recommendation systems are crucial for mitigating information overload and enhancing user experience. However, fragmented and heterogeneous data across different platforms and devices hinders accurate and diverse recommendations. To address this limitation, we propose CIF-HGR, a collaborative and privacy-preserving federated heterogeneous graph recommendation framework tailored for eIoT. CIF-HGR integrates three major components: (1) heterogeneous graph construction and adaptive learning to extract high-quality representations from large-scale eIoT data; (2) a consortium blockchain-based collaboration layer offering privacy-preserving identity authentication and secure session management; and (3) a federated heterogeneous graph recommendation scheme that employs differential privacy and contribution-based incentives to ensure fairness, efficiency, and sustainability.
Extensive experiments confirm that our method outperforms recent baselines in both single- and multi-institution eIoT environments. In the single-institution scenario on Amazon, CIF-HGR improves NDCG@10 from 0.2487 (LightGCN) to 0.2639 and Coverage@10 from 0.1034 to 0.1191. Under federated aggregation, our optimized scheme raises Coverage@10 from 0.1647 (FedKD) to 0.1682 and remains competitive with FedMR’s NDCG@10 (0.2940 vs. 0.2936). Moreover, CIF-HGR maintains an attack success rate below 0.85 at , underscoring its efficacy in balancing accuracy, coverage, and privacy.
{"title":"CIF-HGR: A privacy-preserving and collaborative framework for cross-institutional federated heterogeneous graph recommendation in energy IoT","authors":"Ning Wang , Ya Li , Yuanbang Li","doi":"10.1016/j.compeleceng.2025.110083","DOIUrl":"10.1016/j.compeleceng.2025.110083","url":null,"abstract":"<div><div>In the era of big data and the Energy Internet of Things (eIoT), recommendation systems are crucial for mitigating information overload and enhancing user experience. However, fragmented and heterogeneous data across different platforms and devices hinders accurate and diverse recommendations. To address this limitation, we propose CIF-HGR, a collaborative and privacy-preserving federated heterogeneous graph recommendation framework tailored for eIoT. CIF-HGR integrates three major components: (1) heterogeneous graph construction and adaptive learning to extract high-quality representations from large-scale eIoT data; (2) a consortium blockchain-based collaboration layer offering privacy-preserving identity authentication and secure session management; and (3) a federated heterogeneous graph recommendation scheme that employs differential privacy and contribution-based incentives to ensure fairness, efficiency, and sustainability.</div><div>Extensive experiments confirm that our method outperforms recent baselines in both single- and multi-institution eIoT environments. In the single-institution scenario on Amazon, CIF-HGR improves NDCG@10 from 0.2487 (LightGCN) to 0.2639 and Coverage@10 from 0.1034 to 0.1191. Under federated aggregation, our optimized scheme raises Coverage@10 from 0.1647 (FedKD) to 0.1682 and remains competitive with FedMR’s NDCG@10 (0.2940 vs. 0.2936). Moreover, CIF-HGR maintains an attack success rate below 0.85 at <span><math><mrow><mi>ϵ</mi><mo>=</mo><mn>1</mn><mo>.</mo><mn>0</mn></mrow></math></span>, underscoring its efficacy in balancing accuracy, coverage, and privacy.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110083"},"PeriodicalIF":4.0,"publicationDate":"2025-02-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143369830","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 : 2025-02-08DOI: 10.1016/j.compeleceng.2025.110125
Jiming Yang, Feipeng Da, Ru Hong, Zeyu Cai, Shaoyan Gai
Zero-shot domain adaptation is a specialized area within transfer learning focused on achieving domain adaptation without using any samples from the target domain. This is particularly important when target domain samples are difficult to obtain. The rapid development in generative models, particularly diffusion models, has introduced robust tools for zero-shot domain adaptation tasks. This paper proposes an innovative framework to address domain adaptive semantic segmentation under zero-shot conditions. We introduce a Dynamic Control Fusion Module, which autonomously learns the fusion scales and effectively integrates hidden states with image controls, enhancing generation in complex scenarios. Furthermore, we propose a Semantic and Image-Text Consistency Strategy, designed to impose consistency constraints on both the semantic content and the style of generated images, ensuring closer alignment with the target domain. We perform experiments on Cityscapes, ACDC, and GTAV datasets. The results show that our method improves the quality of generated target domain images and semantic segmentation performance, demonstrating its effectiveness in zero-shot domain adaptation tasks. Overall, our method shows consistent improvements over baseline approaches across the five sub-experiments. Overall, our method demonstrates consistent improvements over baseline approaches across most domain adaptation tasks. Specifically, in the tasks involving adaptation to Night and Snow, it achieves 2.6% and 2.3% higher mIoU compared to the baseline, respectively.
{"title":"Zero-shot domain adaptation with enhanced consistency for semantic segmentation","authors":"Jiming Yang, Feipeng Da, Ru Hong, Zeyu Cai, Shaoyan Gai","doi":"10.1016/j.compeleceng.2025.110125","DOIUrl":"10.1016/j.compeleceng.2025.110125","url":null,"abstract":"<div><div>Zero-shot domain adaptation is a specialized area within transfer learning focused on achieving domain adaptation without using any samples from the target domain. This is particularly important when target domain samples are difficult to obtain. The rapid development in generative models, particularly diffusion models, has introduced robust tools for zero-shot domain adaptation tasks. This paper proposes an innovative framework to address domain adaptive semantic segmentation under zero-shot conditions. We introduce a Dynamic Control Fusion Module, which autonomously learns the fusion scales and effectively integrates hidden states with image controls, enhancing generation in complex scenarios. Furthermore, we propose a Semantic and Image-Text Consistency Strategy, designed to impose consistency constraints on both the semantic content and the style of generated images, ensuring closer alignment with the target domain. We perform experiments on Cityscapes, ACDC, and GTAV datasets. The results show that our method improves the quality of generated target domain images and semantic segmentation performance, demonstrating its effectiveness in zero-shot domain adaptation tasks. Overall, our method shows consistent improvements over baseline approaches across the five sub-experiments. Overall, our method demonstrates consistent improvements over baseline approaches across most domain adaptation tasks. Specifically, in the tasks involving adaptation to Night and Snow, it achieves 2.6% and 2.3% higher mIoU compared to the baseline, respectively.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110125"},"PeriodicalIF":4.0,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143350519","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}
The Random Vector Functional Link (RVFL) is a prominent and widely used approach effective in tackling a wide range of challenging problems in various research fields in the case of regression and classification of real-world problems. An Intuitionistic fuzzy RVFL Classifier (IFRVFLC) boosts the overall classification performance of the RVFL network and enhances its classification accuracy on noisy datasets. On the other hand, ResNet50 architecture offers great potential in the field of artificial intelligence and is used for any object recognition task. The major drawback of ResNet50 architecture is that there is no transparency in the middle layers during the classification process, which makes it challenging to examine the training process. So, to eradicate this disadvantage, we have proposed a hybrid ResNet50-IFRVFLC model. which combines the ResNet50 and IFRVFLC models for the classification of plant species through their leaf image which is one of the biggest challenges in computer vision. Firstly, the ResNet50 model obtains the deep features using textures of the leaf images, and then PCA is applied as a feature reduction technique to extract the important features. The extracted features from PCA are taken as an input to IFRVFLC architecture for leaf image classification. The efficacy of the ResNet50-IFRVFLC model is evaluated by comparing its performance to that of the support vector machine (SVM), intuitionistic fuzzy SVM (IFSVM), twin SVM (TSVM), kernel ridge regression (KRR), RVFL, twin RVFL (TRVFL), RVFL with ε-insensitive Huber loss (ε-HRVFL) and 1-norm TRVFL (TRVFL1norm). Furthermore, extensive statistical analysis has been performed to evaluate the significance of the noted performance differences, including Friedman and post-hoc Nemenyi tests. The experimental results are examined in terms of average accuracy, F1-Score, AUC and G-Mean. The ResNet50-IFRVFLC model achieved 91.23 % mean accuracy, outperforming SVM (88.04 %), TSVM (89.55 %), RVFL (88.84 %), TRVFL (90.31 %), IFSVM (89.47 %), KRR (87.10 %), ε-HRVFL (87.65 %) and TRVFL1norm (90.15 %) for leaf datasets. Out of all the existing baseline models implemented in this study, ResNet50-IFRVFLC has attained the highest classification accuracy of 94.860 % and a maximum F1-score of 0.972 on the leaf datasets.
{"title":"A hybrid approach for plant leaf detection using ResNet50- intuitionistic fuzzy RVFL (ResNet50-IFRVFLC) classifier","authors":"Upendra Mishra , Deepak Gupta , Achyuth Sarkar , Barenya Bikash Hazarika","doi":"10.1016/j.compeleceng.2025.110135","DOIUrl":"10.1016/j.compeleceng.2025.110135","url":null,"abstract":"<div><div>The Random Vector Functional Link (RVFL) is a prominent and widely used approach effective in tackling a wide range of challenging problems in various research fields in the case of regression and classification of real-world problems. An Intuitionistic fuzzy RVFL Classifier (IFRVFLC) boosts the overall classification performance of the RVFL network and enhances its classification accuracy on noisy datasets. On the other hand, ResNet50 architecture offers great potential in the field of artificial intelligence and is used for any object recognition task. The major drawback of ResNet50 architecture is that there is no transparency in the middle layers during the classification process, which makes it challenging to examine the training process. So, to eradicate this disadvantage, we have proposed a hybrid ResNet50-IFRVFLC model. which combines the ResNet50 and IFRVFLC models for the classification of plant species through their leaf image which is one of the biggest challenges in computer vision. Firstly, the ResNet50 model obtains the deep features using textures of the leaf images, and then PCA is applied as a feature reduction technique to extract the important features. The extracted features from PCA are taken as an input to IFRVFLC architecture for leaf image classification. The efficacy of the ResNet50-IFRVFLC model is evaluated by comparing its performance to that of the support vector machine (SVM), intuitionistic fuzzy SVM (IFSVM), twin SVM (TSVM), kernel ridge regression (KRR), RVFL, twin RVFL (TRVFL), RVFL with ε-insensitive Huber loss (ε-HRVFL) and 1-norm TRVFL (TRVFL<sub>1norm</sub>). Furthermore, extensive statistical analysis has been performed to evaluate the significance of the noted performance differences, including Friedman and post-hoc Nemenyi tests. The experimental results are examined in terms of average accuracy, F1-Score, AUC and G-Mean. The ResNet50-IFRVFLC model achieved 91.23 % mean accuracy, outperforming SVM (88.04 %), TSVM (89.55 %), RVFL (88.84 %), TRVFL (90.31 %), IFSVM (89.47 %), KRR (87.10 %), ε-HRVFL (87.65 %) and TRVFL<sub>1norm</sub> (90.15 %) for leaf datasets. Out of all the existing baseline models implemented in this study, ResNet50-IFRVFLC has attained the highest classification accuracy of 94.860 % and a maximum F1-score of 0.972 on the leaf datasets.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110135"},"PeriodicalIF":4.0,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143350520","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 : 2025-02-08DOI: 10.1016/j.compeleceng.2025.110134
Tiancheng Wang , Yi Xu , Di Guo , Xi-Ming Sun
With the emergence of deep learning, its technique has been widely used in remaining useful life (RUL) prediction for turbofans. Due to its complex nature, RUL prediction poses significant challenges such as incomplete life data and the labor-intensive process of data labeling. To address the issue, many studies have turned to semi-supervised learning. However, most of these studies have utilized unlabeled data solely from the complete fault history, overlooking the overhang history, which leads to a notable decrease in prediction accuracy. To tackle this problem, this paper proposes a novel methodology that combines contrast learning with variational autoencoders (VAE). Through a symmetric structure, the proposed approach effectively learns the similarity between labeled and unlabeled data, thereby enhancing the prediction accuracy of variational autoencoders. Additionally, the K-nearest neighbor (KNN) regression algorithm is employed to label the unlabeled data, and screening rules are established to eliminate data with poor labeling effects. The effectiveness and stability of the proposed method are rigorously evaluated through numerous comparative experiments.
{"title":"A contrastive semi-supervised remaining useful life prediction method with incomplete life histories on turbofan","authors":"Tiancheng Wang , Yi Xu , Di Guo , Xi-Ming Sun","doi":"10.1016/j.compeleceng.2025.110134","DOIUrl":"10.1016/j.compeleceng.2025.110134","url":null,"abstract":"<div><div>With the emergence of deep learning, its technique has been widely used in remaining useful life (RUL) prediction for turbofans. Due to its complex nature, RUL prediction poses significant challenges such as incomplete life data and the labor-intensive process of data labeling. To address the issue, many studies have turned to semi-supervised learning. However, most of these studies have utilized unlabeled data solely from the complete fault history, overlooking the overhang history, which leads to a notable decrease in prediction accuracy. To tackle this problem, this paper proposes a novel methodology that combines contrast learning with variational autoencoders (VAE). Through a symmetric structure, the proposed approach effectively learns the similarity between labeled and unlabeled data, thereby enhancing the prediction accuracy of variational autoencoders. Additionally, the K-nearest neighbor (KNN) regression algorithm is employed to label the unlabeled data, and screening rules are established to eliminate data with poor labeling effects. The effectiveness and stability of the proposed method are rigorously evaluated through numerous comparative experiments.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110134"},"PeriodicalIF":4.0,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143350517","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 : 2025-02-08DOI: 10.1016/j.compeleceng.2025.110141
Far Chen Jong, Musse Mohamud Ahmed, Wei Kin Lau, Md Abu Sayed
Sarawak's abundant green energy resources make it ideal for energy transition. However, the dispersed and intermittent nature of these sources poses reliability challenges, exacerbated by a lack of comprehensive integration, monitoring, control, and automation strategies. Therefore, this research paper proposes a novel hybrid green energy systems model, operating stably at 15 kV within a ring topology system. To enhance the model's dynamism, unique algorithms have been developed to stream real-time data from the Grid System Operator and Solcast into the simulation. An innovative Industrial Internet of Things (IIoT) communication framework has been established to connect the simulation model with a Supervisory Control and Data Acquisition (SCADA) platform, addressing the intermittency issues associated with green energy. The research demonstrates the successful implementation of effective monitoring, control, and automation strategies, even with dynamic real-time data. To further validate the framework's applicability to real-world scenarios, a hardware model incorporating a Raspberry Pi 4 and IoT components was successfully integrated with the SCADA system. The effectiveness of these strategies was confirmed by the hardware prototype. This novel framework provides a valuable planning tool for researchers to further analyze the model, contributing to the implementation of a more robust and resilient green energy infrastructure.
{"title":"Modeling novel hybrid green energy systems with IIoT-based real-time dynamic monitoring, control and automation","authors":"Far Chen Jong, Musse Mohamud Ahmed, Wei Kin Lau, Md Abu Sayed","doi":"10.1016/j.compeleceng.2025.110141","DOIUrl":"10.1016/j.compeleceng.2025.110141","url":null,"abstract":"<div><div>Sarawak's abundant green energy resources make it ideal for energy transition. However, the dispersed and intermittent nature of these sources poses reliability challenges, exacerbated by a lack of comprehensive integration, monitoring, control, and automation strategies. Therefore, this research paper proposes a novel hybrid green energy systems model, operating stably at 15 kV within a ring topology system. To enhance the model's dynamism, unique algorithms have been developed to stream real-time data from the Grid System Operator and Solcast into the simulation. An innovative Industrial Internet of Things (IIoT) communication framework has been established to connect the simulation model with a Supervisory Control and Data Acquisition (SCADA) platform, addressing the intermittency issues associated with green energy. The research demonstrates the successful implementation of effective monitoring, control, and automation strategies, even with dynamic real-time data. To further validate the framework's applicability to real-world scenarios, a hardware model incorporating a Raspberry Pi 4 and IoT components was successfully integrated with the SCADA system. The effectiveness of these strategies was confirmed by the hardware prototype. This novel framework provides a valuable planning tool for researchers to further analyze the model, contributing to the implementation of a more robust and resilient green energy infrastructure.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110141"},"PeriodicalIF":4.0,"publicationDate":"2025-02-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143350518","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}
Wind resource assessment is crucial for establishing wind farms and prediction of their economic benefits. The one key problem for wind resource assessment is to estimate the probability distribution of wind speed. In this study, we propose a nonparametric generative approach based generative conditional distribution sampler (GCDS) to sample wind speed data at different locations, which is equivalent to estimating wind speed distribution. The proposed approach can used to fit wind speed data and infer the distribution of wind speed at new locations with no observations. The proposed approach reduces the transmission and accumulation of errors caused by traditional interpolation methods. The analysis results show that the proposed method outperforms other models under key metrics, the improvement is generally over 14.7% for distribution fitting and interpolation fitting.
{"title":"Distribution inference of wind speed at adjacent spaces using generative conditional distribution sampler","authors":"Xutao Li , Guoqing Huang , Weiyang Yu , Rui Yin , Haitao Zheng","doi":"10.1016/j.compeleceng.2025.110123","DOIUrl":"10.1016/j.compeleceng.2025.110123","url":null,"abstract":"<div><div>Wind resource assessment is crucial for establishing wind farms and prediction of their economic benefits. The one key problem for wind resource assessment is to estimate the probability distribution of wind speed. In this study, we propose a nonparametric generative approach based generative conditional distribution sampler (GCDS) to sample wind speed data at different locations, which is equivalent to estimating wind speed distribution. The proposed approach can used to fit wind speed data and infer the distribution of wind speed at new locations with no observations. The proposed approach reduces the transmission and accumulation of errors caused by traditional interpolation methods. The analysis results show that the proposed method outperforms other models under key metrics, the improvement is generally over 14.7% for distribution fitting and interpolation fitting.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110123"},"PeriodicalIF":4.0,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143349381","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 : 2025-02-07DOI: 10.1016/j.compeleceng.2025.110148
Manliang Wang, Bingtuan Gao
The extensive grid-connected photovoltaic (PV) systems have given rise to numerous challenges to the stability of power systems. Flexible power point tracking (FPPT) strategy can bring flexibility and frequency support functionality to PV power generation, which is designed to control the output active power to a given active power command value. As many maximum power point tracking methods under partial shading conditions (PSC) are studied, specific technical methods are still needed to realize FPPT under PSC. This paper proposes a mathematical model for determining the power-voltage (P-V) characteristics of PV array under PSC and a novel model-based FPPT method under PSC. Firstly, an analytical model is established to ascertain the features of the multi-peaks P-V curve of PV array under PSC. Secondly, the mathematical model for accurately calculating the voltage value of each trough point on the multi-peaks P-V curve is proposed, without measuring irradiance and temperature. Comparison between actual voltage value and calculated voltage value and sensitivity analysis have validated the high accuracy of the proposed mathematical model. Finally, according to the above analytical model, a skip-compare-locate FPPT method is proposed, which can realize quick and accurate tracking of any active power command value. Simulation results verify the advanced performance of the proposed FPPT method compared with other existing methods in terms of tracking accuracy, tracking speed, and algorithm complexity, especially under complex PSC with multiple peaks. Experimental results further prove the effectiveness and practicality of the proposed FPPT method.
{"title":"Model-based flexible power point tracking method for photovoltaic systems under partial shading conditions","authors":"Manliang Wang, Bingtuan Gao","doi":"10.1016/j.compeleceng.2025.110148","DOIUrl":"10.1016/j.compeleceng.2025.110148","url":null,"abstract":"<div><div>The extensive grid-connected photovoltaic (PV) systems have given rise to numerous challenges to the stability of power systems. Flexible power point tracking (FPPT) strategy can bring flexibility and frequency support functionality to PV power generation, which is designed to control the output active power to a given active power command value. As many maximum power point tracking methods under partial shading conditions (PSC) are studied, specific technical methods are still needed to realize FPPT under PSC. This paper proposes a mathematical model for determining the power-voltage (P-V) characteristics of PV array under PSC and a novel model-based FPPT method under PSC. Firstly, an analytical model is established to ascertain the features of the multi-peaks P-V curve of PV array under PSC. Secondly, the mathematical model for accurately calculating the voltage value of each trough point on the multi-peaks P-V curve is proposed, without measuring irradiance and temperature. Comparison between actual voltage value and calculated voltage value and sensitivity analysis have validated the high accuracy of the proposed mathematical model. Finally, according to the above analytical model, a skip-compare-locate FPPT method is proposed, which can realize quick and accurate tracking of any active power command value. Simulation results verify the advanced performance of the proposed FPPT method compared with other existing methods in terms of tracking accuracy, tracking speed, and algorithm complexity, especially under complex PSC with multiple peaks. Experimental results further prove the effectiveness and practicality of the proposed FPPT method.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110148"},"PeriodicalIF":4.0,"publicationDate":"2025-02-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143349380","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}
This paper proposes a current optimization model predictive control with common-mode voltage (CMV) reduction (COMPC-CMVR) for three-level T-type inverters to suppress CMV and improve current quality without increasing the losses. The CMV is restricted within udc/6 by excluding voltage vectors (VVs) with high CMV. However, the reduction in VVs reduces the current quality. The neutral point (NP) voltage optimization interval is proposed in the COMPC-CMVR to improve the current control performance, where the grid current is the only control objective when the NP voltage is within the voltage optimization interval. The small and medium VVs are divided into P-type and N-type VVs to balance the NP voltage without weighting factors. On this basis, two novel candidate VVs sets are proposed. The COMPC-CMVR considers only four to seven feasible VVs in each control cycle, which reduces the computational burden. Finally, simulation and experimental results show that COMPC-CMVR performs well in terms of steady-state and transient responses. The COMPC-CMVR can effectively suppress the CMV and improve current quality without increasing the losses.
{"title":"A current optimization model predictive control with common-mode voltage reduction for three-level T-type inverters","authors":"Zhikang Guo, Zhaoxun Li, Weifeng Zhang, Yizhan Jiang, Yu Tian, Xiang Wu, Guojun Tan","doi":"10.1016/j.compeleceng.2025.110151","DOIUrl":"10.1016/j.compeleceng.2025.110151","url":null,"abstract":"<div><div>This paper proposes a current optimization model predictive control with common-mode voltage (CMV) reduction (COMPC-CMVR) for three-level T-type inverters to suppress CMV and improve current quality without increasing the losses. The CMV is restricted within <em>u<sub>dc</sub></em>/6 by excluding voltage vectors (VVs) with high CMV. However, the reduction in VVs reduces the current quality. The neutral point (NP) voltage optimization interval is proposed in the COMPC-CMVR to improve the current control performance, where the grid current is the only control objective when the NP voltage is within the voltage optimization interval. The small and medium VVs are divided into P-type and N-type VVs to balance the NP voltage without weighting factors. On this basis, two novel candidate VVs sets are proposed. The COMPC-CMVR considers only four to seven feasible VVs in each control cycle, which reduces the computational burden. Finally, simulation and experimental results show that COMPC-CMVR performs well in terms of steady-state and transient responses. The COMPC-CMVR can effectively suppress the CMV and improve current quality without increasing the losses.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110151"},"PeriodicalIF":4.0,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143208983","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}
Mixed Integer Linear Programming (MILP) is utilized in behavioral synthesis as a mathematical model to design efficient hardware. However, solving large MILP models poses significant computational challenges due to their NP-hard nature. Paralleling can tackle this challenge by amortizing the execution time, yet unbalanced loads can hinder its effectiveness. In this paper, we address the load balance issue of parallel Branch and Bound (B&B) algorithms, particularly sub-tree parallelism, which exhibit efficiency in solving MILP models derived from behavioral synthesis. The proposed algorithm strategically partitions the original problem into sub-problems by selecting decision variables that appear in a higher number of constraints to prioritize load balance and enhance solver performance. We evaluate the effectiveness of our method using MILP models derived from Mediabench data flow graphs of various sizes. The experimental results indicate that the proposed algorithm achieves speedups ranging from approximately 1 to 13 times, highlighting its efficacy in improving the scalability and efficiency of MILP solving for behavioral synthesis.
{"title":"Load balanced sub-tree decomposition algorithm for solving Mixed Integer Linear Programming models in behavioral synthesis","authors":"Mahmood Fazlali , Mina Mirhosseini , Mahdi Movahedian Moghaddam , Somayyeh Timarchi","doi":"10.1016/j.compeleceng.2025.110104","DOIUrl":"10.1016/j.compeleceng.2025.110104","url":null,"abstract":"<div><div>Mixed Integer Linear Programming (MILP) is utilized in behavioral synthesis as a mathematical model to design efficient hardware. However, solving large MILP models poses significant computational challenges due to their NP-hard nature. Paralleling can tackle this challenge by amortizing the execution time, yet unbalanced loads can hinder its effectiveness. In this paper, we address the load balance issue of parallel Branch and Bound (B&B) algorithms, particularly sub-tree parallelism, which exhibit efficiency in solving MILP models derived from behavioral synthesis. The proposed algorithm strategically partitions the original problem into sub-problems by selecting decision variables that appear in a higher number of constraints to prioritize load balance and enhance solver performance. We evaluate the effectiveness of our method using MILP models derived from Mediabench data flow graphs of various sizes. The experimental results indicate that the proposed algorithm achieves speedups ranging from approximately 1 to 13 times, highlighting its efficacy in improving the scalability and efficiency of MILP solving for behavioral synthesis.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110104"},"PeriodicalIF":4.0,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143208982","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 : 2025-02-06DOI: 10.1016/j.compeleceng.2025.110114
Divyanshi Dwivedi , K. Victor Sam Moses Babu , Pratyush Chakraborty , Mayukha Pal
The reliability and efficiency of electrical distribution systems are required for ensuring an uninterrupted power supply and minimizing operational disruption, as failures could lead to significant power outages and safety hazards. This work proposes a novel approach for the classification of electrical faults in distribution systems, utilizing an advanced machine learning technique combined with the vector visibility graphs (VVG). Initially, electrical signal data from the distribution system are collected and transformed into a visibility network, by mapping multivariate time series data to vector space and establishing visibility criteria between vectors. Also, complex network parameters as features from obtained visibility network. Subsequently, a simplicial complex is constructed from the visibility network to explore the topology and connectivity patterns inherent in the electrical data. The Bron-Kerbosch algorithm is employed to detect maximal cliques within the network, serving as a robust method for identifying intricate relationships and anomalies indicative of faults. Characterization of the simplicial complex is performed using both vector and scalar quantities, to extract meaningful features from the electrical signals. These features are then synthesized to capture the maximum values of vectors, focusing on the most significant attributes for fault classification. Then, the feature set is fed into a support vector machine (SVM) classifier for training and validating to distinguish between fault and no fault conditions. The proposed methodology demonstrates superior performance in fault classification, significantly enhancing an accuracy of 99.51% of fault detection in electrical distribution systems.
{"title":"Simplicial complexes using vector visibility graphs for multivariate classification of faults in electrical distribution systems","authors":"Divyanshi Dwivedi , K. Victor Sam Moses Babu , Pratyush Chakraborty , Mayukha Pal","doi":"10.1016/j.compeleceng.2025.110114","DOIUrl":"10.1016/j.compeleceng.2025.110114","url":null,"abstract":"<div><div>The reliability and efficiency of electrical distribution systems are required for ensuring an uninterrupted power supply and minimizing operational disruption, as failures could lead to significant power outages and safety hazards. This work proposes a novel approach for the classification of electrical faults in distribution systems, utilizing an advanced machine learning technique combined with the vector visibility graphs (VVG). Initially, electrical signal data from the distribution system are collected and transformed into a visibility network, by mapping multivariate time series data to vector space and establishing visibility criteria between vectors. Also, complex network parameters as features from obtained visibility network. Subsequently, a simplicial complex is constructed from the visibility network to explore the topology and connectivity patterns inherent in the electrical data. The Bron-Kerbosch algorithm is employed to detect maximal cliques within the network, serving as a robust method for identifying intricate relationships and anomalies indicative of faults. Characterization of the simplicial complex is performed using both vector and scalar quantities, to extract meaningful features from the electrical signals. These features are then synthesized to capture the maximum values of vectors, focusing on the most significant attributes for fault classification. Then, the feature set is fed into a support vector machine (SVM) classifier for training and validating to distinguish between fault and no fault conditions. The proposed methodology demonstrates superior performance in fault classification, significantly enhancing an accuracy of 99.51% of fault detection in electrical distribution systems.</div></div>","PeriodicalId":50630,"journal":{"name":"Computers & Electrical Engineering","volume":"123 ","pages":"Article 110114"},"PeriodicalIF":4.0,"publicationDate":"2025-02-06","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143208980","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