检索结果-内蒙古大学图书馆

Thermal fault diagnosis of power equipment through improved instance segmentation and deep neural network temperature fitting

引用

Dianji yu Kongzhi Xuebao/electric Machines and Control 2025年第2期29卷 116-125页

作者： Zheng, Hanbo Hu, Junhao Li, Jinheng Liang, Yanshen Huang, Junjie Guangxi Key Laboratory of Power System Optimization and Energy Technology Guangxi University Nanning530004 China Electric Power Research Institute State Grid Hubei Electric Power Co. Ltd. Wuhan430077 China

The malfunctioning equipment displays anomalous temperatures at the fault location, posing a significant risk to the security and stability of the power system. To evaluate the thermal status of power equipment, a method for diagnosing thermal faults was proposed. This approach was implemented through instance segmentation and temperature fitting. Firstly, a dataset was constructed based on the infrared images of power equipment collected at the substation, and an improved instance segmentation model was developed. The enhanced model achieves a 3. 6% improvement in the mean average precision of masks and a 2. 5% improvement in the boxes, while the model’s floating point operation is maintained at only 12. 1 GFLOPs. Secondly, a method for fitting the temperature values of the equipment based on deep neural networks was proposed. This method achieved better fitting accuracy than other methods, with an R-square of 0. 999 7 and a fitting temperature deviation of approximately 3% . Finally, a method for automatically diagnosing thermal abnormalities was proposed in accordance with the current industrial guidelines. This algorithm was founded upon kernel density estimation for the obtainment of temperature parameters. The experimental findings indicate that our algorithm precisely detects and is alert to the fault breaker, thus increasing operational safety. © 2025 Editorial Department of electric Machines and Control. All rights reserved.

关键词： Deep neural networks

来源：评论

学校读者我要写书评

暂无评论

Electrochemical extraction of strontium from molten salts using reactive zincand aluminum electrodes

引用

International Journal of Minerals,Metallurgy and Materials 2025年第4期32卷 892-901页

作者： Yongcheng Zhang Taiqi Yin Lei Zhang Xiaochen Zhang Tao Bo Xiaoli Tan Mei Li Wei Han Zhejiang Key Laboratory of Data-Driven High-Safety Energy Materials and Applications Ningbo Institute of Materials Technology and EngineeringChinese Academy of SciencesNingbo 315201China College of Materials Science and Chemical Engineering Harbin Engineering UniversityHarbin 150001China MOE Key Laboratory of Resources and Environmental System Optimization College of Environmental Science and EngineeringNorth China Electric Power UniversityBeijing 102206China Qianwan Institute of CNITECH Ningbo 315336China

Herein, the electrochemical behaviors of Sr on inert W electrode and reactive Zn/Al electrodes were systematically investig-ated in LiCl–KCl–SrCl2molten salts at 773 K using various electrochemical methods. The chemical reaction potentials of Li and Sr on re-active Zn/Al electrodes were determined. We observed that Sr could be extracted by decreasing the activity of the deposited metal Sr onthe reactive electrode, although the standard reduction potential of Sr(II)/Sr was more negative than that of Li(I)/Li. The electrochemicalextraction products of Sr on reactive Zn and Al electrodes were Zn13Sr and Al4Sr, respectively, with no codeposition of Li *** on the density functional theory calculations, both Zn13Sr and Al4Sr were identified as stable intermetallic compounds with Zn-/Al-rich phases. In LiCl–KCl molten salt containing 3wt% SrCl2, the coulombic efficiency of Sr in the Zn electrode was ~54%. The depolar-ization values for Sr on Zn and Al electrodes were 0.864 and 0.485 V, respectively, exhibiting a stronger chemical interaction between Znand Sr than between Al and Sr. This study suggests that using reactive electrodes can facilitate extraction of Sr accumulated while elec-trorefining molten salts, thereby enabling the purification and reuse of the salt and decreasing the volume of the nuclear waste.

关键词： strontium reductive extraction molten salt depolarization effect reactive electrode

来源：评论

学校读者我要写书评

暂无评论

Molecular Dynamics Study of the Viscosity Characteristics of Soybean Oil-Based Natural Ester Insulating Oil

引用

IEEE Transactions on Dielectrics and electrical Insulation 2025年

作者： Zheng, Hanbo Lan, Hongce Wang, Ke Cai, Shengwei Yang, Tao Li, Yi Wang, Yang Guangxi University Guangxi Key Laboratory of Power System Optimization and Energy Technology Nanning53004 China China Electric Power Research Institute Beijing100192 China State Grid Corporation of China State Grid Henan Electric Power Research Institute Zhengzhou450000 China

Many studies have shown that natural ester insulating oil is an excellent transformer oil with a superior flash point, fire resistance, and renewability compared to mineral insulating oil. However, due to their molecular properties, the high viscosity of natural ester insulating oils results in poor heat transfer, limiting their safe application in transformers. In this paper, the viscosity values of the soybean oil-based natural ester insulating oil are measured experimentally in the temperature range of 293 K to 433 K. The micro-mechanisms affecting the viscosity of natural ester insulating oil are investigated in conjunction with molecular dynamics simulations. We also conducted a correlation analysis between specific heat and viscosity and the simulation results. In addition, the kinematic viscosity of natural ester insulating oils as a function of temperature is also obtained in this paper by the free volume theory. The temperature ranges from 293 K to 433 K, and the functional model closely agrees with the experimental measurements. The research in this paper can be informative for the study and prediction of the viscosity of natural ester insulating oils as well as for the study of thermal conductivity. © 2025 IEEE.

关键词： Fatty acid methyl ester

来源：评论

学校读者我要写书评

暂无评论

Enhancing IC substrate manufacturing through differential geometry and lightweight networks for etching defect detection

引用

Journal of Manufacturing systems 2025年 80卷 902-915页

作者： Yu, Yongxing Huang, Dan Hu, Yueming School of Automation Science and Engineering South China University of Technology Guangzhou510641 China Engineering Research Center of Precision Electronic Manufacturing Equipment Ministry of Education and Guangdong Provincial Engineering Laboratory for Advanced Chip Intelligent Packaging Equipment South China University of Technology Guangzhou510641 China Department of Automation North China Electric Power University Baoding071003 China Baoding key Laboratory of State Detection and Optimization Regulation for Integrated Energy System North China Electric Power University Baoding071003 China

With the advancement of high-density interconnect technology in semiconductor manufacturing, the precision and complexity of integrated circuit (IC) substrates have significantly increased, placing higher demands on quality control. Efficient and accurate detection of complex etching defects, which often occur during manufacturing, has become critical to preventing potential product defects. A defect detection method is proposed that combines a lightweight network with differential geometry tools to address the issue of etching defects in IC substrates. First, an improved deformable model is used to rapidly extract regular circuit trace contours from complex metallographic images, and morphological processing is applied to enhance the details, achieving precise image segmentation. For under-etching defects between circuit traces, differential processing of the original and segmented images is performed to locate abnormal regions. Subsequently, an optimized lightweight network based on MobileNet, termed OMNet, is designed to achieve the rapid identification of under-etching defects in these regions. For etching defects occurring on circuit traces, the DGEtch method employs a high-precision discrete curvature calculation based on the Frenet frame to evaluate angular discontinuities in contours, enabling accurate detection of etching defects. Experimental results demonstrate that the proposed method achieves an average recall rate of over 95% and maintains a precision above 90%. It exhibits high accuracy and stability in detecting etching defects and consistently outperforms existing models, particularly in handling complex mixed defects. This study provides an effective solution for detecting complicated defects in high-density IC substrate manufacturing. © 2025 The Society of Manufacturing Engineers

关键词： Inspection

来源：评论

学校读者我要写书评

暂无评论

Gas Sensitivity Study of Early Diagnostic Markers for Lung Cancer Using Mote2 Single Molecular Membranes Doped with Different Tm Atoms:Based on Density Functional Theory

SSRN

引用

SSRN 2025年

作者： Chen, Yujie Jiang, Wenhao Zhang, Yiyi Chen, Dachang Xu, Min Liu, Jiefeng Jia, Pengfei School of Electrical Engineering Guangxi University Nanning China School of Electrical and Electronic Engineering Wuhan Polytechnic University Wuhan China College of Food Science and Bioengineering Xihua University Chengdu China Guangxi Engineering Technology Innovation Center of Electric Power Transmission and Distribution Network Lightning Protection Guangxi University Nanning China Guangxi Key Laboratory of Power System Optimization and Energy Technology China

Lung cancer has become one of the deadliest and most prevalent cancers worldwide, and the use of gas sensors to detect volatile organic compounds (VOCs) in the exhaled breath of lung cancer patients is gaining increasing popularity. Compared with traditional medical diagnostic methods, this method is cost-effective and less invasive. During our experiments, we employ density functional theory to explore how transition metal (Cu, Pd, Pt)-doped MoTe2 single-molecule membranes respond to VOCs commonly found in the exhalation gas of patients with lung cancer in the early stages of the disease. All three modified systems exhibited excellent thermal stability, and the sorption of VOCs is significantly enhanced compared to the pristine MoTe2, ensuring effective desorption and sensing performance at elevated temperatures. Moreover, the changes in the band gap before and after adsorption are notably distinct, indicating strong gas sensitivity. Among the doped structures, MoTe2-Cu shows the highest adsorption capacity for C5H8, C3H6O, and C3H8O, accompanied by the largest change in the band gap. Due to the varying sensitivities of the three lung cancer biomarker sensors to different gases, cross-sensitivity can be minimised, highlighting the potential for qualitative analysis of VOC gas mixtures. This offers new insights and methods for the early detection and prevention of lung cancer. © 2025, The Authors. All rights reserved.

关键词： Gas mixtures

来源：评论

学校读者我要写书评

暂无评论

Gas Sensitivity Study of Early Diagnostic Markers for Lung Cancer Using Mote2 Single Molecular Membranes Doped with Different Tm Atoms:Based on Density Functional Theory

SSRN

引用

SSRN 2025年

Lung cancer has become one of the deadliest and most prevalent cancers worldwide, and the use of gas sensors to detect volatile organic compounds (VOCs) in the exhaled breath of lung cancer patients is gaining increasing popularity. Compared with traditional medical diagnostic methods, this method is cost-effective and less invasive. During our experiments, we employ density functional theory to explore how transition metal (Cu, Pd, Pt)-doped MoTe2 single-molecule membranes respond to VOCs commonly found in the exhalation gas of patients with lung cancer in the early stages of the disease. All three modified systems exhibited excellent thermal stability, and the sorption of VOCs is significantly enhanced compared to the pristine MoTe2, ensuring effective desorption and sensing performance at elevated temperatures. Moreover, the changes in the band gap before and after adsorption are notably distinct, indicating strong gas sensitivity. Among the doped structures, MoTe2-Cu shows the highest adsorption capacity for C5H8, C3H6O, and C3H8O, accompanied by the largest change in the band gap. Due to the varying sensitivities of the three lung cancer biomarker sensors to different gases, cross-sensitivity can be minimised, highlighting the potential for qualitative analysis of VOC gas *** offers new insights and methods for the early detection and prevention of lung cancer. © 2025, The Authors. All rights reserved.

关键词： Lung cancer

来源：评论

学校读者我要写书评

暂无评论

Architecture design of power system fault calculation based on cloud computing technology

Architecture design of power system fault calculation based ...

引用

2017 IEEE Conference on energy Internet and energy system Integration, EI2 2017

作者： Dongxu, Yang Hua, Wei Hongbo, Wei Key Laboratory of Guangxi Electric Power System Optimization and Energy-saving Technology Guangxi University Guangxi China

ISBN: (纸本)9781538614273

This paper presents an architecture of power system fault calculation based on cloud computing technology to solve the problems of the current fault calculation system, which is caused by the insufficient calculation ability. The existing fault calculation system can not meet the requirements of the calculating efficiency and accuracy with the expansion of power system. The concept, characteristics, advantages, structure of cloud computing and service oriented architecture (SOA), as well as the architecture of the proposed system are described in detail in this paper. Besides, the cloud computing technology and SOA are applied to the fault calculation system, reaching the elastic scaling of the calculation and storage, reliable disaster recovery of the system, and a higher interoperability. To validate the design, the proposed system is applied to an actual power system, and the operation effect has proved its feasibility and superiority. © 2017 IEEE.

关键词： Cloud computing

来源：评论

学校读者我要写书评

暂无评论

An Attention-based CNN-GRU Model for Resident Load Short-term Forecast 5

An Attention-based CNN-GRU Model for Resident Load Short-ter...

引用

5th IEEE Conference on energy Internet and energy system Integration, EI2 2021

作者： Wei, Shangfu Bai, Xiaoqing Key Laboratory of Guangxi Electric Power System Optimization and Energy-saving Technology Guangxi University Nanning China

ISBN: (纸本)9781665434256

Under the background of electricity market, accurate short-term forecast of residential load can effectively strengthen demand side management. However, the randomness and uncertainty of residential electricity consumption increase the difficulty of prediction. The traditional prediction models are facing with some problems that it is difficult to effectively extract high-dimensional features and important information is easy to lose when the length of the time series is too large. Based on these, the paper established an Attention-based CNN-GRU model for resident load short-term forecast. In this model, CNN plays a role in extracting the features of historical power consumption data, GRU neural network combined with Attention mechanism is used to model these extracted features dynamically. Real data of individual household electricity consumption is taken as an example, and simulation experiments are carried out in three scenarios of univariate single step, multivariate single step and multivariate multi-step prediction respectively. The results of the experiment demonstrate that the proposed model achieves reasonable accuracy and robustness. © 2021 IEEE

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

Data-Driven RNN Method for Distribution Network Optimal power Flow Problems 3

Data-Driven RNN Method for Distribution Network Optimal Powe...

引用

3rd power system and Green energy Conference, PSGEC 2023

作者： Zhang, Zixuan Yang, Shiao Jia, Yujing Xiao, Jiawen Bai, Xiaoqing Guangxi University Key Laboratory of Guangxi Electric Power System Optimization and Energy-Saving Technology Guangxi Nanning China

ISBN: (纸本)9798350340099

The optimal power flow (OPF) problem has long been a central focus for researchers in the field of power systems. With the widespread adoption of data-driven approaches, this paper introduces a novel solution to the distribution network OPF problem using a recurrent neural network (RNN) method. By leveraging the power of RNNs, this method offers a promising approach to address the challenges associated with optimizing power flow in distribution networks. To evaluate the effectiveness of the proposed method, a simulation experiment was conducted on a 12-bus distribution network system. The results of the experiment not only confirmed the method's efficacy but also showcased its superior performance compared to existing approaches. This success paves the way for the future application of the RNN method in larger-scale power systems. Overall, this research highlights the significance of data-driven methods in addressing the challenges of power system optimization. The successful application of the RNN method in solving the distribution network OPF problem underscores its potential to revolutionize the field and improve the efficiency and reliability of power systems in the future. © 2023 IEEE.

关键词： Recurrent neural networks

来源：评论

学校读者我要写书评

暂无评论

Short-term Load Forecasting Based on R-SDBN 3

Short-term Load Forecasting Based on R-SDBN

引用

3rd IEEE Conference on energy Internet and energy system Integration, EI2 2019

作者： Jiang, Xiongfeng Wei, Hua Lyu, Zhongliang Lian, Chunjie Guangxi University Key Laboratory of Guangxi Electric Power System Optimization and Energy-saving Technology Nanning China

ISBN: (纸本)9781728131375

This paper proposes a self-organizing deep belief network with robust locally weighted scatterplot smoothing method (R-SDBN) for short-term load forecasting, which can not only organize the structure of short-term load forecasting automatically but also denoise the input load data. The deep belief network (DBN) integrates deep learning and feature learning. It can quickly analyze a large number of data and has a strong data fitting ability. And the robust locally weighted scatterplot smoothing (RLWSS) method is introduced to weaken the noise and correct the outlies of load data. However, the optimal structure of DBN and the optimal span of RLWSS are different given specific applications. In the proposed model, the structure of DBN and the span of RLWSS are selforganized with minimum root mean square error as objective, not selected by experience, during the training process, and they are modified after each forecasting. Taking the actual load of a certain city in China as an example, the root mean square error (RMSE), the mean absolute percentage error (MAPE), and the maximum absolute error (MAE) of the 24hour forecasting is 4.38MW, 6.54%, and 12.48MW respectively. Compared with the backward propagation neural network, support vector machine (SVM), and DBN, R-SDBN has remarkable advantages. © 2019 IEEE.

关键词： Forecasting

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：