Article
Engineering, Electrical & Electronic
Huan Yuan, Xiaonan Wang, Aijun Yang, Dingxin Liu, Xiaohua Wang, Mingzhe Rong
Summary: This study investigated the decomposition products of SF6/air gas mixtures under different pulsed discharge voltages, revealing the main products and their yields increasing with the number of discharge pulse. The production rates of SOF2, SO2, and SOF4 were positively correlated with discharge voltage and the proportion of air in the gas mixtures, while SO2F2 was not affected by these factors. These findings provide a basis for determining discharge failures and the discharge degree of power equipment.
Article
Computer Science, Artificial Intelligence
Tong Gao, Hao Chen
Summary: In this study, a multicycle disassembly-based decomposition algorithm (MCD-DA) is proposed to efficiently solve the training problem of multiclass support vector machine (SVM). MCD-DA constructs a graph model to re-express the constraints in multiclass SVM, partitions the complex feasible region into simple sub-feasible regions, and designs multiple cycle-based disassembly strategies to update the working variables analytically. Experimental results demonstrate that MCD-DA outperforms typical optimization algorithms for more sample cases.
PATTERN RECOGNITION
(2023)
Article
Engineering, Multidisciplinary
Wong Jee Keen Raymond, Chong Wan Xin, Lai Weng Kin, Hazlee Azil Illias
Summary: Traditional machine learning models rely on manual feature extraction for training data and may experience decreased classification accuracy when faced with noise interference. A CNN-based PD classification system using transfer learning was proposed in this study, showing significantly higher classification accuracy under noise contamination compared to traditional models.
Article
Environmental Sciences
Guangxin Liu, Liguo Wang, Danfeng Liu, Lei Fei, Jinghui Yang
Summary: This article proposes a non-parallel SVM model, which improves the classification effect and generalization performance for hyperspectral images by adding an additional empirical risk minimization term and bias constraint.
Article
Engineering, Electrical & Electronic
Jie Gao, Yungang Zhang, Xinhui Li, Guangwen Shi, Yucun Zhang
Summary: This article proposes a quantitative calibration model based on Fourier transform infrared (FTIR) spectroscopy and feature selection algorithm combined with neural network for fast and non-destructive detection of SF6 decomposition products concentration in gas-insulated switchgear (GIS). The results showed that the proposed model achieved high precision and accuracy, and it is of great significance for evaluating the internal insulation status.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Engineering, Civil
Yong Pan, Xiangmo Zhao, Zhigang Xu, Junwei Li, Yifei Li, Li Liu
Summary: The abnormal driving behavior of buses poses a greater security threat, and effectively identifying this behavior has become a challenge in ensuring public transportation safety. This paper introduces an abnormal behavior recognition model for buses based on an improved support vector machine, which has shown a high recognition rate in verification and offers a valuable tool for enhancing the safety of public transportation operations.
JOURNAL OF ADVANCED TRANSPORTATION
(2021)
Article
Chemistry, Multidisciplinary
Matthew Martin Zarachoff, Akbar Sheikh-Akbari, Dorothy Monekosso
Summary: This study introduces a Chainlet-based Ear Recognition algorithm using Multi-Banding and Support Vector Machine (CERMB-SVM) which significantly outperforms PCA-based techniques in ear recognition accuracy. The proposed method divides gray input images into bands, performs Canny edge detection, creates binary edge maps, and generates chainlet histogram vectors for training and testing SVM. The results indicate that the CERMB-SVM method excels in ear recognition, surpassing anchor chainlet technique and state-of-the-art learning-based approaches.
APPLIED SCIENCES-BASEL
(2022)
Article
Engineering, Marine
Wenhe Shen, Jianxi Yao, Xinjue Hu, Jialun Liu, Shijie Li
Summary: With the rapid development of Maritime Autonomous Surface Ships (MASS), building an accurate ship dynamics model using system identification method has become a critical issue. This study proposes a non-parametric and robust two-phase system identification method, which filters the data using an improved complete ensemble empirical mode decomposition and models the ship dynamics using Semblance least square support vector machine (S-LS-SVM) with a state-of-the-art Semblance kernel function. Compared to traditional methods, this method significantly improves the Root Mean Square Error (RMSE) of overall prediction on the test dataset.
Article
Engineering, Electrical & Electronic
Sibasankar Padhy
Summary: This paper proposes a tensor-based approach using MLSVD for hand gesture recognition, achieving high accuracy in inter-session and inter-subject evaluations on multiple databases. The method outperforms state-of-the-art methods in both inter-session and inter-subject evaluations.
IEEE SENSORS JOURNAL
(2021)
Article
Engineering, Electrical & Electronic
Wenchao Miao, Qi Xu, K. H. Lam, Philip W. T. Pong, H. Vincent Poor
Summary: Protection devices are crucial in DC systems, but series arc faults may not be detected by conventional devices, leading to malfunctions and fire hazards. This paper proposes a series arc-fault detection system based on modified EMD technique and SVM algorithm for reliable and efficient operation of DC systems. The effectiveness of arc-fault detection is significantly improved by acquiring accurate arc signatures without predefining various thresholds.
IEEE SENSORS JOURNAL
(2021)
Article
Computer Science, Theory & Methods
Umesh Gupta, Deepak Gupta
Summary: This paper presents two efficient variant models to handle noise and outliers, obtaining solutions by solving a system of linear equations and minimizing the impact of noise. The proposed models demonstrate exceptional generalization performance.
FUZZY SETS AND SYSTEMS
(2022)
Article
Computer Science, Interdisciplinary Applications
Bikram Kumar, Deepak Gupta
Summary: The paper introduces a novel method ULTBSVM which utilizes Universum data to enhance the classification of healthy and seizure EEG signals, showing promising results in experiments.
COMPUTER METHODS AND PROGRAMS IN BIOMEDICINE
(2021)
Article
Computer Science, Information Systems
Jie Sun, Hamido Fujita, Yujiao Zheng, Wenguo Ai
Summary: This paper focuses on multiclass financial distress prediction using SVM and decomposition fusion methods, showing that OVO-SVM outperforms OVR-SVM and ECOC-SVM in overall performance and is preferred. Data preprocessing mechanisms can greatly enhance the model performance, while OVO-SVM is more competitive for predicting financial pseudosoundness and moderate financial distress compared to human expertise.
INFORMATION SCIENCES
(2021)
Article
Automation & Control Systems
Kirandeep Kour, Peter Benner, Sergey Dolgov, Martin Stoll
Summary: This paper proposes a Tensor Train Multi-way Multi-level Kernel (TT-MMK) method, which combines different techniques to achieve more reliable and accurate performance in high-dimensional data classification.
JOURNAL OF MACHINE LEARNING RESEARCH
(2023)
Article
Computer Science, Information Systems
Bo Liu, Ruiguang Huang, Yanshan Xiao, Junrui Liu, Kai Wang, Liangjiao Li, Qihang Chen
Summary: This paper introduces the role of Universum in supervised and semi-supervised learning and incorporates it into TWSVM to improve generalization performance. To enhance generalization performance in complex environments, an adaptive robust Adaboost-based twin support vector machine with universum learning (ARABUTWSVM) is proposed.
INFORMATION SCIENCES
(2022)
Article
Engineering, Environmental
Peng Wu, Yi Li, Song Xiao, Junyi Chen, Ju Tang, Dachang Chen, Xiaoxing Zhang
Summary: The study successfully fabricated a highly sensitive C4F7N gas sensor based on SnO2 nanoparticles and explored its sensing mechanism towards C4F7N. The research provides convincing support for lowering the potential operation risk brought by C4F7N in electrical engineering.
JOURNAL OF HAZARDOUS MATERIALS
(2022)
Article
Chemistry, Physical
Guozhi Zhang, Zengting Wang, Xiaoxing Zhang
Summary: This article explores the sensing and adsorption behaviors of Ru-doped MoS2 monolayer upon SO2 and SOF2 gases in SF6 insulation device. The simulation results show chemisorption for SO2 and physisorption for SOF2. The article also highlights the potential of Ru-MoS2 monolayer as a sensing material for detecting SO2 and scavenging SOF2, as well as the importance of frontier molecular orbital theory in promoting the steady operation of power systems.
Article
Chemistry, Analytical
Yi Wang, Dalin Ding, Yin Zhang, Zian Yuan, Shuangshuang Tian, Xiaoxing Zhang
Summary: This paper proposes a detection technology for C5F10O mixed gas based on infrared spectrum, including mixture ratio detection and trace detection. A quantitative model based on stepwise multiple linear regression (SMLR) is established to improve the detection accuracy. The study also analyzes the overlap of C5F10O with other gases and identifies a suitable band for trace detection. The experimental and simulated infrared spectrum information provided in this research supports the development of detection technology for C5F10O mixed gas insulation equipment.
VIBRATIONAL SPECTROSCOPY
(2022)
Article
Multidisciplinary Sciences
Xiajin Rao, Dajian Li, Xiaoxing Zhang, Xiaofei Xia, Yi Su, Yufeng Lu, Boya Peng
Summary: This paper experimentally studied the partial discharge characteristics of C6F12O mixed gas and found that the breakdown voltage is greatly improved after adding C6F12O. The partial discharge initiation voltage and extinction voltage of the mixed gas gradually increase with the increase of mixing ratio. The research in this paper provides an important reference for the application of C6F12O mixed gas in medium and low voltage equipment.
SCIENTIFIC REPORTS
(2022)
Article
Chemistry, Analytical
Wei Liu, Xinjie Qiu, Xiaoxing Zhang, Shuangshuang Tian, Zian Yuan, Weihao Liu
Summary: The adsorption characteristics of ZSM-5 molecular sieve on C4F7N and its decomposition products were investigated. The results showed that ZSM-5 had a certain adsorption effect on the fluorocarbon gases, with the highest adsorption efficiency for C3F6 and C3F8. The adsorption effect of ZSM-5 on gas molecules depends on their polarity.
Article
Engineering, Electrical & Electronic
Xiaoxing Zhang, Jin Yan, Yin Zhang, Lin Cheng, Chao Bian, Xuan Chen
Summary: In this article, the characteristic peaks of SO2F2 and SOF2 were obtained by comparing simulation and experimental infrared spectrum. A TDLAS experiment platform was built and the concentration of SO2F2 and SOF2 was detected through this platform. The study provides a feasible scheme for the quantitative detection of SO2F2 and SOF2 gases.
IEEE TRANSACTIONS ON DIELECTRICS AND ELECTRICAL INSULATION
(2022)
Article
Engineering, Electrical & Electronic
Zhenwei Chen, Xiaoxing Zhang, Hongtu Cheng, Yin Zhang, Tingli Yang, Yangxiaoge Zhang
Summary: In this study, the quantitative gas detection of CO is carried out based on photoacoustic spectroscopy technology. By optimizing the nonresonant photoacoustic cell and building a detection platform, the study successfully detects the concentration of CO using photoacoustic spectroscopy. The lower detection limits of CO with different background gases are predicted and verified experimentally, demonstrating the accuracy of the predictions. This study provides guidance for the engineering application of CO photoacoustic spectroscopy detection in SF6 equipment.
IEEE TRANSACTIONS ON DIELECTRICS AND ELECTRICAL INSULATION
(2022)
Article
Chemistry, Analytical
Hanting Zhang, Guozhi Zhang, Xiaoxing Zhang, Hanlu Tian, Changyue Lu, Jianben Liu, Yin Zhang
Summary: This paper investigates a flexible built-in high-sensitivity elliptic monopole antenna using new techniques, with a thickness of only 0.28 mm, providing high signal detection performance under different curvature radii.
Article
Chemistry, Physical
Yi Li, Song Xiao, Xiaoxing Zhang, Pan Jia, Shuangshuang Tian, Cheng Pan, Fuping Zeng, Dachang Chen, Yuyue Chen, Ju Tang, Jiaqing Xiong
Summary: In this study, a microfibers membrane with omnidirectional superelasticity, permeability, and superhydrophobicity (SPSM) was developed through synchronous electrospinning and electrospraying techniques. Versatile integration of fibers with conductors was achieved, creating diverse elastic conductors and triboelectric-active materials. A self-cleaning stretchable TENG was demonstrated for materials identifying and hand posture monitoring.
Article
Materials Science, Multidisciplinary
Dajian Li, Xiajin Rao, Xiaoxing Zhang, Boya Peng, Shaoming Pan, Wei Huang, Shouxiao Ma
Summary: This paper calculates the thermodynamic and transport properties of C6F12O at different temperatures, and finds that C6F12O has good conductivity and thermal conductivity characteristics, which are suitable for the use of gas-insulated equipment.
FRONTIERS IN MATERIALS
(2022)
Article
Engineering, Electrical & Electronic
Zhaolun Cui, Chang Zhou, Amin Jafarzadeh, Shengyan Meng, Yanhui Yi, Yufei Wang, Xiaoxing Zhang, Yanpeng Hao, Licheng Li, Annemie Bogaerts
Summary: This study investigates the effects of packing size and surface properties of gamma-Al2O3 pellets on the degradation of SF6 in a packed bed dielectric barrier discharge system. The results show that decreasing packing size improves the degradation performance at low input powers, but decreases efficiency with excessively small packing pellets. Lowering the packing size promotes the generation of SO2 and reduces the yields of S-O-F products. After discharge, the pellet surface becomes smoother with the appearance of S and F elements. SF6 is likely to be adsorbed at the Al-III site over the gamma-Al2O3(110) surface, and it is more easily decomposed than in the gas phase.
Article
Engineering, Electrical & Electronic
Fanchao Ye, Xiaoxing Zhang, Yi Li, Qianqian Wan, Jean-Marc Bauchire, Dunpin Hong, Song Xiao, Ju Tang
Summary: This study evaluated the arc interruption and decomposition behavior of a C4F7N/air gas mixture, finding that it has weaker arc interruption capability compared to SF6. The decomposition of the gas mixture produced a significant amount of gaseous and solid by-products. The acute inhalation toxicity test showed relatively high toxicity of the C4F7N/air mixture, emphasizing the need for safety precautions.
Article
Engineering, Electrical & Electronic
Xiaoxing Zhang, Xiaohan Li, Yi Wang, Fei Meng, Yi Zou, Shuangshuang Tian, Zhaolun Cui
Summary: Thermal degradation can effectively eliminate the greenhouse effect caused by SF6 waste gas emissions, providing an environmentally friendly treatment method. In this study, AlPO4, Zr-3(PO4)(4), and CePO4 were used as catalysts to degrade SF6/air mixture gas at 800 degrees C, with CePO4 showing the best catalytic degradation effect and achieving a degradation rate of 84.7% for SF6.
JOURNAL OF ELECTRICAL ENGINEERING & TECHNOLOGY
(2023)
Article
Chemistry, Analytical
Chengqiang Liao, Lei Zhang, Guozhi Zhang, Changyue Lu, Xiaoxing Zhang
Summary: This paper proposes a resonant cavity monopole UHF antenna sensor based on the Fabry-Perot resonant cavity antenna technology to meet the real demand for broadband full-band high-gain antenna sensors in PD UHF detection test and online monitoring of power equipment. The sensor's Voltage Standing Wave Ratio (VSWR) is optimized using curved flow technology, and the gain is optimized using a slot dual resonant structure. The performance of the sensor is tested using a built PD detection test platform.
Article
Engineering, Electrical & Electronic
Yalong Li, Yi Li, Zhuo Wei, Xiaoxing Zhang, Weijiang Chen, Song Xiao, Ju Tang, Jean-Marc Bauchire, Dunpin Hong
Summary: This study proposes a quantitative detection technology for C5 perfluoroketone (C5-PFK) concentration using ultraviolet differential optical absorption spectroscopy (UV-DOAS). The spectral absorption curve of the C5-PFK molecule is explored based on density functional theory (DFT). The calibration curve demonstrates high linearity, with a linear determination coefficient above 0.9999. The relative error of C5-PFK concentration in gas mixture and the calibration curve can be determined, reaching a maximum of 8.1%. The gas supplement and detection method proposed in this study provide accurate detection of C5-PFK concentration changes in engineering applications.
IEEE TRANSACTIONS ON DIELECTRICS AND ELECTRICAL INSULATION
(2022)