Article
Engineering, Electrical & Electronic
Pietro Romano, Antonino Imburgia, Giuseppe Rizzo, Guido Ala, Roberto Candela
Summary: The usability of a new Direct Current Periodic waveform for partial discharge qualification in HVDC systems is demonstrated through tests conducted on various materials.
IEEE ELECTRICAL INSULATION MAGAZINE
(2021)
Article
Chemistry, Multidisciplinary
Xuewen Yan, Yuanyuan Bai, Wenwen Zhang, Chen Cheng, Jihong Liu
Summary: This paper proposes a method for detecting and recognizing partial discharges in high-voltage equipment using phase-resolved partial discharge analysis technique and convolutional neural network. The algorithm achieves a recognition accuracy exceeding 98% when implemented on a microcontroller.
APPLIED SCIENCES-BASEL
(2023)
Article
Chemistry, Analytical
Lukas Benesl, Petr Mlynek, Michal Ptacek, Vaclav Vycital, Jiri Misurec, Jan Slacik, Martin Rusz, Petr Musil
Summary: This article discusses the use of Power Line Communication (PLC) technology to assess the technical condition of cable routes, highlighting the significance of preventing cable breakdowns and power outages. The methodology for calculating the coefficient is presented with two specific examples that demonstrate the benefits for distribution system operators (DSOs).
Article
Engineering, Multidisciplinary
Yang Zhou, Yiying Liu, Na Wang, Xutao Han, Junhao Li
Summary: This paper demonstrates the application of microfiber coupler sensor (MFCS) and optimized support vector machine (BSO-SVM) algorithm in the pattern recognition of transformer partial discharge (PD) signals. The results show that the combination of MFCS and BSO-SVM achieves high classification accuracy and fast convergence speed, which is of great significance for transformer fault diagnosis.
Article
Engineering, Electrical & Electronic
Chien-Kuo Chang, Hsuan-Hao Chang, Bharath Kumar Boyanapalli
Summary: This study presents a defect-type recognition technique based on pulse sequence analysis (PSA) that does not require voltage signal measurement. The PSA method showed better performance compared to the conventional phase-resolved PD (PRPD) pattern, even when the voltage signal is lost.
IEEE TRANSACTIONS ON DIELECTRICS AND ELECTRICAL INSULATION
(2022)
Review
Engineering, Electrical & Electronic
Shuaibing Li, Binglei Cao, Jin Li, Yi Cui, Yongqiang Kang, Guangning Wu
Summary: Composited cable terminals are crucial for reliable power delivery in power supply systems. The changing operating conditions and various factors can lead to defects or failures, making condition monitoring and defect detection imperative. This review provides an overview of cable terminal condition assessment, including monitoring methods, defect detection approaches, and future research directions.
Article
Engineering, Electrical & Electronic
Zan Wang, Zhongquan Liu, Lili Qiao, Dingdong Qian, Zhongxian Chen, Chaofei Gao, Wei Wang
Summary: An EFPI fiber optic ultrasonic sensor is used for detecting and recognizing partial discharge ultrasonic signals in Gas Insulated Switchgear. It has advantages of high sensitivity and strong anti-interference ability compared to traditional piezoelectric sensors. By setting up PD models and using the EFPI sensor to detect and extract ultrasonic signals, waveform features are used for pattern recognition and analysis using probabilistic neural network and support vector machine algorithms. The EFPI sensor provides prominent features in the detected ultrasonic signal, and both pattern recognition algorithms achieve an average recognition rate of over 85%, with the support vector machine outperforming the probabilistic neural network in recognition effectiveness.
JOURNAL OF ELECTRICAL ENGINEERING & TECHNOLOGY
(2023)
Article
Computer Science, Interdisciplinary Applications
Rajat Srivastava, Vinay Avasthi, R. Krishna Priya
Summary: In this work, a new data-driven model using optimized Convolutional Neural Networks (CNNs) is proposed to recognize the condition of PD pulses of power cables. The model reduces the dimensionality of power cable data using Principal Component Analysis (PCA) and extracts features using technical indicators. A new hybrid optimization algorithm is used to improve the accuracy of the model. The experimental results show that the proposed model outperforms existing models like SVM, CNN, and LSTM.
ADVANCES IN ENGINEERING SOFTWARE
(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
Engineering, Electrical & Electronic
Saliha Abdul Madhar, Armando Rodrigo Mor, Petr Mraz, Rob Ross
Summary: This paper investigates the surface discharge behavior of various dielectric samples under DC, using measurements of material properties and Finite Element simulation to estimate partial discharge defects. The experimental results show a highly plausible behavior of PD defects under DC, with a significant degree of similarity to AC surface discharge behavior. Novel partial discharge fingerprints for surface PD defects are proposed as a tool for defect identification under HVDC.
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS
(2021)
Article
Chemistry, Multidisciplinary
Yuanyuan Sun, Shuo Ma, Shengya Sun, Ping Liu, Lina Zhang, Jun Ouyang, Xianfeng Ni
Summary: A new method for transformer PD pattern recognition using the MobileNets convolutional neural network (MCNN) model was proposed in this study, and the effectiveness and superiority of the model were validated through experimental data.
APPLIED SCIENCES-BASEL
(2021)
Article
Engineering, Electrical & Electronic
Yuhang Yao, Ju Tang, Cheng Pan, Wenbin Song, Yang Luo, Kailai Yan, Qian Wu
Summary: Fingerprint parameters of partial discharge (PD) are commonly used for insulation defect identification. This study found that different defects in XLPE cables have distinct PD characteristics under different voltages and polarities. Applying GA-BPNN method can eliminate the influence of voltage level and polarity, improving defect identification rate.
INTERNATIONAL JOURNAL OF ELECTRICAL POWER & ENERGY SYSTEMS
(2021)
Review
Computer Science, Information Systems
Norfadilah Rosle, Nor Asiah Muhamad, Mohamad Nur Khairul Hafizi Rohani, Mohamad Kamarol Mohd Jamil
Summary: This paper reviews the research on PD patterns and classifiers for XLPE cables, discusses the differences in sensor development based on PD detection in the past 27 years, and concludes that using artificial neural network (ANN) for PD signal pattern recognition performs better in terms of accuracy and repeatability.
Article
Chemistry, Multidisciplinary
Xuewen Yan, Chen Cheng, Chen Zhang, Lei Bai, Wenwen Zhang
Summary: The system designed for monitoring partial discharge caused by insulation faults of switchgears utilizes UHF sensors and IoT technology for online monitoring of discharge signals, with data uploaded to a cloud platform. Experimental results demonstrate that the monitoring system is stable and reliable, meeting practical needs.
APPLIED SCIENCES-BASEL
(2023)
Article
Engineering, Electrical & Electronic
Zerui Li, Kai Zhou, Xiangdong Xu, Pengfei Meng, Yao Fu, Zijin Zeng
Summary: This article proposes a novel partial discharge (PD) detection method based on thermal excitation to eliminate the extinguish phenomenon and detect potential defects in earlier stage. It introduces the physical processes in PD activity and the reasons for discharge extinguishing. The principle and mechanism of the thermal excitation detection method are described and analyzed, and a finite element simulation model is constructed. The effectiveness is validated by performing thermal excitation-based PD test on some defective cables, showing significant improvement in the sensitivity of PD detection for cable accessories and early detection of potential defects.
IEEE TRANSACTIONS ON POWER DELIVERY
(2023)
