An arc fault detection method based on a one-dimensional dilated convolutional neural network

JIANG Huiling; BAI Gali; ZHOU Zheng; DENG Qing; TENG Jie; ZHANG Yue; ZHOU Liang; ZHOU Zhengqing

doi:10.16511/j.cnki.qhdxxb.2023.26.057

PDF(6627 KB)

Journal of Tsinghua University(Science and Technology) ›› 2024, Vol. 64 ›› Issue (3) : 492-501. DOI: 10.16511/j.cnki.qhdxxb.2023.26.057

PUBLIC SAFETY SCIENCE AND TECHNOLOGY

An arc fault detection method based on a one-dimensional dilated convolutional neural network

{{article.zuoZhe_EN}}

Author information +

History +

Abstract

[Objective] The arc fault of the low-voltage distribution system is one of the primary causes of residential fires. Due to the diverse load types and complex connection methods in residential areas, arcs fault exhibit many similar and concealed characteristics, making them difficult to detect. This frequently leads to issues with arc fault protection devices, such as false alarms and missed detections. The conventional detection method, which is based on manually extracting arc fault feature vectors, is incomplete and heavily relies on expert-designed features. Consequently, this impedes the development of highly generalizable models. In addressing these challenges of multiload systems, this paper proposes a method for serial arc fault diagnosis and load recognition based on a one-dimensional dilated convolutional neural network (1D-DCNN). [Methods] First, a series of experiments on multiload arcs fault are conducted using a custom-designed experimental platform. This platform supports both single-load and dual-branch load conditions during testing. Normal and faulty current data on the main bus are sampled under various operating conditions at a rate of 500 kHz. During the data processing phase, the continuous time series data are discretized and normalized based on the half-cycle length. Subsequently, a 1D-DCNN is used to extract features from the high-sampling-rate arc fault current data. Furthermore, the scaled exponential linear unit activation functions and residual connections are introduced to address the challenges of gradient vanishing and network degradation. Moreover, the cyclic padding method is adopted to alleviate boundary effects and enhance the model's robustness to dataset shift. The arc fault detection model is developed by integrating average ensemble learning with a Softmax multiclassifier. Precision, recall, and specificity are used to assess the efficiency of the model. Finally, the accuracy of the proposed model in load classification, load state recognition, and overall accuracy is compared with that of other classical models, providing a comprehensive assessment of its efficacy. [Results] The findings of this method were as follows: (1) The accuracy of arc detection using a recurrent neural network was considerably low, primarily due to gradient vanishing and exploding gradients, making it difficult to effectively train the model. (2) Under the condition of equal parameter count between a 1D-DCNN and 1D-CNN, the dilated convolutional operation expanded the receptive field, resulting in greater accuracy than the 1D-CNN model. (3) The proposed method achieved a remarkable accuracy of 99.67% in detecting arc faults for both single-load and mixed-load scenarios, with an accuracy of 99.95% and an overall accuracy of 99.62%. [Conclusions] This research presents a unique model capable of autonomously learning features from high-sampling-rate current data without requiring manual feature extraction. It efficiently detects arcs fault while identifying the type of faulty load simultaneously. The model outperforms typical convolutional neural networks on validation of the test set, thereby meeting the requirements for arc fault identification. This advancement has major implications for serial arc fault detection and load recognition applications.

Key words

one-dimensional dilated convolutional / series arc fault / branch arc fault / arc fault detection / workload classification

Cite this article

EndNote

Ris (Procite)

Bibtex

Download Citations

JIANG Huiling, BAI Gali, ZHOU Zheng, DENG Qing, TENG Jie, ZHANG Yue, ZHOU Liang, ZHOU Zhengqing. An arc fault detection method based on a one-dimensional dilated convolutional neural network[J]. Journal of Tsinghua University(Science and Technology). 2024, 64(3): 492-501 https://doi.org/10.16511/j.cnki.qhdxxb.2023.26.057

References

[1]张晶, 陈涛, 黄丽达, 等. 接触不良引发发光连接的动态变化[J]. 清华大学学报(自然科学版), 2022, 62(6):1000-1009. ZHANG J, CHEN T, HUANG L D, et al. Dynamics of glowing contacts triggered by poor electrical contact[J]. Journal of Tsinghua University (Science and Technology), 2022, 62(6):1000-1009. (in Chinese)
[2]鲍光海, 江润. 基于磁通不对称分布的串联电弧故障检测研究[J]. 仪器仪表学报, 2019, 40(3):54-61. BAO G H, JIANG R. Research on series arc fault detection based on asymmetrical distribution of magnetic flux[J]. Chinese Journal of Scientific Instrument, 2019, 40(3):54-61. (in Chinese)
[3]KE Y, ZHANG W B, SUO C G, et al. Research on low-voltage AC series arc-fault detection method based on electromagnetic radiation characteristics[J]. Energies, 2022, 15(5):1829.
[4]ZHAO L, ZHOU Y H, CHEN K L, et al. Using spectrum of the light for high speed arcing fault detection[C]//2019 IEEE IAS Electrical Safety Workshop. Jacksonville, USA:IEEE, 2018:19435238.
[5]白辉, 许志红. 基于小波包变换和高阶累积量的电弧故障识别方法[J]. 电力自动化设备, 2020, 40(11):195-202, 224. BAI H, XU Z H. Arc fault identification method based on wavelet packet transform and high-order cumulant[J]. Electric Power Automation Equipment, 2020, 40(11):195-202, 224. (in Chinese)
[6]江军, 李威, 李波, 等. 电流频谱Hurst指数在串联电弧故障检测中的应用[J]. 控制理论与应用, 2022, 39(3):561-569. JIANG J, LI W, LI B, et al. Application of the Hurst index of current frequency spectrum in series arc fault detection[J]. Control Theory & Applications, 2022, 39(3):561-569. (in Chinese)
[7]龙官微, 穆海宝, 张大宁, 等. 基于多特征融合神经网络的串联电弧故障识别技术[J]. 高电压技术, 2021, 47(2):463-471. LONG G W, MU H B, ZHANG D N, et al. Series arc fault detection technology based on multi-feature fusion neural network[J]. High Voltage Engineering, 2021, 47(2):463-471. (in Chinese)
[8]苏晶晶, 许志红. 基于混沌分形理论的故障电弧诊断方法研究[J]. 电机与控制学报, 2021, 25(3):125-133. SU J J, XU Z H. Arc fault diagnosis method based on chaos and fractal theories[J]. Electric Machines and Control, 2021, 25(3):125-133. (in Chinese)
[9]ZHAO H J, LIU J P, LOU J C. Series arc fault detection based on current fluctuation and zero-current features[J]. Electric Power Systems Research, 2022, 202:107626.
[10]GONG Q Y, PENG K, WANG W, et al. Series arc fault identification method based on multi-feature fusion[J]. Frontiers in Energy Research, 2022, 9:824414.
[11]JIANG J, LI W, WEN Z, et al. Series arc fault detection based on random forest and deep neural network[J]. IEEE Sensors Journal, 2021, 21(15):17171-17179.
[12]PARK C J, DANG H L, KWAK S, et al. Detection algorithms of parallel arc fault on AC power lines based on deep learning techniques[J]. Journal of Electrical Engineering & Technology, 2022, 17(2):1195-1205.
[13]张婷, 王海淇, 张认成, 等. 基于自归一化神经网络的电弧故障检测方法[J]. 仪器仪表学报, 2021, 42(3):141-149. ZHANG T, WANG H Q, ZHANG R C, et al. An arc fault detection method based on the self-normalized convolutional neural network[J]. Chinese Journal of Scientific Instrument, 2021, 42(3):141-149. (in Chinese)
[14]ZHANG S, QU N, ZHENG T F, et al. Series arc fault detection based on wavelet compression reconstruction data enhancement and deep residual network[J]. IEEE Transactions on Instrumentation and Measurement, 2022, 71:3508409.
[15]HU C Q, QU N, ZHANG S. Series arc fault detection based on continuous wavelet transform and DRSN-CW with limited source data[J]. Scientific Reports, 2022, 12(1):12809.
[16]林靖怡, 王尧, 李奎, 等. 基于自组织特征映射网络的电弧故障检测方法[J]. 电力自动化设备, 2020, 40(8):210-216. LIN J Y, WANG Y, LI K, et al. Arc fault detection method based on self-organizing feature mapping network[J]. Electric Power Automation Equipment, 2020, 40(8):210-216. (in Chinese)
[17]LI W C, LIU Y L, LI Y, et al. Series arc fault diagnosis and line selection method based on recurrent neural network[J]. IEEE Access, 2020, 8:177815-177822.
[18]WANG Y, HOU L M, PAUL K C, et al. ArcNet:Series AC arc fault detection based on raw current and convolutional neural network[J]. IEEE Transactions on Industrial Informatics, 2022, 18(1):77-86.
[19]BENGIO Y, SIMARD P, FRASCONI P. Learning long-term dependencies with gradient descent is difficult[J]. IEEE Transactions on Neural Networks, 1994, 5(2):157-166.
[20]中华人民共和国国家质量监督检验检疫总局, 中国国家标准化管理委员会. 电弧故障保护电器(AFDD)的一般要求:GB/T 31143-2014[S]. 北京:中国标准出版社, 2014. General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China, Standardization Administration of the People's Republic of China. General requirements for arc fault protection detection devices (AFDD):GB/T 31143-2014[S]. Beijing:Standards Press of China, 2014. (in Chinese)
[21]易清明, 吕人毅, 石敏, 等. 融合多尺度空洞卷积与反卷积的轻量化目标检测[J]. 华南理工大学学报(自然科学版), 2022, 50(12):41-48. YI Q M, Lü R Y, SHI M, et al. Lightweight object detection combined with multi-scale dilated convolution and multi-scale deconvolution[J]. Journal of South China University of Technology (Natural Science Edition), 2022, 50(12):41-48. (in Chinese)
[22]颜铭靖, 苏喜友. 基于三维空洞卷积残差神经网络的高光谱影像分类方法[J]. 光学学报, 2020, 40(16):163-172. YAN M J, SU X Y. Hyperspectral image classification based on three-dimensional dilated convolutional residual neural network[J]. Acta Optica Sinica, 2020, 40(16):163-172. (in Chinese)
[23]江国乾, 周俊超, 武鑫, 等. 基于空洞因果卷积网络的风电机组异常检测[J]. 太阳能学报, 2023, 44(5):368-375.[JP+2]JIANG G Q, ZHOU J C, WU X, et al. Wind turbine anomaly detection based on dilated causal convolution network[J], Acta Energiae Solaris Sinica, 2023, 44(5):368-375. (in Chinese)
[24]KAYHAN O S, VAN GEMERT J C. On translation invariance in CNNs:Convolutional layers can exploit absolute spatial location[C]//2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). Seattle, USA:IEEE, 2020:14262-14273.
[25]KLAMBAUER G, UNTERTHINER T, MAYR A, et al. Self-normalizing neural networks[C]//Advances in Neural Information Processing Systems, California, USA:NIPS, 2017:971-980.
[26]HE K M, ZHANG X Y, REN S Q, et al. Deep residual learning for image recognition[C]//2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). Las Vegas, USA:IEEE, 2016:770-778.
[27]中华人民共和国国家质量监督检验检疫总局, 中国国家标准化管理委员会. 电气火灾监控系统第4部分:故障电弧探测器:GB 14287.4-2014[S]. 北京:中国标准出版社, 2014. General Administration of Quality Supervision, Inspection and Quarantine of the People's Republic of China, Standardization Administration of the People's Republic of China. Electrical fire monitoring systems. Part 4:Arcing fault detectors:GB 14287.4-2014[S]. Beijing:Standards Press of China, 2014. (in Chinese)
[28]KINGMA D P, BA J. Adam:A method for stochastic optimization[C/OL]. (2015-05-09)[2023-02-05]. https://dare.uva.nl/search?identifier=a20791d3-1aff-464a-8544-268383c33a75.