Low voltage AC series arc fault detection method based on parallel deep convolutional neural network

2nd International Symposium on Application of Materials Science and Energy Materials

材料科学;能源学

Yu, Qiongfang^1^2 ; Huang, Gaolu^1 ; Yang, Yi^1

School of Electrical Engineering and Automation, Henan Polytechnic University, Jiaozuo Henan, China^1

Postdoctoral Programme of Beijing Research Institute, Dalian University of Technology, Beijing, China^2

关键词: Connection mode;    Convolutional neural network;    Detection accuracy;    Low voltages;    Normal operations;    Percentage points;    Series arc-fault;    Training data sets;   

Others  :  https://iopscience.iop.org/article/10.1088/1757-899X/490/7/072020/pdf
DOI  :  10.1088/1757-899X/490/7/072020

学科分类：材料科学（综合）

来源: IOP

【 摘 要 】

The complexity and concealment of series arc fault threaten the safety of household power supply system. Detection of series arc faults using threshold and characteristics extracted from current or voltage may affected by load type and connection mode. Based on AlexNet, a new parallel deep convolutional neural network detection method is proposed in this paper. The series arc fault and normal operation current signal data of three types of load totally 7200 sets were collected respectively. Using the collected data, a training data set and four test data sets were constructed to train and test the proposed convolutional neural network. The experimental results show that the detection accuracy of parallel AlexNet is higher than that of AlexNet, with a maximum of 16.75 percentage points. The stability of parallel AlexNet is about twice as high as that of AlexNet, and the maximum is close to four times.