A CNN recognition method for early stage of 10 kV single core cable based on sheath current

• Published in: Electric power systems research Vol. 184; p. 106292
• Main Authors: Chi, Peng, Zhang, Zhe, Liang, Rui, Cheng, Cheng, Chen, Shaokang
• Format: Journal Article
• Language: English
• Published: Amsterdam Elsevier B.V 01.07.2020; Elsevier Science Ltd
• Subjects: 10 kV single core cable; Artificial intelligence; Artificial neural networks; CNN; Computer simulation; Electric currents; Electricity generation; Feature recognition; Mutation; Mutation information; Neural networks; Sheath current; Sheaths; Simulation; State recognition; Waveforms; Wavelet transform; Wavelet transforms; Windows (intervals); Sheath current; Mutation information; CNN; Wavelet transform; State recognition; 10 kV single core cable
• ISSN: 0378-7796; 1873-2046
• DOI: 10.1016/j.epsr.2020.106292

•Analysis of early state recognition of 10 kV cable from the angle of sheath current.•The mutation information in sheath current as the research subject.•Construct the state recognition matrix by waveform and energy characteristics.•A deep convolutional network adapted to cable state recognition.•Four simulation models reflecting cable early state are performed in PSCAD software. Traditional analysis of cable early state recognition is mainly based on one or several threshold values of electrical characteristics, but the calculation of threshold is often affected by measurement accuracy and external disturbance, which inevitably reduces recognition accuracy. The development of artificial intelligence provides a new way to solve this problem. This paper presents a deep convolutional neural network (CNN) recognition method for early state of 10 kV single core cable based on sheath current. Firstly, waveform and energy characteristics which are extracted from the mutation information of sheath current by wavelet transform, are used to construct cable state recognition matrix. The mutational signal is detected by the cumulative sum (CU-SUM) method and intercepted by a set time window. Secondly, a 7-layer deep CNN is constructed according to the features of recognition matrix. Then the CNN model is trained by the adaptive moment estimation (Adam) method to get the recognition model of cable state. Finally, the proposed method is used to recognize cable early state by large number of samples which are obtained from the simulation of four cable states with PSCAD software. Compared with other methods, the results of simulation demonstrate that the proposed method has a high recognition accuracy.