Electric shock identification method based on probabilistic neural network and wavelet analysis
1. College of Mechanical and Electrical Engineering, Xinjiang Agricultural University, Urumqi, Xinjiang 830052, China; 2. College of Mechanical and Electrical Engineering, Xinjiang Vocational University, Urumqi, Xinjiang 830013, China
Abstract: To solve the problem that the existing residual current protection device was difficult to identify electric shock accidents, a new method of electric shock identification based on wavelet high-frequency distribution characteristics and probabilistic neural network(PNN) was proposed. The S-transform was used to analyze the spectral characteristics of the residual current signal including the moment of electric shock, which was found that the high-frequency component of the residual current signal at the time of electric shock had sudden change in amplitude. The wavelet highfrequency distribution of each layer was extracted through the multi-scale frequency window provided by the wavelet multi-resolution analysis. To describe the electric shock accident, the normalized processing of each layer of wavelet highfrequency distribution mutations was used to accumulate and quantify the high-frequency characteristics of the first 5 layers of the residual current signal. Taking full account of the randomness of the time of electric shock accidents, the extracted wavelet features were divided into categories. A PNNbased electric shock accident recognition model was constructed, and the network smoothing parameters in the defined domain were optimized according to the specified step. The mean clustering method was also used to optimize the network structure. The results show that there is significantly sudden change in the amplitude of the residual current signal above 500 Hz at the time of electric shock. After normalization, the cumulative sum of the amplitude mutations of the wavelet high-frequency distribution of each layer can well describe the amplitude sudden increase in the corresponding stage of the wavelet high-frequency distribution of each layer. The optimal smoothing parameter interval of the established PNN network model is from 0.15 to 0.29, and the best corresponding identification rate of electric shock accidents is 95.5%.
李春兰1, 罗杰1, 石砦1, 王长云1, 叶豪2, 王海杨2, 王静1. 基于小波分析和概率神经网络的触电事故识别方法[J]. 江苏大学学报(自然科学版), 2023, 44(1): 75-81.
LI Chunlan1, LUO Jie1, SHI Zhai1, WANG Changyun1, YE Hao2, WANG Haiyang2, WANG Jing1. Electric shock identification method based on probabilistic neural network and wavelet analysis[J]. Journal of Jiangsu University(Natural Science Eidtion)
, 2023, 44(1): 75-81.
LI C L, DU S H, LIAO C Y. Impacts of distribution network unbalanced impedance on action performances of residual current operated device[C]∥Proceedings of the 2008 3rd International Conference on Deregulation and Restructuring and Power Technologies, doi: 10.1109/DRPT.2008.4523757.
LI C L, YE H, WANG C B, et al. Human electric shock law based on pig electric shock experiment[J]. Journal of Jiangsu University (Natural Science Edition), 2019, 40(5): 553-558,565. (in Chinese)
LI C L, XIA L L, WANG C B, et al. Study on chaos detection method of electric shock current based on elliptic domain segmentation[J]. Power System Protection and Control, 2017, 45(15):69-76. (in Chinese)
GUAN H O, LI W K, DU S H, et al. Detection model of biological electric shock current based on HilbertHuang transform[J]. Transactions of the Chinese Society of Agricultural Engineering, 2017, 33(14): 202-209. (in Chinese)
HAN X H, DU S H, SU J, et al. Fault transient feature extraction and fault type identification for electrical shock signals[J]. Power System Technology, 2016, 40(11): 3591-3596. (in Chinese)
LIU Y M, DU S H, SHENG W X. Prediction method of electric shock current based on SVM and neural network fusion feedback[J]. Power System Technology, 2020, 44(5): 1972-1977. (in Chinese)
GUAN H O, LIU M, LI C L, et al. Classification recognition model of electric shock fault based on wavelet packet transformation and quantum neural network[J]. Transactions of the Chinese Society of Agricultural Engineering, 2018, 34(5):183-190. (in Chinese)
LU Q W, WANG T, LI Z R, et al. Detection method of series arcing fault based on wavelet transform and singular value decomposition[J]. Transactions of China Electrotechnical Society, 2017,32(17):208-217. (in Chinese)
[9]
MISHRA D P, SAMANTARAY S R, JOOS G. A combined wavelet and datamining based intelligent protection scheme for microgrid[J]. IEEE Transactions on Smart Grid, 2016, 7(5): 2295-2304.
[10]
HAMMAMI A. Uncertainty principles for the HankelGabor transform[J]. Indian Journal of Pure and Applied Mathematics, 2020, 51(1):251-264.
LIU Q Q, ZHENG C M, HE H K, et al. A variable step size LMS speech denoising algorithm based on wavelet threshold[J]. Journal of Jilin University (Science Edition), 2022, 60(4):943-949. (in Chinese)
ZHANG Y, MA C Z, YANG P, et al. Face feature extraction algorithm based on wavelet transform and improved principal component analysis[J]. Journal of Jilin University (Science Edition), 2021, 59(6): 1499-1503. (in Chinese)
[13]
ZHANG X L, WANG Z L, LU C. Fault diagnosis of gearbox based on the nonlinear output frequency response functions and PNN[J]. Vibroengineering Procedia, 2018, 19:71-75.
[14]
WANG H H, WANG P, LIU T. Power quality disturbance classification using the stransform and probabilistic neural network[J]. Energies, doi:10.3390/en10010107.
[15]
CHEN L, HAN W, HUANG Y H, et al. Online fault diagnosis for photovoltaic modules based on probabilistic neural network[J]. European Journal of Electrical Engineering, 2019, 21(3): 317-325.