Biological Electric-Shock Fault Identification Method Based on Multi-Feature Optimization Selection under Unbalanced Small Sample
Gao Wei1,2, Rao Junmin1, Quan Shengxin1, Guo Moufa1,2
1. College of Electrical Engineering and Automation Fuzhou University Fuzhou 350108 China; 2. Fujian Province University Engineering Research Center of Smart Distribution Grid Equipment Fuzhou 350108 China
Abstract：The existing residual current device (RCD) operates based on the amplitude of the residual current, but if the threshold is not reasonably set, the RCD is prone to reject or misoperate. Therefore, identifying biological electric-shock faults from grounding faults is a crucial approach. Current research only selects one or several features without following proper feature selection rules. Furthermore, machine learning methods require a certain number of samples to train the model to ensure algorithm accuracy and stability. However, obtaining a large number of biological electric-shock samples is challenging during actual experiments, and the algorithm model cannot learn the waveform in real settings. To solve the above problems, a biological electric-shock fault identification method based on multi-feature optimization selection under unbalanced small samples is proposed. Firstly, variational auto-encoders (VAE) is adopted to multiply the electric-shock small sample data collected by experiments to achieve positive and negative sample balance. Due to the complexity and danger of the scenes, it is difficult to obtain the actual electric-shock samples. The problem of small samples will lead to low accuracy and poor effectiveness of the training model, and the unbalanced samples will lead to deviations in the prediction results of the model, resulting in poor identification accuracy of a few types of samples. Therefore, a few samples are enhanced by introducing VAE to improve the effectiveness of the model. Secondly, 23 features which can reflect the dynamic characteristics of the waveform are extracted in time domain, the optimal expression feature group is selected from them by Gaussian kernel Fisher discriminant analysis (GKFDA) and maximal information coefficient (MIC). Through data analysis, various index features can be extracted from the changing forms of biological electric-shock waveforms. The addition of high-quality features will improve the diagnostic accuracy of the classifier to a certain extent, but the introduction of bad and redundant features will increase the running time of the algorithm and reduce the diagnostic accuracy of the classifier. Therefore, GKFDA and MIC are combined to perform feature scoring for each feature, and the optimal expression feature group is selected intuitively and independently based on the scoring results, which could improve the feature quality and reflect the regularity of feature selection. Finally, a forgetting-factor-based online sequential extreme learning machine (FOS-ELM) algorithm is investigated to identify the electric-shock behavior. There are abundant electric-shock scenes in the real environments. The escape behaviors of living objects during electric shock will have a great influence on the electric-shock waveform, which makes it difficult for the traditional off-line classifier to have adaptability. The online sequential extreme learning machine (OS-ELM) has an online learning mechanism that allows online updates for new samples without the historical data. The forgetting factor is introduced to form FOS-ELM, aiming to further solve the shortcoming of slow learning speed of OS-ELM, so that it can quickly adapt to changes of environmental samples with higher learning efficiency. The experimental data of conventional grounding fault and biological electric-shock fault in 12 scenes were collected for the verification of the proposed algorithm. The results show that the diagnosis accuracy of the proposed model can reach 98.75%, among which all 40 conventional grounding fault samples are correctly judged with an accuracy of 100%, while only 1 of 40 actual biological electric-shock fault samples is wrong with an accuracy of 97.5%. From the perspective of time, the average online learning time is 1.378 ms, and the average diagnosis time is only 1.33 ms.
[1] Mitolo M.Shock hazard in the presence of protective residual-current devices[J]. IEEE Transactions on Industry Applications, 2010, 46(4): 1552-1557. [2] 韩晓慧, 杜松怀, 苏娟, 等. 触电信号暂态特征提取及故障类型识别方法[J]. 电网技术, 2016, 40(11): 3591-3596. Han Xiaohui, Du Songhuai, Su Juan, et al.Fault transient feature extraction and fault type identification for electrical shock signals[J]. Power System Technology, 2016, 40(11): 3591-3596. [3] 关海鸥, 杜松怀, 苏娟, 等. 一种触电信号的自动快速检测模型[J]. 电网技术, 2013, 37(8): 2328-2335. Guan Haiou, Du Songhuai, Su Juan, et al.An automatic and quick detection model of electric shock signals[J]. Power System Technology, 2013, 37(8): 2328-2335. [4] 刘永梅, 杜松怀, 盛万兴. 基于SVM-神经网络融合反馈的触电电流检测方法[J]. 电网技术, 2020, 44(5): 1972-1977. Liu Yongmei, Du Songhuai, Sheng Wanxing.Prediction method of electric shock current based on SVM and neural network fusion feedback[J]. Power System Technology, 2020, 44(5): 1972-1977. [5] 赵启承, 虞雁凌. 基于长短期记忆神经网络的生命体触电电流检测[J]. 传感器与微系统, 2022, 41(1): 142-145. Zhao Qicheng, Yu Yanling.Detection of electric shock current of living body based on long-term and short-term memory neural network[J]. Transducer and Microsystem Technologies, 2022, 41(1): 142-145. [6] 王金丽, 刘永梅, 杜松怀, 等. 基于剩余电流固有模态能量特征的生物触电故障诊断模型[J]. 农业工程学报, 2016, 32(21): 202-208. Wang Jinli, Liu Yongmei, Du Songhuai, et al.Bioelectric shock fault diagnosis model based on residual current natural mode energy characteristics[J]. Transactions of the Chinese Society of Agricultural Engineering, 2016, 32(21): 202-208. [7] Wei J C J, Edwards G A, Martin D J, et al. Allometric scaling of skin thickness, elasticity, viscoelasticity to mass for micro-medical device translation: from mice, rats, rabbits, pigs to humans[J]. Scientific Reports, 2017, 7: 15885. [8] Benias P C, Wells R G, Sackey-Aboagye B, et al.Structure and distribution of an unrecognized inter-stitium in human tissues[J]. Scientific Reports, 2018, 8: 4947. [9] Pritchard G.A new model for human body impedance[J]. IEEE Transactions on Power Delivery, 2022, 37(2): 955-959. [10] 黄冬梅, 吴志浩, 孙园, 等. 基于VAE预处理和RP-2D CNN的不平衡负荷数据类型辨识方法[J]. 电力系统及其自动化学报, 2022, 34(10): 66-72, 80. Huang Dongmei, Wu Zhihao, Sun Yuan, et al.Data type identification method for imbalanced load based on VAE preprocessing and RP-2D CNN[J]. Proceedings of the CSU-EPSA, 2022, 34(10): 66-72, 80. [11] Aboalayon K, Faezipour M, Almuhammadi W, et al.Sleep stage classification using EEG signal analysis: a comprehensive survey and new investigation[J]. Entropy, 2016, 18(9): 272. [12] 王毅, 刘黎明, 李松浓, 等. 基于经验小波变换复合熵值与特征融合的故障电弧检测[J]. 电网技术, 2023, 47(5): 1912-1919. Wang Yi, Liu Liming, Li Songnong, et al.Arc fault detection based on empirical wavelet transform composite entropy and feature fusion[J]. Power System Technology, 2023, 47(5): 1912-1919. [13] 刘树鑫, 宋健, 刘洋, 等. 交流接触器触头系统运动分析及故障诊断研究[J]. 电工技术学报, 2021, 36(增刊2): 477-486. Liu Shuxin, Song Jian, Liu Yang, et al.Research on motion analysis and fault diagnosis of contact system of AC contactor[J]. Transactions of China Electrotechnical Society, 2021, 36(S2): 477-486. [14] 崔芮华, 张振, 佟德栓, 等. 基于改进经验小波变换多特征融合的航空交流串联电弧故障检测[J]. 电工技术学报, 2022, 37(12): 3148-3161. Cui Ruihua, Zhang Zhen, Tong Deshuan, et al.Aviation AC series arc fault detection based on improve empirical wavelet transform multi-feature fusion[J]. Transactions of China Electrotechnical Society, 2022, 37(12): 3148-3161. [15] 张立石, 梁得亮, 刘桦, 等. 基于小波变换与逻辑斯蒂回归的混合式配电变压器故障辨识[J]. 电工技术学报, 2021, 36(增刊2): 467-476. Zhang Lishi, Liang Deliang, Liu Hua, et al.Fault identification of hybrid distribution transformer based on wavelet transform and logistic regression[J]. Transactions of China Electrotechnical Society, 2021, 36(S2): 467-476. [16] 崔芮华, 李泽, 佟德栓. 基于三维熵距和熵空间的航空电弧故障检测与分类技术[J]. 电工技术学报, 2021, 36(4): 869-880. Cui Ruihua, Li Ze, Tong Deshuan.Arc fault detection and classification based on three-dimensional entropy distance and entropy space in aviation power system[J]. Transactions of China Electrotechnical Society, 2021, 36(4): 869-880. [17] 王俊, 罗嘉玮, 冯宗琮, 等. 基于近似熵的中性点不接地系统故障区段定位[J]. 电工技术, 2021(3): 33-36. Wang Jun, Luo Jiawei, Feng Zongcong, et al.Fault section location of neutral ungrounded system based on approximate entropy[J]. Electric Engineering, 2021(3): 33-36. [18] 王来, 樊重俊, 杨云鹏, 等. 面向不平衡数据分类的KFDA-Boosting算法[J]. 计算机应用研究, 2019, 36(3): 807-811. Wang Lai, Fan Chongjun, Yang Yunpeng, et al.KFDA-Boosting algorithm oriented to imbalanced data classification[J]. Application Research of Computers, 2019, 36(3): 807-811. [19] 徐华电, 苏建徽, 张军军, 等. 基于谐波特征与核Fisher判别分析的孤岛检测方法研究[J]. 电工技术学报, 2016, 31(3): 25-30. Xu Huadian, Su Jianhui, Zhang Junjun, et al.Research on islanding detection based on harmonic char-acteristics and kernel Fisher discriminant analysis[J]. Transactions of China Electrotechnical Society, 2016, 31(3): 25-30. [20] 李滨, 高枫. 基于虚拟相似日与DA-LSTPNet的地区电网短期负荷预测[J]. 电力系统自动化, 2021, 45(22): 55-64. Li Bin, Gao Feng.Short-term load forecasting for regional power grid based on virtual similar days and dual-stage attention-based long and short time pattern network[J]. Automation of Electric Power Systems, 2021, 45(22): 55-64. [21] 吉兴全, 张朔, 张玉敏, 等. 基于IELM算法的配电网故障区段定位[J]. 电力系统自动化, 2021, 45(22): 157-166. Ji Xingquan, Zhang Shuo, Zhang Yumin, et al.Fault section location for distribution network based on improved electromagnetism-like mechanism algori-thm[J]. Automation of Electric Power Systems, 2021, 45(22): 157-166. [22] 张秦梫, 宋辉, 姜勇, 等. 基于OS-ELM的变压器局部放电模式识别[J]. 高电压技术, 2018, 44(4): 1122-1130. Zhang Qinqin, Song Hui, Jiang Yong, et al.Partial discharge pattern recognition of transformer based on OS-ELM[J]. High Voltage Engineering, 2018, 44(4): 1122-1130. [23] 李扬, 李国庆, 顾雪平, 等. 基于集成OS-ELM的暂态稳定评估方法[J]. 电工技术学报, 2015, 30(14): 412-418. Li Yang, Li Guoqing, Gu Xueping, et al.Transient stability assessment of power systems based on ensemble OS-ELM[J]. Transactions of China Electrotechnical Society, 2015, 30(14): 412-418. [24] 郭威, 于建江, 汤克明, 等. 动态数据流分析的在线超限学习算法综述[J]. 计算机科学, 2019, 46(4): 1-7. Guo Wei, Yu Jianjiang, Tang Keming, et al.Survey of online sequential extreme learning algorithms for dynamic data stream analysis[J]. Computer Science, 2019, 46(4): 1-7. [25] 郭亮, 王祥业, 姜文聪. 基于生成对抗网络的注入电流式热声成像逆问题研究[J]. 电工技术学报, 2021, 36(增刊1): 22-30. Guo Liang, Wang Xiangye, Jiang Wencong.The study on the inverse problem of applied current thermo-acoustic imaging based on generative adversarial network[J]. Transactions of China Electrotechnical Society, 2021, 36(S1): 22-30. [26] 徐佳宁, 倪裕隆, 朱春波. 基于改进支持向量回归的锂电池剩余寿命预测[J]. 电工技术学报, 2021, 36(17): 3693-3704. Xu Jianing, Ni Yulong, Zhu Chunbo.Remaining useful life prediction for lithium-ion batteries based on improved support vector regression[J]. Transactions of China Electrotechnical Society, 2021, 36(17): 3693-3704. [27] 关海鸥, 刘梦, 李春兰, 等. 基于小波包变换和量子神经网络的触电故障类型识别模型[J]. 农业工程学报, 2018, 34(5): 183-190. Guan Haiou, Liu Meng, Li Chunlan, 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. [28] 蔡智萍, 郭谋发, 魏正峰. 基于BP神经网络的低压配电网生命体触电识别方法研究[J]. 电网技术, 2022, 46(4): 1614-1623. Cai Zhiping, Guo Moufa, Wei Zhengfeng.Research on recognition method of living body shock in low-voltage distribution network based on BP neural network[J]. Power System Technology, 2022, 46(4): 1614-1623.
2024, Vol. 39 
