Page 91 - 《应用声学》2025年第1期
P. 91

第 44 卷 第 1 期          方春华等: 高压电缆终端铅封缺陷超声图像卷积神经网络识别                                           87


                 Zhou Changcheng, Zhao Haijun, Ren Ping.  Improve-  Fang Chunhua, Hu Dongsan, Tao Yuning, et al. Ultra-
                 ment on lead sealing technology of high-voltage cable ac-  sonic phased array flexible coupling detection of lead-
                 cessories[J]. Northeast Electric Power Technology, 2021,  sealing defects in high voltage cable terminals[J]. High
                 42(2): 23–26, 29.                                 Voltage Engineering, 2022, 48(1): 29–37.
              [6] 鲁宁. 高压 XLPE 电缆击穿的制造因素分析及有效控制 [J].             [16] 方春华, 胡冻三, 郭凯歌, 等. 高压电缆终端铅封缺陷超声检
                 电力工程技术, 2017, 36(4): 43–47, 58.                   测方法研究 [J]. 中国测试, 2022, 48(3): 118–123.
                 Lu Ning. Manufacture factor analysis and effective con-  Fang Chunhua, Hu Dongsan, Guo Kaige, et al. Research
                 trol of high voltage XLPE cable breakdown[J]. Electric  on ultrasonic testing method for lead sealing defects of
                 Power Engineering Technology, 2017, 36(4): 43–47, 58.  high voltage cable terminals[J]. China Measurement &
              [7] 罗俊华, 邱毓昌, 杨黎明. 10 kV 及以上电力电缆运行故障统                Test, 2022, 48(3): 118–123.
                 计分析 [J]. 高电压技术, 2003, 29(6): 14–16.            [17] Masci J, Meier U, Ciresan D, et al. Steel defect clas-
                 Luo Junhua, Qiu Yuchang, Yang Liming. Operation fault  sification with max-pooling convolutional neural net-
                 analysis of CLPE power cable above 10 kV[J]. High Volt-
                                                                   works[C]//The 2012 International Joint Conference on
                 age Engineering, 2003, 29(6): 14–16.
                                                                   Neural Networks (IJCNN). June 10–15, 2012. Brisbane,
              [8] 曹京荥, 查显光, 陈杰, 等. 220 kV 电缆线路终端故障及仿真
                                                                   Australia. IEEE, 2012.
                 分析 [J]. 电力工程技术, 2018, 37(6): 151–155.
                                                                [18] Faghih-Roohi S, Hajizadeh S, Nunez A, et al. Deep con-
                 Cao Jingying, Zha Xianguang, Chen Jie, et al. Terminal
                                                                   volutional neural networks for detection of rail surface de-
                 fault and simulation analysis of 220 kV cable line[J]. Elec-
                                                                   fects[C]//2016 International Joint Conference on Neural
                 tric Power Engineering Technology, 2018, 37(6): 151–155.
                                                                   Networks (IJCNN). July 24-29, 2016. Vancouver, BC,
              [9] 杨金海, 刘雪锋, 余磊. 高压电缆铅封裂纹缺陷检测方法研
                                                                   Canada. IEEE, 2016.
                 究 [J]. 大众用电, 2021, 36(8): 41–42.
                                                                [19] 马海辉, 余小玲, 吕倩, 等. 一维卷积神经网络在往复式压缩机
             [10] 谭杨宝. 110 kV 电缆终端故障分析及防范措施 [J]. 中国高新
                                                                   气阀故障诊断中的应用 [J]. 西安交通大学学报, 2022, 56(4):
                 科技, 2018(15): 105–107.
                                                                   101–108.
             [11] 吴广宁. 电气设备状态监测的理论与实践 [M]. 北京: 清华大
                                                                   Ma Haihui, Yu Xiaoling, Lyu Qian, et al. Application of
                 学出版社, 2005: 190–207.
                                                                   one-dimensional convolutional neural network in fault di-
             [12] 陈海燕, 李亮, 夏正武, 等. 复合绝缘子内部缺陷超声相控阵
                                                                   agnosis of reciprocating compressor air valve[J]. Journal
                 柔性耦合检测 [J]. 高电压技术, 2019, 45(4): 1274–1280.
                                                                   of Xi’an Jiaotong University, 2022, 56(4): 101–108.
                 Chen Haiyan, Li Liang, Xia Zhengwu, et al. Inspection
                                                                [20] 常海涛, 苟军年, 李晓梅. Faster R-CNN 在工业 CT 图像
                 of internal defects of composite insulators by ultrasonic
                 phased array method based on flexible coupling[J]. High  缺陷检测中的应用 [J]. 中国图象图形学报, 2018, 23(7):
                 Voltage Engineering, 2019, 45(4): 1274–1280.      1061–1071.
             [13] 马君鹏, 孙兴涛, 李硕, 等. 基于超声导波的盆式绝缘子缺陷                  Chang Haitao, Gou Junnian, Li Xiaomei. Application
                                                                   of faster R-CNN in image defect detection of indus-
                 检测及定位 [J]. 高电压技术, 2019, 45(12): 3941–3948.
                 Ma Junpeng, Sun Xingtao, Li Shuo, et al. Detection and  trial CT[J]. Journal of Image and Graphics, 2018, 23(7):
                 location for defects of basin-type insulator based on ul-  1061–1071.
                 trasonic guided wave[J]. High Voltage Engineering, 2019,  [21] 常亮, 邓小明, 周明全, 等. 图像理解中的卷积神经网络 [J]. 自
                 45(12): 3941–3948.                                动化学报, 2016, 42(9): 1300–1312.
             [14] 邓红雷, 何战峰, 陈力. 复合绝缘子脱粘缺陷的超声导波检                    Chang Liang, Deng Xiaoming, Zhou Mingquan, et al.
                 测 [J]. 高电压技术, 2019, 45(1): 196–202.               Convolutional neural networks in image understanding[J].
                 Deng Honglei, He Zhanfeng, Chen Li.  Inspection of  Acta Automatica Sinica, 2016, 42(9): 1300–1312.
                 debonding defects of composite insulators by ultrasonic  [22] 卓维, 张磊. 深度神经网络的快速学习算法 [J]. 嘉应学院学
                 guide wave[J]. High Voltage Engineering, 2019, 45(1):  报, 2014, 32(5): 13–17.
                 196–202.                                          Zhuo Wei, Zhang Lei. A fast learning algorithm for deep
             [15] 方春华, 胡冻三, 陶玉宁, 等. 高压电缆终端铅封缺陷超声相                  neural network[J]. Journal of Jiaying University, 2014,
                 控阵柔性耦合检测 [J]. 高电压技术, 2022, 48(1): 29–37.          32(5): 13–17.
   86   87   88   89   90   91   92   93   94   95   96