基于深度學(xué)習(xí)的變電站六氟化硫儀表智能識別方法
陳如風(fēng)1，吳翊穎1，林燁2，蘭茜2，張皓駿2
(1 國網(wǎng)福建省電力有限公司福州供電公司，福建 福州 350004；
2 福州大學(xué) 電氣工程與自動化學(xué)院，福建 福州 350108)
摘 要：變電站的現(xiàn)場因受污漬、光照、拍攝角度等干擾因素影響，遠(yuǎn)程巡視系統(tǒng)拍攝圖像中的儀表信息弱化，指針信息缺失，導(dǎo)致儀表的識別準(zhǔn)確率較低。針對該問題，采用了 YOLOv5 的目標(biāo)檢測框架，設(shè)計了交叉融合的特征金字塔網(wǎng)絡(luò)，增強(qiáng)了基于 YOLOv5 的目標(biāo)檢測網(wǎng)絡(luò)對儀表位置信息的提取能力；針對儀表圖像模糊、傾斜等導(dǎo)致指針割裂問題，采用 U-Net 語義分割網(wǎng)絡(luò)來識別儀表圖像中的指針，實現(xiàn)了干擾環(huán)境下的儀表指針生成。實驗表明，提出的基于深度學(xué)習(xí)的六氟化硫(SF6)儀表智能識別算法在變電站復(fù)雜環(huán)境中表現(xiàn)出了較強(qiáng)的識別能力，指針識別準(zhǔn)確率由原來的63%提升至96%。
關(guān)鍵詞： 變電站；遠(yuǎn)程巡視；深度學(xué)習(xí)；六氟化硫儀表；智能識別；交叉融合；特征金字塔網(wǎng)絡(luò)
中圖分類號：TM63 ；TM764.1 文獻(xiàn)標(biāo)識碼：B 文章編號：1007-3175(2025)10-0061-05
Intelligent Recognition Method for SF6 Instrument in
Substation Based on Deep Learning
CHEN Ru-feng1, WU Yi-ying1, LIN Ye2, LAN Xi2, ZHANG Hao-jun2
（1 Fuzhou Power Supply Company, State Grid Fujian Electric Power Co., Ltd, Fuzhou 350004, China;
2 College of Electrical Engineering and Automation, Fuzhou University, Fuzhou 350108, China）
Abstract: Due to the influence of interference factors such as stains, light, and shooting angles at the substation site, the instrument information in the images captured by the remote inspection system is weakened, and the pointer information is missing, resulting in a relatively low recognition accuracy of the instruments. Aiming at this problem, the target detection framework of YOLOv5 was adopted, and a cross-fusion feature pyramid network was designed, which enhanced the ability of the target detection network based on YOLOv5 to extract the position information of instruments. To address the problem of pointer fragmentation caused by blurred and tilted instrument images, the U-Net semantic segmentation network is adopted to identify the pointers in the instrument images, achieving the generation of instrument pointers in the interference environment. Experiments show that the proposed intelligent recognition algorithm for sulfur hexafluoride (SF6) instrument based on deep learning has demonstrated strong recognition capabilities in the complex environment of substations, with the accuracy rate of pointer recognition increasing from the original 63% to 96%.
Key words: substation; remote inspection; deep learning; SF6 instrument; intelligent recognition; cross-fusion; feature pyramid network
參考文獻(xiàn)
[1] 丁思奎，李健. 變電站巡檢機(jī)器人應(yīng)用中存在的問題分析及解決方案[J]. 電工電氣，2016(2) ：57-58.
[2] 余福. 基于深度學(xué)習(xí)的變電站指針儀表讀數(shù)識別方法研究[D]. 吉林：東北電力大學(xué)，2022.
[3] 付世雄. 基于圖像處理的指針式儀表示數(shù)識別[D]. 哈爾濱：哈爾濱工業(yè)大學(xué)，2022.
[4] 錢玉寶，王紫涵，邱騰煌. 指針式儀表讀數(shù)識別的研究現(xiàn)狀與發(fā)展[J] . 電子測量技術(shù)，2024，47(8) ：110-119.
[5] 荊永菊，薄樹奎，郝曉玉，等. 基于圖像處理的指針式儀表讀數(shù)識別[J] . 信息技術(shù)與信息化，2024(12) ：27-31.
[6] LIN Ye, XU Zhezhuang, WU Yiying, et al.A multitask network for occluded meter reading with synthetic data generation technology[J].Advanced Engineering Informatics，2025，64 ：51-64.
[7] ZHOU Chuanhua, ZHOU Jiayi, YU Cai, et al.Multichannel sliced deep RCNN with residual network for text classification[J].Chinese Journal of Electronics，2020，29(5) ：880-886.
[8] 南曉虎，丁雷. 深度學(xué)習(xí)的典型目標(biāo)檢測算法綜述[J].計算機(jī)應(yīng)用研究，2020，37(S2) ：15-21.
[9] LIN Ye, XU Zhezhuang, YUAN Meng, et al.Pointer generation and main scale detection for occluded meter reading based on generative adversarial network[J].Measurement， 2024，234，114836-114850.
