张勇,宋爱波,苏猛猛,王天予,王清未,陈锐
ZHANG Yong,SONG Aibo,SU Mengmeng,WANG Tianyu,WANG Qingwei,CHEN Rui

• 1:国家电网有限公司华北分部调度控制中心
• 2:东南大学计算机科学与工程学院

摘要(Abstract)：

电气接线图是表示电气设备的位置、描述其接线与配线方式的工程图纸，需要转换成XML文件格式以用于智能电网系统调度管理，这一转换过程被称为矢量化。针对人工矢量化效率低、费时费力的问题，提出了一种电气接线图矢量化技术解决方法。首先给出了融合注意力机制的YOLOv3电气图元检测方法，实现了对各类电气图元的精准识别与定位；然后提出基于多尺度特征提取的文本区域检测算法，实现对图纸中电气文本标注区域的定位，再结合CRNN模型对其内容进行识别；最后提出基于模板匹配的矢量化成图策略，以图元组为单位，对图纸中电气元素间的关联关系进行分析。该方法已有实际部署运行案例，能高效、准确完成电气接线图纸的矢量化工作，满足电力系统的应用需求。
Electrical wiring diagrams, which represent the positions of electrical equipment and describe their wiring and cabling methods, need to be converted into XML for smart grid system scheduling and management. This conversion process is known as vectorization. To address the inefficiencies and time-consuming nature of manual vectorization, this paper proposes a solution for vectorizing electrical wiring diagrams. First, the paper introduces an attention mechanism-enhanced YOLOv3 method for precise recognition and localization of graphic elements in electrical wiring diagrams. Next, it proposes a text region detection algorithm based on multi-scale feature extraction to locate electrical text annotation areas on the diagrams, and uses a CRNN model to recognize their content. Finally, the paper presents a vectorization strategy based on template matching, which analyzes the relationships between electrical elements on the diagrams at the graphic element group level. This method has been practically deployed and can efficiently and accurately complete the vectorization of electrical wiring diagrams, meeting the application needs of power systems.

关键词(KeyWords)： 电气接线图;矢量化;目标检测;文本识别;模板匹配
electrical wiring diagram;vectorization;object detection;text recognition;template matching

Abstract:

Keywords:

基金项目(Foundation): 国家自然科学基金（62061146001）;; 国家电网科技部项目（5108202340042A-1-1-ZN）

作者(Author): 张勇,宋爱波,苏猛猛,王天予,王清未,陈锐
ZHANG Yong,SONG Aibo,SU Mengmeng,WANG Tianyu,WANG Qingwei,CHEN Rui

DOI: 10.19585/j.zjdl.202408004

参考文献(References)：

• [1]李衡，朱理，郑洁，等.基于Prophet算法的配电网线路峰值负荷预测[J].浙江电力，2022,41(3):20-26.LI Heng,ZHU Li,ZHENG Jie,et al.Peak load forecasting of distribution network lines based on prophet algorithm[J].Zhejiang Electric Power,2022,41(3):20-26.
• [2]曹文斌，潘武略，戚宣威，等.海上风电高抗匝间保护误动分析及对策建议[J].浙江电力，2023,42(1):54-62.CAO Wenbin,PAN Wulue,QI Xuanwei,et al.Maloperation analysis and suggested countermeasures for inter-turn protection of high-voltage shunt reactors in offshore wind power transmission[J].Zhejiang Electric Power,2023,42(1):54-62.
• [3]周丹阳，黄晓燕.极端天气下基于RF-LSTM的配电网馈线薄弱环节识别方法[J].浙江电力，2022,41(12):71-78.ZHOU Danyang,HUANG Xiaoyan. An identification method for weak links in distribution network feeders based on RF-LSTM under extreme weather[J].Zhejiang Electric Power,2022,41(12):71-78.
• [4]程乐峰，余涛，张孝顺，等.机器学习在能源与电力系统领域的应用和展望[J].电力系统自动化，2019,43(1):15-31.CHENG Lefeng,YU Tao,ZHANG Xiaoshun,et al.Machine learning for energy and electric power systems:state of the art and prospects[J].Automation of Electric Power Systems,2019,43(1):15-31.
• [5] JULCA-AGUILAR F D,HIRATA N S T.Symbol detection in online handwritten graphics using faster R-CNN[C]//2018 13th IAPR International Workshop on Document Analysis Systems(DAS).April 24-27,2018.Vienna.IEEE,2018:151-156.
• [6] WU W K,CHEN C Y,LEE J S. Embedded YOLO:faster and lighter object detection[C]//Proceedings of the2021 International Conference on Multimedia Retrieval.Taipei.ACM,2021.
• [7] ELYAN E,JAMIESON L,ALI-GOMBE A.Deep learning for symbols detection and classification in engineering drawings[J].Neural Networks,2020,129:91-102.
• [8]张烨，冯玲，穆靖宇，等.输电线路绝缘子覆冰厚度图像识别算法[J].电力系统自动化，2016,40(21):195-202.ZHANG Ye,FENG Ling,MU Jingyu,et al.Image identification algorithm of icing thickness for insulator in power transmission lines[J].Automation of Electric Power Systems,2016,40(21):195-202.
• [9]刘思言，王博，高昆仑，等.基于R-FCN的航拍巡检图像目标检测方法[J].电力系统自动化，2019,43(13):162-168.LIU Siyan,WANG Bo,GAO Kunlun,et al.Object detection method for aerial inspection image based on regionbased fully convolutional network[J].Automation of Electric Power Systems,2019,43(13):162-168.
• [10]周自强，纪扬，苏烨，等.基于迁移学习卷积神经网络的电缆隧道锈蚀识别算法[J].中国电力，2019,52(4):104-110.ZHOU Ziqiang,JI Yang,SU Ye,et al.A hybrid transfer learning/CNN algorithm for cable tunnel rust recognition[J].Electric Power,2019,52(4):104-110.
• [11] DE P,MANDAL S,DAS A,et al.Detection of electrical circuit elements from documents images[C]//SPIE Proceedings","Document Recognition and Retrieval XXII.San Francisco,California,USA.SPIE,2015.
• [12]赵锴，李继东，黄佳，等.基于Cascade R-CNN的配电网鸟巢检测[J].浙江电力，2021,40(3):73-78.ZHAO Kai,LI Jidong,HUANG Jia,et al.Bird nest detection in distribution network based on cascade R-CNN[J].Zhejiang Electric Power,2021,40(3):73-78.
• [13]冯双，陈佳宁，汤奕，等.基于SPWVD图像和深度迁移学习的强迫振荡源定位方法[J].电力系统自动化，2020,44(17):78-87.FENG Shuang,CHEN Jianing,TANG Yi,et al.Location method of forced oscillation source based on SPWVD image and deep transfer learning[J].Automation of Electric Power Systems,2020,44(17):78-87.
• [14] SHI B G,BAI X,BELONGIE S.Detecting oriented text in natural images by linking segments[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).July 21-26,2017.Honolulu,HI.IEEE,2017.
• [15] ZHOU X Y,YAO C,WEN H,et al.EAST:an efficient and accurate scene text detector[C]//2017 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).July 21-26,2017.Honolulu,HI.IEEE,2017.
• [16] LIAO M H,ZHANG J,WAN Z Y,et al.Scene text recognition from two-dimensional perspective[J]. Proceedings of the AAAI Conference on Artificial Intelligence,2019,33(1):8714-8721.
• [17] LI H,WANG P,SHEN C H. Towards end-to-end text spotting with convolutional recurrent neural networks[J].ArXiv e-Prints,2017:arXiv:1707.03985.
• [18]刘欣，宁新福，金翼，等.配电网分布式储能的分层优化配置方法[J].浙江电力，2023,42(5):95-104.LIU Xin,NING Xinfu,JIN Yi,et al.A hierarchical optimal configuration method for distributed energy storage in distribution networks[J]. Zhejiang Electric Power,2023,42(5):95-104.
• [19]王梓耀，罗庆全，萧文聪，等.电网厂站接线图人工智能识别关键方法[J].电力系统自动化，2023,47(2):115-124.WANG Ziyao,LUO Qingquan,XIAO Wencong,et al.Key method of wiring diagram recognition with artificial intelligence for power plant and substation[J]. Automation of Electric Power Systems,2023,47(2):115-124.
• [20]苏运光，罗俊，熊浩，等.基于存量图形拓扑分析的厂站接线图自动生成[J].电力系统自动化，2020,44(3):162-166.SU Yunguang,LUO Jun,XIONG Hao,et al. Automatic generation of substation wiring graph based on inventory graph topology analysis[J].Automation of Electric Power Systems,2020,44(3):162-166.
• [21]高玮，彭群生.基于二维视图特征的三维重建[J].计算机学报，1999,22(5):481-485.GAO Wei,PENG Qunsheng. 3d reconstruction based on2d view feature[J].Chinese Journal of Computers,1999,22(5):481-485.