摘要
传统检测模型中变电站设备运行工况识别目标与背景分界模糊,导致变电站涉及短路损耗工况检测效率较低,针对该问题,提出了基于图像识别技术的检测模型算法。将图像转化为二值图像,提取二值图像特征矢量,采用支持向量机识别变电站设备运行工况。结合图像识别技术有效分离出待识别目标与背景,以计算机集成制造系统CIM资产为基准,建立运行待识别目标工况检测模型,由此完成变电站设备运行工况检测模型算法研究。通过实验对比结果可知,该模型最高检测效率可达到97%,能够有效反映变电站设备运行工况,进而减轻调度工作人员的工作强度,提高检测效果,对变电站在线检测系统研发具有参考价值。
In the traditional detection model,the boundary between the target and the background of substation equipment is blurred,which leads to the low detection efficiency.In order to solve this problem,a detection model algorithm based on image recognition technology is proposed.The image is transformed into binary image,the feature vector of binary image is extracted,and the support vector machine is used to identify the operating conditions of substation equipment.Combined with image recognition technology,the target and background to be identified are effectively separated.Based on the CIM asset of computer integrated manufacturing system,the working condition detection model of the target to be identified is established,and the algorithm research of substation equipment operating condition detection model is completed.The experimental results show that the maximum detection efficiency of the model can reach 97%,which can effectively reflect the operating conditions of substation equipment,and then reduce the working intensity of dispatching staff and improve the detection effect.It has reference value for the research and development of substation online detection system.
作者
王铭民
许栋栋
陈曦
高建
Wang Mingmin;Xu Dongdong;Chen Xi;Gao Jian(State Grid Jiangsu Electric Power Co.,Ltd,Nanjing 210024,China;North China Electric Power University,Beijing 102206,China)
出处
《科技通报》
2019年第12期91-95,共5页
Bulletin of Science and Technology
关键词
图像识别技术
变电站设备
运行工况
检测模型
二值图像
image recognition technology
substation equipment
operation mode
detection model
two value image
