摘要
为实现输电线路覆冰形状分类识别和参数化描述的全自动过程,针对现有研究中冰形定义不准确、描述不统一的不足,提出一种基于偏心椭圆方程的常见冰形统一参数描述方法,该方法考虑了覆冰表面不对称性和导线扭转的影响.对于不规则的真实覆冰,建立了冰形自动化识别系统,通过内点法计算最优冰形参数,并采用预先学习的支持向量机模型对识别结果进行评定.选取实际覆冰图像对上述方法进行验证,所得结果表明覆冰重叠率可达90%,能较好地反映导线覆冰截面特征,为输电线路智能运维提供参考.
To realize the automatic classification recognition and parametric description of icing shape on transmission line,and solve the inaccurate definition and inconsistent description of icing shape in the existing research,the common icing shapes are described by unified parameters based on the eccentric ellipse equation,which considers the influence of asymmetric icing surface and conductor torsion.An automatic icing shape recognition system is established for the irregular actual icing shape to calculate the optimal icing shape parameters by the interior point method.The pre-learned support vector machine model is used to assess the recognition results.The proposed method is verified by using actual icing images,and the results show that the overlap rate reaches 90%,which can reflect the cross-section characteristics of conductor icing.Also,it will provide reference for the intelligent operation and maintenance of transmission lines.
作者
楼文娟
陈卓夫
王礼祺
LOU Wenjuan;CHEN Zhuofu;WANG Liqi(College of Civil Engineering and Architecture, Zhejiang University, Hangzhou 310058, China)
出处
《大连理工大学学报》
CAS
CSCD
北大核心
2021年第4期375-382,共8页
Journal of Dalian University of Technology
基金
国家自然科学基金资助项目(51838012).
关键词
输电线路
自动识别
参数描述
覆冰形状
智能监测
transmission line
automatic recognition
parametric description
icing shapes
intelligent monitoring