摘要
内部过热是开关柜运维过程中的常见缺陷。现有的开关柜内部接触式测温技术在可靠性与安全性方面不足,存在着推广的局限;外部非接触式红外巡检则具有难定量诊断缺陷的问题。本文以KYN28-12A型630 A的自冷型开关柜作为研究对象,建立其温度流体场仿真模型,通过流线分析寻找合适的表面测温点,并通过正交计算得出不同工况、缺陷下开关柜负荷电流,将表面测点温度与内部热点温度之间的关系作为反演样本,最后基于支持向量机回归方法完成样本训练,形成反演计算用的降阶模型并进行了现场验证。研究成果实现了在无需内置温度传感器条件下开关柜内部热点温度的定量测算方法,在实施环节具有安全性与便捷性,尤其适用于大量难以停电改造的存量开关柜。
Internal overheating is a common defect in the operation and maintenance of switchgear cabinet.The existing switchgear contact temperature measurement technology is not reliable and safe,which results in limitations in its popularization;the external non-contact infrared inspection is difficult to diagnose defects quantitatively.The paper takes the KYN28-12A 630 A self-cooling switchgear cabinet as the research object,establishes its temperature fluid field simulation model,finds appropriate surface temperature measurement points through streamline analysis,and obtains the relationship between the load current,surface temperature and internal hot spot under different working conditions and defects as the inversion samples in orthogonal calculation,then trains the samples based on the support vector machine regression to form a reduced-order model for inversion calculation.The method is verified through field measurement.The research realizes the quantitative measurement method of the internal hot spot temperature of the switchgear cabinet without built-in temperature sensors,which is safe and convenient in the implementation process,and is especially suitable for a large number of stock switchgear cabinets that are difficult to modification without power outage.
作者
陈隽
任劼帅
何松
吴泳聪
李劲彬
牛博瑞
CHEN Jun;REN Jieshuai;HE Song;WU Yongcong;LI Jinbin;NIU Borui(Electric Power Research Institute of Hubei Electric Power Company Co.,Ltd.,Wuhan 430077,China;School of Electrical Engineering and Automation,Wuhan University,Wuhan 430072,China)
出处
《电工电能新技术》
CSCD
北大核心
2023年第12期88-96,共9页
Advanced Technology of Electrical Engineering and Energy
基金
国网湖北省电力有限公司科技项目(B3153221001J)。
关键词
开关柜
热点温度
在线监测
反演计算
温度流体场
支持向量机回归
switchgear cabinet
hot spot temperature
online monitoring
inversion calculation
temperature fluid field
support vector machine regression
