摘要
GIS设备导体触头温升过高会引发重大事故,光纤光栅技术在电气设备温度在线监测方面具有广泛的应用前景。针对光纤光栅温度传感器无法直接安装在导体触头表面进行温度测量的难题,提出了一种基于测量多点外壳温度及环境温度,从而间接计算导体触头温度的方法。建立了单相GIS隔离开关三维有限元计算模型,得到GIS内部温度分布规律,确定了多个光纤光栅温度传感器的最优安装位置。进行GIS隔离开关温升实验,将实验数据与仿真结果进行了比较,验证了仿真计算模型的准确性。最后利用人工神经网络算法对环境温度、外壳温度和导体触头温度进行了曲线拟合,计算结果与实测数据对比表明本方法具有较高准确度。该在线测温方法简单有效,无需破坏现有GIS设备的结构,具有较好的应用前景。
In the GIS equipment, if the temperature of the conductor contact is too high, it will lead to a major accident. The on-line temperature monitoring technology based on Fiber Bragg Grating has broad application prospects. For the situations that the FBG temperature sensor cannot be fixed on the surface of conductor contacts, a method to indirectly obtain the temperature of the contacts by directly measuring the multipoint shell temperature and the environmental temperature was proposed in this paper. The three-dimensional heat transfer model of the GIS disconnecting switch was established and the temperature distribution around the contact section was acquired. Based on the simulation results, the installation positions of the FBG temperature sensors were determined. Then experimental studies were taken to validate the accuracy of the simulation model. Finally, the neural network algorithm was used to deal with the experimental temperature data of contacts, multipoint shell and environment, and results showed a high accuracy. The method is simple and effective, without destroying the structure of the GIS equipment, which would have a good application prospect.
出处
《电工技术学报》
EI
CSCD
北大核心
2015年第12期298-306,共9页
Transactions of China Electrotechnical Society
基金
国家高技术发展研究计划(863计划)资助项目(2011AA05A121)
国家自然科学基金(51277061和51420105011)
关键词
GIS
触头温度
多点分布式
光纤光栅
有限元模型
神经网络算法
GIS,contact temperature,multipoint-distributed,fiber bragg grating,finite element model,neural network algorithm