Contact Temperature Monitoring and Layout Optimization of Shell Temperature Sensor of 500kV GIS

Contact temperature of GIS is the most direct index reflecting contact thermal fault information, so it is necessary to monitor its temperature in real time. At present, it is impossible to install sensors on the contacts to monitor their temperature. So, in this paper, an inversion calculation method is presented for contact temperature of GIS using the monitoring values of multi-point shell temperature and environmental temperature and an optimal layout method of the multi-point sensors installed on the shell is proposed. First, the temperature rise test was carried out on the 500kV real size GIS to obtain the distribution characteristics of internal temperature field. Then, support vector regression machine was used to gain the corresponding relationship between multi-point shell temperature, environmental temperature and contact temperature, and the actual measurement of multi-point shell temperature and environmental temperature was used to obtain the contact temperature by inversion calculation. By comparing the calculated and measured results, the average absolute monitoring error was 0.47%. Finally, the genetic algorithm was used to optimize the layout of the shell temperature sensors, and the number of sensors was reduced from 25 to 5. After optimization, the average absolute monitoring error was 2.07%, which saved the cost and improved the speed of inversion calculation.