摘要
为了在特高压长间隙放电特性试验研究中选择出较优的数据统计方法,采用多级试验法对特高压猫头真型塔边相I串对塔身间隙进行了5000μs长波前操作冲击放电特性试验,获得了不同电压等级试验电压。并采用直观法、加权平均法、最小二乘法3种方法统计了试验数据,求取了50%放电电压和标准偏差的估计值,对比了3种方法统计结果的差异,提出了在特高压长间隙放电电压及变异系数研究中较适用的方法。研究表明,加权平均法和最小二乘法有较好的计算精确度,均可适用于长间隙冲击电压试验数据统计,其中,在该试验数据统计中,前者50%放电电压估计值U^50和标准偏差估计值σ^与后者相比仅相差0.18%和-2.36%;直观法统计误差较明显,不推荐使用。
In order to obtain one better test data statistical method,to ensure the insulation strength,and to optimize the tower design of UHVAC transmission line in China,we researched the statistical methods of 50% flashover voltage U50 and the standard deviation σ for long air-gap test results,and analyzed the flashover characteristics of the tower gap.Moreover,5000 μs long wave-front switching impulse flashover was tested by multistage method,which was mainly about the air-gap from conductor to tower body for UHV cat-head tower side phase ＇I＇ insulator string.The direct vision,weighted mean and least square methods were used to analyze the test results so as to obtain the 50% flashover voltage and standard deviation of the sample data.The differences of the results from three methods were compared,and the more appropriate method was suggested.
出处
《高电压技术》
EI
CAS
CSCD
北大核心
2011年第9期2171-2177,共7页
High Voltage Engineering
基金
国家电网公司1000kV级交流特高压输变电工程关键技术重大科研项目(SGKJ[2007]70)~~
关键词
特高压
长波前操作冲击
多级法
标准偏差
最小二乘法
加权平均法
ultra high voltage（UHV）
long wave-front switching impulse
multistage method
standard deviation
least square method
weighted mean method
