基于可移动特高频天线阵列的变电站站域放电源检测与定位研究

Detection and Localization of Discharge Sources in Substation Area Using Movable UHF Antenna Array

基于特高频天线阵列的局部放电检测方法已被应用于敞开式变电站(air-insulated substations,AIS)站域的局部放电巡检,但目前装置存在定位精度低等缺点。该文针对变电站局部放电去噪方法、传感器阵列优化布置、定位算法及检测系统开发等方面开展了深入研究。针对电晕噪声、手机通讯信号及非脉冲噪声,提出了合理选取检测频带、硬件陷波器及能量陡度阈值等去噪方法;通过克拉美罗下界分析及定位均方根误差仿真对天线阵列布置进行了优化,选取为4×1m的菱形天线阵;采用基于最大似然估计的概率定位法求解时差定位方程组,综合多次局部放电信号定位结果,可降低时差估计误差的影响,提高了定位准确度。开发了变电站局部放电检测与定位系统,包括超宽带全向双锥天线、信号调理电路及信号高速采集分析单元,用于110kV AIS站域局部放电检测,测试结果验证了系统的有效性及定位准确性。

Partial discharge （PD） detection method based on ultra-high-frequency （UHF） antenna array has been applied to PD monitoring in air-insulated substations （AIS）. In previously developed instruments, the estimated range is generally with low localization accuracy. We mainly focus on several problems such as de-noising methods, configuration optimization of antenna array, localization algorithm and development of whole system, etc. For on-site background noise of corona discharge, mobile phone communication signals and other non-impulse noises, several kinds of de-noising methods such as optimization of detection bandwidth, hardware band-stop filter and analysis of energy steepness, were proposed to reduce the influence of different kinds of noises. The configuration of antenna array was optimized by comparing the Cramér-Rao lower bound （CRLB） and mean-square error （MSE） distribution, and the antenna array is chosen as diamond array with size of 4 × 1m. The probability-based localization （PBL） algorithm via maximum likelihood estimation was adopted to solve the time difference equation group. The PBL reasonably integrates the time differences of continuous signal sequences, which can effectively reduce the influence of time difference of arrival （TDOA） estimation errors and improve the localization accuracy. Finally, a comprehensive system was developed, including four-antenna array, signal conditioning unit and high-speed acquisition unit. The system was applied for PD detection and location in on-site 110kV substations, and the results prove the effectiveness of the system.