基于经验小波变换的电晕电流降噪方法

Corona Current De-Noising Method Based on Empirical Wavelet Transform

随着电压等级升高和输电距离增加,在特高压直流输电线路上因电晕电流引起的电能损失和环境问题更加严重。深入分析电晕电流特性,对抑制电晕效应、实现长距离高效输电具有重要意义。电晕电流的复杂性导致时域和频域特性分析面临诸多困难,因此提出了基于经验小波变换的电晕电流降噪方法。首先,针对超长实测电晕电流数据,提出分段经验小波变换;其次,根据电晕电流的复杂频谱特性,提出了改进的频谱极大值搜索方法;最后,结合分段经验小波变换的分解结果,重构出了降噪后的电晕电流。实验结果表明,提出方法可以有效地抑制不同电压等级下实测电晕电流中的多种噪声成分,实现超长电晕数据的高效降噪处理,为电晕电流特性研究提供了崭新方法,为准确计算电能损失、探索电晕放电规律提供理论和技术支撑。

With increase of transmission voltage level and transmission distance, power loss and environmental problems caused by corona current on UHV DC transmission line become more serious. Detailed analysis of corona current is of great importance for suppressing corona effects and achieving efficient large-scale power transmission. Complexity of corona current leads to difficulties in analysis of time and frequency characteristics. This paper proposes an empirical wavelet transform （EWT） based de-noising method for corona current. Firstly, piecewise EWT is proposed for practically measured overlong corona current. Secondly, local maximum searching method is improved according to spectral characteristics of corona current. Finally, de-noised corona current is reconstructed based on decomposition result ofpiecewise EWT. Experimental results show that the proposed method suppresses various noise components of the measured corona current with different voltage levels effectively and realizes efficient overlong corona data processing. This paper provides a new method for corona current research and offers theoretical and technical support for accurate power loss calculation and exploration of corona discharge rules.