论文部分内容阅读
文中提出一种用于变压器局部放电在线监测信号的去噪方法,该方法基于尺度相关的最优小波分解去噪算法和脉冲序列分析算法。通过双电流传感器分别耦合流过每一相高压套管和高压套管末屏接地线的脉冲电流,首先采用逐层最优小波分解算法有效地抑制白噪声且减小对脉冲波形造成的畸变,再利用脉冲序列分析算法分析双传感器输出的脉冲序列,达到去除脉冲干扰的目的。实验结果表明,该方法能够鉴别出外部的脉冲干扰,有效地提取出变压器内部的局部放电信号,具有很好的应用价值。
This paper presents a de-noising method for transformer partial discharge on-line monitoring signal, which is based on the scale-dependent optimal wavelet decomposition de-noising algorithm and pulse sequence analysis algorithm. The dual-current sensor is used to couple the pulse current flowing through the grounding wire of the high-voltage bushing and the high-voltage bushing at each phase respectively. Firstly, the optimal wavelet decomposition algorithm is used to effectively suppress the white noise and reduce the distortion caused by the pulse waveform. Then the pulse sequence analysis algorithm is used to analyze the pulse sequence output by the dual sensor to achieve the purpose of removing the pulse interference. The experimental results show that this method can identify the external pulse interference and effectively extract the partial discharge signal inside the transformer, which has a good application value.