基于NGO-VMD-HHT的电缆局部放电信号特征量提取方法

Feature extraction method of cable partial discharge signal based on NGO-VMD-HHT

电缆的局部放电量可以在很大程度上反映出电缆设备的绝缘状态,针对在电缆局部放电检测过程中,局部放电特征信息提取较为困难、提取准确性低等问题,提出基于北方苍鹰优化算法(NGO)-变分模态分解(VMD)-希尔伯特-黄变换(HHT)的特征提取方法。首先利用NGO迭代寻找在进行VMD时需要设置的最佳分解参数,得到最优分解层数与惩罚因子;然后通过VMD对采集到的信号数据进行分解,得到多个基于最优分解参数的模态分量;最后利用希尔伯特边际谱理论分析信号的频谱特性,再结合信号的统计特征与熵特征构建局部放电信号的特征量集合。实验结果表明,所提基于NGO-VMD-HHT的电缆局部放电特征提取方法能够有效分解局部放电信号并构建相应特征量集合。

The phenomenon of partial discharge within a cable predominantly mirrors the insulation condition of cable-based equipment.Addressing a range of challenges encompassing the intricate extraction of partial discharge characteristic data and the inherent limitations pertaining to extraction precision during cable partial discharge detection,a feature extraction method based on northern goshawk optimization(NGO)-variational mode decomposition(VMD)-Hilbert-Huang transform(HHT)is proposed.Initially,the iterative application of the NGO is employed to determine the optimal configuration parameters essential for conducting VMD.Consequently,the optimal count of decomposition layers and the corresponding penalty factor are ascertained through this process.Then,the collected partial discharge signal is decomposed by VMD to obtain multiple modal components based on the optimal decomposition parameters.Finally,the Hilbert marginal spectrum theory is used to extract the characteristic components.The experimental results show that the feature extraction method of cable partial discharge based on NGO-VMD-HHT proposed in this paper can effectively decompose the partial discharge signal and accurately extract the corresponding feature quantity.