郑艳艳,朱永利,高佳程.基于SSA与ICA的变压器局部放电混合信号分离[J].电测与仪表,2020,57(22):84-90. ZHENG Yanyan,ZHU Yongli,GAO Jiacheng.Hybrid Signal Separation of Transformer Partial Discharge Based on SSA and ICA[J].Electrical Measurement & Instrumentation,2020,57(22):84-90.
基于SSA与ICA的变压器局部放电混合信号分离
Hybrid Signal Separation of Transformer Partial Discharge Based on SSA and ICA
When the partial discharge （PD） signal of transformer is a hybrid signal mixing multiple signals of different PD types , in order to correctly diagnose discharge types, the hybrid signal needs to be separated. When the hybrid signal has a high signal-to-noise ratio (SNR), independent component analysis (ICA) can be used to effectively separate independent source signals. However, when the SNR is low, the performance of ICA is significantly affected. However, the PD signal of the transformer has serious noise, and the periodic narrowband noise is particularly serious. Aiming at the case of severe narrowband noise in hybrid signal, a method for separating the hybrid signal of transformer partial discharge based on singular spectrum analysis （SSA） and ICA is proposed. The method first performs singular spectrum analysis on the hybrid signal with noise for suppressing periodic narrowband noise; Then, the ICA is used to separate the independent source signals from the hybrid signal after noise reduction. The simulation results show that the method of this partial discharge paper can effectively separate the independent source signals from the hybrid signal with low SNR.
