中文摘要：

为了提高时频分布的分析精度和有效性，笔者根据局部放电脉冲信号的非平稳特性，提出使用联合时频分布对其进行处理，从而分析脉冲信号的时频特性。针对目前在非平稳信号处理中广泛使用、属于Cohen类的Wigner-Ville分布（WVD），笔者以仿真脉冲为对象，分别研究了其原始WVD、平滑伪Wigner分布（SPWVD）以及重排平滑伪Wigner分布（RSPWVD）。分析结果表明：WVD能够对局部放电脉冲信号引发的短时变化具有高度的敏感性，更适合表征非平稳信号的时变信息；SPWVD能消除双线性函数引起的交义干扰项，但是以牺牲分布的时频分辨率为代价的：RSPWVD具有很好的时频聚集性，易于提取单个脉冲波形的时频特征参数，从而用于放电类型的模式识别以及深入研究绝缘系统中的局部放电机理。

英文摘要：

According to the unstability of partial discharge pulse signals, a joint time-frequency analysis method was put forward to process the signals to obtain their characteristics in time-frequency domain. Since the wigner-ville distribution which belongs to the Cohen class time-frequency energy distribution is commonly used at present for processing unstable signals, a simulated PD pulse was chosen as the object to investigate its original WVD, smoothing pseudo Wigner distribution （SPWVD） and rearranged-smoothing-pseudo Wigner distribution （RSPWVD）. The results show that WVD is highly sensitive to the short-variation of unstable PD signal, which is more suitable to extract time-varying information. SPWVD can remove the cross-terms of interference caused by the bilinear function but reduce the time-frequency resolution. For having good time-frequency concentration, RSPWVD provides a powerful tool to evaluate parameters for single PD pulse shape, which can be used in PD pattern recognition and in-depth study of PD mechanism in electrical insulation system.