基于小波能谱的电力暂态信号分类识别方法

Electric Power Transient Signals Classification and Recognition Method Based on Wavelet Energy Spectrum

为区分故障性暂态信号与非故障性暂态信号,在研究电力系统各种暂态信号及其特点的基础上,建立了500kV输电线路EMTDC仿真模型,并产生了单相断路器操作、电容投切、接地短路、一次电弧故障、非故障性雷击和故障性雷击6种类型的暂态信号。利用多分辨分析的小波变换提取上述各暂态信号的频带能量特征和局部能量特征,总结得到了能量统计图。分析了各暂态信号的能量分布特点,并从保护的角度出发提出了暂态信号分类识别的判据。大量的仿真试验验证了该方法的可行性和有效性。

To distinguish fault transient signals from non-fault transient signals, on the basis of researching various transient signals and their characteristics the authors establish an EMTDC model to simulate 500kV transmission line. Using this model, six kinds of transient signals such as manipulation of single phase circuit breaker, switching on/off of capacitor bank, transmission line grounding fault, primary arcing fault, lightning stroke without fault and lightning stroke fault are simulated. By use of wavelet transform with multi-resolution analysis both frequency band energy features and local energy features of above-mentioned transient signals are extracted, thus the energy statistical charts of these transient signals are obtained. The energy distribution features of these transient signals are analyzed, in the view of protection the criterion for classification and recognition of transient signals are proposed. The feasibility and effectiveness of the proposed method are verified by a lot of simulation results.