【摘要】 为了能够为各类电能扰动事件的本质研究和有效治理提供准确、可靠的依据,提出基于时频原子变换(timefrequency transform,TFT)和改进型自组织映射神经网络(improved self-organizing map,ISOM)的电能质量扰动在线监测与识别新方法。TFT具有自适应复带通滤波特性,其频窗中心与频窗半径解耦及频窗宽度不受中心频率的约束,可以灵活调整。通过设置恰当的频域窗口,TFT可有效抑制邻近频率分量的相互干扰,且有较好的动态响应速度。TFT能准确监测电力系统波形中电能质量事件,并为类型识别提供物理意义明确、指标具体的实时模式特征。依据TFT提取的特征向量,ISOM可准确识别单一或同时存在的多重电能质量事件,并对其严重程度进行直观表达,能动态反映电能质量事件各自的发展变化轨迹。仿真验证结果表明了所提出方法的有效性和优越性。更多还原

【Abstract】 In order to supply accurate and reliable basis for the essential research and effective management of various power disturbance events,a new method for power disturbance events online monitoring and identifying was presented in this paper based on time frequency transform(TFT) and improved self-organizing map(ISOM).TFT has adaptive complex band-pass filtering capability and decouples the center and width of frequency window.The frequency window width can be flexibly adjusted from the constraints of central frequency.The interference between adjacent frequency components can be effectively inhibited by adjusting the scale parameter to set the appropriate frequency domain window and also it has a good dynamic response speed.TFT provides clear physical meaning,also specific index of the instantaneous model for multiple power disturbance events analyse and identification.According to the eigenvector extracted by TFT,ISOM can accurately identify single or multiple power quality events existing in power disturbance waveforms,and has intuitionistic expression of their severity,and is also able to reflect multiple power quality event developments in their respective track dynamicly.Simulation results demonstrated the effectiveness and superiority of this method.更多还原