基于多属性决策改进S变换的电压暂降检测与识别

Voltage Sag Detection and Identification Based on Multi-attribute Decision Making Improved S Transform

电压暂降的检测和识别是解决电压暂降相关问题的基础和关键。在S变换的高斯窗函数上增加调节因子,通过定义起止时刻误差、暂降深度误差、局部标准差、峰度和能量5个指标得到其与改进S变换高斯窗调节因子的关系曲线,应用最优组合赋权的多属性决策法对该5项指标进行赋权,从而确定调节因子的取值。应用改进S变换提取5种不同电压暂降源的相关特征指标,将指标数据输入BP神经网络进行分类识别。仿真结果表明,改进后的S变换能更加精确地提取电压暂降的特征指标数据,最后的分类识别正确率更高。

The detection and identification of voltage sag is the foundation and key to solve the problems related to voltage sag.The adjustment factor is added to the Gauss window function of S transform.By defining the start and end time error,sag depth error,local standard deviation,kurtosis and energy,the relationship curve between them and the adjustment factor of improved S-transform Gaussian window is obtained.The optimal combination of multiple attribute decision-making method is used to weight the five indicators,so as to determine the value of the adjustment factor.The improved S-transform is used to extract the relevant characteristic indexes of five different voltage sag sources,and the index data is input into the BP neural network for classification and recognition.The simulation results show that the improved S-transform can more accurately extract the characteristic index data of voltage sag,and the final classification and recognition rate is higher.