变压器局部放电超高频信号多尺度网格维数的提取与识别

Multi-Scale Grid Dimension Extraction and Recognition of Ultra-High Frequency Signals of Transformer Partial Discharge

提出了一种基于小波包多尺度分析和网格维数的变压器局部放电超高频信号模式识别的新方法,采用小波包多尺度变换提取局部放电超高频信号在多尺度上的小波系数,通过改进差盒计数法计算多尺度小波系数的网格维数,并将多尺度网格维数作为特征量用于局部放电超高频信号的识别。采用了4种典型放电模型产生局部放电,并采用3阶Hilbert分形天线检测局部放电超高频信号,提取的信号特征量输入径向基函数神经网络进行分类识别,识别正确率最低为70%。

A new approach for pattern recognition of ultra-high frequency （UHF） signals of transformer partial discharge （PD） based on multi-scale analysis of wavelet packet and grid dimensionality is proposed, in which the multi-scale transform of wavelet packet is used to extract wavelet coefficients of UHF PD signals and by means of improving difference box-counting method the grid dimensionality of multi-scale wavelet coefficients is calculated and the multi-scale grid dimensionality is taken as the characteristic quantity that is used in the recognition of UHF PD signals. Four typical discharge models are adopted to generate PD and a third-order Hilbert fractal antenna is used to detect UHF PD signals; the extracted characteristic quantities are input into radial basis function （RBF） neural network for the classification and recognition. Experiment results show that the lowest correct recognition rate of the proposed method is 70%.