基于线调频小波变换的谐波快速可视化识别方法

Rapid and visual harmonic identification based on Chirplet transform

线调频小波变换是傅里叶变换、短时傅里叶变换和小波变换的一般形式。在介绍高斯线调频小波变换算法的基础上,提出用线调频小波变换可视化地检测和识别出谐波的特性,即根据对电能质量谐波扰动信号的线调频小波变换矩阵的等高线图,可视化地区分出谐波是固定不变、线性时变或非线性时变,先识别谐波的大致类型,为选取合适的信号分析工具提供保证。仿真结果表明,线调频小波变换的等高线图可以很好地检测出电能质量信号扰动发生的时刻和持续时间。

Chirplet transform is the general form of Fourier transform,short- time Fourier transform and Wavelet transform. The algorithm of Chirplet transform is introduced and used to detect and identify harmonics. Harmonics can be visually classified to non-variable,linear time-varying and nonlinear time-varying according to Chirplet transform matrix contour figures of power quality harmonic signals,based on which an appropriate signal analysis tool is thus selected. Simulation results show that,the Chirplet transform matrix contour figures can effectively detect the beginning time and duration of power quality signal disturbance.