电能扰动检测和识别线调频三角样条小波方法  

Power quality disturbance detection and identification using Chirplet transform with trigonometric spline kernels

作　　者：胡国胜[1,3] 朱锋峰[2,3] 任震[2] 

机构地区：[1]广东科学技术职业学院 [2]华南理工大学电力学院 [3]华南理工大学理学院

出　　处：《继电器》2006年第2期46-52,共7页Relay

基　　金：国家自然科学基金(50077008);广东省自然科学基金(033044)

摘　　要：线调频小波变换是傅里叶变换和小波变换的一般情况,它具有更灵活的时频局部化特性。但连续线调频小波变换实质上是积分的数值计算,传统的方法是用多项式插值函数代替被积函数求积分。对电能扰动信号来说,计算的精度不高,影响检测效果。文中用连分式有理插值函数替代被积函数,提高了积分精度。最后,以三角样条小波函数作为线调频小波变换的母函数,并用线调频三角样条小波变换来检测和识别电能扰动,它的等高线可以清楚地显示出干扰发生时刻和持续时间。Chirplet transforms are generalized cases of Fourier transforms and wavelet transforms and have more flexible time-frequency characteristics than wavelet transforms. But the calculation of continued Chirplet transform is naturally numerical calculation of integration. Traditionally, the integrand of Chirplet transform is replaced by polynomial interpolation function and the precision is not high enough for detecting power quality disturbance signals. In this paper, the new calculation of Chirplet transform is presented, that is, to enhance calculation precision, continued fraction rational interplant function is replaced to integrand. Finally, trigonometric spline wavelet function is regarded as mother function of Chirplet transform, and Chirplet transforms are used to detect and identify power quality disturbances. The contours of Chirplet transforms can clearly show the location information of disturbances and lasting time. This project is supported by National Natural Science Foundation of China（ No. 50077008） and National Natural Science Foundation of Guangdong Province （ No. 033044）.

关 键 词：电能质量 干扰 线调频小波变换 三角样条小波 连分式有理插值 

分 类 号：TM714[电气工程—电力系统及自动化]