用Prony方法进行时频分解及其与经典方法的比较  

Comparison of time-frequency decomposition using Prony method and classical methods

作　　者：张旭俊 张宇 Zhang Xujun;Zhang Yu(Electric Power Research Institute of State Grid Jiangxi Electric Power Limited Company,Nanchang 330096,China)

机构地区：[1]国网江西省电力有限公司电力科学研究院,南昌330096

出　　处：《电测与仪表》2020年第14期148-152,共5页Electrical Measurement & Instrumentation

基　　金：国网江西省电力有限公司科技项目(521820180013)。

摘　　要：文章用Prony方法对采样数据系列进行时频分解,并用自动识别主模态项数的方法对其进行改进,使其容易收敛求解。因Prony分解可用矩形窗分段分割减少点数,由此可得出振荡衰减的正弦函数组合,时频分解比较清晰、简洁。所提出的Prony方法也与其它经典方法进行图例比较,结果表明用Prony方法进行时频分解可弥补经典方法的某些不足。用列主元求解法方程R时,可自动确定主模态项数,利于Prony分解获得最简结果。This paper introduces the time-frequency decomposition of the sampled data series using the Prony method,and improves it by automatically identifying the number of main mode terms,so that it can be easily converged and solved.Since Prony decomposition can adopt rectangular window segmentation to reduce the number of points,the combination of sine functions with oscillation attenuation can be obtained.Time-frequency decomposition is relatively clear and concise.Prony method has been compared with other classical methods in illustration,the results show that the time-frequency decomposition using Prony method can make up for some shortcomings of the classical methods.When solving R of normal equation with column pivot elements,the number of main mode terms can be determined automatically,and it is helpful for Prony decomposition to obtain the simplest result.

关 键 词：Prony分解 FFT分解 小波矩阵分析 时频分解 

分 类 号：TM11[电气工程—电工理论与新技术]