STFT时频分析算法研究及其在列车振动分析中的应用  

Research of STFT Time-Frequency Analysis Algorithm and Its Application in Train Vibration Analysis

作　　者：王晓华[1] 彭程[2] 梁静[2] 

机构地区：[1]西南交通大学机械工程学院,成都610031 [2]西南交通大学牵引动力国家重点实验室,成都610031

出　　处：《噪声与振动控制》2011年第1期65-68,共4页Noise and Vibration Control

摘　　要：时频联合分析是处理非平稳信号的有效方法。由于其计算过程中产生的时频表示矩阵数据量大,导致可处理的原始数据的长度有限,只能处理较短的时间信号。而样本数据不够或采样时间太短又无法全面反映系统的特性。以使用短时傅里叶变换来分析某动车地板在加速过程中的时频特性为例。为解决这一问题,提出一种方法,首先根据分析频带对原始数据进行重采样,再对频率值进行合理的设置,大大增加可处理原始数据的长度。以某动车地板在加速过程中的时频特性分析为例,用短时傅里叶变换进行分析,验证此方法的有效性。The joint time-frequency analysis is an effective way to deal with non-stationary signal.Since its time-frequency matrix generated in the process of computation is very large,this method can only deal with short-time signals.However,the characteristics of the system can not be fully reflected if the sample data is not sufficiently long or the sampling time for signal collection is too short.In order to solve this problem,a novel method is proposed.At first,the original data is re-sampled according to the required frequency range.Then,the frequency values are reasonably set.As a result,the length of the original data can be greatly increased.Taking the analysis of the time-frequency characteristics of the train floor in accelerating process as an example with the use of short-time Fourier transform,the effectiveness of this method is verified.

关 键 词：振动与波 时频联合分析 STFT 重采样 优化算法 

分 类 号：U260.143[机械工程—车辆工程]