基于FRFT的高速列车安全性态评估数据特征分析

发布时间：2018-06-23 19:44

本文选题：分数阶傅里叶变换 + 高速列车监测数据　；参考：《西南交通大学》2013年硕士论文

【摘要】：随着我国高铁技术的飞速发展,列车车辆的速度也在不断提高,而在列车速度提高的同时也恶化了列车的动态环境,使得列车的轮轨作用力增加、各个部件的振动和蛇行运动加剧。同时随着列车服役时间的增加,列车零部件磨损加快导致其性能参数快速蜕变,严重影响了列车的运行品质。因此如何提取高速列车监测数据的有效特征,并快速准确的估计出高速列车安全服役性能,已成为目前高速列车安全预警领域亟待解决的问题。高速列车监测数据具有非线性和非平稳性,分数阶傅里叶变换(Fractional Fourier Transform, FRFT)是傅里叶变换的一种推广形式,能够同时表征信号的时域和频域信息,在非平稳信号处理领域得到广泛的应用。因此本文采用分数阶傅里叶变换提取信号的分数域特征,并运用分数域特征实现列车运行状态的评估,主要研究工作如下：分析了短时傅里叶变换、小波变换、魏格纳-威利分布(WVD)和Randon-Wigner分布等经典时频分析方法的优缺点,并深入研究了新时频分析方法分数阶傅里叶变换的基本原理和相关理论知识。提取高速列车监测数据的分数域特征,首先对高速列车监测数据进行分数阶傅坐叶变换,将数据变换到分数阶傅里叶空间,然后对变换后的三维数据进行侧面投影,得到不同分数阶下的信号峰值曲线,最后计算峰值曲线的统计特征。研究了高速列车转向架的结构和相关动力学分析。运用分数阶傅里叶变换提取高速列车监测数据特征,研究速度对仿真数据各个工况特征分布情况的影响。对所有通道单工况的仿真数据进行仿真,并运用支持向量机对四种单一工况进行分类,通过分析不同通道不同速度下四种工况的识别率情况,统计出对各个工况具有较好分类效果的通道。对试验监测数据进行仿真和分类统计,进一步验证通道和特征的有效性。通过比较两种单一工况及其混合工况在横向和垂向上的特征分布情况,来探讨了多故障工况与单一工况之间的关联。对参数渐变工况进行仿真,研究列车从原车正常变化到三种完全故障过程中的特征分布情况。综上所述,本文运用分数阶傅里叶变换对高速列车不同工况的数据进行仿真分析,实验结果证实了分数域特征对高速列车故障识别的可行性与有效性。
[Abstract]:With the rapid development of high-speed rail technology in China, the speed of train vehicles is also increasing, while the speed of the train has also deteriorated the dynamic environment of the train, which makes the wheel-rail force of the train increase. The vibration and snake motion of each component are intensified. At the same time, with the increase of train service time, the wear of train parts is accelerated, which leads to the rapid transformation of its performance parameters, which seriously affects the running quality of the train. Therefore, how to extract the effective features of high-speed train monitoring data and estimate the safe service performance of high-speed train quickly and accurately has become an urgent problem in the field of high-speed train safety warning. High speed train monitoring data are nonlinear and non-stationary. Fractional Fourier transform (FRFT) is a generalized form of Fourier transform, which can represent the time domain and frequency domain information of the signal at the same time. It is widely used in the field of non-stationary signal processing. Therefore, fractional Fourier transform is used to extract the fractional domain feature of the signal, and the fractional domain feature is used to evaluate the train running state. The main research work is as follows: the short time Fourier transform, wavelet transform are analyzed. The advantages and disadvantages of the classical time-frequency analysis methods, such as WVD and Randon-Wigner distribution, are discussed. The basic principle and relevant theoretical knowledge of the new time-frequency analysis method, fractional Fourier transform, are studied. The fractional domain feature of the high-speed train monitoring data is extracted. Firstly, the fractional Fourier transform is carried out on the high-speed train monitoring data, and the data is transformed into the fractional Fourier space, and then the transformed 3D data is side-projected. The peak curve of signal with different fractional order is obtained, and the statistical characteristics of the peak curve are calculated at last. The structure and dynamic analysis of high-speed train bogies are studied. The fractional-order Fourier transform is used to extract the features of high-speed train monitoring data, and the influence of speed on the distribution of simulation data under different operating conditions is studied. The simulation data of all channels are simulated, and the support vector machine is used to classify the four single working conditions, and the recognition rate of the four working conditions under different channels and different speeds is analyzed. The channel which has good classification effect for each working condition is counted out. The experimental monitoring data are simulated and classified to verify the validity of the channel and features. By comparing the horizontal and vertical characteristic distributions of two kinds of single working conditions and their mixing conditions, the relationship between the multi-fault conditions and the single working conditions is discussed. The characteristic distribution of train from normal change to three kinds of complete faults is studied by simulation of parameter gradient condition. To sum up, the fractional Fourier transform is used to simulate and analyze the data of high-speed train under different working conditions. The experimental results confirm the feasibility and effectiveness of the fractional domain feature in fault identification of high-speed train.
【学位授予单位】：西南交通大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：U298.1

【参考文献】