基于二维分块自适应阈值小波的地铁塞拉门数据压缩  

Segmented adaptive wavelet threshold for data compression of metro vehicles plug door

作　　者：刘洋[1] 刘东[1] 李城汐 谢乐 李振伟 蒋启龙[1] LIU Yang;LIU Dong;LI Chengxi;XIE Le;LI Zhenwei;JIANG Qilong(School of Electrical Engineering,Southwest Jiaotong University,Chengdu 610000,China)

机构地区：[1]西南交通大学电气工程学院,四川成都610000

出　　处：《中南大学学报（自然科学版）》2021年第4期1213-1221,共9页Journal of Central South University:Science and Technology

基　　金：国家自然科学基金资助项目(61531016)。

摘　　要：针对地铁塞拉门进行PHM故障诊断时海量数据难以存储的问题,提出一种基于二维分块自适应阈值小波压缩算法。为更有效降低数据周期间冗余,该方法以一次塞拉门完整开关门过程为1个周期,将多周期原始数据排列成二维矩阵,经归一化处理形成二维灰度图像,之后根据各部分图像特点提取其局域最大值以及最小值进行自适应分块,通过预设阈值和压缩率等自动调整各块二维小波阈值放大系数,最终使用二维小波完成数据压缩。以江苏某地地铁车辆段段内测试数据为例,使用不同地铁塞拉门状态数据进行实测。研究结果表明:当信噪比为43 dB、均方误差约为0.7%时,压缩比可达到1.5%,同时面对幅值变化较大的波形,使用本文算法在保证压缩率的同时,对幅值较小的区域可保留足够的细节信息,验证了算法的可行性与稳定性。In order to deal with the large amount of data generated during the PHM fault diagnosis of the subway sliding door,a compression method based on two-dimensional block adaptive threshold wavelet was proposed.In order to reduce the redundancy in the periodic data more effectively,the method used a two-dimensional grayscale representation of the original one-dimensional data according to the cycle of opening and closing the door once to form a two-dimensional grayscale image.The grayscale image was adaptively partitioned,and the wavelet threshold amplification factor of each block was automatically adjusted through preset thresholds and compression ratios.Finally,the data were compressed using a two-dimensional wavelet.Taking the test data in a subway depot in a certain place in Jiangsu Province as an example,different subway plug door status data were used for actual measurement.The results show that when the signal to noise ratio is 43 dB and the mean square error is about 0.7%,a compression ratio of 1.5%can be achieved.In the face of waveforms with large amplitude changes,the algorithm used in this work can retain enough detailed information for areas with small amplitudes while ensuring the compression rate,which verifies the feasibility and stability of the algorithm.

关 键 词：地铁塞拉门 数据压缩 二维小波变换 自适应分块 动态阈值 

分 类 号：TU375[建筑科学—结构工程]