论文部分内容阅读
在分析了旋转机械振动信号的特点和小波变换在信号奇异性检测上的特性后 ,提出了利用小波系数表征信号的奇异性特征 ,及用信号的频谱来表征信号的整体特征。而用这二类数据表征信号时的数据量远远小于振动时域信号的数据量。因此本文提出了利用这二类信号对振动信号进行数据压缩的方法。通过仿真计算和对实际数据的计算证明 ,该方法既可以得到较高的信号压缩比又保留了信号的局部特征 ,有着很好的信号重构性。
After analyzing the characteristics of rotating machinery vibration signals and the characteristics of wavelet transform in signal singularity detection, this paper presents the singularity characteristics of signal using wavelet coefficients and the signal spectrum to characterize the overall characteristics of signals. The amount of data used to characterize the signal with these two types of data is much smaller than the amount of data in the vibration time domain signal. Therefore, this paper presents the use of these two types of signals on the vibration signal data compression method. Through the simulation and the calculation of the actual data, it is proved that this method not only can obtain higher signal compression ratio but also retains the local characteristics of the signal, has a good signal reconstruction.