【摘要】 通过振动测试设备采集到的信号因外界干扰往往呈现出非线性、非平稳性和非光滑性等特点,并常常会偏离其真实值,给后续分析带来了误差甚至错误。因此振动信号的预处理在信号分析中非常重要。信号预处理是对采样信号提取特征量的首要步骤,它直接决定着后续诊断技术的使用效果及精度。讨论了振动信号预处理的两种方法——消除多项式趋势项和数据平滑处理,并基于MATLAB软件平台进行算法实现。以两著名的地震波为例,对上述算法进行验证。结果表明,处理后的信号质量有明显的改善。更多还原

【Abstract】 A vibration signal collected by vibration test equipment often shows nonlinear,non-balanced and unsmoothed,and deviates from the true value because of outside noise,thus the error and wrong take place in the sequent analysis result.Therefore,the pre-processing of vibration signal is very important in the signal analysis.The pre-processing of vibration signal is the first step of signal eigenvalue extraction and directly determines effect and accuracy of application of subsequent diagnosis technology.This paper discusses on two methods of vibration signal pre-processing,the eliminating polynomial trend item and data smooth processing.Moreover,an algorithm program is realized based on MATLAB.To take an example for two well-known seismic waves,this paper verifies the above algorithm.The results show the processed signal has been improved obviously.更多还原