基于LP与小波包的复杂机电系统监测序列混沌降噪方法  

Chaotic noise reduction method for complex electromechanical system monitoring sequence based on LP and wavelet packet

作　　者：谢军太[1] 冯龙飞 高建民[1] 高智勇[1] 高旭[1] XIE Juntai;FENG Longfei;GAO Jianmin;GAO Zhiyong;GAO Xu(State Key Laboratory of Manufacturing Systems Engineering,Xi’an Jiaotong University,Xi’an 710049,China)

机构地区：[1]西安交通大学机械制造系统工程国家重点实验室,西安710049

出　　处：《振动与冲击》2020年第7期1-7,共7页Journal of Vibration and Shock

基　　金：国家自然科学基金(51375375)。

摘　　要：针对流程工业生产系统监测数据存在强噪声和混沌性的特点,提出了一种局部投影方法(Local Projection Method)与小波包方法相结合的信号降噪方法。该方法先利用局部投影方法从动力学系统嵌入流形的角度进行多次迭代降噪,并根据关联维数来判定迭代终止;再利用小波包方法从频率的角度进行降噪,抑制高频噪声的干扰,取得了较好的降噪效果。用Lorenz时间序列进行仿真验证,对仿真时间序列加入不同程度的噪声,对比分析小波包、局部投影与该方法降噪后的相空间、SNR值和最大Lyapunov指数,证明了该方法对于中高强度噪声具有更好的降噪效果。并将该方法用于某压缩机组的实际监测数据降噪,评估三种方法的降噪效果,进一步验证了该方法的优越性。Aiming at features of strong noise and chaos existing in monitoring data of process industrial production systems,a signal de-noising method based on the local projection(LP)method and wavelet packet was proposed.Firstly,LP method was used to perform several iterations for a signal’s de-noising from the point of view of dynamic system embedding manifold,and judge iteration termination using correlation dimension number.Then,the wavelet packet method was used to do noise reduction for a noisy signal from the point of view of frequency,suppress inference of high-frequency noise,and obtain better de-noising effect.Lorenz time series was used to do simulation verification.Different levels of noise were added to the time series,and phase space,SNR and the maximum Lyapunov exponent obtained with wavelet packet,LP and the proposed method,respectively were analyzed contrastively.The results showed that the proposed method has better noise reduction effect for medium-high intensity noise.Finally,the proposed method was used to reduce noise of actual monitoring data of a certain compressor unit,and the three methods’noise reduction effects were evaluated to further verify the superiority of the proposed method.

关 键 词：局部投影 小波包变换 关联维数 降噪 

分 类 号：TH17[机械工程—机械制造及自动化]