自适应滤波算法在纱疵信号去噪中的应用  

Application of Adaptive Filtering Algorithm in Defect Detection of Yarn Signal

作　　者：卢斌 何勇[1] 刘传群 彭达 LU Bin;HE Yong;LIU Chuan-qun;PENG Da(College of Mechanical Engineering,Donghua University,Shanghai 201620,China)

机构地区：[1]东华大学机械工程学院,上海201620

出　　处：《自动化与仪表》2018年第3期46-51,共6页Automation & Instrumentation

摘　　要：LMS算法在语音增强和信号预测等信号处理中有着广泛应用,在纺纱系统清纱过程中纱疵信号信噪比低,普通滤波算法无法检测出湮没在复杂噪声中的纱疵信号,而LMS自适应滤波算法能很好地滤去信号中的噪声。LMS自适应滤波算法的步长是影响其性能的重要因素,普通的固定步长LMS算法在系统稳态误差和收敛速度之间存在顾此失彼的缺点,为了降低这个矛盾并且提高算法的性能,该文将LMS算法中固定步长因子μ(n)设为与误差信号e(n)的相关的函数,从而引出一种变步长自适应滤波LMS算法,并使用Matlab对算法中的参数值进行分析与调整,Matlab仿真结果显示该算法能较好地协调系统的收敛速度和稳态误差性能。针对纱线信号的特征,将该算法用来对纱线信号进行初步处理,纱疵信号去噪率达到86%,为纱疵的后续精确识别和处理提供基础。:LMS algorithm has been widely used in signal processing such as voice enhancement and signal prediction.The signal-to-noise ratio of the yarn defect in the spinning system is low,and the general filtering algorithms can not detect the yarn defect in annihilation in complex noise.The LMS adaptive filtering algorithm can solve this problem well.The commonly used fixed-step LMS algorithm has a contradiction between system convergence rate and steady-state error.In order to reduce this contradiction and improve the performance of the algorithm,in this paper,the fixed step factorμ(n)in the LMS algorithm is set as a function related to the error signal e(n)and a variable step size LMS algorithm is proposed.And the parameters of the algorithm are analyzed and adjusted by using Matlab.The results of Matlab simulation show that the algorithm can coordinate the convergence speed and steady-state error performance of the system.According to the characteristics of the yarn signal,the algorithm is used to initialize the yarn signal,and the denoising rate reaches 86%,which provides a basis for the subsequent identification and processing of yarn defects.

关 键 词：自适应滤波 变步长自适应滤波算法 收敛速度 纱线信号 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]