CAJ | 학술논문

针对滚动轴承疲劳损伤初期声发射信号微弱而易湮没于噪声中难以有效检测的问题，提出一种基于二次相关加权阈值的声发射信号小波包降噪算法。通过计算原始信号各小波包分解成分自相关函数与噪声自相关函数的互相关系数（二次相关），对“3σ”阈值加权得到小波包分层阈值，实现对声发射信号的降噪。仿真分析与实测声发射信号分析表明，该方法能够有效去除干扰噪声，凸显声发射冲击信号。
침대곤동축승피로손상초기성발사신호미약이역인몰우조성중난이유효검측적문제，제출일충기우이차상관가권역치적성발사신호소파포강조산법。통과계산원시신호각소파포분해성분자상관함수여조성자상관함수적호상관계수（이차상관），대“3σ”역치가권득도소파포분층역치，실현대성발사신호적강조。방진분석여실측성발사신호분석표명，해방법능구유효거제간우조성，철현성발사충격신호。
Weak acoustic emission (AE)signals in the early stage of fatigue of rolling bearings may be annihilated in noise,and they are difficult to detect effectively.Aiming at this problem,a new method was proposed to reduce noise for these AE signals. The method was based on a quadratic correlation weighted threshold of wavelet packet transformation.The quadratic correlation was defined as the cross correlation coefficients among autocorrelation functions of wavelet packet decomposition components of the original AE signals and the autocorrelation of noise.The threshold for wavelet packet was obtained by weighting the thresholds of “3σ”with the quadratic correlation coefficients to realize the noise reduction of the AE signals.The analysis of simulated and measured acoustic emission signals showed that this method can effectively remove the interference of noise,and highlight the impulses in acoustic emission signals.