弧焊过程电信号小波软阈值消噪

在ＣＯ２弧焊过程电信号的实际测试过程中，噪声的存在常是难以避免的，消除噪声是信号分析中重要的环节。小波变换具有很好时、频域性，以不同的小波变换尺度，可将分解成不同的频率分量。对于连续信号函数，随着尺度的增大，小波系数也增大；对于噪声，其小波变换系数随尺芳的增大而减小。据此，可消除信号中的噪声成分。论文重点分析了小波软阈值法信号消噪方法，通过对实测弧焊过程电信号的消噪处理，对比分析了低通滤波、粗糙小

De-noising in Electric Signals of Arc Welding Process via Wavelet Soft threshold

During the practical measuring of the electric signals of CO 2 welding, the noise always can't be avoided. De noising is a key link in analyzing signals. Wavelets have good time frequency characteristics. The signals can be decomposed into different frequency components with different wavelet scales.For continuous signals, wavelet transform coefficient will increase in direct ratio with the scale. For noise, the coefficient will decrease in inverse ratio with the scale. On the basis of the principle, noise may been removed from original signal. The de noising via wavelet soft threshold is emphatically analyzed in this paper. The de noising results of three kinds of methods for the practically measured electric signals of arc welding process are given out. De noising by the traditional low pass filter and wavelet crude filter, the signals will show serious distortion in the breaking position. De noising via wavelet soft threshold can keep the break position of signals out of distortion as well as eliminate the noise in the signals. So this method can improve the effect of extracting the characteristic information from the signals.