Research on signal processing method of mine wire rope damage detectio
-
摘要: 采用电磁检测法检测矿用钢丝绳受损情况时,检测信号中含有大量噪声,且存在尖峰和突变干扰,增大了损伤识别难度,需要对原始检测信号进行降噪处理。常用的傅里叶变换无法处理运行中的钢丝绳检测信号,而小波变换因存在平移不变性较差、频带混叠等问题而影响检测准确度。提出了基于双树复小波变换的矿用钢丝绳损伤检测信号处理方法。首先采用Q平移法构造双树复小波高低通滤波器,对原始信号进行3层双树复小波分解,得到高低频信号分量;然后采用最小极大方差软阈值方法对分解信号进行降噪处理;最后对降噪信号进行重构。在实验室环境下搭建了钢丝绳损伤检测试验平台,对基于双树复小波变换的钢丝绳损伤检测信号处理方法的降噪性能进行验证,结果表明:该方法可有效减少检测信号中的尖峰和突变数量,使信号平稳,降噪效果优于经典小波变换,且增大了奇异点处信号峰值,有利于后续特征提取。Abstract: When using electromagnetic detection method to detect the damage of mine wire rope, the detection signal contains a lot of noise. Moreover, there are spikes and mutations interference, which increase the difficulty of damage identification. Therefore, it is necessary to reduce the noise of the original detection signal. The commonly used Fourier transform cannot process the operating wire rope detection signal. The wavelet transform has problems of poor translation invariance and frequency aliasing, which affect the detection accuracy. This paper proposes a method for mine wire rope damage detection based on dual-tree complex wavelet transform. Firstly, a dual-tree complex wavelet high and low pass filter is constructed by Q-shift method, and the original signal is decomposed by 3-layer dual-tree complex wavelet to obtain the high and low frequency signal components. Secondly, a soft threshold method with minimax variance is used to reduce the noise of the decomposed signal. Finally, the noise reduction signal is reconstructed. A wire rope damage detection test platform is built in the laboratory environment to verify the noise reduction performance of the wire rope damage detection signal processing method based on dual-tree complex wavelet transform. The results show that the method can reduce the number of spikes and mutations in the detection signal effectively and make the signal stable. The noise reduction effect is better than that of classical wavelet transform. The method increases the signal peak value at the singularity point, which is beneficial to the subsequent feature extraction.
点击查看大图
计量
- 文章访问数: 83
- HTML全文浏览量: 18
- PDF下载量: 8
- 被引次数: 0