基于训练字典的稀疏分解应用于动液面声波信号去噪

Application of Sparse Decomposition Based on Training Dictionary to the Signal of Acoustic Wave Denoising of Dynamic Liquid Level

为实现油井动液面声波信号的有效去噪,获得清晰的反射波信号,本文根据过完备字典的稀疏分解理论,利用KSVD算法对动液面样本数据进行训练,并结合OMP算法构造动液面信号过完备训练字典,完成对动液面声波信号的有效去噪声。并对比小波硬阈值降噪效果,验证了本文研究的基于KSVD算法训练字典的稀疏分解对解决油井动液面声波信号去噪问题的有效性。

In order to realize the effective denoising of the acoustic signal of the oil well dynamic liquid level and obtain the clear reflection wave signal, this article is based on the over-complete dictionary of sparse decomposition theory, train dynamic liquid level sample data with KSVD algorithm, and combined with OMP algorithm to construct dynamic liquid level signal over-complete training dictionary. It compares the denoising effect with wavelet hard threshold denosing, verifies this sparse decomposition training over-complete dictionary of this paper based on the KSVD algorithm. It shows notable effectiveness in denoising signal of reflection acoustic wave in oil well dynamic liquid level.