基于盲源分离的抽油机噪声压制方法  

Method for noise suppression of pumping units based on blind source separation

作　　者：赵彦青[1] 萧蕴诗[1] 

机构地区：[1]同济大学电子信息工程学院,上海201804

出　　处：《计算机应用》2014年第A01期349-351,共3页journal of Computer Applications

基　　金：国家自然科学基金资助项目(40872090)

摘　　要：地震勘探领域中,采集的地震记录中往往混有抽油机噪声,严重干扰地震有效信号。针对该问题,提出一种基于最大信噪比(SNR)盲源分离(BSS)的抽油机噪声压制方法。该方法采用盲源分离的原理,结合地震数据信噪模型,分离出地震有效信号及抽油机噪声。模型试算与实际资料应用表明,该方法能够对抽油机噪声进行有效压制,提高地震资料的信噪比,为后续的地震资料处理提供了基础保障。In seismic exploration, seismic data are usually mixed with the noise from pumping units, which will seriously interfere with effective seismic signals. To address this issue, a method for noise suppression of pumping units, which is based on Blind Source Separation （BSS） of maximum Signal-to-Noise Ratio （SNR） algorithm, was proposed. This method can separate effective seismic data with the noise from pumping units based on the fundamentals of BSS and seismic signal-noise models. The applications of model and seismic data show that it can, not only suppress the noise from pumping units, but also improve the SNR of seismic data, thus offering a solid base for subsequent seismic data processing.

关 键 词：地震资料 抽油机 噪声压制 盲源分离 信噪比 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置] P315.61[自动化与计算机技术—控制科学与工程]