地震勘探数据采集及噪声压制方法的研究

发布时间：2018-02-13 01:24

本文关键词： 地震勘探抽取滤波器现场可编程门阵列小波阈值去噪　出处：《中国地质大学(北京)》2015年硕士论文　论文类型：学位论文

【摘要】：地震勘探作为一种重要的物理勘探手段,在油田的开发中起着重要的作用。野外地震数据采集和地震资料的数字处理技术是地震勘探的两个重要环节。只有得到高质量的地震剖面图,才能为石油勘探提供可靠的依据。实际的野外数据采集过程经常会受到各种噪声的干扰(规则干扰波和随机干扰波),这些都直接影响地震解释工作的准确性。本文分析了地震勘探仪器中调制器的量化噪声,为设备的选型提供了理论依据。针对调制器输出的高速码流,设计一套基于FPGA的数字抽取滤波器。研究了随机噪声的降噪算法,提出了一种新的降噪算法。主要内容有:1)建立了Sigma-delta ADC调制器的线性等效模型,具体分析了一阶和二阶调制器的量化信噪比。推导了高阶调制器量化信噪比和过采样率的关系。结果表明:采用四阶Sigma-delta ADC转换器进行采集可以实现120 d B以上的高精度的地震勘探技术。2)选用Xilinx公司的xc5vlx85-1ff676作为硬件开发平台,设计了一套基于FPGA多速率降采样数字滤波器系统。首先选择合适的滤波器结构,接着使用VHDL和Verilog硬件描述语言编写了滤波器代码。其中设计FIR滤波器时借助了Xilinx公司的开发软件ISE中自带IP核,极大的提高了开发流程,并且合理的运用芯片资源。主要实现两个功能:第一就是实现512倍的降采样,第二就是可以滤除带外噪声。3)介绍了地震勘探中常见噪声的特点,针对随机噪声研究了小波阈值消噪理论,提出了一种新的阈值函数消噪算法,经过测试,能够很好的实现去除随机噪声的效果。
[Abstract]:Seismic exploration as an important means of physical exploration, Field seismic data acquisition and digital processing of seismic data are two important links in seismic exploration. Only high-quality seismic profiles can be obtained. In order to provide reliable basis for petroleum exploration, the actual field data acquisition process is often disturbed by various kinds of noise (regular jamming wave and random interference wave), which directly affect the accuracy of seismic interpretation. In this paper, the quantization noise of modulator in seismic exploration instrument is analyzed. A set of digital decimation filter based on FPGA is designed for the high speed bitstream output of modulator, and the noise reduction algorithm of random noise is studied. A new noise reduction algorithm is proposed. The linear equivalent model of Sigma-delta ADC modulator is established. The quantization signal-to-noise ratio of first and second order modulators is analyzed in detail, and the relation between quantization signal-to-noise ratio and oversampling rate of high-order modulators is deduced. The results show that the acquisition of four-order Sigma-delta ADC converters can achieve 120dB or more. High-precision seismic exploration technology. 2) xc5vlx85-1ff676 of Xilinx Company is chosen as hardware development platform. A digital filter system based on FPGA multi-rate downsampling is designed. Then we use VHDL and Verilog hardware description language to write the filter code, in which the FIR filter is designed with the help of the IP core in the development software ISE of Xilinx Company, which greatly improves the development process. And reasonable use of chip resources. The main realization of two functions: the first is to achieve 512 times of downsampling, the second is to filter out of the band noise. 3) introduced the characteristics of common noise in seismic exploration. In this paper, wavelet threshold denoising theory is studied for random noise, and a new threshold function denoising algorithm is proposed. After testing, the effect of removing random noise can be achieved well.
【学位授予单位】：中国地质大学(北京)
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：P631.44

【参考文献】