Font Size: a A A

The Application Of Adaptive Time-frequency Peak Filtering Based On Cross-correlation In Seismic Exploration

Posted on:2016-12-16Degree:MasterType:Thesis
Country:ChinaCandidate:X JinFull Text:PDF
GTID:2180330467497446Subject:Electronic and communication engineering
Abstract/Summary:PDF Full Text Request
Seismic exploration technology is a main means of the exploration of mineral resourcessuch as oil, natural gas. It mainly uses artificial technology to blast seismic wave. According tothe different law of seismic wave propagation, we can determine whether seismic recordcontains mineral resources. In the process of seismic exploration, it’s inevitably mixed withhumanities, environment noise and equipment noise. These noises can directly affect thejudgment of seismic exploration results. In order to improve the accuracy of seismicexploration, so noise attenuation in the seismic record is very important.The time-frequency peak filtering (TFPF) is a very effective method to suppress seismicrandom noise. Time-frequency peak filtering method is a kind of unbiased estimation algorithm,which has advantages in the time-frequency domain methods of suppressing random noise. Butin seismic signal processing, due to the uncertainties of random noise and seismic signal of thenonlinear effect, time-frequency peak filtering algorithm has an estimated error for the effectivesignal. Signal amplitude attenuation is serious, even causes signal waveform distortion, whichleads to poor time-frequency peak filtering section in effective residual signal.In order to improve the estimation accuracy of time-frequency peak filtering algorithm, wepropose adaptive TFPF method based on the cross-correlation characteristics of seismic records.We introduce a constraint of residual signal into time-frequency peak filtering algorithm byadding a filtering operator and weighted coefficients in fidelity term. Based on the coherence ofseismic event, we apply the adjacent multi-channel cross-correlation to adjust weightedcoefficients adaptively. Adaptive TFPF algorithm has bigger weighted coefficients for effectivesignals to maintain the signal amplitude, and smaller weight coefficients for the random noisesuppression. We apply this method to synthetic seismic data and field common-shot pointrecord. Compared with the traditional TFPF method, adaptive TFPF method based oncross-correlation has a better ability to reserve the amplitude of the signal, and improves thesignal-to-noise ratio of seismic record.
Keywords/Search Tags:Seismic signal processing, Cross-correlation, Weighted coefficient, Seismic random noise
PDF Full Text Request
Related items