A Study On Borehole Acoustic Reflection Imaging And Geological Application

The acoustic reflection imaging technology has recently become an important topic in the characteristic of geological structures using well logging data and plays an important role in the comprehensive evaluation of oil and gas reservoirs,geothermal reservoirs and reservoir production design.Borehole reflection imaging uses the weak reflection signals in array acoustic log.To attack this problem,this paper has carried out in-depth and detailed research on numerical simulation,signal processing method and field application analyses.The studies focus on developing the signal processing techniques to improve the reflected wave data quality and interpret the imaging result.The techniques include the imaging noise reduction methods and quantitative attribute analysis methods.The major contribution of thesis consists of the following three parts.In the first part,we study the extraction method of weak reflected wave signal.Firstly,the problems of extracting and characterizing the reflected wave in the array acoustic log data are analyzed.On this basis,the band-pass filtering,data interpolation,and other preprocessing methods are performed to eliminate effect of random noises and bad data gaps.Using the apparent slowness difference between reflected wave and borehole wave,the linear-prediction based wave separation method and the residual direct wave suppression method are studied.To improve the existing method,a dispersion-based linear prediction wave field separation method is developed to effectively eliminate dispersion effect of the borehole-dipole waves and further suppress the direct borehole wave in the data.Median filtering and f-k filtering method were used to further filter residual borehole waves.According to the characteristics of logging data acquisition,the up and down-going reflections are obtained from the common source gather(CSG)and the common receiver gather(CRG)data,respectively.The characteristics difference between borehole and reflected waves are studied and a reflected wave extraction method combining the array wave separation processing is used.This method can well preserve the formation reflected waves.In the end,a set of far borehole weak reflection signal extraction processing workflow is formed.Using the workflow and the CSG and CRG data gathers,the up and down-going reflection wavefields are obtained respectively,which provides the data for the imaging processing.The second part studies the data processing methods for improving SNR(signal to noise ratio)in the imaging processing.Firstly,considering the low SNR characteristic of reflected wave data,computing power limitation of the computer,and requirement of imaging quality,the efficient post-stack frequency-wavenumber f-k wave migration method is used.The influence of the dip moveout(DMO)correction in the frequency-wavenumber domain on the common midpoint(CMP)acoustic log data is studied.The processing method for the CMP data is integrated with the dip stacking and the NMO + CMP processing.On the basis of the workflow,an adaptive Wiener filter and an anisotropic diffusion filtering method are applied to the resulting data.At the same time,a velocity-guided median filtering(VGMF)is applied to the reflected waves to enhance the reflected waves and suppress the impulsive and other coherent noise.Moreover,we also study the adaptive Wiener filtering method and the anisotropic diffusion filtering method to eliminate the other imaging artifacts.After these processes,the image quality and interpretation accuracy are significantly improved.The third part analyses and interprets the reflection image data.Firstly,an attribute recognition and quantitative analysis method for acoustic reflection imaging is proposed.The statistical amplitude attribute and the analysis of the complex analytical signal are applied to the image data.At the same time,an instantaneous wave number attribute method is used to separate the up-dip and down-dip reflections.By combining the instantaneous amplitude attribute with the azimuthal dipole acoustic imaging,an efficient and accurate method for inverting reflector orientation is developed.This inversion method allows for obtaining geometric parameters of the reflector(i.e.strike,dip,and distance from the well).Based on the research results of the thesis work,a numerical simulation software and acoustic reflection imaging software are implemented.The cross well geologic body,near-well high angle geologic body,the cave,vug system and other types of geological bodies can all be modeled and their imaging results can be used to guide field data interpretation.Finally,the data processing workflow is applied to various field example,including the fracturing evaluation of hydraulic fracturing,the horizontal well reservoir boundary recognition,the unconventional oil and gas fracture analysis,and the near-well fault detection,etc.,with good application results.The applications provide technical support and the basis for the near-well reservoir architecture delineation.