Research And Application Of Inspection Method For Low-level Faults

In the late stage of oilfield development,the low-level faults are of great significance for the research of the remaining oil distribution and the contradiction between injection and production.In Shengli,Daqing and other oil fields in China,these low-level faults can play a good role in lateral sealing of thin reservoirs and control the enrichment of remaining oil.On the conventional seismic profiles,the faults show the discontinuity,abrupt change of the shape,and sudden increase or decrease in the number of the same phase axis,what's more,they are always appear accompanying with the abnormal waves such as diffraction waves,scattering waves,and so on.Due to the limit of the seismic frequency band and signal-to-noise ratio,it is more difficult to apply conventional methods for small fault interpretation.Generally speaking,only a few wells may encounter low-level faults.Due to the limited horizontal extension range of the stratum in wells,it is difficult to determine the strike,dip,dipping direction,through well-to-well stratigraphic comparison.At present,many scholars at home and abroad have produced a series of fault prediction and recognition technologies.However,due to the large influence of noise on the traditional fault detection methods,it is difficult to accurately predict the low-level faults.Previous researches focused on fault identification methods,such as coherence cube and edge detection.However,there is a lack of systematic research on the seismic response characteristics of faults with different dip angles and different fault distances.This will make the detection and identification of low-level faults lack effective theoretical basis,and reduce the credibility of detection.In this paper,we firstly study the base pursuit algorithm based on compressed sensing and the complex field nonlinear anisotropic diffusion filtering.Obtain high-precision reflection coefficient profiles through the base-pursuit inversion,thus the vertical resolution and lateral continuity of the profiles were improved.Then the Shock filter was added to the anisotropic diffusion term in complex domain,and the linear field anisotropic diffusion filter was developed from the linear field to the nonlinear field.The real and imaginary parts of the complex field anisotropic diffusion filter include different potential information,the real part can be used for filtering,denoising,while the imaginary part can be used for edge(or small fault)detection.The anisotropic diffusion filtering and denoising capability of complex field is analyzed by numerical simulation and actual data.The imaginary part information is used to detect faults.Then we analyzed the effects of body curvature properties,ant tracking techniques,and base-tracking inversion methods on fault recognition.These methods can identify medium-or small-scale faults with high signal-to-noise ratios in mesoscale or seismic data,but the low-level fault recognition is more difficult.Therefore,it is a good method for the low-order faults identification to combine the body curvature property,ant tracking technology,and the base tracking algorithm with the complex field nonlinear anisotropic diffusion filter.Finally,the complex field nonlinear anisotropic diffusion filtering and base-tracking dipole inversion,the body curvature method are comprehensively analyzed.Several methods are complemented and verified by each other each other and verify each other,it is an effective method for the comprehensive prediction and recognition of low-order faults.