Multi-level Complex-domain Anisotropic Diffusion Filtering Based On Frequency Expanding

Fractured reservoirs have strong heterogeneity,anisotropy and complex reservoir space characteristics.If the resolution of seismic data is lower than the signal to noise,the exploration of fractured oil and gas reservoirs is more difficult.There is a certain correlation between cracks and faults.In this paper,the faults and the cracks around the faults are combined(hereinafter referred to as "fractures")for research.In the research process,two issues were mainly considered:(1)The problem of accurately describing the boundary of the fractured body.The development of interference wave near the fractured body,how to keep the boundary information of the fractured body during noise suppression is worth studying.(2)The longitudinal resolution of the fractured body,that is,how to eliminate or reduce the influence of wavelet continuity on the longitudinal distribution of the fractured body.Specifically,the following research work has been carried out and corresponding results have been achieved:(1)After a series of comparisons and discussions,the multi-order complex domain anisotropic diffusion filtering method is studied.In order to solve the defect of poor edge retention ability of anisotropic diffusion filtering,Guy Gilboa developed anisotropic diffusion filtering into the complex domain.The real part and the imaginary part obtained after the anisotropic diffusion filtering in the complex domain represent different information.The real part is equivalent to Gaussian convolution of the image and can be used for filtering and denoising.The imaginary part is equivalent to the Gaussian convolution of the image.Laplace transform can be used for edge detection.If anisotropic diffusion filtering is performed on the extracted real part information again,the edges can be extracted on the basis of filtering,and the signal-to-noise ratio can be further improved.Through numerical simulation and actual data analysis,the feasibility of multi-order complex anisotropic diffusion filtering is proved.(2)According to convolution theory,seismic records are the convolution of reflection coefficient and wavelet.The fault fault is only reflected in the reflection coefficient.If we denoise the seismic record,the denoising process will affect the reflection coefficient and wavelet.The denoising effect is not ideal,and it is easy to distort the fault.In this paper,the reflection coefficient profile is obtained by base tracking,and the seismic records are synthesized with wavelets after multi-order complex anisotropic diffusion filtering.The results show that this method is better than filtering the profile directly to protect the fault boundary and improve the signal-to-noise ratio.(3)The importance of low-frequency and high-frequency information in seismic data is analyzed.First,through wavelet extension,we can find that if we widen the low frequency while maintaining the mid-to-high frequency,the main lobe amplitude of the wavelet is strengthened and the side lobe amplitude is weakened in the time domain,which can effectively improve the longitudinal resolution of the seismic record;Compared with non-broadband and wideband single-channel records and fault models,it can be found that the seismic records after extension are more concentrated in the stratum,which can effectively compress the time length of seismic wavelets.Third,by extending the actual seismic profile It can be found that the seismic profile rich in low-frequency information after extension has higher longitudinal resolution;(4)Through numerical simulation and analysis of the denoising ability of the actual data,the effectiveness of extension frequency and multi-order anisotropic diffusion filtering in the complex domain is proved.In addition,geometric attributes such as volume curvature attributes and ant tracking technology can also be combined with multi-order anisotropic diffusion filtering based on extension frequency to highlight boundary information.Therefore,it is possible to effectively detect and identify broken fractures and provide reliable data for seismic interpretation.