Quantitative Analysis For Seismic Attribute Of Carbonate Caves Base On The Physical Model

The carbonate reservoir in China has a great potential of exploration and development.The pore and cave reservoir has the characteristics of deep burial and complicated distributions,it's difficult to make exploration and development and quantitative predictions.So the quantitative analysis for seismic attribute of carbonate caves base on the physical model is very important for reservoir prediction and reserves estimation.In this paper,on the basis of the three-dimensional physical model data of carbonate reservoir for Halahatang area in Tarim Basin,the seismic response characteristics of the buried hill karst reservoir and the interval karst reservoir are analyzed.By using seismic attribute technology and three-dimension visualization technology,the quantitative relationship curves between the apparent volume and the actual volume for the cave reservoir of the buried hill karst area and the interval karst area are established,and the differences of the relationship curves between the apparent volume and the actual volume for the caves reservoir of the buried hill karst area and the interval karst area are analyzed,these curves are affected by different factors such as different overlying layers,different fillings and different cave size.Through the examples of synthetic seismic records,the time-frequency energy spectrum of sparse constraint inverse spectral decomposition has higher time and frequency resolution,better time-frequency focusing property and more anti-noise ability than the traditional spectral decomposition methods such as STFT,S-transform and CWT,which can effectively reflect the local characteristics of signal and be not affected by the length and shape of time window.The time-frequency phase spectrum of sparse constraint inverse spectral decomposition not only clearly reflects the distribution of time-frequency energy,but also contains the effective phase information.The combination of the time-frequency energy spectrum and phase spectrum is used for the identification of the multi-caves combinations in the carbonate reservoir.This method not only can reflect the strong amplitude properties of the cave reservoir,but also can directly the combination types and the number of cave,which provides a good basis for the subsequent reservoir prediction and the quantitative estimation of the cave reservoir volume.In addition,according to the sparse property of sparse constraint inverse spectral decomposition and the reconfigurability of the seismic signal,the method isused for the random noise attenuation,and the random noise of the physical model data and the actual data of carbonate reservoir can be effectively removed.The method of Q estimation dependent on the frequency is very sensitive to the result of spectral estimation,which makes the accurate spectral estimation more favorable for the Q estimation.The problem of the time-frequency resolution exists in estimation of Q values by using the traditional spectral decomposition methods,but the sparse constraint inverse spectral decomposition just makes up for this defect,the Q estimation method based on sparse constraint inverse spectral decomposition is proposed.The advantages and effectiveness of proposed method are demonstrated by the examples of the synthetic seismic data.This proposed method is applied to the cave reservoir in carbonate rock with different gas content and different porosity,and get some understanding which can provide the basis for reservoir oil-gas possibility evaluation.