Study Of The Priori Information Construction And Seismic Stochastic Inversion Method

Seismic stochastic inversion is a nonlinear inversion method combining priori information construction and optimization theory.Compared with the deterministic inversion,stochastic inversion based on statistical priori information can effectively integrate the high-frequency information of well-logging data and have a higher resolution.However,the conventional seismic stochastic inversion method is running the defects of slow computing speed and poor lateral continuity.In this paper,we study fast priori information construction methods – the geostatistical priori information construction based on Fast Fourier Transform-Moving Average(FFT-MA)spectrum simulation and Gradual Deformation Method(GDM),and the mixed priori information construction based on fractal theory and kriging interpolation technology,as well as the improved perturbation methods and optimization strategies to improve the computational speed of seismic stochastic inversion.By combining with deterministic inversion method,to some extent it can improve the lateral continuity of stochastic inversion results.Seismic stochastic inversion is formulated in the Bayesian theory framework.Under the premise of obtaining high quality seismic data and well-logging data,we can make full use of the known geological information,seismic data and well-logging data to obtain the inversion results by combining geostatistics,fractal theory with seismic inversion technology.So the inversion results are more in line with the constraints of geological structure.Stochastic simulation technology can characterize quantificationally the uncertainty of reservoir parameter models.Then use the seismic inversion and nonlinear optimization algorithm for the screening of simulation results.It provides reliable reservoir parameters for seismic interpretation and has very important practical significance.Stochastic inversion methods we proposed directly use the exact Zoeppritz equation to calculate synthetic seismogram,which can reduce the error caused by the Zoeppritz approximate equations,So the inversion results are more accurate.An important task of reservoir exploration and evaluation is to identify the reservoir fluids.In this paper,we extract the elastic parameters,petrophysical parameters and fluid factors from the limited seismic data,well-logging data and geological background,which can be more comprehensive for the reservoir exploration and development.Rock physics theory is very important for getting the qualitative evaluation of rock skeleton and pore fluid properties.So we can use rock physics theory to build the relationship between petrophysical parameters,fluid factors and seismic data.And then realize the seismic stochastic inversion of petrophysical parameters and fluid factors.In addition,we apply the direct inversion method,which avoids the error accumulation and reduces the uncertainty transmission.And the synthetic data and practical application show that the methods we proposed not only have theoretical significance,but also certain practical value.