The Study Of The Sweet Spot Identification Method Based On Stochastic Inversion

With the rapid development of petroleum industry and the discovery of the sedimentary characteristics of reservoir with different geological characteristics,the low permeability,strong reservoir heterogeneity sandstone reservoir exploration and development has become a topic in recent years.The "sweetness" refers to local existence high permeability reservoir under the condition of the the common low permeability relatively and is the most attention target for oil and gas exploration and development in the low permeability reservoir.Markov chain monte carlo method(MCMC)is a kind of stochastic inversion,it can get a lot sampling from the posterior probability distribution.It can obtain unknown parameter estimates and all kinds of uncertainty information related to inversion parameters.It is a appropriate method for "sweetness" identification.But conventional MCMC method has certain limitation.Due to the limited computing time Settings or inappropriate search strategy,MH algorithm cannot fully search a complex parameter space.In order to solve the problem,this paper draw lessons from Quantum annealing algorithm and proposes quantum MH algorithm to improve the efficiency and precision of the inversion.It reduces the unnecessary transfer of states in preliminary stage of the algorithm to make Markov chain convergence in the posterior probability distribution quickly and steadily by introducing the concept of Hamilton.Under the framework of bayesian theory,we propose pre-stack inversion method based on quantum MH algorithm.We can build the prior distribution and the likelihood function by seismic data,well logging data and geological data and inverse the parameters we need by quantum MH sampling algorithm.The Pre-stack inversion method based on quantum MH algorithm can make Markov chain of the inversion parameters converge to a posteriori probability distribution quickly and steadily and we can obtain important parameters support for reservoir identification.In view of the Sweet plot recognition,elastic impedance inversion method based on quantum MH algorithm is proposed in this paper.Due to Sweet plot and non-Sweet plot have certain physical differences,so some elastic parameters of them have certain differences.So we can use rock physics method to find sensitive elastic parameters for Sweet plot.Then we inverse the sensitive elastic inversion to realize Sweet Plot identification.