Fast structurally constrained pre-stack seismic inversion and reservoir lithology identification
Pre-stack seismic inversion focuses on the variation of amplitude with offsets/angles,which is a key technique for quantitative prediction of reservoirs.The stability,continuity and fidelity of conventional pre-stack seismic inversion are limited for complex structures such as high-dipping and faults.Based on this,we developed a fast structurally constrained pre-stack seismic inversion and reservoir lithology identification method.Firstly,we take into account the reflection features of seismic data,because the reflection features can represent the trend of seismic reflection events.The seismic reflection features can be characterized by the dip angles,which can be obtained from the poststack seismic data.Secondly,we construct the structural feature constraints and objective functions without rearranging the seismic gathers.On the one hand,the target generalized function is characterized by the lithology sensitivity factor,which makes it possible to identify the lithology directly.On the other hand,the Hadamard product operator is introduced to avoid the rearrangement of seismic gathers,which greatly improves the computational cost.Then,the Alternating Direction Method of Multipliers(ADMM)algorithm is used to solve the objective function.Finally,numerical experiments and practical application show that this method can effectively improve the noise resistance of pre-stack seismic inversion and the reliability of reservoir lithology identification,especially in complex geology.
NETL
NSTL
万方数据
