Seismic inversion initial model construction method based on deep learning

Abstract

The invention discloses a seismic inversion initial model construction method based on deep learning, and the method comprises the steps: taking a logging longitudinal wave impedance curve, a logging transverse wave impedance curve, a logging longitudinal and transverse wave velocity ratio curve and a logging density curve as learning targets, and employing a depth feedforward neural network algorithm to fuse a conventional inversion initial model and seismic attributes so as to construct an inversion initial model with low-frequency information and stratigraphic structure information at the same time. The initial model constructed by the method has better consistency with the stratum occurrence characteristics, the problem of time penetration in the traditional initial model can be effectively solved, high-density horizon interpretation data is not needed, the workload of horizon interpretation can be greatly reduced, and the inversion result corresponding to the initial model constructed by the method better conforms to geological knowledge.

Description

technical field [0001] The invention belongs to the field of oil and gas exploration and development, and in particular relates to a method for constructing an initial model of seismic inversion based on deep learning. Background technique [0002] Seismic inversion is the main method for reservoir prediction and hydrocarbon detection, and the initial model required for inversion is one of the key factors affecting the inversion effect. The existing seismic inversion initial model construction method uses layers and faults to construct an isochronous sequence frame, and uses an appropriate interpolation algorithm to extrapolate the low-frequency information in the logging along the sequence frame to obtain the initial model required for the inversion. The existing technical schemes for seismic inversion initial model construction mainly have the following problems: 1) The existing technical schemes are only suitable for sedimentary facies dominated by vertical accretion. Th...

AI Technical Summary

Problems solved by technology

The existing technical schemes for seismic inversion initial model construction mainly have the following problems: 1) The existing technical schemes are only suitable for sedimentary facies dominated by vertical accretion. The initial model constructed by the technical scheme will appear "time-traveling" phenomenon, which will lead to contradictions between the inversion results and the stratigraphic occurrence

2) When a complex fault system develops in the study area, the existing technical solutions require high-density horizon interpretation data to ensure the accuracy of the isochronous sequence frame, which will greatly increase the workload and difficulty of interpretation, and affect the accuracy of seismic inversion. progress and effect

Images