Reservoir fracture prediction method and system

Abstract

The invention discloses a reservoir fracture prediction method and system. The method comprises steps: multiple fractures are generated randomly in a work area range, and feature parameters of the multiple fractures are acquired; based on the feature parameters of the multiple fractures, a discrete fracture mesh model is established; based on the discrete fracture mesh model, production history numerical simulation is carried out, and production dynamic parameter simulation values are acquired; based on the production dynamic parameter simulation values, a covariance matrix for the productiondynamic parameter simulation values and actual production dynamic parameters is acquired, and Kalman gains are further acquired; based on the Kalman gains, the feature parameters of the multiple fractures are adjusted; and in a condition of convergence of the feature parameters of the multiple fractures, the converged feature parameters are used as the feature parameters of the reservoir fractures.

Description

technical field [0001] The invention relates to the field of seismic exploration, and more specifically, to a reservoir fracture prediction method and system. Background technique [0002] In the field of petroleum exploitation technology, in the process of water injection, gas, polymer development and other enhanced recovery factors, reservoir fractures have a great impact on fluid flow and directly affect the development effect: the natural main fractures of the reservoir often become water channeling, gas Inaccurately predicting the location of natural fractures in the reservoir will lead to a decline in the effect of water injection and gas injection, and even force the production well to be shut down. [0003] At present, there are three main methods for identifying the position, trend, and conductivity of natural fractures in reservoirs: first, it is judged based on the results of imaging logging. This method can identify fractures accurately, but only the fracture sur...

AI Technical Summary

Problems solved by technology

[0004] The inventors found that the existing crack prediction techniques based on the ensemble Kalman filter method are all based on conventional structured grids (orthogonal grids or corner grids, etc.), and such grids are effective in dealing with fracture types. The reservoir can only be simplified, and the conductivity of the fracture must be equivalent to the matrix grid. Therefore, it is impossible to accurately and easily simulate the fractured reservoir, and the corresponding fracture inversion results are not accurate enough.

Images