CN-122021411-A - Inversion method of shaft flow state

CN122021411ACN 122021411 ACN122021411 ACN 122021411ACN-122021411-A

Abstract

The embodiment of the application provides an inversion method of a well bore flow state. The method comprises the steps of constructing a shaft hydraulic model, collecting well-killing parameters of a well-killing site in real time, obtaining an initial state parameter set through the shaft hydraulic model, adopting an integrated Kalman filtering algorithm to generate a plurality of state parameter subsets corresponding to the initial state parameter set, correcting each state parameter subset according to the plurality of state parameter subsets and the well-killing parameters to obtain corrected each state parameter subset, inverting a shaft flowing state according to each corrected state parameter subset by the shaft hydraulic model to obtain shaft pressure distribution data and gas-containing rate distribution data, and obtaining inversion results according to a casing pressure value, a vertical pressure value and shaft pressure distribution data, or obtaining inversion results according to slurry pond increment and gas-containing rate distribution data. The method can accurately acquire the pressure distribution data and the gas content distribution data of the shaft.

Inventors

YANG HONGWEI
LI JUN
LONG ZHENYU
CHANG SIWEI
ZHANG GENG
ZHANG HUI
LIAN WEI
An Jintao
WANG DIAN

Assignees

中国石油大学(北京)

Dates

Publication Date: 20260512
Application Date: 20260108

Claims (10)

1. A method for inverting a flow state of a well bore, the method being applied to an electronic device and comprising: Constructing a shaft hydraulic model; Collecting well control parameters of a well control site in real time, wherein the well control parameters comprise a casing pressure value, a vertical pressure value and a mud pit increment; acquiring an initial state parameter set through the shaft hydraulic model; generating a plurality of state parameter subsets corresponding to the initial state parameter set by adopting an integrated Kalman filtering algorithm; correcting each state parameter subset according to the state parameter subsets and the well control parameters to obtain corrected state parameter subsets; inverting the flow state of the shaft by the shaft hydraulic model according to the corrected state parameter subsets to obtain shaft pressure distribution data and gas fraction distribution data; acquiring an inversion result according to the casing pressure value, the vertical pressure value and the wellbore pressure distribution data; or acquiring the inversion result according to the increment of the slurry pool and the gas fraction distribution data.
2. The method of claim 1, wherein the constructing a wellbore hydraulics model comprises: acquiring shaft structure data, physical parameters of fluid and well control process parameters; and constructing a wellbore hydraulics model based on multiphase flow theoretical data and a drift flow model according to the wellbore structure data, the physical parameters of the fluid and the well control process parameters, wherein the multiphase flow theoretical data is used for representing dynamic behaviors of the fluid in different phases, and the drift flow model is used for representing motion differences among the fluid in different phases.
3. The method of claim 1, wherein the obtaining an initial set of state parameters by the wellbore hydraulics model comprises: Acquiring shaft structure data; dividing the well bore into a plurality of grid nodes according to the well bore structure data; the wellbore hydraulics model adopts a finite difference method to calculate initial state parameters of each grid node and obtain an initial state parameter set.
4. The method of claim 1, wherein correcting each subset of state parameters based on the plurality of subsets of state parameters and the kill parameter to obtain corrected subsets of state parameters comprises: Calculating an average value of the plurality of state parameter subsets; obtaining covariance matrixes of the state parameter subsets according to the state parameter subsets and the average value; and obtaining corrected state parameter subsets according to the covariance matrix and the well control parameters.
5. The method of claim 4, wherein said deriving a covariance matrix for said plurality of state parameter subsets from said plurality of state parameter subsets and said average value comprises: In the formula, A covariance matrix representing the plurality of state parameter subsets, The average value is represented by a value of the average, Representing the j-th state parameter of the 1 st state parameter subset at time t, Representing the j-th state parameter of the 2 nd subset of state parameters at time t, Representing the jth state parameter of the mth state parameter subset at time t, Indicating the deviation value of the jth state parameter at time t.
6. The method of claim 4, wherein obtaining corrected subsets of state parameters based on the covariance matrix and the kill parameters comprises: In the formula, Representing the value of the kalman gain, Representing the covariance matrix, wherein H represents an intermediate operator, and R represents the covariance matrix of the well control parameter; Representing a matrix of the corrected plurality of state parameter subsets at time t, Representing the j-th state parameter of the i-th state parameter subset at time t, A j-th well control parameter indicating the added observation noise at the t-th time; representing the j-th state parameter of the corrected i-th state parameter subset at time t +1, A model operator representing the wellbore hydraulic model, Representing the error, Q represents the error covariance matrix of the wellbore hydraulic model.
7. The method of claim 1, wherein generating the plurality of state parameter subsets corresponding to the initial state parameter set using an integrated kalman filter algorithm comprises: In the formula, Representing the j-th state parameter of the i-th state parameter subset at a time t equal to 1, A model operator representing the wellbore hydraulic model, At time t is 0, adopting an integrated Kalman filtering algorithm, generating a j state parameter of the i state parameter subset according to the initial state parameter set, Representing errors, Q represents an error covariance matrix of the wellbore hydraulic model, E represents an identity matrix, and m represents the number of the plurality of state parameter subsets.
8. The method of claim 1, wherein the obtaining inversion results from the casing pressure values, the stand pressure values, and the wellbore pressure profile data comprises: acquiring a pressure value in a shaft annulus at the current moment from the shaft pressure distribution data, wherein the shaft annulus is an annular space formed between an inner sleeve of the shaft and a drill rod; Acquiring a pressure value in a wellhead drill rod from the wellbore pressure distribution data; if the casing pressure value is consistent with the pressure value in the annular space of the shaft, and the vertical pressure value is consistent with the pressure value in the wellhead drill rod, the inversion result is correct inversion; if the casing pressure value is inconsistent with the pressure value in the well bore annulus, and the vertical pressure value is inconsistent with the pressure value in the well head drill rod, the inversion result is an inversion error.
9. The method of claim 1, wherein the obtaining the inversion result from the mud pit increment and the gas fraction distribution data comprises: Calculating the increment of the mud pit to be verified according to the gas content distribution data; If the increment of the mud pit to be verified is consistent with the increment of the mud pit, the inversion result is that the inversion is correct; if the increment of the mud pit to be verified is inconsistent with the increment of the mud pit, the inversion result is an inversion error.
10. The method of claim 9, wherein the formula for calculating the mud pit increment to be verified based on the gas fraction distribution data is: In the formula, Indicating the increment of the mud pit to be verified at the t-th moment, Representing the current gas volume in the shaft at the t moment; representing the cumulative volume of the exhaust gas history, Representing the gas content of the p-th grid node of the well bore at the t-th moment, Representing the annulus cross-sectional area of the p-th mesh node of the wellbore, The grid step length of the grid nodes is represented, and N represents the total number of the grid nodes of the shaft; to represent the discharge flow, n is the number of time steps; For the time step size of the time step, Is the gas phase flow rate at the ground.

Description

Inversion method of shaft flow state Technical Field The application relates to the technical field of well drilling well control, in particular to a method for inverting a well shaft flowing state. Background In petroleum drilling engineering, development of deep wells and ultra-deep wells faces challenges of complex wellbore flow conditions and high well control risks. Dynamic changes in the gas-liquid two-phase flow within the wellbore directly affect the bottom hole pressure balance, possibly leading to serious safety risks and environmental hazards upon overflow or blowout accidents. The throttling well control technology is used as a core means of well control, and the bottom hole pressure is dynamically maintained to be slightly higher than the stratum pressure by adjusting the opening of a throttling valve and the casing pressure of a wellhead in real time so as to prevent blowout from being out of control. However, wellbore flow conditions during well killing are affected by a number of factors, which can result in dramatic changes in wellbore pressure and gas fraction distributions. When the traditional hydraulic model calculates the pressure and flow parameters of a well bore, the fluid state is assumed to change slowly, and the dynamic response requirement under the sudden working condition is difficult to adapt. In addition, field operators rely on experience to judge the state of the well bore, and the well control scheme is easy to fail due to information lag or misjudgment. Therefore, there is a need for an inversion method of wellbore flow state, which can accurately obtain wellbore pressure distribution data and gas content distribution data by obtaining directly observable data and reversely calculating unknown data. Disclosure of Invention The embodiment of the application provides an inversion method of a shaft flowing state, which is used for achieving the effect of accurately acquiring shaft pressure distribution data and gas content distribution data. In a first aspect, an embodiment of the present application provides a wellbore flow state inversion method, applied to an electronic device, including constructing a wellbore hydraulic model; the method comprises the steps of collecting well-killing parameters of a well-killing site in real time, wherein the well-killing parameters comprise a casing pressure value, a vertical pressure value and a slurry pool increment, acquiring an initial state parameter set through a well-shaft hydraulic model, generating a plurality of state parameter subsets corresponding to the initial state parameter set through an integrated Kalman filtering algorithm, correcting each state parameter subset according to the plurality of state parameter subsets and the well-killing parameters to obtain corrected state parameter subsets, inverting a well-shaft flowing state according to each corrected state parameter subset through the well-shaft hydraulic model to obtain well-shaft pressure distribution data and gas-containing rate distribution data, and acquiring inversion results according to the casing pressure value, the vertical pressure value and the well-shaft pressure distribution data, or acquiring inversion results according to the slurry pool increment and the gas-containing rate distribution data. In one possible implementation, constructing a wellbore hydraulics model includes obtaining wellbore structural data, physical parameters of a fluid, and well killing process parameters, and constructing a wellbore hydraulics model based on multiphase flow theory data and a drift flow model according to the wellbore structural data, the physical parameters of the fluid, and the well killing process parameters, wherein the multiphase flow theory data is used for representing dynamic behaviors of the fluid in different phases, and the drift flow model is used for representing motion differences between the fluid in different phases. In one possible implementation, acquiring the initial state parameter set through the wellbore hydraulics model comprises acquiring wellbore structure data, dividing the wellbore into a plurality of grid nodes according to the wellbore structure data, and calculating the initial state parameters of each grid node by the wellbore hydraulics model through a finite difference method to acquire the initial state parameter set. In one possible implementation, correcting each state parameter subset according to a plurality of state parameter subsets and well killing parameters to obtain corrected each state parameter subset, wherein the correcting each state parameter subset comprises calculating an average value of the plurality of state parameter subsets, obtaining covariance matrixes of the plurality of state parameter subsets according to the plurality of state parameter subsets and the average value, and obtaining corrected each state parameter subset according to the covariance matrixes and the well killing parameters. In one possi