This application claims priority to Chinese Patent Application Ser. No. CN202211384610.7 filed on 7 Nov. 2022.
The present invention relates to the technical field of oil and gas field development, and in particular, to a method and system for predicting water flooding recovery of fault block reservoirs considering a whole process optimization.
Fault block reservoirs are widely developed in China, among which the proven ones have huge geological reserves and great development value. However, fault block reservoirs, characterized by unique geological structures, high fault development, complex oil-water system, and strong reservoir heterogeneity, are difficult to explore and develop. At present, crude oil is widely exploited by water flooding development at home and abroad. A number of development dements of water flooding fault block reservoirs have entered into high water cut stage, the water flooding effects of which vary greatly, with some up to 60% or more and individual ones less than 10%. Therefore, it is a key for optimal planning and adjustment of fields to accurately predict water flooding recovery in different development dements. At present, water flooding recovery prediction is mainly performed by core analysis, water flooding characteristic curve, production decline, and empirical formula. Core analysis is based on core model flooding experiments, and fails to accurately simulate fault characteristics due to the size limitation of core models. Water flooding characteristic curve and production decline are based on the analysis of water flooding development effect of integral reservoirs, without considering the influence of specific parameters of fault block reservoirs, such as fault block area, fault density, and water body multiples. Empirical formula, depending on specific reservoir types, is limited in generalization. Therefore, it is urgent to provide a method for predicting water flooding recovery of fault block reservoirs because that the existing recovery prediction methods fail to accurately grasp the potential space of water flooding development of the fault block reservoirs.
In view of the deficiencies of the prior art and the development characteristics of fault block reservoirs, the present invention proposes a method for predicting water flooding recovery of fault block reservoirs considering a whole process optimization. Water flooding recovery of fault block reservoirs can be predicted more accurately in the present invention, which is of great significance for evaluating and understanding the development potential of old oilfields such as water flooding fault block reservoirs and making reasonable development and optimization schemes, effectively assisting optimal planning and deployment of fields.
The present invention further provides a system for predicting water flooding recovery of fault block reservoirs considering a whole process optimization.
Technical solutions of the present invention are:
A method for predicting water flooding recovery of fault block reservoirs considering a whole process optimization includes the following steps:
Further preferably, in step (1), the selected reservoir static physical parameters include underground crude oil viscosity μo, effective formation thickness h, permeability k, inter-layer permeability ratio Vm, and variation coefficient of permeability Vr; the selected production dynamic parameters include well spacing density Wd and production multiples PV; and the selected characteristic parameters of the fault block reservoirs include fault block area A, fault density df, and water body multiples N.
Further preferably, the variance D in step (2) is calculated by formula (I):
In formula (I), D represents the variance; n represents the number of samples; Xi represents the water flooding recovery corresponding to the i value of a certain influencing factor, %; and X represents an average value of n water flooding recovery, %.
Further preferably, the power function relationship, logarithmic function relationship and polynomial function relationship in step (3) are shown in formula (II), formula (III) and formula (IV), respectively:
R=aX
b (II)
R=a ln(X)+b (III)
R=a
0
+a
1
X+a
2
X
2
+a
3
X
3
+ . . . +a
n
X
n+ . . . (IV)
In formula (II), formula (III) and formula (IV), R represents the recovery, %; X represents the master parameters; and a, b, an represents undetermined coefficient of the functional relationship with a subscript n=0, 1, 2, 3 . . . .
A specific implementation process of step (4) preferred according to the present invention includes:
A specific implementation process of step (5) preferred according to the present invention includes:
Further preferably, the correlation model between the water flooding recovery of the fault block reservoirs and all the master parameters in step (6) is shown in formula (V):
R=a
1 ln(A)+a2 log(PV)+a3 ln(μo)+a4(df)+a5kb+a6Vmc+a7Vr2+a8Vr+a9 (V)
In formula (V), R represents the recovery, %; A represents the fault block area, km2; PV represents the production multiples; μo represents the underground crude oil viscosity, mPa·s; df represents the fault density, bar/km2; k represents the permeability, 10−3 μm2; Vm represents the inter-layer permeability ratio; and Vr represents the variation coefficient of permeability.
A system for predicting water flooding recovery of fault block reservoirs considering a whole process optimization includes:
Beneficial effects of the present invention are:
The present invention will be further explained in details with reference to the accompanying drawings and specific embodiments, whereby the above and other objects, features and advantages of the present invention would be more obvious and understandable.
A method for predicting water flooding recovery of fault block reservoirs considering a whole process optimization, as shown in
(1) Determining Influencing Factors in Water Flooding Recovery of Fault Block Reservoirs
Geological information of fault block reservoirs was collected in a target block; reservoir static physical parameters and production dynamic parameters were determined in combination with field production data; and fault block characteristic parameters affecting water flooding recovery of the fault block reservoirs were determined based on basic characteristics and actual development thereof.
(2) Screening Master Parameters of Water Flooding Recovery of Fault Block Reservoirs
Values of each of the influencing factors were changed using single-factor analysis method; the water flooding recovery of the fault block reservoirs was calculated by using numerical simulators of water flooding; significance of the influencing factors was analyzed with a variance D as an evaluating criterion for primary and secondary influencing factors, where the greater the variance D was, the higher the significance of the influencing factors was; and the influencing factors with variances greater than 1 were selected as master parameters of the water flooding recovery of the fault block reservoirs.
(3) Determining a Single-Factor Correlation Between the Water Flooding Recovery of the Fault Block Reservoirs and the Master Parameters
A correlation between flooding recovery of the fault block reservoirs and each of the master parameters was determined using non-linear regression method, where the correlation included power function relationship, logarithmic function relationship, and polynomial function relationship.
(4) Designing Multi-Factor Orthogonal Experimental Schemes Based on the Master Parameters
The number of levels and values of each of the master parameters were determined in combination with allowable range of parameters in fault block reservoir field, and an appropriate orthogonal experimental design table was selected.
(5) Performing a Whole-Process Water Flooding Optimization for Each of the Orthogonal Experimental Schemes
With the maximum water flooding recovery as a target, separate-layer injection and production was optimized at the moment of production; well-type conversion was optimized when comprehensive water cut of the reservoirs reaches 90%; and injection and production adjustment was optimized when the comprehensive water cut of the reservoirs reaches 95%.
(6) Establishing a Prediction Model for the Water Flooding Recovery of the Fault Block Reservoirs
A formula of a correlation model between the water flooding recovery of the fault block reservoirs and all the master parameters was determined according to the single-factor correlation between water flooding recovery and the master parameters determined in step (3); unknown parameters of the formula of the correlation model between the water flooding recovery and all the master parameters were determined using least square method based on results of the whole process optimization of orthogonal experiments, so as to obtain a prediction model of the water flooding recovery of the fault block reservoirs.
(7) Calculating the Water Flooding Recovery of the Fault Block Reservoirs
The water flooding recovery of the fault block reservoirs was calculated according to the prediction model of the water flooding recovery of the fault block reservoirs established in step (6).
The method for predicting water flooding recovery of fault block reservoirs considering a whole process optimization according to Embodiment 1, the differences are:
In step (1), geological information of M fault block reservoirs, a target block, was collected; reservoir static physical parameters and production dynamic parameters were determined in combination with field production data; and fault block characteristic parameters affecting water flooding recovery of the fault block reservoirs were determined based on basic characteristics and actual development thereof.
M fault block belongs to a fan-open fault block reservoir with a dip angle among 2 to 5°, the updip direction of which is blocked by two intersecting faults and open to one side. The plane shape is like a fan. Based on the field data and the characteristics of block development, the selected reservoir static physical parameters include underground crude oil viscosity μo, effective formation thickness h, permeability k, inter-layer permeability ratio Vm, and variation coefficient of permeability Vr; the selected production dynamic parameters include well spacing density Wd and production multiples PV; and the selected characteristic parameters of the fault block reservoirs include fault block area A, fault density df, and water body multiples N.
A model of reservoir numerical simulation is established according to the actual data of M fault block field, which is divided into 62×145×5=44950 grids using corner grid system. The wells are arranged parallel to the oil-water boundary line according to the characteristics of fault blocks and considering the actual situation in the field, with a well pattern of determinant. The model of reservoir numerical simulation is shown in
The variance D in step (2) is calculated by formula (I):
In formula (I), D represents the variance; n represents the number of samples; Xi represents the water flooding recovery corresponding to the i value of a certain influencing factor, %; and X represents an average value of n water flooding recovery, %.
The influencing factors in water flooding recovery of fault block reservoirs and calculation results of primary and secondary factors thereof are shown in Table 1:
As shown in Table 1, the primary and secondary factors of each of the influencing factors are: underground crude oil viscosity>production multiples>variation coefficient of permeability>permeability>fault block area>inter-layer permeability ratio>fault density>water body multiples>well spacing density>effective formation thickness. The influencing factors with variances greater than 1 are selected as master parameters of the water flooding recovery of the fault block reservoirs, including: underground crude oil viscosity, production multiples, variation coefficient of permeability, permeability, fault block area, inter-layer permeability ratio, and fault density.
The power function relationship, logarithmic function relationship and polynomial function relationship in step (3) are shown in formula (II), formula (III) and formula (IV), respectively:
R=aX
b (II)
R=a ln(X)+b (III)
R=a
0
+a
1
X+a
2
X
2
+a
3
X
3
+ . . . +a
n
X
n+ . . . (IV)
In formula (II), formula (III) and formula (IV), R represents the recovery, %; X represents the master parameters; and a, b, an represents undetermined coefficient of the functional relationship with a subscript n=0, 1, 2, 3 . . . .
In this embodiment, a correlation between flooding recovery of the fault block reservoirs and each of the master parameters is determined using non-linear regression method, where there is a power function relationship of the water flooding recovery with the permeability and inter-layer permeability ratio; there is a logarithmic function relationship of the water flooding recovery with the fault block area, production multiples and underground crude oil viscosity; there is a quadratic polynomial function relationship between the water flooding recovery and the variation coefficient of permeability; and there is a linear relationship between the water flooding recovery and the fault density.
A specific implementation process of step (4) includes:
Firstly, the upper and lower limits of values of parameters required for orthogonal experiment analysis were determined based on distribution intervals of the reservoir static physical parameters and fault block characteristic parameters determined in step (1) obtained from field tests.
Secondly, the upper and lower limits of values of the required production dynamic parameters were determined according to distribution intervals of the production dynamic parameters used in developed reservoirs of the same type. Several levels (5) were taken for each of the parameters in the orthogonal experiment analysis, and values of each of the levels were uniformly sampled between the upper and lower limits of the values of each of the parameters.
Finally, the orthogonal experimental design table was determined and the multi-factor orthogonal experimental schemes were compiled according to the determined number of the master parameters and several values (5) of the levels of each of the parameters.
In this embodiment, the number of levels of each of the master parameters is determined to be five in combination with relevant research data and the value range of the actual parameters of reservoir fields, the values thereof being shown in Table 2.
On this basis, an orthogonal experimental design table with 7 influencing factors and 5 levels is selected. The obtained multi-factor orthogonal experimental scheme is shown in Table 3, which is the orthogonal experimental scheme for the master parameters of water flooding recovery of fault block reservoirs and the results of the whole process optimization thereof.
A specific implementation process of step (5) includes:
Multi-layer reservoirs were divided into two sets of layer series of development longitudinally at the moment of water flooding into production according to the orthogonal experiment schemes for each of the fault block reservoirs; specifically, different combinations of the multi-layer reservoirs were calculated by using numerical simulators of water flooding reservoirs, where the combination of the multi-layer reservoirs corresponding to the scheme with the maximum water flooding recovery was a preferred implementation of the separate-layer injection and production.
Producing wells were converted into water injection wells every other when the comprehensive water cut of the reservoirs reached 90%, that is, the original line of producing wells was converted into the producing wells and water injection wells arranged at intervals.
Injection rates of each of the water injection wells and liquid producing rates of each of the producing wells were calculated as adjustable variables by using the numerical simulators of water flooding reservoirs when the comprehensive water cut of the reservoirs reached 95%, where the combination of the injection rates of each of the water injection wells and the liquid producing rates of each of the producing wells corresponding to the scheme with the maximum water flooding recovery was a preferred injection and production scheme. The numerical simulation results of water flooding recovery of fault block reservoirs based on the whole process optimization are shown in Table 3.
The correlation model between the water flooding recovery of the fault block reservoirs and all the master parameters in step (6) is shown in formula (V):
R=a
1 ln(A)+a2 log(PV)+a3 ln(μo)+a4(df)+a5kb+a6Vmc+a7Vr2+a8Vr+a9 (V)
In formula (V), R represents the recovery, %; A represents the fault block area, km2; PV represents the production multiples; μo represents the underground crude oil viscosity, mPa·s; df represents the fault density, bar/km2; k represents the permeability, 10−3 μm2; Vm represents the inter-layer permeability ratio; and Vr represents the variation coefficient of permeability.
The verification results of the prediction model, with goodness of fit up to 98.95%, are shown in
A system for predicting water flooding recovery of fault block reservoirs considering a whole process optimization includes:
Number | Date | Country | Kind |
---|---|---|---|
202211384610.7 | Nov 2022 | CN | national |