Offshore oilfield subsection water injection parameter design method based on communication unit delimitation

Abstract

An offshore oilfield segmented water injection parameter design method based on communication unit demarcation comprises the specific steps that S1, an offshore loose sandstone longitudinal multilayer oil reservoir inter-well connectivity model is constructed, an inter-well communication network is constructed, and the bottom hole pressure of each well and the water content of each sand prevention section are calculated; matching the production dynamics calculated by the model with the actual historical production dynamics by optimizing the characteristic parameters, realizing the inversion solution of the model parameters, and determining a historical fitting objective function suitable for the connected model; and S2, determining a water injection splitting coefficient, water injection efficiency and a fluid channeling index of each layer of water injection well of the oil reservoir according to the inter-well connectivity model of the offshore loose sandstone longitudinal multi-layer oil reservoir constructed in the S1. Aiming at the problem that a production well of the offshore loose sandstone reservoir is easy to generate water channeling or water logging and the like, in combination with a real-time production allocation and injection allocation production algorithm, accurate inversion of the connectivity model and real-time prediction of the production dynamics of an oil-water well are realized according to geological characteristics, development dynamics and construction of the connectivity model of the offshore loose sandstone reservoir.

Description

Technical Field

[0001] The present invention relates to the technical field of oilfield development, and more specifically to a method for designing segmented water injection parameters for offshore oilfields based on the delineation of connected units. Background Art

[0002] Due to the loose argillaceous cementation, easy sand production, and severe reservoir heterogeneity of sandstone reservoirs, after long-term high-speed water injection development, the main seepage areas in the reservoir plane and between layers are fixed, and problems such as water channeling, water flooding, and dead oil areas are prominent. The utilization efficiency of injected water is low, resulting in poor oil recovery and water flooding development effect in the oilfield.

[0003] Currently, there are inversion models such as related analysis models, multiple linear regression models, capacitance models, and system analysis models that use oilfield injection-production dynamic data to study inter-well connectivity, but there are still some problems. For example, it is impossible to conduct connectivity analysis and calculation by layer, and it is difficult to characterize the inter-layer heterogeneity; the connectivity parameters inverted by the model lack clear geological significance; it can only fit and predict production fluid data, and cannot perform inversion calculations by integrating other oil-water dynamic indicators such as water cut, reducing the reliability of the inversion results. Therefore, we invented the design of segmented water injection parameters for offshore oilfields based on the delineation of connected units. Summary of the Invention

[0004] The present invention overcomes the deficiencies in the prior art and provides a method for designing segmented water injection parameters for offshore oilfields based on the delineation of connected units.

[0005] The object of the present invention is achieved by the following technical solutions.

[0006] The method for designing segmented water injection parameters for offshore oilfields based on the delineation of connected units specifically includes the following steps:

[0007] S1. Construct an inter-well connectivity model for a vertically multi-layered reservoir of loose sandstone in the offshore area, construct an inter-well connectivity network, calculate the bottom hole pressure of each well and the water cut of each sand control section, and make the production dynamics calculated by the model match the actual historical production dynamics by optimizing characteristic parameters to realize the inversion solution of model parameters and determine the historical fitting objective function applicable to this connectivity model;

[0008] S2. According to the inter-well connectivity model for a vertically multi-layered reservoir of loose sandstone in the offshore area constructed in S1, determine the water injection splitting coefficient, water injection efficiency, and cross-flow index of the injection wells in each layer of the reservoir, and determine the inter-well connectivity after water injection development in the oilfield according to the water injection efficiency and cross-flow index to judge whether the water injection scheme for each layer section needs to be adjusted.

[0009] The specific steps of S1 include:

[0010] S11. Simplify and represent the reservoir system as an interconnected node network considering geological characteristics, and calculate the conductivity for calculating the inter-well flow rate.

[0011] q ij = T ij (p i - p j )

[0012] In the formula, q ij is the flow rate between well i and well j, m 3 / d; T ij is the inter-well conductivity, m 3 / Mpa; p i and p j are the pressures at two well points respectively, Mpa;

[0013] S12. Establish the material balance equation,

[0014]

[0015] In the formula, N l is the number of all layers of well i, layer; N w is the number of all upstream wells of well i, well; T ijk is the inter-well conductivity between well i and well j in the k-th layer, m 3 / (d·Mpa); p i , p j are the bottom-hole pressures of well i and well j respectively, Mpa; q i is the flow rate of well i, positive for injection and negative for production, m 3 / d; C tk is the comprehensive compressibility of the k-th layer, Mpa -1 ; V ijk is the connected volume between well i and well j in the k-th layer, m 3 ;

[0016] S13. Using the implicit difference method, the unknowns within the time step are calculated by solving an algebraic equation set at each time step,

[0017]

[0018] S14. Calculate the water cut of well i in the k-th layer and the bottom-hole pressures of each well. The oil saturation at each node can be obtained by using the saturation tracking monitoring method, and then:

[0019]

[0020] In the formula: is the water cut of well i in the k-th layer at time n, %; is the water cut of well i in the k-th layer from the direction of well j at time n, %; q ijk is the flow rate between well i and well j in the k-th layer, m 3 / d;

[0021] S15. Calculate the dynamic indexes of water flooding development based on the established connectivity model, and make the calculated production dynamics of the model coincide with the actual historical production dynamics by optimizing and adjusting the characteristic parameters of the connectivity units between wells, so as to realize the inversion solution of the model parameters;

[0022] Determine the historical matching objective function applicable to the connectivity model as follows,

[0023]

[0024] Satisfy:

[0025] m≥0

[0026]

[0027] In the formula, O(m) is the historical matching objective function; m is the vector composed of inversion parameters, which is the vector composed of connectivity volume, conductivity and relative permeability parameters; d obs is the actual observed value; G(m) is the vector composed of the production dynamic data calculated and generated by the present invention; C d is the dynamic covariance matrix; V R is the total effective pore volume of the reservoir, m 3 ;

[0028] S16. Use the EnOpt algorithm to obtain the approximate gradient of the objective function in S15;

[0029] S17. According to the gradient obtained in S16, use the projection gradient method to iteratively solve the constrained problem, and its iterative formula is,

[0030]

[0031] In the formula: m l+1 is the reservoir parameter matrix of the (l + 1)-th iteration; γ is the search step size; T is the projection matrix.

[0032] The specific steps of S16 include:

[0033] S161. Calculate the mean value of the parameter set and the covariance matrix ∑ θ ,

[0034]

[0035]

[0036] S162. Update the parameter set using the gradient information of the objective function, and the gradient of the objective function is approximately

[0037]

[0038] In the formula, J i is the objective function value of the i-th set member, and

[0039]

[0040] is the mean value of the objective function values;

[0041] S163. Repeat the above steps until the objective function converges or reaches the preset number of iterations to obtain the approximate gradient.

[0042] In S2, after determining the mobility in the connected unit using the upstream weight method and calculating the transverse and longitudinal splitting coefficients, the water injection efficiency of the injection well in each sand control section is obtained.

[0043] The specific steps of S2 include:

[0044] S21. Using the inverted connectivity model, according to the seepage theory, the productivity index of the inter-well connected unit is expressed as

[0045]

[0046] In the formula: J ijk is the productivity index between wells i and j in the k-th layer, m 3 / (d·Mpa); λ ik is the mobility at well i, 10 -3 μm 2 / (mPa·s); λ ijk is the mobility in the connected unit between wells i and j in the k-th layer, 10 -3 μm 2 / (mPa·s); L ijk is the inter-well distance between wells i and j in the k-th layer, m; r ik is the wellbore radius of well i in the k-th layer, m; s ik is the skin factor of well i in the k-th layer; the superscripts n and n - 1 represent the n-th and (n - 1)-th time steps respectively;

[0047] S22. Determine the mobility in the connected unit from the mobilities at the two end nodes of the connected unit by the upstream weight method,

[0048]

[0049] In the formula, λ jk is the mobility at well j, 10 -3 μm 2 / (mPa·s); K ijk is the average permeability between wells i and j in the k-th layer, 10 -3 μm2 ; K ro 、K rw are the relative permeabilities of oil and water respectively, 10 -3 μm 2 ; S wik and S wjk are the water saturations of Well i and Well j in the k-th layer; μ ok 、μ wk are the viscosities of oil and water in the k-th layer respectively, mPa·s;

[0050] S23. Calculate the total productivity index of Well i and the vertical splitting coefficient in the k-th layer,

[0051]

[0052] In the formula, A ik is the splitting coefficient of Well i in the k-th layer; J ik is the productivity index of Well i in the k-th layer, m 3 / (d·Mpa); J i is the total productivity index of Well i, m 3 / (d·Mpa);

[0053] S24. Determine the planar splitting coefficient between Well i and Well j in the k-th layer by the product of the inter-well conductivity and the pressure difference as,

[0054]

[0055] In the formula, A ijk is the splitting coefficient of Well i in the k-th layer in the direction of Well j; q ijk is the split liquid volume of Well i in the k-th layer in the direction of Well j within the connected unit, m 3 / d; q ik is the split liquid volume of Well i in the k-th layer, m 3 / d;

[0056] S25. Set N l virtual nodes for the water well according to the number of vertically perforated layers. Considering the total water injection volume of the water well and the water absorption ratio of the inspection, matching, testing and adjustment section, the splitting of the injection dynamics of each section can be realized, forming a new simulation method for multi-layer injection allocation measures. Furthermore, the accurate distribution of the planar water injection splitting coefficient of each layer of the reservoir can be obtained. Then the water injection volume of Well i in the k-th layer is:

[0057]

[0058] In the formula: σ ik is the water absorption ratio of Well i in the k-th layer during testing and adjustment; q i is the water injection volume of Well i, m 3 ;

[0059] S26. Calculate the water injection efficiency of the water injection well in each sand control section:

[0060]

[0061] In the formula, e wis is the water injection efficiency of well i in the s-th section, %; f wjk is the water cut of well j in the k-th layer, %; N s is the number of perforated layers in the s-th section;

[0062] S27. Let NI be the comprehensive index of inter-well crossflow, which is used to characterize the degree of crossflow;

[0063] NI = 0.5Z 注采井组 + 0.5max(Z 对应采油井 )

[0064] In the formula: Z 注采井组 is the evaluation index of the injection-production well group, and Z 对应采油井 is the evaluation index of the corresponding production well;

[0065] S28. Judge the severity of crossflow according to the value of the comprehensive index of inter-well crossflow.

[0066] When the comprehensive index is in the range of 0 - 0.48, the production status of the injection-production well group is normal and no water injection adjustment is required;

[0067] When the comprehensive index is in the range of 0.48 - 0.72, the degree of crossflow is medium, and water control measures can be carried out for the well group;

[0068] When the comprehensive index is in the range of 0.72 - 1, the degree of crossflow is strong, and the well group must carry out water control measures and adjust water injection.

[0069] max(Z 对应采油井 ) in S27 is the direction of the most serious water crossflow.

[0070] The beneficial effects of the present invention are as follows: Based on the design of segmented water injection parameters for offshore oilfields delimited by connected units, mainly aiming at the problem that water crossflow or water flooding is likely to occur in production wells of offshore unconsolidated sandstone reservoirs, combined with the real-time production allocation and injection algorithm, the accurate inversion of the connectivity model and the real-time prediction of the production dynamics of oil and water wells are realized. Based on the principle of inter-well connectivity, according to the geological characteristics, development dynamics, etc. of offshore unconsolidated sandstone reservoirs, an inter-well connectivity model for horizontal and vertical multi-layer offshore unconsolidated sandstone reservoirs is constructed, and the inter-well connectivity is quantitatively characterized in the sand control section, and then data such as the plane and longitudinal water injection splitting coefficients, water injection efficiency, and crossflow index of injection wells in each layer of the reservoir are obtained, which is convenient for guiding oilfield development. Description of the Drawings

[0071] Figure 1 It is a schematic diagram of the fitting result of cumulative oil production in the block in the embodiment of the present invention;

[0072] Figure 2 Schematic diagram of the fitting result of formation water cut in the embodiment of the present invention;

[0073] Figure 3 Schematic diagram of the simulation calculation error of multiple well sections in the embodiment of the present invention;

[0074] Figure 4 Schematic diagram of the water injection efficiency of a single well in the embodiment of the present invention;

[0075] Figure 5 Schematic diagram of the complete process of well group adjusted water injection in the embodiment of the present invention. Detailed implementation manners

[0076] Embodiment

[0077] The method for designing the segmented water injection parameters of an offshore oilfield based on the delineation of connected units specifically includes the following steps:

[0078] S1. Construct an inter-well connectivity model for a longitudinal multi-layer reservoir of unconsolidated sandstone in the offshore area, construct an inter-well connectivity network, calculate the bottom hole pressure of each well and the water cut of each sand control section, and make the production performance calculated by the model match the actual historical production performance by optimizing the characteristic parameters, so as to realize the inversion solution of the model parameters and determine the historical fitting objective function applicable to this connectivity model;

[0079] S2. According to the inter-well connectivity model of the longitudinal multi-layer reservoir of unconsolidated sandstone in the offshore area constructed in S1, determine the water injection splitting coefficient, water injection efficiency, and crossflow index of the injection wells in each layer of the reservoir, and determine the inter-well connectivity after the water injection development of the oilfield according to the water injection efficiency and crossflow index, so as to judge whether the water injection scheme for each layer section needs to be adjusted.

[0080] The specific steps of S1 include:

[0081] S11. Simplify and represent the reservoir system as an interconnected node network considering geological characteristics, and calculate the conductivity for calculating the inter-well flow rate,

[0082] q ij = T ij (p i - p j )

[0083] In the formula, q ij is the flow rate between well i and well j, m 3 / d; T ij is the inter-well conductivity, m 3 / Mpa; p i and p j are the pressures of two well points respectively, Mpa;

[0084] In this step, the reservoir system is simplified and characterized as an interconnected node network considering a series of complex geological features such as well points, faults, and water bodies. The inter-well connectivity network is characterized by parameters such as conductivity, connected volume, bottom-hole pressure, and saturation. The conductivity is used to characterize the fluid flow capacity, so as to calculate the inter-well flow rate through the conductivity.

[0085] S12. Establish the material balance equation.

[0086]

[0087] In the formula, N l is the number of all layers of Well i, layer; N w is the number of all upstream wells of Well i, well; T ijk is the conductivity between Well i and Well j in the kth layer, m 3 / (d·Mpa); p i , p j are the bottom-hole pressures of Well i and Well j respectively, Mpa; q i is the flow rate of Well i, positive for injection and negative for production, m 3 / d; C tk is the comprehensive compressibility of the kth layer, Mpa -1 ; V ijk is the connected volume between Well i and Well j in the kth layer, m 3 ;

[0088] In this step, according to the data obtained in S11, the mass conservation of inter-well fluid flow is described, and the material balance equation is established. The material balance equation represents the relationship between the total amount of fluid flowing into the reservoir, the total amount of fluid flowing out of the reservoir, and the change in the amount of fluid in the reservoir within a certain period of time.

[0089] S13. Using the implicit difference method, the unknowns within the time step are calculated by solving an algebraic equation set at each time step.

[0090]

[0091] This step is used to improve the stability and accuracy of numerical calculations when dealing with fluid flow and the material balance equation.

[0092] S14. Calculate the water cut of Well i in the kth layer and the bottom-hole pressure of each well. The oil saturation at each node can be obtained by using the saturation tracking and monitoring method, and then:

[0093]

[0094] In the formula: is the water cut of Well i in the kth layer at time n, %; is the water cut of Well i in the kth layer from the direction of Well j at time n, %; qijk is the flow rate between Well i and Well j in the k-th layer, m 3 / d;

[0095] S15. Calculate the dynamic indicators of water flooding development based on the established connectivity model, and make the model calculation production dynamics coincide with the actual historical production dynamics by optimizing and adjusting the characteristic parameters of the connectivity units between wells, so as to realize the inversion solution of the model parameters;

[0096] Determine the historical matching objective function applicable to the connectivity model as follows,

[0097]

[0098] Satisfy:

[0099] m ≥ 0

[0100]

[0101] In the formula, O(m) is the historical matching objective function; m is the vector composed of inversion parameters, which is the vector composed of connectivity volume, conductivity and relative permeability parameters; d obs is the actual observed value; G(m) is the vector composed of the production dynamic data calculated and generated by the present invention; C d is the dynamic covariance matrix; V R is the total effective pore volume of the reservoir, m 3 ;

[0102] In this step, according to the Bayesian theory, a framework is provided to combine the prior information and the observed data for parameter estimation, and the error function is incorporated into the Bayesian framework to achieve the purpose of determining the historical matching objective function applicable to the connectivity model.

[0103] S16. Use the EnOpt algorithm to obtain the approximate gradient of the objective function in S15;

[0104] The existing gradient-free local production optimization algorithms are the least squares method (OLS), the ensemble optimization algorithm (EnOpt), and the stochastic perturbation approximation method (SPSA). Since there are many large-scale optimization problems in the connectivity units in this scheme, the EnOpt algorithm is used to obtain the approximate gradient.

[0105] S17. According to the gradient obtained in S16, use the projected gradient method to iteratively solve this constrained problem, and its iterative formula is,

[0106]

[0107] In the formula: m l+1 is the reservoir parameter matrix of the (l + 1)-th iteration; γ is the search step size; T is the projection matrix.

[0108] The specific steps of S16 include:

[0109] S161. Calculate the mean of the parameter set and the covariance matrix ∑ θ ,

[0110]

[0111] S162. Update the parameter set using the gradient information of the objective function. The gradient of the objective function is approximated as

[0112]

[0113] where J i is the objective function value of the i-th set member, is the mean of the objective function values;

[0114]

[0115] where α is the step factor;

[0116] S163. Repeat the above steps until the objective function converges or reaches the preset number of iterations to obtain the approximate gradient.

[0117] In S2, the mobility in the connected unit is determined by the upstream weight method. After calculating the transverse and longitudinal splitting coefficients, the water injection efficiency of the injection well in each sand control section is obtained.

[0118] The specific steps of S2 include:

[0119] S21. Using the inverted connectivity model, according to the seepage theory, the productivity index of the inter-well connected unit is expressed as

[0120]

[0121] where: J ijk is the productivity index between wells i and j in the k-th layer, m 3 / (d·Mpa); λ ik is the mobility at well i, 10 -3 μm 2 / (mPa·s); λ ijk is the mobility in the connected unit between wells i and j in the k-th layer, 10 -3 μm 2 / (mPa·s); L ijk is the inter-well distance between wells i and j in the k-th layer, m; r ik is the wellbore radius of well i in the k-th layer, m; s ik is the skin factor of well i in the k-th layer; the superscripts n and n - 1 represent the n-th and (n - 1)-th time steps respectively;

[0122] S22. Determine the mobility within the connected unit based on the mobilities at the two end nodes of the connected unit through the upstream weight method.

[0123]

[0124] In the formula, λ jk is the mobility at well j, 10 -3 μm 2 / (mPa·s); K ijk is the average permeability between wells i and j in the k-th layer, 10 -3 μm 2 ; K ro , K rw are the relative permeabilities of oil and water respectively, 10 -3 μm 2 ; S wik and S wjk are the water saturations of wells i and j in the k-th layer; μ ok , μ wk are the viscosities of oil and water in the k-th layer respectively, mPa·s;

[0125] S23. Calculate the total productivity index of well i and the vertical splitting coefficient in the k-th layer.

[0126]

[0127] In the formula, A ik is the splitting coefficient of well i in the k-th layer; J ik is the productivity index of well i in the k-th layer, m 3 / (d·Mpa); J i is the total productivity index of well i, m 3 / (d·Mpa);

[0128] S24. Determine the planar splitting coefficient between wells i and j in the k-th layer by the product of the inter-well conductivity and the pressure difference as

[0129]

[0130] In the formula, A ijk is the splitting coefficient of well i in the k-th layer in the direction of well j; q ijk is the split liquid volume of well i in the direction of well j within the connected unit in the k-th layer, m 3 / d; q ik is the split liquid volume of well i in the k-th layer, m 3 / d;

[0131] S25. Set the water wells to N according to the number of vertically perforated layers. lFor a virtual node, considering the total injection volume of the water well and the water absorption ratio of the inspection, matching, testing and adjusting section, the splitting of the injection dynamics of each section can be realized, forming a new simulation method for multi-layer water injection with separate injection measures, and then obtaining the accurate distribution of the plane water injection splitting coefficient of each layer of the reservoir. The injection volume of Well i in the k-th layer is:

[0132]

[0133] In the formula: σ ik is the water absorption ratio of Well i in the k-th layer measured and adjusted; q i is the injection volume of Well i, m 3 ;

[0134] S26. Calculate the water injection efficiency of the injection well in each sand control section:

[0135]

[0136] In the formula, e wis is the water injection efficiency of Well i in the s-th section, %; f wjk is the water cut of Well j in the k-th layer, %; N s is the number of layers perforated in the s-th section;

[0137] S27. Let NI be the comprehensive index of inter-well crossflow, which is used to characterize the degree of crossflow;

[0138] NI = 0.5Z 注采井组 + 0.5max(Z 对应采油井 )

[0139] In the formula: Z 注采井组 is the evaluation index of the injection-production well group, and Z 对应采油井 is the evaluation index of the corresponding production well;

[0140] S28. Judge the severity of crossflow according to the value of the comprehensive index of inter-well crossflow,

[0141] When the comprehensive index is in the range of 0 - 0.48, the production status of the injection-production well group is normal and no water injection adjustment is required;

[0142] When the comprehensive index is in the range of 0.48 - 0.72, the degree of crossflow is medium and the well group can carry out water control measures;

[0143] When the comprehensive index is in the range of 0.72 - 1, the degree of crossflow is strong and the well group must carry out water control measures and adjust water injection.

[0144] max(Z 对应采油井 ) in S27 is the direction of the most serious water crossflow.

[0145] Such as Figure 1 and Figure 2As shown in the figure, in this embodiment, taking Block X in the Bohai Oilfield as an example, according to the above steps, considering the reservoir geological parameters, based on the principle of inter-well connectivity, using the production dynamic parameters of the target block as the fitting index, adjusting the model conductivity and connectivity volume for automatic history matching, the fitting accuracy reaches 94.2%. Figure 1 The cumulative oil production of the block shown in Figure 2 and the water cut fitting results of the block shown in are relatively good.

[0146] Furthermore, select the simulation calculation error statistics of 15 segments of 4 wells, namely C4, C7, C38, and C54, which were used for water absorption profile testing in 2020. As Figure 3 shown, the relative errors are all kept within 10%, indicating that the simulation calculation of this method has a good correspondence with the on-site measured data, verifying the accuracy of the method of considering building the model based on the equivalent of the sand control section of the injection well and splitting the calculation based on the production index.

[0147] Based on the inter-well connectivity model of the combined well pattern in Block X after history matching, perform simulation calculations using the water injection splitting coefficient of the injection well, the liquid production of each production well, etc. As Figure 4 shown, obtain the average water injection efficiency of the block. As Figure 5 shown, in the last step, dynamically adjust the water injection plan through the above content, and the complete process ends.

[0148] To sum up, based on the connection unit method, this solution can transform the complex three-dimensional field of the reservoir into a one-dimensional connected field, inversely calculate the water injection efficiency and crossflow degree of each layer section, determine the inter-well connectivity after the oilfield water injection development, and then optimize the water injection plan for each layer section to achieve stable oil production and water control.

[0149] The method for adjusting the injection-production relationship of the reservoir applying this solution has achieved a good development effect of increasing oil production and controlling water in the block.

[0150] The above has described the embodiments of the present invention in detail, but the content described is only the preferred embodiments of the present invention and cannot be considered as limiting the scope of implementation of the present invention. All equivalent changes and improvements made according to the scope of the application of the present invention should still fall within the scope covered by the patent of the present invention.

Claims

1. A method for designing segmented water injection parameters for offshore oilfields based on the delimitation of connected units, characterized in that, The specific steps include: S1. Construct an inter-well connectivity model for the longitudinal multi-layered offshore unconsolidated sandstone reservoir, build an inter-well connectivity network, calculate the bottom-hole pressure of each well and the water cut of each sand control section, make the production dynamics calculated by the model match the actual historical production dynamics by optimizing the characteristic parameters, realize the inversion solution of the model parameters, and determine the historical fitting objective function applicable to this connectivity model; S2. According to the inter-well connectivity model of the longitudinal multi-layered offshore unconsolidated sandstone reservoir constructed in S1, determine the water injection splitting coefficient, water injection efficiency, and crossflow index of each layer of the injection well in the reservoir, and determine the inter-well connectivity after water injection development of the oilfield according to the water injection efficiency and crossflow index to judge whether the water injection scheme for each section needs to be adjusted.

2. The method for designing the segmented water injection parameters of an offshore oilfield delimited based on connection units according to claim 1, wherein The specific steps of S1 include: S11. Simplify and characterize the reservoir system as an interconnected node network considering geological characteristics, and calculate the conductivity for calculating the inter-well flow rate, qij = Tij(pi - pj) where q ij is the flow rate between well i and well j, m 3 / d; T ij is the inter-well conductivity, m 3 / Mpa; p i and p j are the pressures at two well points respectively, Mpa; S12. Establish a material balance equation, Where, N l is the total number of layers in Well i, layer; N w is the total number of upstream wells of Well i, well; T ijk is the conductivity between Well i and Well j in the k-th layer, m 3 / (d·Mpa); p i , p j are the bottom-hole pressures of Well i and Well j respectively, Mpa; q i is the flow rate of Well i, positive for injection and negative for production, m 3 / d; C tk is the comprehensive compressibility of the k-th layer, Mpa -1 ; V ijk is the connected volume between Well i and Well j in the k-th layer, m 3 ; S13. Use the implicit difference method to calculate the unknowns within the time step by solving an algebraic equation set at each time step, S14. Calculate the water cut of Well i in Layer k, the bottom-hole pressure of each well, and the oil saturation at each node can be obtained by using the saturation tracking monitoring method, and then obtain: Wherein: is the water cut at the nth moment in Well i of Layer k, %; is the water cut from the direction of Well j at the nth moment in Well i of Layer k, %; q ijk is the flow rate between Well i and Well j in Layer k, m 3 / d; S15. Calculate the dynamic indexes of water flooding development based on the established connectivity model, make the production dynamics calculated by the model match the actual historical production dynamics by optimizing and adjusting the characteristic parameters of each inter-well connectivity unit, and realize the inversion solution of the model parameters; Determine the historical fitting objective function applicable to the connectivity model as follows, Satisfy: m≥0 Where, O(m) is the history matching objective function; m is a vector composed of inversion parameters, which is a vector composed of connected volume, conductivity, and relative permeability parameters; d obs is the actual observed value; G(m) is a vector composed of the production performance data calculated and generated by the present invention; C d is the dynamic covariance matrix; V R is the total effective pore volume of the reservoir, m 3 ; S16. Use the EnOpt algorithm to obtain the approximate gradient of the objective function in S15; S17. According to the gradient obtained in S16, use the projected gradient method to iteratively solve this constrained problem, and its iterative formula is, ml +1 = ml - γT * ▽l(ml) where: m l+1 is the reservoir parameter matrix for the (l + 1)-th iteration; γ is the search step size; T is the projection matrix.

3. The method for designing segmented water injection parameters of an offshore oilfield based on connected units as claimed in claim 2, wherein The specific steps of S16 include: S161. Calculate the mean of the parameter set and the covariance matrix ∑ θ , S162. Update the parameter set using the gradient information of the objective function, and the approximate gradient of the objective function is, where J i is the objective function value of the i-th set member, is the mean of the objective function values; where α is the step factor; S163. Repeat the above steps until the objective function converges or reaches the preset number of iterations to obtain the approximate gradient.

4. The method for designing segmented water injection parameters of an offshore oilfield based on connectivity units according to claim 1, characterized in that: In S2, use the upstream weight method to determine the mobility within the connectivity unit, calculate the transverse and longitudinal splitting coefficients, and then obtain the water injection efficiency of the injection well in each sand control section.

5. The method for designing segmented water injection parameters of an offshore oilfield based on connected units as claimed in claim 1, wherein The specific steps of S2 include: S21. Using the inverted connectivity model, according to the seepage theory, the productivity index of the inter-well connectivity unit is expressed as, Where: J ijk is the productivity index between wells i and j in the k-th layer, m 3 / (d·Mpa); λ ik is the mobility at well i, 10 -3 μm 2 / (mPa·s); λ ijk is the mobility within the connected unit between wells i and j in the k-th layer, 10 -3 μm 2 / (mPa·s); L ijk is the distance between wells i and j in the k-th layer, m; r ik is the wellbore radius of well i in the k-th layer, m; s ik is the skin factor of well i in the k-th layer; the superscripts n and n - 1 represent the n-th and (n - 1)-th time steps respectively; S22. Determine the mobility within the connectivity unit from the mobilities at both ends of the connectivity unit by the upstream weight method, Where λ jk is the mobility at well j, 10 -3 μm 2 / (mPa·s); K ijk is the average permeability between wells i and j in the k-th layer, 10 -3 μm 2 ; K ro , K rw are the relative permeabilities of oil and water respectively, 10 -3 μm 2 ; S wik and S wjk are the water saturations of wells i and j in the k-th layer; μ ok , μ wk are the viscosities of oil and water in the k-th layer respectively, mPa·s; S23. Calculate the total productivity index of Well i and the longitudinal splitting coefficient in Layer k, Where A ik is the splitting coefficient of Well i in the k-th layer; J ik is the productivity index of Well i in the k-th layer, m 3 / (d·Mpa); J i is the total productivity index of Well i, m 3 / (d·Mpa); S24. Determine the planar splitting coefficient between Well i and Well j in Layer k by the product of the inter-well conductivity and the pressure difference as, Where A ijk is the splitting coefficient of Well i in the direction of Well j at the k-th layer; q ijk is the split liquid volume of Well i in the direction of Well j within the connected unit at the k-th layer, m 3 / d; q ik is the split liquid volume of Well i at the k-th layer, m 3 / d; S25. Set N l virtual nodes according to the number of longitudinally perforated layers of the water well. Considering the total water injection volume of the water well and the water absorption ratio of the inspection, matching, testing and adjustment section, the splitting of the injection dynamics of each section can be realized, forming a new simulation method for multi-layer separate injection measures. Furthermore, the accurate distribution of the plane water injection splitting coefficient of each layer of the reservoir can be obtained. Then, the water injection volume of well i in the kth layer is: Where: σ ik is the water injection absorption ratio of Well i in the k-th layer; q i is the water injection volume of Well i, m 3 ; S26. Calculate the water injection efficiency of the injection well in each sand control section: where e wis is the water injection efficiency of well i in the s-th section, %; f wjk is the water cut of well j in the k-th layer, %; N s is the number of layers perforated in the s-th section; S27. Let NI be the comprehensive index of inter-well crossflow, which is used to characterize the degree of crossflow; NI = 0.5Z 注采井组 + 0.5 max(Z 对应采油井 ) Where: Z 注采井组 is the evaluation index of the injection-production well group, and Z 对应采油井 is the evaluation index of the corresponding oil production well; S28. Judge the severity of crossflow according to the value of the comprehensive index of inter-well crossflow, When the comprehensive index is in the range of 0 - 0.48, the production status of the injection-production well group is normal and no water injection adjustment is required; When the comprehensive index is in the range of 0.48 - 0.72, the degree of crossflow is medium, and water control measures can be carried out for the well group; When the comprehensive index is in the range of 0.72 - 1, the degree of crossflow is strong, and the well group must carry out water control measures and adjust water injection.

6. The method for designing segmented water injection parameters of an offshore oilfield based on connected units according to claim 5, characterized in that; In S27, max(Z 对应采油井 ) is the water channeling direction with the most severe channeling.

Images