A method for inverse calculation of pressure-measured mobility based on reservoir numerical simulation

Abstract

The present invention belongs to the field of exploration technology and engineering - well logging engineering, and specifically relates to a method for inverse calculation of pressure - flow mobility based on reservoir numerical simulation. The present invention collects the actual test pressure data obtained by pumping pressure measurement, establishes a numerical simulation model for pumping pressure measurement, obtains the simulated pressure dynamic data through simulation calculation, then optimizes the fluid parameters, fits the simulated pressure dynamic data and the actual test pressure data, and finally, based on the optimized fluid parameters, further inverse - calculates to obtain the fluid mobility, making the obtained fluid mobility closer to the real situation.

Description

Technical Field

[0001] The present invention belongs to the field of exploration technology and engineering - well logging engineering. Specifically, the present invention relates to a method for inverse calculation of pressure - flow mobility based on reservoir numerical simulation. Background Art

[0002] The wireline formation testing technology is a new technology for directly, quickly, and accurately judging the properties of formation fluids and studying reservoirs and oil and gas reservoirs. For some problems that are difficult to accurately interpret the fluid properties, oil - water interface, etc. by conventional well logging data, the above problems can be well solved by the new wireline formation testing well logging technology. At the same time, it can also obtain reservoir fluid, formation pressure and other data in a timely, fast, and accurate manner. The wireline formation testing technology has been successfully applied in many oilfields at home and abroad, mainly including: oil and gas exploration, fluid phase analysis, oil - gas - water identification, oil - water interface prediction, reservoir productivity prediction and productivity evaluation, reservoir permeability evaluation and interpretation, etc. Among them, the accurate evaluation of fluid mobility is the key to reservoir effectiveness evaluation, reservoir productivity and oil - water interface prediction, and accurate oil - gas - water identification.

[0003] Establishing the seepage model of the testing process is the basis for the interpretation of wireline formation testing data. Currently, the main interpretation models are mainly based on the seepage of single - phase fluids, that is, the fluid in the formation pores is assumed to be a single - phase fluid, and considering different influencing factors of the testing process, the relationship between formation and instrument parameters and the testing pressure response is obtained by analytical or numerical methods. Now, the commonly used methods for analyzing wireline formation testing data include the quasi - steady - state pressure - drop method, spherical - flow pressure - build - up method, radial - flow pressure - build - up method, formation flow analysis method, automatic fitting analysis method, neural network method, least - squares method and other methods. Its main principle is to use the double - logarithmic curves of pressure and pressure build - up, apply the fitting optimization algorithm, and through the fitting of the measured curve and the template curve in the interpretation chart, reveal the reservoir permeability and fluid mobility, and judge the pollution status of the reservoir. It needs to rely on professional well - testing interpretation software, and has the characteristics of a relatively complex model and high requirements for the quality of testing data.

[0004] Currently, the published related research mainly focuses on the wireline formation testing interpretation method based on the analytical model, and the research on the pressure - flow mobility interpretation method based on reservoir numerical simulation technology is relatively less. At the same time, the existing numerical simulation methods have not fully combined intelligent optimization algorithms to achieve accurate interpretation of fluid mobility. The present invention intends to propose an interpretation method of pressure - flow mobility based on numerical simulation by fully considering the physical process of pressure - pumping in the actual reservoir. Summary of the Invention

[0005] In view of the deficiencies of the prior art, the present invention discloses a method for inverse calculation of pressure-measured mobility based on reservoir numerical simulation. The present invention collects the actual test pressure data obtained by pump-pressure measurement, establishes a numerical simulation model for pump-pressure measurement, obtains simulated pressure dynamic data through simulation calculation, then optimizes the fluid parameters, fits the simulated pressure dynamic data and the actual test pressure data, and finally calculates the fluid mobility based on the optimized fluid parameters, making the obtained fluid mobility closer to the real situation.

[0006] The technical problems to be solved by the present invention are realized by the following technical solutions: A method for inverse calculation of pressure-measured mobility based on reservoir numerical simulation, comprising the following steps:

[0007] S1. By collecting test point information, obtain a dynamic data set for historical fitting of pump-pressure measurement, and obtain actual test pressure data;

[0008] S2. Establish a numerical simulation model for pump-pressure measurement, and obtain simulated pressure dynamic data through simulation;

[0009] S3. Based on the actual test pressure data obtained in step S1, use the simulated pressure dynamic data calculated in step S2 to define the minimization problem shown in formula (1), and by optimizing and adjusting the fluid parameters in the simulation process in step S2, make the calculated value conform to the actual value to achieve the fitting of the measured pressure dynamic data;

[0010]

[0011] In the formula: is the objective function, reflecting the gap between the calculated value and the actual value; is the fluid parameter vector; and C d are respectively the test pressures of each block obtained from actual tests and their error covariance matrix; is the vector of pressure dynamic data calculated by using reservoir numerical and simulation models;

[0012] S4. Inverse calculate the fluid mobility according to the final fluid parameters optimized in step S3.

[0013] Preferably, in step S4 of the present invention, the calculation formula of the fluid mobility is: fluid mobility = permeability / fluid viscosity.

[0014] Preferably, in step S3 of the present invention, a particle swarm optimization algorithm is used for solution.

[0015] Preferably, in step S1 of the present invention, select test points with a test time of more than 10 s, a constant pump pumping flow rate, and a stable pressure recovery value within a preset time as the data points for interpretation and inversion.

[0016] Preferably, in step S1, for the case where the actual measured pressure data is missing, linear interpolation is used for completion.

[0017] For the outliers in the actual measured pressure data, they are deleted and then obtained by linear interpolation.

[0018] For the case where the smoothness of the curve formed by the actual measured pressure data does not meet the requirements, exponential smoothing is used for processing.

[0019] Preferably, in step S3, according to the numerical simulation model of the pressure measuring pump pumping reservoir established in step S2, numerical simulations of the pressure measuring pump pumping are respectively carried out to obtain the pressure and pressure recovery curves during the pressure measuring pump pumping process, and the fluid mobility is calculated by using the steady-state or unsteady-state Darcy flow formula.

[0020] By comparing and analyzing the influence of the changes of each fluid parameter during the calculation process on the fluid mobility, the main controlling factors affecting the fluid mobility in the fluid parameters are determined as the fluid parameter vector in formula (1).

[0021] Preferably, the pressure recovery well test curve in double logarithmic coordinates is used to analyze the pressure propagation range during the pump pumping pressure measurement process to determine the nature of the measured fluid mobility.

[0022] Preferably, in step S2, the numerical simulation model of the pump pumping pressure measurement includes three modules: model geological and fluid property input, model generation, and model operation.

[0023] Module 1: Based on geological, fluid, and technological parameters such as different oil and gas reservoirs, fluid physical properties, and probe sizes, a fine-grid geological model is constructed by using the multi-layer radial logarithmic transformation method, and a multi-phase and multi-component fluid model is established considering the fluid component composition to simulate the flow pattern during the pressure measuring pump pumping process.

[0024] Module 2: Establish a numerical simulation model of the pump pumping pressure measurement reservoir that truly reflects specific information.

[0025] Module 3: Run the established numerical simulation model and output the theoretical dynamic curves such as the pump pumping fluid volume and pressure.

[0026] Preferably, the present invention further includes step 5: Inverting the reservoir permeability according to the final fluid parameters optimized in step S3 to realize the evaluation of the reservoir physical properties.

[0027] In view of the problems in the current mobility interpretation of the main control factors and the unclear pressure propagation range during the pump - off pressure measurement in oil and gas wells, as well as the low accuracy of mobility interpretation, the present invention innovatively proposes a method system for fitting measured pressure dynamic data and inverting mobility based on reservoir numerical simulation technology. This method is easy to understand, simple and fast to design, and can quickly fit the pump - off pressure measurement dynamic data under complex working conditions of oil and gas reservoirs according to the reservoir conditions of the target block, and also has certain guiding significance for the comprehensive evaluation of reservoir effectiveness.

[0028] Compared with the prior art, the beneficial effects of the present invention are as follows: Based on the actual test pressure data obtained from on - site pump - off pressure measurement, the pressure dynamic data calculated by using the established reservoir numerical simulation model of pump - off pressure measurement is used. By optimizing and adjusting fluid parameters, the calculated value is made to conform to the actual value to achieve the fitting of the measured pressure dynamic data. Then, the fluid mobility is inversely calculated by using the finally fitted corresponding fluid parameters, so that the obtained fluid mobility is closer to the actual situation and more accurate, providing reliable reference data for on - site operations. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] Figure 1 It is a model diagram of gas - in - place saturation of a deep condensate gas reservoir in the embodiment;

[0030] Figure 2 It is a schematic diagram of the double - logarithmic curve of the pressure derivative recovery in pump - off pressure measurement in the embodiment;

[0031] Figure 3 It is a schematic diagram of the fitting curve of pump - off pressure measurement dynamic data in the embodiment;

[0032] Figure 4 It is a schematic diagram of the change of the objective function value during the fitting process of pump - off pressure measurement dynamic data in the embodiment. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0033] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention.

[0034] A method for inverting the mobility of pressure measurement based on reservoir numerical simulation includes the following steps:

[0035] S1. By collecting test point information, obtain a dynamic data set for historical fitting of pump - off pressure measurement to obtain actual test pressure data.

[0036] S2. Establish a numerical simulation model of pump - off pressure measurement, and obtain simulated pressure dynamic data through simulation.

[0037] S3. Based on the actual test pressure data obtained in step S1, using the simulated pressure dynamic data calculated in step S2, define the minimization problem shown in formula (1). By optimizing and adjusting the fluid parameters in the simulation process in step S2, make the calculated value conform to the actual value to achieve the fitting of the measured pressure dynamic data.

[0038]

[0039] In the formula: is the objective function, reflecting the gap between the calculated value and the actual value; is the fluid parameter vector; and C d are the test pressures and their error covariance matrices of each block obtained from actual tests respectively; is the pressure dynamic data vector calculated by using reservoir numerical and simulation models.

[0040] S4. Invert and calculate the fluid mobility according to the final fluid parameters optimized in step S3.

[0041] In step S4, the calculation formula of fluid mobility is: fluid mobility = permeability / fluid viscosity.

[0042] In step S3, the particle swarm optimization algorithm is used for solution.

[0043] In step S1, select the test points with a test time of more than 10 s, a constant pumping flow rate, and a stable pressure recovery value in a relatively short time as the data points for interpretation and inversion.

[0044] In step S1, for the situation where the actual test pressure data is missing, linear interpolation is used for completion.

[0045] For the outliers in the actual test pressure data, they are deleted and then obtained by linear interpolation.

[0046] For the situation where the smoothness of the curve formed by the actual test pressure data does not meet the requirements, exponential smoothing is used for processing.

[0047] In step S3, according to the pressure measurement pump pumping reservoir numerical simulation model established in step S2, conduct pressure measurement pump pumping numerical simulations respectively to obtain the pressure and pressure recovery curves during the pressure measurement pump pumping process, and calculate the fluid mobility using the steady-state or non-steady-state Darcy flow formula.

[0048] By comparing and analyzing the influence of the changes of each fluid parameter on the fluid mobility during the calculation process, determine the main control factors affecting the fluid mobility in the fluid parameters as the fluid parameter vector in formula (1).

[0049] Analyze the pressure propagation range during the pumping pressure measurement process using the pressure buildup well test curve in double logarithmic coordinates to determine the properties of the measured fluid mobility.

[0050] In step S2, the pumping pressure measurement numerical simulation model includes three modules: model geology and fluid property input, model generation, and model operation.

[0051] Module 1: Based on different geological, fluid, and process parameters, use the multi-layer radial logarithmic transformation method to construct a fine-grid geological model, consider the fluid component composition, and establish a multi-phase and multi-component fluid model to simulate the flow pattern during the pressure measurement pumping process.

[0052] Module 2: Establish a pumping pressure measurement reservoir numerical simulation model that truly reflects specific information.

[0053] Module 3: Run the established numerical simulation model and output theoretical dynamic curves such as pumping fluid volume and pressure.

[0054] The pressure measurement mobility inversion calculation method based on reservoir numerical simulation is characterized in that it further includes step 5: Invert and calculate the reservoir permeability according to the final fluid parameters optimized in step S3 to realize the evaluation of reservoir physical properties.

[0055] Example

[0056] Step S1, Preprocessing of dynamic data of pumping pressure measurement in oil and gas wells

[0057] Due to the large differences in the properties of the reservoir during pumping pressure measurement, abnormal test points such as tight points and dry points appear, resulting in the pressure continuing to rise or dropping rapidly after the pump is stopped, making it difficult to reach a stable state, and the pressure recovery value decreases or shows no recovery trend as the number of pumping times increases; in addition, due to complex and harsh working conditions such as high temperature, high pressure, and multiphase flow during construction, abnormal points such as seal loss points, overpressure points, and thick mud cake points will appear, resulting in no obvious change in the pumping test pressure, a long test time, and irregular changes in the pressure recovery curve, making it difficult to accurately calculate and invert the fluid mobility.

[0058] To address the above problems and avoid mobility interpretation errors, select test points with a long test time (generally more than 10 s), a constant pumping flow rate, and a stable pressure recovery value within a relatively short time (for example, the preset time is 10 minutes) as the data points for interpretation and inversion, and perform operations such as supplementing missing data, correcting abnormal values, data translation, and curve smoothing on the abnormal points. Specifically, for missing data, linear interpolation is used to complete it; for abnormal values, they can be deleted and then obtained by linear interpolation; for the case where the curve is not smooth enough, exponential smoothing can be used for processing to improve the data quality and form a dynamic data set applied to the history matching of pumping pressure measurement.

[0059] Step S2: Establishment of a numerical simulation model for an oil and gas reservoir during the pump pressure measurement process

[0060] Based on the current characteristics of cable formation testing technology, an oil reservoir numerical simulation method is used to establish an oil reservoir numerical simulation model that can reflect the actual process of pressure measurement by pump pumping. By inputting parameters such as pump pumping flow rate and time for calculation, dynamic pressure data can be obtained.

[0061] The numerical simulation model for pump pressure measurement includes three modules: input of model geology and fluid properties, model generation, and model operation. Module 1: Based on geological, fluid, and technological parameters such as different oil and gas reservoirs, fluid physical properties, and probe sizes, a fine-grid geological model is constructed using the multi-layer radial logarithmic transformation method. Considering the fluid component composition, a multi-phase and multi-component fluid model is established to simulate the flow pattern during the pump pressure measurement process. Module 2: Quickly establish a numerical simulation model for pump pressure measurement in an oil reservoir that truly reflects specific information. Module 3: Run the established numerical simulation model, which can directly output theoretical dynamic curves such as pump fluid volume and pressure.

[0062] Such as Figure 1 , a diagram of the numerical simulation model for pump pressure measurement in an oil and gas reservoir in the South China Sea is given.

[0063] Step S3: Establishment of a fitting model for pressure measurement dynamic data based on the particle swarm algorithm

[0064] First, analysis of the pressure propagation range based on the pressure recovery curve

[0065] The pressure recovery well test curve in double logarithmic coordinates is used to analyze the pressure propagation range during the pump pressure measurement process. To determine the main influencing area during the pump pressure measurement process and reveal the properties of the measured fluid mobility, the pressure propagation range during the pump pressure measurement process is analyzed.

[0066] The results show that due to the short pressure measurement time and small displacement, the pressure wave propagation is only limited to the range of 10 - 30 cm, and mainly within 10 cm. After the pressure measurement, the pressure quickly returns to the original state. Therefore, the mobility obtained during the pressure measurement stage is mainly the mobility of the mud contamination zone, and it is difficult to obtain the in-situ true mobility of the deep reservoir. In addition, based on the formation pressure obtained in Step S2, the pressure derivative curve during the pressure recovery stage is plotted in double logarithmic coordinates, such as Figure 2As shown (the abscissa is time and the ordinate is the derivative of the pressure difference), by analyzing the curve characteristics, it can be found that the pressure derivative curves under different pollution radii can all be divided into the early well storage and skin effect section and the near-spherical flow section, indicating that the pressure wave in the polluted zone reservoir has gradually reached a pseudo-steady state. As the range of the polluted zone increases, the time for the appearance of near-spherical flow increases. According to the pressure data measured in the field well example BD21-1-10, the analysis of the pressure build-up curve shows that the pressure build-up curve presents an obvious near-spherical flow pattern in the later stage, which is consistent with the shape of the numerical simulation recovery curve, proving that the pressure propagation range is limited within the mud pollution zone, and the measured fluid mobility is the fluid mobility of the mud filtrate. What is proved here is that the fluid pumped for pressure measurement is the mud filtrate, and the fluid mobility is the mobility of the mud filtrate. This provides a basis for the explanation of the fluid mobility pumped by the pump later, helps to determine the approximate value range of fluid parameters, and improves the optimization efficiency.

[0067] Secondly, analysis of the main controlling factors of pressure dynamic changes based on the numerical simulation model

[0068] To determine the main controlling factors of the model pressure during the interpretation of the pressure measurement mobility, set the reservoir permeability to 1 - 2.75 mD, the mud filtrate viscosity to 0.205 - 0.8 mPa·s, the change range of the pumping displacement of the pressure measurement pump to 10 - 60 cc / min, and the change range of the pressure measurement time to 3 - 9 s. According to the numerical simulation model of the pressure measurement pumping reservoir established in step S2, conduct numerical simulations of the pressure measurement pumping respectively, obtain the pressure and pressure build-up curves during the pressure measurement pumping process, use the steady-state or non-steady-state Darcy seepage formula to calculate the fluid mobility, and determine the main controlling factors of the fluid mobility through comparative analysis.

[0069] The expression of the steady-state or non-steady-state Darcy seepage formula is:

[0070] k / μ = Q / (G*Δp)

[0071] In the formula, G is the shape coefficient of the pumping probe, cm; Q is the fluid flow rate, cm 3 / s; Δp is the pressure difference, MPa; k / μ is the fluid mobility, μm 2 / cP; k is the permeability, μm 2 ; μ is the fluid viscosity, cP.

[0072] The results show that there are significant differences in mobility under different permeability and mud filtrate viscosity scenarios (Table 1). Therefore, it is the main controlling factor for the measured pressure mobility. However, the differences in fluid mobility values under different measured pressure displacement and measured pressure time scenarios are relatively small, indicating that the influence on the calculation error of mobility is relatively small. Therefore, during the fitting process of the measured dynamic curve, attention should be focused on the settings of reservoir permeability and fluid viscosity parameters to achieve rapid fitting and inversion. The determined main controlling factors mainly lay the foundation for the parameters of the following fitting inversion. After determining the main controlling factors, it shows that these parameters have a relatively large impact on the fitting results. Therefore, they are the parameters that need to be optimized and adjusted.

[0073] Table 1 shows the comparison of the influence of different factors on the measured pressure mobility

[0074]

[0075] Finally, a fitting model for measured pressure dynamic data based on the particle swarm algorithm is established

[0076] Based on the actual test pressure data obtained by sorting in step S1, the pressure dynamic data is calculated using the pump pumping measured pressure reservoir numerical simulation model established in step S2. The minimization problem shown in formula (1) is defined. By optimizing and adjusting the fluid viscosity and reservoir permeability parameters, the optimal values are selected within the range of 0.1 - 10 to make the calculated value consistent with the actual value, so as to achieve the fitting of the measured pressure dynamic data.

[0077]

[0078] In the formula: is the objective function, reflecting the gap between the calculated value and the actual value; is the fluid parameter vector, which includes but is not limited to permeability and viscosity parameters; and C d are the test pressures of each block obtained from the actual test and their error covariance matrix respectively. The error covariance matrix here reflects the covariance matrix formed by the fitting error and the fitting variables. It is an inherent mathematical term and there are mature calculation methods, which are prior arts; is the pressure dynamic data vector calculated using the reservoir numerical and simulation model, which is a function of the flow rate parameter vector and is specifically determined according to the reservoir numerical and simulation model. It is a prior art and will not be elaborated here.

[0079] For the optimization problem shown in Equation (1), the particle swarm optimization algorithm (PSO) is adopted here for solution. In this algorithm, a possible solution, that is, a fitting result, is assumed to be a particle. Then each particle can be regarded as an individual in the D-dimensional search space. The current position of the particle corresponds to a candidate solution of the optimization problem, that is, a possible fitting result. The optimal solution found by each particle individually is called the individual extreme value, that is, the optimal solution obtained by a certain particle through search iteration. The optimal individual extreme value in the population is regarded as the current global optimal solution, that is, the optimal solution within the entire search space. Through continuous iterative search and calculation, the velocities and positions of each particle are updated until the optimal solution that meets the termination condition, that is, reaches the maximum number of iterative steps, is obtained. The number of iterative steps is generally less than 100 steps. Finally, the fitting of the dynamic data of the pump pressure measurement and the simulation data is realized.

[0080] Step S4: Fitting of the dynamic pressure measurement data and interpretation and inversion of the fluid mobility

[0081] According to the fitting model of the dynamic pressure measurement data based on the particle swarm algorithm established in Step S3, taking Well Wenchang 9-7-3d as an example, the dynamic data of the pump pressure measurement of this well are sorted out, and a fine reservoir numerical simulation model of the pump pressure measurement based on the radial grid is established. Taking the reservoir permeability and fluid viscosity as independent variables, according to Equation (1), the particle swarm algorithm is used for iterative optimization, and the iterative time step is set to 50 steps to realize the fitting of the dynamic data of the pump pressure measurement process. Figure 3 、 Figure 4 They are the fitted curve and the value of the fitting objective function respectively.

[0082] Table 2 shows the variation of the fitting parameters with the fitting iterative steps

[0083]

[0084] Table 2 is the data record of the variation of the fitting parameters with the fitting iterative steps. The more stable the objective function is, the smaller the objective function is, and the higher the fitting degree is. The results show that the accuracy of the fitting result of the dynamic pressure measurement curve of this well reaches more than 90%. After 50 steps, the inverted fitting mobility is 4.42 mD / mPa·s. Specifically, the permeability and fluid viscosity inverted by Step S3 are adopted, and then according to the formula: mobility = permeability / viscosity, the calculation is obtained. The error between it and the true reservoir mobility of 4.25 mD / mPa·s is 3.9%, which verifies the reliability and accuracy of the inversion and interpretation of the fluid mobility based on the fine reservoir numerical simulation method proposed by the present invention.

[0085] The present invention combines reservoir numerical simulation technology and particle swarm intelligent optimization fitting algorithm, and proposes a reservoir fluid mobility interpretation inversion method considering the pump pressure measurement process. First, considering the actual reservoir and construction technology, a reservoir numerical simulation model is established by using numerical simulation method, and the main control factors of pressure measurement mobility interpretation are determined by analyzing the variation law of pressure. Secondly, combined with the particle swarm intelligent fitting optimization algorithm, a pressure measurement mobility interpretation inversion method is constructed. Then, the measured pressure data is fitted to invert the reservoir fluid mobility. Finally, according to the viscosity of mud filtrate, physical property parameters such as reservoir permeability are obtained to realize the evaluation of the reservoir.

[0086] The specific steps for constructing the pressure measurement mobility interpretation inversion method include: (1) First, input the sorted pressure measurement dynamic data (pressure and flow rate), and determine the initial input permeability and fluid viscosity data according to well logging, physical property analogy and other data; (2) Use the numerical simulator, combined with the particle swarm algorithm, to fit the measured pressure and flow rate data by adjusting the reservoir physical properties and permeability data, so as to realize the interpretation inversion of the pressure measurement mobility.

[0087] The prior art uses a fixed instrument coefficient to directly calculate the fluid mobility according to the unsteady and steady state area integration method, and the calculation error may be large. While the present invention is based on the pump pressure measurement data and uses the fitting form to inversely calculate the fluid mobility, and the calculation accuracy is relatively high.

[0088] In addition to inverting the fluid mobility, the present invention can also obtain the reservoir permeability to realize the evaluation of the reservoir physical properties. The process is that according to the inversely obtained fluid mobility, the physical property parameters such as the reservoir permeability can be obtained through simple calculation, and the relationship is: permeability / fluid viscosity = fluid mobility.

Claims

1. A pressure measurement flow inversion calculation method based on reservoir numerical simulation, characterized in that: The following steps are involved: S1. By collecting the test point information during the oil and gas well pressure test and pumping process, a dynamic data set for pumping pressure test history matching is obtained to obtain actual test pressure data; In step S1, test points with a test time of more than 10 seconds, a constant pumping flow rate, and a stable pressure recovery value within a preset time are selected as data points for interpretation and inversion; In step S1, linear interpolation is used to complete the missing actual test pressure data; For outliers in the actual test pressure data, linear interpolation is used after deletion; If the smoothness of the curve formed by the actual test pressure data does not meet the requirements, exponential smoothing is used for processing; S2. Establish a numerical simulation model for pumping pressure measurement, and obtain simulated pressure dynamic data through simulation; S3, according to the hydraulic pumping reservoir numerical simulation model established in step S2, respectively perform hydraulic pumping numerical simulation, obtain the hydraulic pumping process pressure and pressure recovery curve, and calculate the fluid mobility using the steady-state or unsteady-state Darcy flow formula; By comparing and analyzing the influence of the changes of various fluid parameters on the fluid flow rate during the calculation process, the main controlling factors affecting the fluid flow rate among the fluid parameters are determined as the fluid parameter vector in formula (1); The pressure build-up test curve in double logarithmic coordinates is used to analyze the pressure propagation range during the pumping pressure test process and determine the nature of the fluid mobility. Based on the actual test pressure data obtained in step S1, the simulated pressure dynamic data calculated in step S2 is used to define the minimization problem shown in formula (1), and the calculated value is made consistent with the actual value by optimizing and adjusting the fluid parameters in the simulation process in step S2, so as to achieve the fitting of the measured pressure dynamic data; (1) Where: is the objective function, reflecting the gap between the calculated value and the actual value; is the fluid parameter vector; and They are the test pressure of each block and its error covariance matrix obtained from the actual test; The pressure dynamic data vector is obtained by using the reservoir numerical and simulation model calculation; In step S3, a particle swarm optimization algorithm is used to solve the problem; S4. Invert and calculate the fluid mobility according to the final fluid parameters optimized in step S3.

2. The pressure measurement flow inversion calculation method based on reservoir numerical simulation according to claim 1 is characterized in that: In step S4, the fluid fluidity is calculated as follows: fluid fluidity=permeability / fluid viscosity.

3. The pressure measurement flow inversion calculation method based on reservoir numerical simulation according to claim 1 is characterized in that: In step S2, the pumping pressure measurement numerical simulation model includes three modules: model geological and fluid property input, model generation, and model operation; Module 1: Based on different geological, fluid and process parameters, a multi-layer radial logarithmic transformation method is used to construct a fine grid geological model, consider the composition of fluid components, and establish a multiphase and multicomponent fluid model to simulate the flow morphology during the pressure measurement pumping process; Module 2: Establishing and generating a numerical simulation model for pumping and pressure testing reservoirs that truly reflects specific information; Module three runs the established numerical simulation model and outputs the theoretical dynamic curves of pumped fluid volume and pressure.

4. The pressure measurement flow inversion calculation method based on reservoir numerical simulation according to claim 1 is characterized in that: The method further includes step 5: inverting and calculating the final fluid parameters optimized in step S3 to obtain the reservoir permeability, so as to evaluate the reservoir physical properties.

Images