Multi-phase pipe flow parameter online prediction device and method based on three-fluid model

Abstract

The invention provides a multi-phase pipe flow parameter online prediction device and method based on a three-fluid model. The device can be used for collecting multiphase flow parameters, and prediction and output of the multiphase flow parameters are completed through a built-in data processing system. The multiphase pipe flow parameter on-line prediction method comprises the steps that multiphase flow parameters of an implementation device are collected in real time through a data collection card, initial boundary conditions of multiphase flow model calculation are generated through a data processing system, a finite difference method is used for dispersing a three-fluid control equation, a Newton-Raphson method is used for conducting steady-state calculation, and the multiphase pipe flow parameter on-line prediction result is obtained. A steady-state calculation result is used as an initial value of a transient model, online prediction of multiphase flow pipe flow parameters is achieved, and hydraulic and thermal change trends are given and compared with data collected in real time; according to the method provided by the invention, the flow parameters at any position of the multiphase flow pipeline can be calculated, and online prediction and real-time comparison of key parameters such as pressure, temperature, flow and liquid holdup of the multiphase pipeline with any component can be realized.

Description

Technical Field

[0001] The invention relates to a device and method for online prediction of multiphase pipe flow parameters based on a three-fluid model, and belongs to the field of multiphase flow measurement. Background Art

[0002] Oil is one of the most important energy sources in the world and is known as the "blood" of industry. Oil extraction usually includes multiple stages. In the early stage of extraction, the produced products are generally crude oil and its associated gas; in the middle and late stages of extraction, a large amount of primary water will also be produced. Therefore, gas-liquid two-phase flow or oil-gas-water three-phase flow often appears in the production of onshore and offshore oil and gas fields, which are collectively referred to as multiphase flow. There are generally two ways to transport multiphase flow, one is gas-liquid separation and the other is mixed transportation. Obviously, for offshore oil fields and marginal oil fields, they are far away from the land or gathering and transportation stations, and the cost of gas-liquid separation is very high. Therefore, most of the existing oil and gas field transmission pipelines adopt mixed transportation, using pipelines to mix the produced products at the wellhead to the central processing station and then to the downstream. This practice can reduce the pipeline investment cost by 10% to 40%. Mixed transportation technology has been widely used because of its advantages of enabling oil and gas fields to be put into development quickly and reducing initial investment.

[0003] Accurately measuring multiphase flow parameters in pipelines is the key basis for the application of mixed transportation technology, and it is also of great significance to the production and operation of oil and gas fields. Oil and gas field managers generally measure parameters such as pressure, temperature, flow, and liquid holdup by setting various transmitters at key nodes - these parameters can well reflect the flow conditions in the pipeline. For example, when the pressure or flow at the node suddenly increases or decreases, it can be determined whether the pipeline is blocked or leaking; when the temperature drops, hydrates may be generated in the pipeline; when the liquid holdup suddenly increases, slug flow may form in the pipeline. However, a single oil and gas field gathering and transportation pipeline is generally several kilometers or even tens of kilometers long. If the pipeline is divided into small sections every 100 meters, due to construction costs, only one or two key nodes may be set in the middle of the pipeline, and the points without data collection are many times or even dozens of times more than the key nodes. When the parameters of non-data collection nodes are abnormal, the abnormal parameters need to spread to the key nodes before they can be detected. It is difficult for managers to judge and deal with them in a short time, and there is a certain lag, which often leads to accidents. Therefore, oil and gas fields urgently need a method that can scientifically and economically measure or predict the parameters at these non-critical nodes to provide a basis for production operations and safety decisions. In view of the difficulty and importance of multiphase flow parameter measurement, a large number of studies have been carried out both at home and abroad, and some measurement devices and measurement methods have been proposed.

[0004] Patent ZL201410033941.5 proposed a two-phase flow measurement system based on distributed conductivity sensors, which can measure the phase content at the cross section of the pipeline; Patent ZL201310406254.9 announced an online device for measuring the three-phase flow of oil, gas and water in unconventional natural gas; Patent ZL201580061077.2 proposed a tomography device, multiphase flow monitoring system and corresponding method for monitoring mixed phase samples; Patent ZL201080029666.X announced a multiphase fluid measurement device and method for measuring the composition and flow of oil, gas and water in wellbore fluid; Patent ZL201510694791.7 proposed a gas-liquid two-phase flow metering device and method for measuring the mass flow rate and dryness of gas-liquid two-phase. Although the above patents can complete the parameter measurement of multiphase flow, they are all parameters of a local position in the measured pipeline or equipment, and cannot realize the supplement of parameters at non-measurement points.

[0005] Patent ZL201811479567.6 proposes a transient calculation method for multiphase flow based on spatial conservation, which can use limited data points to complete the calculation of parameters along the multiphase flow pipeline and supplement the parameter values ​​of non-measured nodes. However, it does not have an online calculation function and can only perform offline calculations based on historical data. For application scenarios such as oil and gas fields that require real-time data calculations, this method has certain limitations. Summary of the invention

[0006] The purpose of the present invention is to overcome the above-mentioned shortcomings and provide a device and method for online prediction of multiphase pipe flow parameters based on a three-fluid model. The present invention is based on the three-fluid model, separately models oil, gas and water, takes the key node parameters measured in real time in the pipeline as boundary conditions, and online solves the flow parameters such as pressure, temperature, flow rate and liquid holdup of each phase in the model to realize the supplementation and prediction of multiphase pipe flow parameters.

[0007] In order to achieve the above object, the technology of the present invention is implemented by the following scheme:

[0008] An implementation device for online prediction of multiphase pipe flow parameters includes a cyclic curved pipeline, a pressure, temperature, and flow sensor is arranged at the inlet, two pressure sensors and a temperature sensor are arranged at the curved pipe in sequence to monitor the parameters of the intermediate nodes of the pipeline, and a pressure, temperature, and flow sensor is also arranged at the outlet. The implementation device is provided with two data acquisition cards to receive the analog signals of each sensor in real time, and convert them into digital signals for transmission to the data processing system, such as Figure 1As shown; the data processing system includes a data access module, a boundary condition processing module, a multiphase flow parameter steady-state calculation module, a multiphase flow parameter transient prediction module, a parameter online comparison module and a result post-processing module, which are used to realize the real-time calculation, comparison and result output of multiphase flow parameters, such as Figure 2 shown.

[0009] The present invention also includes an online prediction method for multiphase pipe flow parameters based on a three-fluid model, comprising the following steps:

[0010] Step 1: The real-time pressure, temperature and flow rate data of the multiphase flow are measured by various pressure, temperature and flow rate sensors and transmitted to the data processing system via the data acquisition card;

[0011] Step 2: The data processing system reads the flow parameter digital signal converted by the data acquisition card in real time through the data access module, and outputs it to the boundary condition processing module. According to the pipeline parameters and fluid components pre-defined in this module, the model boundary conditions are constructed, and the boundary conditions are transmitted to the multiphase flow parameter steady-state calculation module;

[0012] Step 3: The steady-state calculation module of multiphase flow parameters starts calculation. According to the model boundary conditions input into the model at time j, based on the state equation and the transient three-fluid model, a steady-state three-fluid model is established to solve the flow parameters at any node i in the pipeline at time j, and obtain the change trends of parameters such as pressure, temperature, flow rate, and liquid holdup along the flow direction, so as to supplement the data at non-critical nodes, and output the calculation results to the transient prediction module of multiphase flow parameters as the initial value of transient prediction;

[0013] Step 4: Transient prediction: Take the steady-state data at each node i calculated at time j as the initial value, apply the transient three-fluid model, calculate the flow parameters under the j+1 time layer, and give the trend of pressure, temperature, flow rate, liquid holdup and other parameters at any node i of the pipeline over time to predict future parameters. After the prediction is completed, output the results to the parameter online comparison module to compare the real-time collected data with the predicted data.

[0014] Step 5: Data display and export: All data used and calculated in the data processing system are displayed on the display screen in the form of images through the result post-processing module, or exported to an external system.

[0015] The third step specifically comprises the following steps:

[0016] (1) Calculation of fluid physical parameters

[0017] Preferably, the PR state equation is used to calculate the density, viscosity, enthalpy, entropy and other thermophysical parameters of the mixed fluid, the constructed boundary condition data is read in and transferred to the PR equation, and the thermophysical parameters are solved using the dichotomy method.

[0018] (2) Establishment of three-fluid simulation model

[0019] Preferably, a three-fluid model is used to calculate the fluid parameters in the pipeline. The three-fluid model of the present invention is as follows: Figure 3 As shown, equations are established for three continuous phases and six dispersed phases, including nine continuity equations, three momentum equations, three energy equations and one phase fraction equation.

[0020] The continuity equation is as follows:

[0021]

[0022] The momentum equation is as follows:

[0023]

[0024] The energy equation is as follows:

[0025]

[0026] The phase fraction equation is as follows:

[0027] ∑α k =1 (4)

[0028] Where α k ——phase fraction of k phase;

[0029] ρ k ——density of phase k, kg / m 3 ;

[0030] u k ——Flow rate of phase k, m / s;

[0031] P k ——pressure of phase k, Pa;

[0032] T k ——temperature of phase K, K;

[0033] τ k ——Shear stress of k phase, N / m 2 ;

[0034] h k ——enthalpy of phase k, J;

[0035] F ki ——friction resistance between phases k, Pa·m;

[0036] F kw ——Friction resistance between k phase and tube wall, Pa·m;

[0037] λ k ——Thermal conductivity of phase k, W / (m 2 K);

[0038] ——turbulent viscosity of phase k, mPa·s;

[0039] ——the turbulent Planck number of the k-phase;

[0040] ——heat transfer rate from phase k to component m, W / (m·K);

[0041] g——acceleration due to gravity, m / s 2 ;

[0042] h——fluid height, m.

[0043] θ——pipeline inclination, rad.

[0044] (3) Establishment and discretization of steady-state simulation model

[0045] The steady-state equation does not consider the time term, and the above equations (1) to (4) can be simplified to:

[0046]

[0047]

[0048]

[0049] Along the pipeline flow direction x, the pipeline is divided into n grids. Assuming the grid length Δx, the central difference method is applied to the above equations (5) to (7) to discretize them, and the following equation is obtained:

[0050]

[0051]

[0052]

[0053] F 1～9 are the 9 equations obtained after discretization of the continuity equation, F 10～12 are the three equations obtained after the momentum equation is discretized, F 13～15 The three equations obtained after discretization of the energy equation and one phase fraction equation are shown in Table 1. The hydraulic equation F 1～12 With the thermodynamic equation F 13～15 Separate and rewrite into the following matrix form, we get:

[0054] S s (x)=0 (11)

[0055] R s (x)=0 (12)

[0056] Where S s ——Steady-state hydraulic equation F 1～12 The matrix S s (x)=[F1(x),F2(x),...,F 12 (x)] T , where x is the multiphase flow parameter to be determined (pressure, temperature, flow rate, etc.);

[0057] R s ——Steady-state thermodynamic equation F 13～15 The matrix R s (x)=[F 13 (x),F 14 (x),F 15 (x)] T .

[0058] (4) Model solution

[0059] Applying Taylor expansion to equations (11) and (12), ignoring higher-order terms, we obtain:

[0060] S′ s (x)Δx=-S s (x) (13)

[0061] R′ s (x)Δx=-R s (x) (14)

[0062] Where S′ s (x)——Jacobian matrix of steady-state hydraulic equations;

[0063] R′ s (x)——Jacobian matrix of the steady-state thermodynamic equations;

[0064] Δx s ——Steady-state correction value matrix.

[0065] Read the current flow parameters and boundary conditions, use the Newton-Raphson method to solve equation (13), and obtain the hydraulic parameters such as pressure, flow rate, and liquid holdup of the multiphase flow. Substitute the hydraulic parameters obtained from equation (13) into equation (14) to inversely solve the fluid temperature. Assuming that the calculation error is ε, when equation (15) is satisfied, the steady-state calculation is completed and the parameters are output. Otherwise, substitute the temperature solved by the thermodynamic equation into the hydraulic equation and solve it again until the convergence condition is met. See the flowchart of the steady-state calculation. Figure 4 .

[0066] ε=|T S -T R |<10 -5 (15)

[0067] Where T S ——The fluid temperature assumed when calculating the hydraulic equation, K;

[0068] T R ——Fluid temperature obtained by solving the thermodynamic equation, K.

[0069] The fourth step specifically comprises the following steps:

[0070] (1) Obtaining the initial value of transient calculation

[0071] The multiphase flow parameters such as pressure, temperature, flow rate, and liquid holdup calculated by the steady-state model in step 3 are used as the initial conditions of the transient model;

[0072] (2) Discretization of transient three-fluid model

[0073] The transient three-fluid model is shown in equations (1) to (4). The flow plane is meshed along the pipeline flow direction x and the time growth direction t. The time step is Δt and the space step is still Δx. The time term is discretized using forward difference and the space term is discretized using the central difference method. The following equation is obtained:

[0074]

[0075]

[0076]

[0077] Similarly, we get 9 continuity equations, 3 momentum equations, 3 energy equations and 1 phase fraction equation, which can be rewritten into matrix form:

[0078] S t (x) = 0 (19)

[0079] R t (x) = 0 (20)

[0080] Where S t ——Transient hydraulic equation E 1～12 The matrix S t (x)=[E1(x),E2(x),...,E 12 (x)] T , where x is the multiphase flow parameter to be determined (pressure, temperature, flow rate, etc.);

[0081] R t ——Transient thermodynamic equation E 13～15 The matrix R t (x) = [E 13 (x),E 14 (x),E 15 (x)] T ;

[0082] (3) Solution of transient three-fluid model

[0083] Applying Taylor expansion to equations (21) and (22), ignoring higher-order terms, we obtain:

[0084] S′ t (x)Δx=-S t (x) (21)

[0085] R′ t (x)Δx=-R t (x) (22)

[0086] Where S′ t (x)——Jacobian matrix of transient hydraulic equations;

[0087] R′ t (x)——Jacobian matrix of transient thermodynamic equations;

[0088] Δx t ——Transient correction value matrix.

[0089] Input the steady-state calculation results of step 3, apply the Newton-Raphson method to solve equation (21) first, then solve equation (22). When the convergence condition (15) is met, stop the iteration and get the transient calculation results. The flowchart of transient calculation is shown in Figure 5 .

[0090] The beneficial features of the present invention are:

[0091] (1) Applicable to any fluid. The three-fluid model of the present invention is not limited to the fluid composition when applied. In addition to being able to predict the flow parameters of oil, gas and water three-phases, it can also be applied to the prediction of flow parameters of natural gas condensate, wet natural gas, gas and liquid, etc., and has a wide range of applications.

[0092] (2) The flow parameters of non-critical nodes are supplemented. The measured parameters of the critical nodes in the implementation device of the present invention are used as boundary conditions, and the three-fluid model is used to perform steady-state solution of the flow parameters, and the values ​​of parameters such as pressure, temperature, flow rate and liquid holdup at any position in the pipeline are obtained, which provides a reliable basis for the design, operation and optimization of the pipeline.

[0093] (3) The online prediction of multiphase pipe flow parameters is realized. Based on the steady-state calculated values, the transient three-fluid model is used to predict future flow parameters, providing a scientific basis for the formulation of pipeline production plans. BRIEF DESCRIPTION OF THE DRAWINGS

[0094] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required in the embodiments or the prior art will be briefly introduced below. The drawings described below are some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0095] Figure 1 It is a schematic diagram of the multiphase pipe flow parameter online prediction device of the present invention;

[0096] Figure 2 It is a schematic diagram of a data processing system of a multiphase pipe flow parameter online prediction device of the present invention;

[0097] Figure 3 A heat and mass transfer model established for the multiphase pipe flow parameter prediction method of the present invention;

[0098] Figure 4 A flow chart of steady-state calculation of the multiphase pipe flow parameter prediction method of the present invention;

[0099] Figure 5 A flow chart of transient calculation of the multiphase pipe flow parameter prediction method of the present invention;

[0100] Figure 6 is a steady-state calculation result of multiphase pipe flow parameters in an embodiment of the present invention;

[0101] Figure 7 The comparison results of multiphase pipe flow parameters in the embodiments of the present invention;

[0102] Figure 8 The transient pressure prediction result of multiphase pipe flow parameters in an embodiment of the present invention;

[0103] Fig. 9 is a transient temperature prediction result of multiphase pipe flow parameters in an embodiment of the present invention;

[0104] Fig.10 It is the prediction result of transient liquid holdup of multiphase pipe flow parameters in the embodiment of the present invention. DETAILED DESCRIPTION

[0105] In order to make the purpose, technical solution and advantages of the embodiments of the present invention clearer, the technical solution of the present invention will be fully described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0106] In view of the development of mixed transportation technology, the oil and gas field gathering and transportation system needs a more accurate method to obtain multiphase pipe flow parameters. However, the existing methods can only measure the key node parameters, making it difficult to achieve online prediction of multiphase pipe flow parameters.

[0107] The embodiment of the present invention provides an implementation device for online prediction of multiphase pipe flow parameters, see Figure 1 , including a continuous elbow 101, a data acquisition card 1 102, a data acquisition card 2 104, a data processing system 103, an inlet pressure sensor 105, a temperature sensor 106, a flow sensor 107, pipe middle pressure sensors 108, 110, and a temperature sensor 109; the data processing system 103 is composed of multiple modules, including a data access module 201, a boundary condition processing module 202, a multiphase flow parameter steady-state calculation module 203, a multiphase flow parameter transient prediction module 204, a parameter online comparison module 205, and a result post-processing module 206. The device can be used for the collection of multiphase flow parameters, and the prediction and output of multiphase flow parameters are completed through the built-in data processing system. The data processing system is shown in FIG. Figure 2 .

[0108] In addition, an online prediction method for multiphase pipe flow parameters based on a three-fluid model is also included, which specifically includes the following contents:

[0109] Step 1: The real-time pressure, temperature and flow rate data of the multiphase flow are measured by various pressure, temperature and flow rate sensors and transmitted to the data processing system via the data acquisition card;

[0110] Step 2: The data processing system reads the flow parameter digital signal converted by the data acquisition card in real time through the data access module, and outputs it to the boundary condition processing module. According to the pipeline parameters (such as pipe diameter, pipe length, wall thickness, insulation layer thickness, thermal conductivity, etc.) and fluid components (such as gas components, oil components, etc.) pre-defined in this module, the model boundary conditions are constructed, and the boundary conditions are transmitted to the multiphase flow parameter steady-state calculation module;

[0111] During the production process of oil and gas fields, the output at the wellhead often fluctuates, so unstable multiphase flow will be formed in the pipeline. The spatial distribution of multiphase flow parameters can be calculated using a steady-state model, but it cannot reflect the fluctuation of flow parameters over time. However, steady-state calculations can provide initial values ​​for transient predictions, so steady-state calculations of multiphase flow parameters must be performed first, followed by transient predictions.

[0112] Step 3: Steady-state calculation of multiphase flow parameters: According to the model boundary conditions input into the model at time j, a steady-state three-fluid model is established based on the state equation and the transient three-fluid model, and the flow parameters at any node i in the pipeline at time j are solved to obtain the change trends of parameters such as pressure, temperature, flow rate, and liquid holdup along the flow direction, so as to supplement the data at non-critical nodes, and output the calculation results to the transient prediction module of multiphase flow parameters as the initial value of transient prediction;

[0113] Step 4: Transient prediction: Take the steady-state data at each node i calculated at time j as the initial value, apply the transient three-fluid model, calculate the flow parameters under the j+1 time layer, and give the trend of pressure, temperature, flow rate, liquid holdup and other parameters at any node i of the pipeline over time to predict future parameters. After the prediction is completed, output the results to the parameter online comparison module to compare the real-time collected data with the predicted data.

[0114] Step 5: Data display and export: All data used and calculated in the data processing system are displayed on the display screen in the form of images through the result post-processing module, or exported to an external system.

[0115] The third step specifically comprises the following steps:

[0116] (1) Calculation of fluid physical parameters

[0117] Preferably, the PR state equation is used to calculate the density, viscosity, enthalpy, entropy and other thermophysical parameters of the mixed fluid. The expression of the PR equation is as follows:

[0118]

[0119]

[0120]

[0121] For multiphase fluid mixtures, the Zudkevitch and Joffe mixing rules apply:

[0122]

[0123]

[0124] Where P is system pressure, Pa;

[0125] T——system temperature, K;

[0126] ρ——density of gas or liquid phase, kmol / m 3 ;

[0127] R——gas constant, R=8.314kJ / (kmol·K);

[0128] p c ——critical pressure, Pa;

[0129] V——molar volume of gas phase, m 3 / kmol;

[0130] T c ——critical temperature, K;

[0131] ω——eccentricity factor;

[0132] K ij ——is the binary interaction coefficient of components i, j.

[0133] The constructed boundary conditions are input into the PR equation, and the initial thermophysical parameters of the fluid are solved using the dichotomy method.

[0134] (2) Establishment of three-fluid simulation model

[0135] Preferably, a three-fluid model is used to calculate the fluid parameters in the pipeline, such as Figure 3 As shown. Different from the traditional two-fluid model, the three-fluid model models oil, gas, and water separately, taking into account the interface characteristics of oil and water, and considering the movement of any two dispersed phases in the continuous phase (for example, the gas phase contains dispersed oil and water phases, the oil phase contains dispersed gas and water phases, etc.), and also considers the heat transfer characteristics of the three media of oil, gas, and water in the same cross section, and uses three energy equations to describe their temperatures respectively, so that more accurate parameter prediction results can be obtained.

[0136] Equations were established for three continuous phases and six dispersed phases, including nine continuity equations, three momentum equations, three energy equations and one phase fraction equation.

[0137] The continuity equation is as follows:

[0138]

[0139] The momentum equation is as follows:

[0140]

[0141] The energy equation is as follows:

[0142]

[0143] The phase fraction equation is as follows:

[0144] ∑α k =1 (9)

[0145] Where α k ——phase fraction of k phase;

[0146] ρ k ——density of phase k, kg / m 3 ;

[0147] u k ——Flow rate of phase k, m / s;

[0148] P k ——pressure of phase k, Pa;

[0149] T k ——temperature of phase K, K;

[0150] τ k ——Shear stress of k phase, N / m 2 ;

[0151] h k ——enthalpy of phase k, J;

[0152] F ki ——friction resistance between phases k, Pa·m;

[0153] F kw ——Friction resistance between k phase and tube wall, Pa·m;

[0154] λ k ——Thermal conductivity of phase k, W / (m 2 K);

[0155] ——turbulent viscosity of phase k, mPa·s;

[0156] ——the turbulent Planck number of phase k;

[0157] ——heat transfer rate from phase k to component m, W / (m·K);

[0158] g——acceleration due to gravity, m / s 2 ;

[0159] h——fluid height, m.

[0160] θ——pipeline inclination, rad.

[0161] (3) Establishment and discretization of steady-state simulation model

[0162] The steady-state equation does not consider the time term, and the above equations (1) to (4) can be simplified to:

[0163]

[0164]

[0165]

[0166] Along the pipeline flow direction x, the pipeline is divided into n grids. Assuming the grid length Δx, the central difference method is applied to the above equations (5) to (7) to discretize them, and the following equation is obtained:

[0167]

[0168]

[0169]

[0170] F 1～9 are the 9 equations obtained after discretization of the continuity equation, F 10～12 are the three equations obtained after the momentum equation is discretized, F 13～15 The three equations obtained after discretization of the energy equation and one phase fraction equation are shown in Table 1. The hydraulic equation F 1～12 With the thermodynamic equation F 13～15 Separate and rewrite into the following matrix form, we get:

[0171] S s (x) = 0 (16)

[0172] R s (x) = 0 (17)

[0173] Where S s ——Steady-state hydraulic equation F 1～12 The matrix S s (x)=[F1(x),F2(x),...,F 12 (x)] T , where x is the multiphase flow parameter to be determined (pressure, temperature, flow rate, etc.);

[0174] R s ——Steady-state thermodynamic equation F 13～15 The matrix R s (x)=[F 13 (x),F 14 (x),F 15 (x)] T ;

[0175] (4) Model solution

[0176] Applying Taylor expansion to equations (11) and (12), ignoring higher-order terms, we obtain:

[0177] S′ s (x)Δx=-S s (x) (18)

[0178] R′ s (x)Δx=-R s (x) (19)

[0179] Where S′ s (x)——Jacobian matrix of steady-state hydraulic equations;

[0180] R′ s (x)——Jacobian matrix of the steady-state thermodynamic equations;

[0181] Δx s ——Steady-state correction value matrix.

[0182] Read the current flow parameters and boundary conditions, use the Newton-Raphson method to solve equation (13), and obtain the hydraulic parameters such as pressure, flow rate, and liquid holdup of the multiphase flow. Substitute the hydraulic parameters obtained from equation (13) into equation (14) to inversely solve the fluid temperature. Assuming that the calculation error is ε, when equation (15) is satisfied, the steady-state calculation is completed and the parameters are output. Otherwise, substitute the temperature solved by the thermodynamic equation into the hydraulic equation and solve it again until the convergence condition is met. See the flowchart of the steady-state calculation. Figure 4 .

[0183] ε=|T S -T R |<10 -5 (20)

[0184] Where T S ——The fluid temperature assumed when calculating the hydraulic equation, K;

[0185] T R ——Fluid temperature obtained by solving the thermodynamic equation, K.

[0186] The fourth step specifically comprises the following steps:

[0187] (1) Obtaining the initial value of transient calculation

[0188] The multiphase flow parameters such as pressure, temperature, flow rate, and liquid holdup calculated by the steady-state model in step 3 are used as the initial conditions of the transient model;

[0189] (2) Discretization of transient three-fluid model

[0190] The transient three-fluid model is shown in equations (1) to (4). The flow plane is meshed along the pipeline flow direction x and the time growth direction t. The time step is Δt and the space step is still Δx. The time term is discretized using forward difference and the space term is discretized using the central difference method. The following equation is obtained:

[0191]

[0192]

[0193]

[0194] Similarly, we get 9 continuity equations, 3 momentum equations, 3 energy equations and 1 phase fraction equation, which can be rewritten into matrix form:

[0195] S t (x) = 0 (24)

[0196] R t (x) = 0 (25)

[0197] Where S t ——Transient hydraulic equation E 1～12 The matrix S t (x)=[E1(x),E2(x),...,E 12 (x)] T , where x is the multiphase flow parameter to be determined (pressure, temperature, flow rate, etc.);

[0198] R t ——Transient thermodynamic equation E 13～15 The matrix R t (x) = [E 13 (x),E 14 (x),E 15 (x)] T ;

[0199] (3) Solution of transient three-fluid model

[0200] Applying Taylor expansion to equations (21) and (22), ignoring higher-order terms, we obtain:

[0201] S′ t (x)Δx=-S t (x) (26)

[0202] R′ t (x)Δx=-R t (x) (27)

[0203] Where S′ t (x)——Jacobian matrix of transient hydraulic equations;

[0204] R′ t (x)——Jacobian matrix of transient thermodynamic equations;

[0205] Δx t ——Transient correction value matrix.

[0206] Input the steady-state calculation results of step 3, apply the Newton-Raphson method to solve equation (21) first, then solve equation (22). When the convergence condition (15) is met, stop the iteration and get the transient calculation results. The flowchart of transient calculation is shown in Figure 5 .

[0207] Example

[0208] In order to further illustrate the present invention, a specific embodiment is used to illustrate the usability and innovation of the present invention, which specifically includes the following contents:

[0209] The basic parameters of the multiphase pipe flow parameter online prediction implementation device in the present invention are shown in Table 1. The total length of the circulating elbow is 100m, the pipe diameter is 88.9mm, and the wall thickness is 6.3mm. Due to the small pipe diameter, the multiphase flow friction loss at the elbow can be ignored. Therefore, when performing boundary condition processing, the circulating elbow can be approximated as a straight pipe section with a length of 100m, and it does not affect the final prediction result.

[0210] In this embodiment, the fluid physical properties used are shown in Table 2. The initial inlet flow rate of the implementation device is 2.5 kg / s, and the outlet pressure is 3.5 MPa. At 360 seconds (6 minutes), the flow rate is increased to 5 kg / s, and the multiphase flow parameter prediction method of the present invention is used to predict and compare the flow parameters after the flow rate is increased.

[0211] Table 1 Pipeline parameters

[0212] parameter unit value Pipe length m 100 Pipe diameter mm 88.9 Pipe wall thickness mm 6.3 Absolute pipe roughness mm 0.045 Overall heat transfer coefficient W / (m·K) 0.14 Pipeline inclination ° 5

[0213] Table 2 Fluid composition in pipeline

[0214] Components Measurement value / % Components Measurement value / % <![CDATA[C1]]> 27.57 <![CDATA[C6]]> 1.58 <![CDATA[C2]]> 2.15 <![CDATA[C7]]> 2.45 <![CDATA[C3]]> 3.09 <![CDATA[C8]]> 3.34 <![CDATA[iC4]]> 0.60 <![CDATA[C9]]> 3.57 <![CDATA[nC4]]> 1.54 <![CDATA[C 10 ]]> 2.43 <![CDATA[iC5]]> 0.77 <![CDATA[C 11+ ]]> 26.34 <![CDATA[nC5]]> 0.99 <![CDATA[CO2]]> 0.13 <![CDATA[N2]]> 0.37 <![CDATA[H2O]]> 23.08

[0215] Step 1: The real-time pressure, temperature and flow rate data of the multiphase flow are measured by various pressure, temperature and flow rate sensors and transmitted to the data processing system via the data acquisition card;

[0216] Step 2: The boundary condition processing module receives the collected real-time parameters, takes the real-time measurement values ​​of the flow sensor 107 and the temperature sensor 106 as the model inlet flow and temperature boundaries, and the real-time measurement value of the pressure sensor 113 as the model outlet pressure boundary, and combines the pre-defined pipeline parameters and fluid components to transmit the data to the multiphase flow parameter steady-state calculation module;

[0217] Step 3: Start steady-state calculation, using the pressure measurement data of pressure sensors 105, 108, 110, the temperature measurement data of temperature sensors 109, 112, and the flow measurement data of flow sensor 111 as verification data to verify the accuracy of steady-state calculation. The calculation and comparison results are shown in Figure 6 ;

[0218] Step 4: Transient prediction. Read the steady-state calculation results as the initial value of transient prediction, drive the transient model, and realize the online prediction of flow parameters. And send it to the parameter online comparison module to realize the comparison between the predicted parameters and the measured parameters. The comparison results are shown in Figure 7 In addition, the predicted pressure, temperature, and liquid holdup data are re-exported and plotted through the result post-processing module to obtain the pressure online prediction results. Figure 8 , the temperature online prediction results are shown in Fig. 9 , the online prediction results of liquid holdup are shown in Fig.10 .

[0219] From the above description, it can be seen that this embodiment is based on an implementation device for online prediction of multiphase pipe flow parameters and an online prediction method for multiphase pipe flow parameters based on a three-fluid model, which realizes the supplement of multiphase flow parameters, including pressure, temperature, liquid holdup and other data. It can also use the transient model to calculate the change trend of multiphase flow parameters after the inlet and outlet flow rates change, and realize online prediction, which is of certain significance for guiding actual production.

[0220] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A device and method for online prediction of multiphase pipe flow parameters based on a three-fluid model, characterized in that: It includes a continuous elbow 101, a data acquisition card 1 102, a data acquisition card 2 104, a data processing system 103, an inlet pressure sensor 105, a temperature sensor 106, a flow sensor 107, mid-pipeline pressure sensors 108 and 110, and a temperature sensor 109; the data processing system 103 is composed of multiple modules, including a data access module 201, a boundary condition processing module 202, a multiphase flow parameter steady-state calculation module 203, a multiphase flow parameter transient prediction module 204, a parameter online comparison module 205, and a result post-processing module 206. The device can be used for the collection of multiphase flow parameters, and the prediction and output of multiphase flow parameters are completed through the built-in data processing system.

2. An online prediction device and method for multiphase pipe flow parameters based on a three-fluid model, characterized in that: The following steps are involved: Step 1: The real-time pressure, temperature and flow rate data of the multiphase flow are measured by various pressure, temperature and flow rate sensors and transmitted to the data processing system via the data acquisition card; Step 2: The data processing system reads the flow parameter digital signal converted by the data acquisition card in real time through the data access module, and outputs it to the boundary condition processing module. According to the pipeline parameters (such as pipe diameter, pipe length, wall thickness, insulation layer thickness, thermal conductivity, etc.) and fluid components (such as gas components, oil components, etc.) pre-defined in this module, the model boundary conditions are constructed, and the boundary conditions are transmitted to the multiphase flow parameter steady-state calculation module; During the production process of oil and gas fields, the output at the wellhead often fluctuates, so unstable multiphase flow will be formed in the pipeline; the spatial distribution of multiphase flow parameters can be calculated using the steady-state model, but it cannot reflect the fluctuation of flow parameters over time. However, the steady-state calculation can provide initial values ​​for transient prediction, so the steady-state calculation of multiphase flow parameters must be performed first, followed by transient prediction; Step 3: Steady-state calculation of multiphase flow parameters: According to the model boundary conditions input into the model at time j, a steady-state three-fluid model is established based on the state equation and the transient three-fluid model, and the flow parameters at any node i in the pipeline at time j are solved to obtain the change trends of parameters such as pressure, temperature, flow rate, and liquid holdup along the flow direction, so as to supplement the data at non-critical nodes, and output the calculation results to the transient prediction module of multiphase flow parameters as the initial value of transient prediction; Step 4: transient prediction: take the steady-state data at each node i calculated at time j as the initial value, apply the transient three-fluid model, calculate the flow parameters under the j+1 time layer, give the trend of the pressure, temperature, flow rate, liquid holdup and other parameters at any node i of the pipeline over time, and realize the prediction of future parameters. After the prediction is completed, the results are output to the parameter online comparison module to realize the comparison between the real-time collected data and the predicted data; Step 5: Data display and export: All data used and calculated in the data processing system are displayed on the display screen in the form of images through the result post-processing module, or exported to an external system; The third step specifically comprises the following steps: (1) Calculation of fluid physical parameters Preferably, the PR state equation is used to calculate the thermophysical parameters such as density, viscosity, enthalpy, entropy, etc. of the mixed fluid, the constructed boundary conditions are input into the PR equation, and the initial thermophysical parameters of the fluid are solved by the dichotomy method; (2) Establishment of three-fluid simulation model Preferably, a three-fluid model is used to calculate the fluid parameters in the pipeline; equations are established for three continuous phases and six dispersed phases, including nine continuity equations, three momentum equations, three energy equations and one phase fraction equation; The continuity equation is as follows: The momentum equation is as follows: The energy equation is as follows: The phase fraction equation is as follows: You k =1 (4) Where α k ——phase fraction of k phase; ρ k ——density of phase k, kg / m 3 ; u k ——Flow rate of phase k, m / s; P k ——pressure of phase k, Pa; T k ——temperature of phase K, K; τ k ——Shear stress of k phase, N / m 2 ; h k ——enthalpy of phase k, J; F ki ——friction resistance between phases k, Pa·m; F kw ——Friction resistance between k phase and tube wall, Pa·m; λ k ——Thermal conductivity of phase k, W / (m 2 K); ——turbulent viscosity of phase k, mPa·s; ——the turbulent Planck number of the k-phase; ——heat transfer rate from phase k to component m, W / (m·K); g——acceleration due to gravity, m / s 2 ; h——fluid height, m; θ——pipeline inclination, rad; (3) Establishment and discretization of steady-state simulation model The steady-state equation does not consider the time term, and the above equations (1) to (4) can be simplified to: Along the pipeline flow direction x, the pipeline is divided into n grids. Assuming the grid length Δx, the central difference method is applied to the above equations (5) to (7) to discretize them, and the following equation is obtained: F 1～9 are the 9 equations obtained after discretization of the continuity equation, F 10～12 are the three equations obtained after the momentum equation is discretized, F 13～15 The three equations obtained after discretization of the energy equation and one phase fraction equation are shown in Table 1. The hydraulic equation F 1～12 With the thermodynamic equation F 13～15 Separate and rewrite into the following matrix form, we get: S s (x)=0 (11) R s (x)=0 (12) Where S s ——Steady-state hydraulic equation F 1～12 The matrix S s (x)=[F1(x),F2(x),...,F 12 (x)] T , where x is the multiphase flow parameter to be determined (pressure, temperature, flow rate, etc.); R s ——Steady-state thermodynamic equation F 13～15 The matrix R s (x)=[F 13 (x),F 14 (x),F 15 (x)] T ; (4) Model solution Applying Taylor expansion to equations (11) and (12), ignoring higher-order terms, we obtain: S′ s (x)Δx=-S s (x) (13) R′ s (x)Δx=-R s (x) (14) Where S′ s (x)——Jacobian matrix of steady-state hydraulic equations; R′ s (x)——Jacobian matrix of the steady-state thermodynamic equations; Δx s ——Steady-state correction value matrix; Read the current flow parameters and boundary conditions, use the Newton-Raphson method to solve equation (13), and obtain the hydraulic parameters such as pressure, flow rate, and liquid holdup of the multiphase flow. Substitute the hydraulic parameters obtained by equation (13) into equation (14) and inversely solve the fluid temperature. Assuming that the calculation error is ε, when equation (15) is satisfied, the steady-state calculation is completed and the parameters are output. Otherwise, substitute the temperature solved by the thermodynamic equation into the hydraulic equation and solve it again until the convergence condition is met. ε=|T S -T R |<10 -5 (15) Where T S ——The fluid temperature assumed when calculating the hydraulic equation, K; T R ——Fluid temperature obtained by solving the thermodynamic equation, K; The fourth step specifically comprises the following steps: (1) Obtaining the initial value of transient calculation The multiphase flow parameters such as pressure, temperature, flow rate, and liquid holdup calculated by the steady-state model in step 3 are used as the initial conditions of the transient model; (2) Discretization of transient three-fluid model The transient three-fluid model is shown in equations (1) to (4). The flow plane is meshed along the pipeline flow direction x and the time growth direction t. The time step is Δt and the space step is still Δx. The time term is discretized using forward difference and the space term is discretized using the central difference method. The following equation is obtained: Similarly, we get 9 continuity equations, 3 momentum equations, 3 energy equations and 1 phase fraction equation, which can be rewritten into matrix form: S t (x)=0 (19) R t (x)=0 (20) Where S t ——Transient hydraulic equation E 1～12 The matrix S t (x)=[E1(x),E2(x),...,E 12 (x)] T , where x is the multiphase flow parameter to be determined (pressure, temperature, flow rate, etc.); R t ——Transient thermodynamic equation E 13～15 The matrix R t (x) = [E 13 (x),E 14 (x),E 15 (x)] T ; (3) Solution of transient three-fluid model Applying Taylor expansion to equations (21) and (22), ignoring higher-order terms, we obtain: S′ t (x)Δx=-S t (x) (21) R′ t (x)Δx=-R t (x) (22) Where S′ t (x)——Jacobian matrix of transient hydraulic equations; R′ t (x)——Jacobian matrix of transient thermodynamic equations; Δx t ——transient correction value matrix; Input the steady-state calculation results of step 3, apply the Newton-Raphson method to first solve equation (21), then solve equation (22), and when the convergence condition (15) is met, stop the iteration and obtain the transient calculation results.

Images