A computational fluid dynamics based turbine flow meter design method

Abstract

The application discloses a turbine flowmeter design method based on computational fluid dynamics, and belongs to the technical field of sensors. Step 1, a preliminary structure design of the turbine flowmeter is performed according to performance index requirements; step 2, a fluid domain model of the turbine flowmeter for fluid analysis is established according to the preliminary structure of the turbine flowmeter; step 3, a pretreatment of computational fluid dynamics analysis is performed on the fluid domain model of the turbine flowmeter, so that a grid model of the turbine flowmeter is obtained; step 4, numerical simulation calculation is performed on the grid model of the turbine flowmeter based on computational fluid dynamics; step 5, characteristic curves of the turbine flowmeter are extracted in the numerical calculation results, and the performance index of the turbine flowmeter is evaluated; if the performance index is not satisfied, the turbine flowmeter is fed back to step 1 in combination with internal flow characteristics of the turbine flowmeter in the calculation results, so that the turbine flowmeter is guided to perform structure optimization adjustment. The application effectively improves the design efficiency of the turbine flowmeter.

Description

Technical Field

[0001] This invention relates to the field of sensor technology, and more specifically to a design method for a turbine flow meter based on computational fluid dynamics. Background Technology

[0002] As a core instrument for high-precision volumetric flow measurement, the turbine flow meter's technological development is rooted in the deep integration of fluid dynamics and precision mechanical sensing. Since the mid-20th century, with the increasing demands for industrial automation and energy metering, the turbine flow meter has gradually evolved from a laboratory prototype into standardized industrial equipment, widely used in key fields such as natural gas transportation, petrochemicals, water treatment, and pharmaceuticals.

[0003] Its working principle is based on the law of conservation of angular momentum: when fluid passes through the impeller, it generates a driving torque, which drives the turbine to rotate, and its angular velocity is proportional to the volumetric flow rate; the impeller blades cut the magnetic field of the magnetoelectric sensor, generating a pulse frequency signal that is linearly related to the flow rate, thus realizing non-contact high-resolution measurement. This mechanism gives it significant advantages such as fast response, high repeatability, and wide rangeability (up to 10:1 to 50:1).

[0004] In the design phase of turbine flow meters, traditional numerical calculation methods for turbine rotor speed require manual and repeated adjustments of the speed based on the rotor's resultant torque during transient calculations until the torque reaches equilibrium. This method not only consumes significant computational resources and is time-consuming, but also often relies on experimental data, resulting in high experimental costs and making it difficult to quickly conduct numerical calculations for different inflow conditions. Summary of the Invention

[0005] In view of the above, this invention proposes a turbine flowmeter design method based on computational fluid dynamics (CFD). Introducing mature CFD technology into the design phase of the turbine flowmeter helps shorten the development and design cycle, reduce design costs, and improve the reliability and stability of the designed flowmeter. This method relies on CFD to perform numerical simulations on the designed turbine flowmeter, analyze the calculation results, and evaluate whether the flowmeter's performance meets the performance indicators (such as flow measurement range, pressure loss, pipe diameter, etc.). If the performance indicators do not meet the requirements, the design of the turbine flowmeter needs to be optimized and verified through numerical calculations using CFD until the designed turbine flowmeter meets the performance requirements. Finally, the turbine flowmeter that meets the performance requirements is then manufactured.

[0006] This invention provides a design method for a turbine flow meter based on computational fluid dynamics, comprising: Step 1: Conduct preliminary structural design of the turbine flow meter based on performance requirements; Step 2: Establish a fluid domain model of the turbine flow meter for fluid analysis based on the preliminary structure of the turbine flow meter; Step 3: Perform preprocessing of the turbine flowmeter fluid domain model for computational fluid dynamics analysis to obtain the mesh model of the turbine flowmeter; Step 4: Perform numerical simulation calculations on the mesh model of the turbine flow meter based on computational fluid dynamics; Step 5: Extract the characteristic curve of the turbine flow meter from the numerical calculation results and evaluate the performance index of the turbine flow meter. If the performance index is not met, combine the internal flow characteristics of the turbine flow meter in the calculation results and feed them back to Step 1 to guide the structural optimization and adjustment of the turbine flow meter.

[0007] Preferably, step 1 includes the following steps: Step 1.1: Based on the pipe diameter, pressure loss, and flow measurement range requirements, determine the basic dimensions of each component of the turbine flow meter, as well as the hub diameter, blade tip diameter, hub thickness, number of blades, and blade thickness parameters of the turbine in the turbine flow meter. Step 1.2: Determine the turbine flow meter housing size, interface type, and internal flow channel diameter parameters based on the requirements of pipe diameter, pressure rating, and interface type. Step 1.3: Select the material properties of each component of the turbine flow meter based on the material properties of the measured medium, the temperature and pressure of the medium, and the weight parameters of the flow meter.

[0008] Preferably, step 2 includes the following steps: Step 2.1: Import the preliminary structural drawing of the turbine flow meter into the 3D CAD software. Based on the internal flow channel dimensions of the turbine flow meter housing, initially establish the turbine flow meter fluid domain, including the dimensions of each component of the turbine flow meter. Step 2.2: In the 3D CAD software, remove the space occupied by solid components from the initially established turbine flowmeter fluid domain to obtain the turbine flowmeter fluid domain used in computational fluid dynamics calculations.

[0009] Preferably, step 3 includes the following steps: Step 3.1: Import the obtained turbine flowmeter fluid domain into the mesh generation software. Based on the flow field calculation requirements and the complexity of the fluid domain, mesh generation is performed using a combination of structured and unstructured meshes. Step 3.2: Refine the mesh in areas with large curvature on the model surface, and appropriately enlarge the mesh size in areas with gentle curvature on the model surface, balancing computational accuracy and efficiency. Step 3.3: After the mesh is generated, check the mesh quality to ensure that it meets the requirements of CFD numerical calculation, and finally obtain a mesh model that can be directly used for subsequent numerical simulation.

[0010] Preferably, step 4 includes the following steps: Step 4.1: Set the basic conditions for numerical simulation calculation; Step 4.2: Select the turbulence model for numerical simulation calculation; Step 4.3: Set the boundary conditions for numerical calculation; Step 4.4: After completing the model selection and relevant boundary condition settings for the numerical simulation, initialize the flow field to provide a reasonable initial flow field distribution for iterative calculation; after initialization, perform steady-state numerical simulation calculations and use the steady-state flow field calculation results as the initial flow field for subsequent transient calculations, and obtain a fully developed and convergent flow field distribution through steady-state calculations. Step 4.5: Perform transient calculations on the turbine flow meter based on UDF to obtain the rotor motion characteristics.

[0011] Preferably, step 4.1 includes the following steps: In numerical calculations, the liquid medium is approximated as an incompressible fluid. The minute heat exchange during the flow process is ignored, and the physical properties of the corresponding fluid medium are directly given. The basic governing equations of the fluid medium are as follows: (4.1) (4.2) In the formula , Coordinate system components; , These are the velocity components in the coordinate system; It is time; It is the density of the liquid; For pressure; The kinematic viscosity of the fluid.

[0012] Preferably, step 4.2 includes the following steps: Considering the high-speed rotating turbine, components with significant variations in turbine blade surface curvature, and complex flows near the blade wall, the numerical calculation solution mode is set to double-precision solution, and the turbulence model is selected from viscous fluid dynamics. The turbulence model has the following transport equation: Turbulent kinetic energy k-transport equation: (4.3) Turbulent frequency equation: (4.4) Turbulent eddy viscosity coefficient μ t : (4.5) Turbulent kinetic energy generation term P k : (4.6) The modulus S of the dependent variable tensor: (4.7) (4.8) In the formula: Let be the fluid density; U be the fluid velocity vector; μ be the fluid dynamic viscosity; F1 and F2 be mixing functions used to realize the near-wall region. Model and far-field region The model's smooth transition is achieved by using F1 to control the switching between the near-wall and mainstream regions, and F2 to assist in correcting the characteristics of the far-field region. , These are empirical constants corresponding to the effective diffusion coefficients of turbulent kinetic energy and turbulent eddy frequency, respectively, and their values ​​are obtained by weighting the mixing function F1, i.e. , ; The standard values ​​of the constants in the above formulas in turbulence models are usually: , , , , , , , , .

[0013] Within the discrete framework of the finite volume method, the fundamental governing equations of the fluid and the aforementioned equations are solved iteratively. The transport equations corresponding to the turbulence model ultimately yield stable numerical solutions for key physical quantities at each grid node and element within the flow field. These key physical quantities include at least velocity, pressure, and temperature.

[0014] Preferably, step 4.3 includes the following steps: In the numerical calculation of the turbine flow meter, the energy equation is ignored, and only the flow control equation is considered. The walls of the computational domain are all set as adiabatic, no-slip wall boundary conditions; the outlet boundary uses a free outflow boundary condition; and the inlet uses a velocity inlet boundary condition. The inlet velocity is determined based on the given flow conditions, and the calculation formula is shown in the following equation: (4.9) In the formula: v is the flow velocity at the inlet; Q is the volumetric flow rate at the inlet; and r is the radius at the inlet.

[0015] Preferably, step 4.5 includes the following steps: (a) Read the numerical calculation results of the steady-state flow field; its turbulence model is: Turbulent model, boundary conditions as given in step 4.3.

[0016] (b) Switch the calculation mode from steady state to transient state and configure the relevant settings for transient calculation; (c) Load the written User Defined Function program, set the flow domain corresponding to the turbine rotor as the rotation domain, set its motion mode as mesh motion, and set the rotation motion mode as the loaded UDF program; set the transient time step as the time taken for the turbine flow meter rotor to rotate 2°, and ensure that the rotor rotates at least 30 times in the total calculation time; after completing the above settings, perform transient calculation of the turbine flow meter.

[0017] The transient numerical simulation calculation is complete. The corresponding calculation results are saved for subsequent calculation and analysis of the turbine flow meter. The principle behind the User Defined Function program is as follows: When fluid flows through the rotor, the impact of the fluid on the rotor generates a driving torque Td. When the driving torque exceeds the sum of all deceleration torques, the rotor begins to rotate. During the acceleration and deceleration of the rotor, the torque balance equation of the rotor is as follows: (4-10) In the formula: ω is the angular velocity of the rotor, rad / s; J is the moment of inertia of the rotor, kg·m. 2 ;T r T is the viscous deceleration torque on the blade surface. t T is the deceleration torque of the blade tip clearance; hh T is the viscous frictional deceleration torque at the rotor hub end face; hp The viscous frictional deceleration torque on the outer circumferential surface of the rotor hub; T m For magnetic deceleration torque; T b For bearing deceleration torque; When the rotor speed is stable, the rotor's angular acceleration is zero. At this time, the rotor's torque balance equation is as follows: (4-11) In the formula: Various deceleration torques are represented in equation (4-10). In establishing the flow domain model of the turbine flow meter, the resistance of the bearings to the flow sensor rotor and the magnetic reluctance generated by the speed sensor on the turbine flow meter's motion are ignored. Therefore, the resistance of the turbine flow meter rotor in numerical calculations is mainly generated by fluid viscosity. The torque equation of the turbine flow meter rotor is shown below: (4-12) In summary, the turbine rotor speed of the turbine flow meter under different incoming flow conditions can be calculated using the following formula: (4-13) In the formula: The rotor speed is calculated for the nth time. This refers to the rotor speed calculated in the (n-1)th time. Ultimately, the output turbine flow meter speed It is calculated by the following formula: (4-14) In the formula: The rotor speed is calculated for the nth time. This is the rotor speed calculated for the (n-1)th time.

[0018] Preferably, step 5 includes the following steps: Step 5.1: After the numerical calculation converges and reaches a steady state, post-processing and result analysis are carried out on the internal flow field and rotor motion characteristics of the turbine flow meter; key performance parameters of the rotor are monitored, including at least the fluid torque, rotational speed and inlet / outlet pressure difference; the rotor motion characteristics and flow response law under different incoming flow conditions are quantitatively analyzed to determine whether the sensor characteristic curve meets the design requirements. Step 5.2: If the calculation results show that the sensor performance does not meet the relevant index requirements, the key physical quantities in the computational domain can be extracted with the help of CFD post-processing software, including at least the velocity field, pressure field, turbulent kinetic energy and vorticity. The internal flow state can be intuitively displayed through cloud maps, vector maps and streamline diagrams to identify typical features, including at least flow separation, vortex structure and local energy loss, to clarify the performance shortcomings and provide a clear direction for subsequent structural improvement and optimization design.

[0019] The beneficial effects of this invention are: 1) This invention utilizes a user-defined function (UDF) in the transient numerical calculation of turbine flow meters to automatically solve for the rotational speed based on the turbine rotor torque balance equation. This program can directly simulate the real-time rotation of the turbine rotor under different inflow conditions and accurately obtain the rotor's steady-state speed through automatic iteration, eliminating the need for manual intervention and repeated calculations. This method significantly optimizes the turbine flow meter design process, greatly shortens the design cycle, reduces R&D costs, and effectively improves the design efficiency of turbine flow meters.

[0020] 2) Through systematic analysis of the flow field structure and macroscopic performance, this invention can provide data support and theoretical basis for the structural optimization, error suppression and performance improvement of turbine flow meters, effectively improving design efficiency and product reliability. Attached Figure Description

[0021] The features and advantages of the invention will be more clearly understood by referring to the accompanying drawings, which are schematic and should not be construed as limiting the invention in any way. In the drawings: Figure 1 This is a shaft system diagram of a certain type of turbine flow meter; Figure 2 This is a schematic diagram of the assembly of the housing and the turbine flow meter; Figure 3 This is a diagram of the fluid domain model. Figure 4 This is a half-section of the fluid domain model; Figure 5 This is a diagram of the fluid domain mesh model; Figure 6 This is a schematic diagram of the surface mesh division within the fluid domain; Figure 7 This is a flowchart of transient calculation and a schematic diagram of UDF calculation strategy; Figure 8 This is a schematic diagram of the calculated coefficient curve of a certain type of turbine flow meter; Figure 9 This is a schematic diagram of the calculated pressure loss curve of the model turbine flow meter; Figure 10 This is a pressure distribution cloud map of a certain type of turbine flow meter; Figure 11 This is a velocity distribution cloud map of a certain type of turbine flow meter; Figure 12 Schematic diagram of grid densification for flow sensors; Figure 13 The overall flowchart of the method of this invention. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0023] This invention proposes a turbine flow meter design method based on computational fluid dynamics, comprising the following steps: Step 1: Conduct preliminary structural design of the turbine flow meter based on performance requirements; The common performance indicators of turbine flow meters include the following: Flow measurement range: 300L / h~3000L / h; Interface type: threaded interface or flange interface; Measuring medium: RP-3 aviation kerosene (or other common fuel oil); Medium temperature: ambient temperature; Medium pressure: Atmospheric pressure; Weight requirement: The turbine flow meter must weigh no more than 2kg (excluding the meter head).

[0024] Pressure loss requirement: not greater than 300 kPa; Pipe diameter: DN12; (i) Withstand voltage rating: PN40; During the design process of turbine flow meters, preliminary design is typically carried out based on the aforementioned performance requirements. The specific preliminary design process is as follows: Based on requirements such as pipe diameter, pressure loss, and flow measurement range, determine the basic dimensions of each component of the turbine flow meter, as well as the relevant parameters such as the hub diameter, blade tip diameter, hub thickness, number of blades, and blade thickness of the turbine, the core component of the turbine flow meter.

[0025] The dimensions of the turbine flow meter housing, the interface type, and the inner flow channel diameter are determined based on requirements such as pipe diameter, pressure rating, and interface type.

[0026] The materials of each component of the turbine flow meter are determined based on indicators such as the measured medium, the temperature and pressure of the medium, and the weight of the flow meter.

[0027] The preliminary design of the turbine flow meter can be completed through the above process. The shaft system of a certain type of turbine flow meter is shown below. Figure 1 As shown. The assembly of the completed housing with the turbine flow meter is as follows. Figure 2 As shown.

[0028] Step 2: Establish a fluid domain model of the turbine flow meter for fluid analysis based on the preliminary structure of the turbine flow meter; Establishing a fluid domain model of the turbine flowmeter for computational fluid dynamics (CFD) analysis is a crucial first step in performing CFD analysis. The core of establishing this model is abstracting the space through which the fluid can pass from the physical entity; this space constitutes the turbine flowmeter's fluid domain model. The specific process is as follows: (1) In the three-dimensional CAD software, based on the internal flow channel size of the turbine flow meter housing, the turbine flow meter fluid domain including the size of each component of the turbine flow meter is initially established.

[0029] (2) In 3D CAD software, the space occupied by solid components such as turbine rotors, guide vanes, transition parts, and shafts is removed from the initially established turbine flowmeter fluid domain using a "Boolean subtraction operation". The remaining portion of the initially established turbine flowmeter fluid domain is then the turbine flowmeter fluid domain that can be used for computational fluid dynamics calculations. The fluid domain model of a certain type of turbine flowmeter is shown below. Figures 3-4 Place.

[0030] Step 3: Perform preprocessing of the turbine flowmeter fluid domain model for computational fluid dynamics analysis to obtain the turbine flowmeter mesh model; specifically: The obtained turbine flowmeter fluid domain is imported into the mesh generation software. Based on the flow field calculation requirements and the complexity of the fluid domain, a combination of structured and unstructured meshes is used for mesh generation (or a single mesh type is selected according to the actual working conditions). The choice of mesh type is mainly related to the available computing resources and the complexity of the fluid domain model.

[0031] Structured meshes are typically hexahedral meshes, and dividing the fluid domain into hexahedral meshes has many advantages: high mesh quality, fast generation speed, and less computational resources required under the same computational conditions. However, structured meshes are only suitable for models with simple and regular geometry, and their geometric adaptability is poor, making it difficult to handle complex and irregular structures; when performing structured mesh generation on complex models, the operation is cumbersome, the workload is large, and the time consumption is long.

[0032] Unstructured meshes are mostly tetrahedral meshes, which have strong geometric adaptability, simple mesh generation process, and can quickly adapt to complex shapes. However, under the same conditions, unstructured meshes consume more computational resources and have a relatively slower solution speed.

[0033] Therefore, in practical numerical simulations, a more suitable mesh generation method should be selected based on the complexity of the fluid domain model and available computing resources. For this patent, since the turbine flowmeter fluid domain model is relatively complex and computing resources are sufficient, an unstructured mesh can also be used to partition the fluid domain model.

[0034] Mesh refinement is applied to critical areas with large surface curvature, such as the area around the turbine rotor and the turning points of the flow channel, while the mesh size is appropriately enlarged in areas with gentle flow, such as the main flow channel, to balance computational accuracy and efficiency.

[0035] "Regions with large surface curvature" and "regions with gentle flow" refer to the fluid domain model. The mesh is refined in areas of the fluid domain model where the surface curvature changes significantly. This increases the number of meshes to ensure the quality of the generated mesh cells and the accuracy of numerical calculations. Generally, it is based on experience to determine which part of the model needs to be refined.

[0036] The fluid domain model mesh image of the turbine flow meter in this patent is as follows: Figure 12 As shown, the turbine rotor has a significantly larger surface curvature compared to other components of the flow sensor. To ensure the quality of mesh generation and the accuracy of subsequent numerical calculations, the rotor of the flow sensor is meshed.

[0037] Step 4: Perform numerical simulation calculations on the mesh model of the turbine flowmeter based on computational fluid dynamics (CFD). Numerical calculations using CFD software such as AnsysFluent, Ansys CFX, OpenFoam, and STAR-CCM+ are conducted. The core essence of this approach is to solve the governing equations governing fluid flow using numerical methods, thereby obtaining discrete numerical solutions for each physical quantity in the flow field. Fluid flow processes strictly adhere to fundamental physical laws such as mass conservation, momentum conservation, and energy conservation. The corresponding governing equations (such as the Navier-Stokes equations for incompressible or compressible fluids) form the theoretical basis for numerical calculations.

[0038] During the calculation process, reasonable boundary conditions that conform to actual working conditions (such as velocity / pressure settings at the inlet boundary, pressure settings at the outlet boundary, no-slip / slip conditions at the wall boundary, etc.), fluid material properties, and suitable turbulence models (such as...) are input into the software. Model, (Models, etc.) to construct a complete mathematical boundary value problem. In fluid mechanics software, the finite volume method is used to discretize the continuous computational domain into a large number of interconnected grid cells. The discretized control equations are then discretized and iteratively solved. Finally, the numerical distribution results of key physical quantities such as velocity, pressure, and temperature on each grid node and cell within the flow field are output, providing data support for subsequent flow field characteristic analysis, operating condition optimization, and other research.

[0039] After mesh generation, the mesh quality is checked (e.g., mesh distortion rate, orthogonality) to ensure it meets the requirements of CFD numerical calculations, ultimately resulting in a mesh model that can be directly used for subsequent numerical simulations. The mesh model for a certain type of turbine flowmeter is shown below. Figures 5-6 As shown.

[0040] After mesh generation, evaluating the mesh quality is crucial for ensuring the stability, accuracy, and convergence of numerical calculations. Poor-quality meshes can easily lead to computational divergence, increased numerical dissipation, result distortion, and even computational interruption. Commonly used mesh quality evaluation methods include checking for negative volumes in elements to avoid basic calculation errors; determining aspect ratios to control the degree of element stretching and prevent a sharp drop in local solution accuracy; analyzing skewness (orthogonality) to evaluate the degree of deviation of element shape from the ideal form, which is a core indicator affecting discretization accuracy; and checking warpage and Jacobian ratio to determine the degree of element distortion and ensure the smoothness and reliability of the flow field solution. Only by comprehensively judging the mesh quality as qualified (without poor-quality mesh elements) through the above indicators can the accuracy and reliability of subsequent numerical simulation results be ensured.

[0041] In the design and development method of the turbine flow meter proposed in this invention, the industrial CAE software Ansys Fluent is used to conduct numerical simulation calculations of the turbine flow meter. Specifically, it includes the following steps: Step 4.1: Set the basic conditions for numerical simulation calculation; The turbine flow meter in this invention is mainly used for liquid flow measurement. In numerical calculations, the liquid medium can be approximated as an incompressible fluid. During the flow measurement process of the turbine flow meter, the temperature change of the fluid inside the flow sensor is very small, and the temperature difference between the external environment and the fluid medium is not significant. Therefore, the minute heat exchange during the flow process can be ignored. Thus, in numerical calculations, the density and dynamic viscosity of the corresponding fluid medium can be directly given. For example, RP-3 aviation kerosene under normal temperature and pressure conditions has a density of 778 kg / m³. 3 The dynamic viscosity is 0.00138 Pa·s. Therefore, the basic governing equations for the fluid medium are shown in equations (4.1) to (4.2): (4.1) (4.2) In the formula , Coordinate system components; , These are the velocity components in the coordinate system; It is time; It is the density of the liquid; For pressure; The kinematic viscosity of the fluid.

[0042] Step 4.2: Select the turbulence model for numerical simulation calculation; After determining the basic fluid control equations by setting the material properties of the fluid, it is still necessary to select the appropriate turbulence model based on the type of solution. In the numerical calculation of turbine flow meters, it is necessary to consider the complex flow near the wall of components with large surface curvature changes, such as high-speed rotating turbines and turbine blades. Therefore, the solution mode of the numerical calculation needs to be set to double-precision solution to improve the accuracy of the numerical calculation. The turbulence model selected is from viscous fluid dynamics. Turbulence model. It uses a mixing function to... Model and The combined model can adapt to different incoming flow conditions, has good reliability, converges quickly, and is more accurate for numerical simulations containing flow separation phenomena. Its transport equations are shown in (4.3) to (4.8).

[0043] Turbulent kinetic energy k-transport equation: (4.3) Turbulent frequency equation: (4.4) Turbulent eddy viscosity coefficient μ t : (4.5) The relevant formulas in the transport equations of the turbulent model are shown below: Turbulent kinetic energy generation term P k (4.6) The modulus S of the dependent variable tensor (4.7) (4.8) In the formula: Let be the fluid density; U be the fluid velocity vector; μ be the fluid dynamic viscosity; F1 and F2 be mixing functions used to realize the near-wall region. Model and far-field region The model's smooth transition is achieved by using F1 to control the switching between the near-wall and mainstream regions, and F2 to assist in correcting the characteristics of the far-field region. The mixing functions F1 and F2 and their related formulas are shown below: in: , These are empirical constants corresponding to the effective diffusion coefficients of turbulent kinetic energy and turbulent eddy frequency, respectively, and their values ​​are obtained by weighting the mixing function F1, i.e. , ; The remaining variables are empirical parameters, and commonly used values ​​are given here. The standard values ​​of the constants in the above formulas in turbulence models are usually: , , , , , , , , .

[0044] Within the discrete framework of the finite volume method, the fundamental governing equations of the fluid in equations 4.1 and 4.2, as well as the above-mentioned equations, are solved iteratively. The transport equations corresponding to the turbulence model ultimately yield stable numerical solutions for key physical quantities such as velocity, pressure, and temperature at each grid node and element within the flow field, providing solid data support for subsequent research. A stable numerical solution refers to a flow field iterative calculation reaching a convergent state, where the residuals of each physical quantity fall below the convergence threshold and no longer change significantly with the number of iterations. At this point, the distributions of physical fields such as velocity, pressure, and temperature satisfy the governing equations and the transport equations of the turbulence model, demonstrating numerical convergence and stability, validity, and reliability, and can be used for subsequent analysis and conclusions.

[0045] Step 4.3: Set the boundary conditions for numerical calculation; In the numerical calculation of the turbine flow meter, since the heat exchange effect during the flow process is weak, the energy equation is ignored, and only the flow control equation is considered. The walls of the computational domain are all set as adiabatic no-slip wall boundary conditions; the inlet adopts a velocity inlet boundary, and the inlet velocity is determined according to the given flow conditions. The calculation formula is shown in equation (4.7). (4.9) In the formula: v is the flow velocity at the inlet; Q is the volumetric flow rate at the inlet; and r is the radius at the inlet.

[0046] Because the internal flow mechanism of a turbine flow meter is complex, its pressure loss is difficult to determine in advance. Therefore, a free outflow boundary condition is adopted at the outlet boundary.

[0047] Step 4.4: After completing the model selection and boundary condition settings for the numerical simulation, the flow field needs to be initialized. Flow field initialization aims to provide a reasonable initial flow field distribution for iterative calculations, avoiding calculation divergence or slow convergence due to an unreasonable initial field, and ensuring the stable progress of the numerical iteration process. After initialization, input the specific values ​​of the boundary conditions, conduct steady-state numerical simulation calculations, and use the steady-state flow field calculation results as the initial flow field for subsequent transient calculations. Obtaining a fully developed and convergent flow field distribution through steady-state calculations can effectively avoid calculation instability or slow convergence caused by excessive deviation between the initial flow field and actual operating conditions, thereby improving the stability and computational efficiency of transient calculations.

[0048] Numerical computational fluid dynamics (CFD) calculations in fluid dynamics software such as Ansys Fluent, Ansys CFX, OpenFoam, and STAR-CCM+ essentially involve solving the governing equations governing fluid flow using numerical methods (selection of the numerical model and related settings) to obtain discrete numerical solutions for various physical quantities in the flow field. Fluid flow processes strictly adhere to fundamental physical laws such as conservation of mass, momentum, and energy, and the corresponding governing equations (such as the Navier-Stokes equations for incompressible or compressible fluids) form the theoretical basis for numerical calculations.

[0049] Step 4.5: Perform transient calculations on the turbine flow meter based on UDF to obtain the rotor motion characteristics. The transient calculation process for the turbine flow meter is as follows: Figure 7 As shown.

[0050] (a) Read the numerical calculation results of the steady-state flow field, that is, the flow field distribution calculated in steady state, and its turbulence model is: The turbulent model uses the boundary conditions given above. There's no need to reset the boundary conditions here; the transient calculation is performed iteratively based directly on the steady-state flow field calculation results. The transient calculation involves introducing a time variable into the governing equations and specifying an appropriate time step to achieve a numerical simulation of the flow field's evolution over time.

[0051] (b) Switch the calculation mode from steady state to transient state and prepare the relevant settings for transient calculation, i.e., step (c); (c) Load the written User Defined Function program, set the flow domain corresponding to the turbine rotor as the rotation domain, set its motion mode to mesh motion, and set the rotational motion mode to the loaded UDF program. Set the transient time step to the time required for the turbine flowmeter rotor to rotate 2°, and ensure that the rotor rotates at least 30 revolutions in the total calculation time. After completing the above settings, perform transient calculations for the turbine flowmeter.

[0052] The transient numerical simulation is complete, and the corresponding calculation results are saved for subsequent calculation and analysis of the turbine flow meter. Software such as Flunt allows setting relevant monitoring parameters before calculation, and the calculation results of these monitoring parameters will be output in real time as text during the iteration process. The calculation results here are the flow field distribution at this moment, the calculation results of the monitoring parameters, the rotor speed, and other calculation results.

[0053] The process of building a User Defined Function program includes the following steps: Traditional numerical methods for calculating turbine rotor speed require manual and repeated adjustments of the speed based on the rotor's net torque during transient calculations until the torque reaches equilibrium. This method not only consumes significant computational resources and is time-consuming, but also often relies on experimental data, resulting in high experimental costs and making it difficult to quickly conduct numerical calculations for different inflow conditions.

[0054] This invention utilizes a user-defined function (UDF) to automatically solve for the rotational speed in the transient numerical calculation of turbine flow meters, based on the turbine rotor torque balance equation. This program can directly simulate the real-time rotation of the turbine rotor under different inflow conditions and accurately obtain the rotor's steady-state speed through automatic iteration, eliminating the need for manual intervention and repeated calculations. This method significantly optimizes the turbine flow meter design process, greatly shortens the design cycle, reduces R&D costs, and effectively improves the design efficiency of turbine flow meters.

[0055] The UDF program involved in this invention is developed based on the principle of conservation of angular momentum of the turbine rotor.

[0056] When fluid flows over the rotor, the impact of the fluid on the rotor generates a driving torque (Td). When the driving torque exceeds the sum of all deceleration torques, the rotor begins to rotate. During the rotor's acceleration and deceleration processes, the rotor's torque balance equation is as follows: (4-10) In the formula: ω is the angular velocity of the rotor, rad / s; J is the moment of inertia of the rotor, kg·m2; Tr is the viscous deceleration torque on the blade surface; Tt is the deceleration torque at the blade tip clearance; Thh is the viscous friction deceleration torque at the rotor hub end face; Thp is the viscous friction deceleration torque at the outer circumferential surface of the rotor hub; Tm is the magnetic deceleration torque; Tb is the bearing deceleration torque.

[0057] When the rotor speed is stable, the rotor's angular acceleration is zero. At this time, the rotor's torque balance equation can be expressed as follows: (4-11) In the formula: Various deceleration torques are represented in equation (4-10). In establishing the flow domain model of the turbine flow meter, the rotor and bearing in the flow sensor rotor are approximated as a single unit, ignoring the resistance of the bearing to the flow sensor rotor. Furthermore, the magnetic reluctance generated by the speed sensor in the turbine flow meter has a negligible impact on the turbine flow meter's motion; therefore, the resistance of the turbine flow meter rotor in numerical calculations is mainly generated by fluid viscosity, and the torque equation of the turbine flow meter rotor can be expressed as follows: (4-12) In summary, the turbine rotor speed of a turbine flow meter under different incoming flow conditions can be calculated using the following formula: (4-13) In the formula: The rotor speed is calculated for the nth time. This is the rotor speed calculated for the (n-1)th time.

[0058] Ultimately, the output turbine flow meter speed It is calculated using the following formula.

[0059] (4-14) In the formula: The rotor speed is calculated for the nth time. This is the rotor speed calculated for the (n-1)th time.

[0060] Step 5: Extract the characteristic curve of the turbine flow meter from the numerical calculation results, as shown below. Figure 8-9 Evaluate the performance indicators of the turbine flow meter; if it does not meet the performance indicators, then combine the calculation results with the relevant information. Figure 10-11 The internal flow characteristics of the turbine flow meter, i.e. the internal flow state, are fed back to step 1 to guide the turbine flow meter to make structural optimization adjustments.

[0061] Specifically, after the numerical calculations converge and reach a stable state, post-processing and result analysis are performed on the internal flow field distribution and rotor motion characteristics (rotor speed, torque, and other phase pipe parameters output by the software in real time during the iteration process) of the turbine flowmeter. Key performance parameters such as the fluid torque, speed, and inlet / outlet pressure difference of the rotor are monitored. The rotor motion characteristics and flow response under different inflow conditions are quantitatively analyzed to determine whether the sensor characteristic curves meet the design requirements. The pressure loss curve and flow characteristic curve of a certain type of turbine flowmeter obtained through the above calculation process are shown below. Figures 8-9 As shown.

[0062] Post-processing and results analysis include the following steps: The instrument coefficient of a turbine flow meter is defined as follows: (1) In the formula: f is the frequency of the turbine flow meter rotor, which is obtained by numerical calculation of the rotor speed; Q is the volumetric flow rate through the flow sensor, which is determined by the flow conditions.

[0063] The measurement accuracy of a turbine flow meter can be evaluated by the linearity error E of the instrument coefficient, and its calculation formula is shown below.

[0064] (2) In the formula: (K) i ) max The maximum instrument coefficient for each flow condition; (K) i ) min This represents the minimum number of instruments required for each flow condition. For turbine flow meters, the smaller the linearity error of the instrument coefficient, the higher the measurement accuracy.

[0065] The pressure loss of a turbine flow meter is a key indicator of the energy loss of a flow sensor. Under the same flow conditions, the smaller the pressure loss, the smaller the energy loss of the flow sensor. The pressure loss of a flow sensor can be calculated by the following formula. (3) In the formula: P in The pressure at the inlet can be obtained from the output of numerical calculations; P out The pressure at the outlet can also be obtained from the results output in numerical calculations.

[0066] Through the above calculations, the instrument characteristic curve of the turbine flow meter can be obtained, such as... Figure 8 As shown; the pressure loss curve of the flow sensor can be obtained, such as Figure 9 As shown.

[0067] If the calculation results show that the sensor performance does not meet the relevant performance requirements, CFD post-processing software can be used to extract key physical quantities such as velocity field, pressure field, turbulent kinetic energy, and vorticity within the computational domain. The internal flow state can be visually displayed through contour maps, vector maps, and streamline diagrams, identifying typical characteristics such as flow separation, vortex structures, and local energy losses. This clarifies performance shortcomings and provides a clear direction for subsequent structural improvements and optimization designs. The velocity distribution of the internal flow field of a certain type of turbine flowmeter is shown below. Figures 10-11 As shown.

[0068] A systematic analysis of the flow field structure and macroscopic performance can provide data support and theoretical basis for the structural optimization, error suppression and performance improvement of turbine flow meters, effectively improving design efficiency and product reliability.

[0069] As described above, the design and development method of the turbine flow meter based on the present invention, and its numerical calculation process, from the construction of the computational domain, setting of boundary conditions, selection of turbulence model, initialization of flow field and steady-state-transient coupled solution, to automatic iteration of rotor speed based on UDF, and then to flow field post-processing and performance analysis, form a complete, efficient and automated turbine flow meter simulation and design optimization system.

[0070] This process not only effectively solves the problems of large computational load, long cycle, and reliance on experimental data in traditional methods, but also accurately captures the internal flow mechanism and quickly evaluates the sensor's pressure loss, flow characteristics, and dynamic response performance. Through systematic analysis of the calculation results, a reliable basis can be provided for structural improvement, performance enhancement, and solution iteration, significantly improving the R&D efficiency of turbine flow meters, reducing design costs, and providing solid numerical support for their engineering applications and performance optimization.

[0071] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A design method for turbine flowmeters based on computational fluid dynamics, characterized in that, include: Step 1: Conduct preliminary structural design of the turbine flow meter based on performance requirements; Step 2: Establish a fluid domain model of the turbine flow meter for fluid analysis based on the preliminary structure of the turbine flow meter; Step 3: Perform preprocessing of the turbine flowmeter fluid domain model for computational fluid dynamics analysis to obtain the mesh model of the turbine flowmeter; Step 4: Perform numerical simulation calculations on the mesh model of the turbine flow meter based on computational fluid dynamics; Step 5: Extract the characteristic curve of the turbine flow meter from the numerical calculation results and evaluate the performance index of the turbine flow meter. If the performance index is not met, combine the internal flow characteristics of the turbine flow meter in the calculation results and feed them back to Step 1 to guide the structural optimization and adjustment of the turbine flow meter.

2. The turbine flowmeter design method based on computational fluid dynamics according to claim 1, characterized in that, Step 1 includes the following steps: Step 1.1: Based on the pipe diameter, pressure loss, and flow measurement range requirements, determine the basic dimensions of each component of the turbine flow meter, as well as the hub diameter, blade tip diameter, hub thickness, number of blades, and blade thickness parameters of the turbine in the turbine flow meter. Step 1.2: Determine the turbine flow meter housing size, interface type, and internal flow channel diameter parameters based on the requirements of pipe diameter, pressure rating, and interface type. Step 1.3: Select the material properties of each component of the turbine flow meter based on the material properties of the measured medium, the temperature and pressure of the medium, and the weight parameters of the flow meter.

3. The turbine flow meter design method based on computational fluid dynamics according to claim 2, characterized in that, Step 2 includes the following steps: Step 2.1: Import the preliminary structural drawing of the turbine flow meter into the 3D CAD software. Based on the internal flow channel dimensions of the turbine flow meter housing, initially establish the turbine flow meter fluid domain, including the dimensions of each component of the turbine flow meter. Step 2.2: In the 3D CAD software, remove the space occupied by solid components from the initially established turbine flowmeter fluid domain to obtain the turbine flowmeter fluid domain used in computational fluid dynamics calculations.

4. The turbine flow meter design method based on computational fluid dynamics according to claim 3, characterized in that, Step 3 includes the following steps: Step 3.1: Import the obtained turbine flowmeter fluid domain into the mesh generation software. Based on the flow field calculation requirements and the complexity of the fluid domain, mesh generation is performed using a combination of structured and unstructured meshes. Step 3.2: Refine the mesh in areas with large curvature on the model surface, and appropriately enlarge the mesh size in areas with gentle curvature on the model surface, balancing computational accuracy and efficiency. Step 3.3: After the mesh is generated, check the mesh quality to ensure that it meets the requirements of CFD numerical calculation, and finally obtain a mesh model that can be directly used for subsequent numerical simulation.

5. The turbine flow meter design method based on computational fluid dynamics according to claim 4, characterized in that, Step 4 includes the following steps: Step 4.1: Set the basic conditions for numerical simulation calculation; Step 4.2: Select the turbulence model for numerical simulation calculation; Step 4.3: Set the boundary conditions for numerical calculation; Step 4.4: After completing the model selection and relevant boundary condition settings for the numerical simulation, initialize the flow field to provide a reasonable initial flow field distribution for iterative calculation; after initialization, perform steady-state numerical simulation calculations and use the steady-state flow field calculation results as the initial flow field for subsequent transient calculations, and obtain a fully developed and convergent flow field distribution through steady-state calculations. Step 4.5: Perform transient calculations on the turbine flow meter based on UDF to obtain the rotor motion characteristics.

6. The turbine flow meter design method based on computational fluid dynamics according to claim 5, characterized in that, Step 4.1 includes the following steps: In numerical calculations, the liquid medium is approximated as an incompressible fluid. The minute heat exchange during the flow process is ignored, and the physical properties of the corresponding fluid medium are directly given. The basic governing equations of the fluid medium are as follows: (4.1) (4.2) In the formula , Coordinate system components; , These are the velocity components in the coordinate system; It is time; It is the density of the liquid; For pressure; The kinematic viscosity of the fluid.

7. The turbine flowmeter design method based on computational fluid dynamics according to claim 6, characterized in that, Step 4.2 includes the following steps: Considering the high-speed rotating turbine, components with significant variations in turbine blade surface curvature, and complex flows near the blade wall, the numerical calculation solution mode is set to double-precision solution, and the turbulence model is selected from viscous fluid dynamics. The turbulence model has the following transport equation: Turbulent kinetic energy k-transport equation: (4.3) Turbulent frequency equation: (4.4) Turbulent eddy viscosity coefficient μ t : (4.5) Turbulent kinetic energy generation term P k : (4.6) The modulus S of the dependent variable tensor: (4.7) (4.8) In the formula: Let be the fluid density; U be the fluid velocity vector; μ be the fluid dynamic viscosity; F1 and F2 be mixing functions used to realize the near-wall region. Model and far-field region The model's smooth transition is achieved by using F1 to control the switching between the near-wall and mainstream regions, and F2 to assist in correcting the characteristics of the far-field region. , These are empirical constants corresponding to the effective diffusion coefficients of turbulent kinetic energy and turbulent eddy frequency, respectively, and their values ​​are obtained by weighting the mixing function F1, i.e. , ; Within the discrete framework of the finite volume method, the fundamental governing equations of the fluid and the aforementioned equations are solved iteratively. The transport equations corresponding to the turbulence model ultimately yield stable numerical solutions for key physical quantities at each grid node and element within the flow field. These key physical quantities include at least velocity, pressure, and temperature.

8. The turbine flow meter design method based on computational fluid dynamics according to claim 7, characterized in that, Step 4.3 includes the following steps: In the numerical calculation of the turbine flow meter, the energy equation is ignored, and only the flow control equation is considered. The walls of the computational domain are all set as adiabatic, no-slip wall boundary conditions; the outlet boundary uses a free outflow boundary condition; and the inlet uses a velocity inlet boundary condition. The inlet velocity is determined based on the given flow conditions, and the calculation formula is shown in the following equation: (4.9) In the formula: v is the flow velocity at the inlet; Q is the volumetric flow rate at the inlet; and r is the radius at the inlet.

9. The turbine flow meter design method based on computational fluid dynamics according to claim 8, characterized in that, Step 4.5 includes the following steps: (a) Read the numerical calculation results of the steady-state flow field; its turbulence model is: Turbulent model, with boundary conditions as given in step 4.3; (b) Switch the calculation mode from steady state to transient state and configure the relevant settings for transient calculation; (c) Load the written User Defined Function program, set the flow domain corresponding to the turbine rotor as the rotation domain, set its motion mode as mesh motion, and set the rotation motion mode as the loaded UDF program; set the transient time step to the time taken for the turbine flow meter rotor to rotate 2°, and ensure that the rotor rotates at least 30 times in the total calculation time; after completing the above settings, perform transient calculation of the turbine flow meter. The transient numerical simulation calculation is complete. The corresponding calculation results are saved for subsequent calculation and analysis of the turbine flow meter. The principle behind the User Defined Function program is as follows: When fluid flows through the rotor, the impact of the fluid on the rotor generates a driving torque Td. When the driving torque exceeds the sum of all deceleration torques, the rotor begins to rotate. During the acceleration and deceleration of the rotor, the torque balance equation of the rotor is as follows: (4-10) In the formula: ω is the angular velocity of the rotor, rad / s; J is the moment of inertia of the rotor, kg·m. 2 ;T r T is the viscous deceleration torque on the blade surface. t T is the deceleration torque of the blade tip clearance; hh T is the viscous frictional deceleration torque at the rotor hub end face; hp The viscous frictional deceleration torque on the outer circumferential surface of the rotor hub; T m For magnetic deceleration torque; T b For bearing deceleration torque; When the rotor speed is stable, the rotor's angular acceleration is zero. At this time, the rotor's torque balance equation is as follows: (4-11) In the formula: Various deceleration torques are represented in equation (4-10). In establishing the flow domain model of the turbine flow meter, the resistance of the bearings to the flow sensor rotor and the magnetic reluctance generated by the speed sensor on the turbine flow meter's motion are ignored. Therefore, the resistance of the turbine flow meter rotor in numerical calculations is mainly generated by fluid viscosity. The torque equation of the turbine flow meter rotor is shown below: (4-12) In summary, the turbine rotor speed of the turbine flow meter under different incoming flow conditions can be calculated using the following formula: (4-13) In the formula: The rotor speed is calculated for the nth time. This refers to the rotor speed calculated in the (n-1)th time. Ultimately, the output turbine flow meter speed It is calculated by the following formula: (4-14) In the formula: The rotor speed is calculated for the nth time. This is the rotor speed calculated for the (n-1)th time.

10. The turbine flowmeter design method based on computational fluid dynamics according to claim 1, characterized in that, Step 5 includes the following steps: Step 5.1: After the numerical calculation converges and reaches a steady state, post-processing and result analysis are carried out on the internal flow field and rotor motion characteristics of the turbine flow meter; key performance parameters of the rotor are monitored, including at least the fluid torque, rotational speed and inlet / outlet pressure difference; the rotor motion characteristics and flow response law under different incoming flow conditions are quantitatively analyzed to determine whether the sensor characteristic curve meets the design requirements. Step 5.2: If the calculation results show that the sensor performance does not meet the relevant index requirements, the key physical quantities in the computational domain can be extracted with the help of CFD post-processing software, including at least the velocity field, pressure field, turbulent kinetic energy and vorticity. The internal flow state can be intuitively displayed through cloud maps, vector maps and streamline diagrams to identify typical features, including at least flow separation, vortex structure and local energy loss, to clarify the performance shortcomings and provide a clear direction for subsequent structural improvement and optimization design.

Images