A Simulation Method for Predicting Resin Stirring Time in HP-RTM Process

Abstract

This invention discloses a simulation method for predicting resin stirring time in the HP-RTM process, comprising the following steps: constructing a physical model of a resin mixing tank, and dividing the fluid domain in the physical model into a rotational domain and a stationary domain according to the structure and size parameters of the agitator set in the mixing tank; performing steady-state simulation of the resin stirring process in the physical model using a VOF model, and determining that the flow field of resin stirring has reached stability; adding a tracer to the bottom of the stable flow field of resin stirring to replace resin precipitation by marking unit regions, and setting its initial position and concentration, and then performing transient simulation using the SIMPLE algorithm; setting multiple monitoring points in the flow field of the transient simulation, and obtaining the time for uniform mixing of resin precipitation when the concentration of resin precipitation replaced by tracer shown by multiple monitoring points meets the 95% mixing principle.

Description

Technical Field

[0001] This invention relates to the field of high-pressure resin transfer molding technology, specifically to a method for predicting and simulating resin stirring time in the HP-RTM process. Background Technology

[0002] High-pressure resin transfer molding (HP-RTM) is a derivative of traditional resin transfer molding (RTM). HP-RTM equipment generally operates in a cyclical manner, with the mixing tank constantly running. However, uncontrollable factors such as power outages can cause the equipment to stop working. If the resin is left at room temperature for too long, sedimentation and impurity aggregation can occur. The resin mixing tank, as part of the HP-RTM equipment, functions to agitate the resin, ensuring that impurities and sediments are mixed evenly.

[0003] However, there is no specific parameter standard for how long it takes to achieve a uniform mixing state after the equipment is started. The time is controlled based on work experience. The uncertainty of the mixing state can easily lead to the resin not entering the mold in a relatively ideal uniform mixing state or the resin being over-mixed. Both of these can easily lead to deviations in product performance. Summary of the Invention

[0004] The purpose of this invention is to provide a simulation method for predicting resin stirring time in the HP-RTM process, so as to solve the technical problems caused in the actual production process due to the lack of precise time control of resin sedimentation and stirring uniformity in the shutdown state in the existing high-pressure resin transfer molding process.

[0005] To solve the above-mentioned technical problems, the present invention specifically provides the following technical solution:

[0006] A simulation method for predicting resin stirring time in the HP-RTM process includes the following steps:

[0007] Step 100: Construct a physical model of the resin mixing tank, and divide the fluid domain in the physical model into a rotational domain and a stationary domain according to the structure and size parameters of the agitator set in the mixing tank.

[0008] Step 200: Use the VOF model to perform a steady-state simulation of the resin stirring process in the physical model, and determine that the flow field of the resin stirring has reached a stable state.

[0009] Step 300: Add a tracer to the bottom of the stable flow field of resin stirring to replace resin precipitation by marking the unit region, and set its initial position and concentration, and then use the SIMPLE algorithm to perform transient simulation.

[0010] Step 400: Set up multiple monitoring points in the transient simulation flow field. When the concentration of resin precipitate replaced by tracer shown by multiple monitoring points meets the 95% mixing principle, obtain the time when the resin precipitate is uniformly mixed.

[0011] As a preferred embodiment of the present invention, the method for dividing the fluid domain in the physical model into a rotational domain and a stationary domain based on the structure and dimensional parameters of the agitator installed in the mixing tank includes:

[0012] By performing Boolean operations on the physical model, the interface between the stirrer and the space inside the mixing tank is extracted, and the remaining area obtained by suppressing the solid model of the stirrer in the physical model is the fluid domain.

[0013] The rotational domain is determined based on the grid division of the agitators in the same layer, and the rest of the fluid domain is set as the stationary domain.

[0014] As a preferred embodiment of the present invention, in the process of suppressing the remaining region obtained by the stirrer entity model in the physical model to be a fluid domain, the top surface of the fluid domain is defined as a free liquid surface.

[0015] As a preferred embodiment of the present invention, the stirrer includes a rotating shaft and stirring paddles, and a plurality of stirring paddles are evenly distributed along the same circumference of the rotating shaft to form a stirring body, and a plurality of stirring bodies are provided on the rotating shaft.

[0016] Before the mesh division of the stirring impeller of the agitator determines the rotation domain, the number of distributions of the stirring impeller of the agitator is increased, and a boundary layer is inserted between the stirring impeller and the fluid domain.

[0017] As a preferred embodiment of the present invention, the specific method for achieving stability by steady-state simulation and determining the flow field of resin stirring includes:

[0018] By setting monitoring points in the flow field and determining that the velocity at the monitoring points no longer increases with the number of iterations;

[0019] Obtain the vertical velocity distribution cloud map of the model to obtain the maximum stirring velocity of the flow field; calculate the blade tip velocity using the formula, and make it the same as the maximum stirring velocity of the flow field.

[0020] As a preferred embodiment of the present invention, the structure and dimensional parameters of the stirrer include, but are not limited to: the distance between two adjacent stirring bodies, the diameter of the stirring body, the diameter of the rotational domain formed by the stirring bodies, and the distance from the bottommost stirring body to the bottom of the mixing tank.

[0021] As a preferred embodiment of the present invention, the stirring and mixing time data corresponding to different parameter values ​​of each experimental parameter are analyzed by controlling a single variable.

[0022] Select at least three sets of orthogonal experiments from each experimental parameter to determine the optimal combination of experimental parameters.

[0023] In a preferred embodiment of the present invention, the diameter of the rotating domain is set to be the diameter of the stirring body plus 10 mm.

[0024] As a preferred embodiment of the present invention, fluid inflow boundary conditions and fluid outflow boundary conditions are constructed for the rotating domain, and the stationary domain and the rotating domain exchange data through the fluid inflow boundary conditions and fluid outflow boundary conditions.

[0025] Compared with the prior art, the present invention has the following advantages:

[0026] This invention utilizes a combination of steady-state and transient simulations of resin mixing processes, replacing resin precipitation with tracers, to simulate the state of the mixing tank and agitator during the mixing process. The study reveals the relationship between relevant parameters of the mixing tank and the mixing time, effectively reducing the problem of operators mixing too little time and realizing the transformation from "mixing experience" to "data-driven". Attached Figure Description

[0027] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely exemplary, and those skilled in the art can derive other embodiments based on the provided drawings without creative effort.

[0028] Figure 1 This is a schematic diagram of the physical model structure of the mixing tank according to an embodiment of the present invention;

[0029] Figure 2 This is a schematic diagram of the structure for dividing the fluid domain in the physical model of an embodiment of the present invention;

[0030] Figure 3 This is a schematic diagram of the vertical velocity distribution of the flow field simulated by the physical model in an embodiment of the present invention.

[0031] Figure 4 This is a graph showing the change in the number of velocity iterations at the monitoring points according to an embodiment of the present invention.

[0032] Figure 5 This is a schematic diagram illustrating the construction of the tracer deposition phase according to an embodiment of the present invention;

[0033] Figure 6 This is a schematic diagram of the overall process of an embodiment of the present invention. Detailed Implementation

[0034] 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, and 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.

[0035] Currently, there are many numerical simulation software programs available, such as Ansys Fluent, COMSOL Multiphysics, OpenFOAM, CFX, and SIMULIA Abaqus, all of which can perform numerical simulations of agitators. Among them, Ansys Fluent is widely used in academia and industry, boasts a rich user community and documentation support, and offers various turbulence models to adapt to different flow characteristics. Its multiphase flow model is suitable for handling reactions during mixing, and its post-processing capabilities are also quite powerful. This patent uses Ansys Fluent as an example to illustrate the mixing simulation method.

[0036] The simulation theory of the stirrer in this embodiment mainly involves the following aspects:

[0037] 1. Use the Navier-Stokes equations to describe fluid flow, considering the nonlinearity, viscosity, and incompressibility of the flow.

[0038] 2. Since the stirring process is usually turbulent, the k-ε or k-ω model is used to simulate the energy transfer and eddy characteristics of turbulence. The k-ε model describes turbulence by solving for turbulent kinetic energy (k) and turbulent kinetic energy dissipation rate (ε), and is suitable for most industrial flow conditions. This model is relatively simple to calculate and has high stability, but its predictive ability for low Reynolds number flows and strongly rotating flows is poor, and its simulation of turbulent behavior near the wall is not accurate enough. The k-ω model describes turbulence by solving for turbulent kinetic energy (k) and turbulent kinetic energy frequency (ω), and is suitable for low Reynolds number flows and boundary layer flows. This model simulates flow characteristics near the wall well and is suitable for complex boundary layer flows. It has strong predictive ability for strongly rotating flows, but its numerical value is unstable under certain flow conditions. This patent uses the more stable k-ε model for solution.

[0039] 3. The VOF (Volume of Fluid) model is suitable for simulating multiphase flow, especially when the fluid interfaces are significant and immiscible, such as the separation process of precipitation and liquid. This model can effectively capture the flow characteristics of two phases and is suitable for use in complex multiphase flow scenarios. This patent aims to examine the distribution of precipitation in resin, therefore assuming that the interface between the two phases is significant and immiscible, and thus the VOF model can be used for multiphase flow simulation.

[0040] 4. Precisely set the boundary conditions of the reactor, such as inlet and outlet flow rates and agitator speed, to ensure the realism and accuracy of the simulation. The model involved in this patent has no inlet or outlet; therefore, only the agitator speed and the rotational speed need to be set, which will be explained in detail in the flowchart below.

[0041] 5. Regardless of whether a pressure-based or density-based solver is used, fluid flow requires solving the momentum equation, continuity equation, and turbulence equation, followed by solving the energy equation and composition equation based on the operational settings. The difference lies in the solution method: the pressure-based separate solver solves the momentum equations in the U, V, and W directions separately, and then solves the continuity equation. This leads to the need to ignore other conditions when calculating some equations, resulting in a faster solution speed. The pressure-based coupled solver solves the momentum and continuity equations in one set of equations, and then solves the energy and composition equations separately. The density-based coupled solver solves all equations together, and finally solves the turbulence equation separately, achieving the highest accuracy, but it is less likely to converge and requires a longer time. Regardless of the solver used, the mass conservation equation and momentum conservation equation must be satisfied. This patent uses a pressure-based solver for this purpose.

[0042] mass conservation equation:

[0043]

[0044] In the formula: u, v, w — velocity vectors along the x, y, z directions.

[0045] Momentum conservation equation:

[0046]

[0047] In the formula: f i —The component of the volumetric force per unit mass in the i-direction; τ mn —The projection component of viscous stress in the n direction onto a plane perpendicular to the m axis.

[0048] like Figures 1 to 6 As shown, this invention provides a simulation method for predicting resin stirring time in the HP-RTM process, comprising the following steps:

[0049] Step 100: Construct a physical model of the resin mixing tank, and divide the fluid domain in the physical model into a rotational domain and a stationary domain according to the structure and size parameters of the agitator set in the mixing tank.

[0050] Step 200: Use the VOF model to perform a steady-state simulation of the resin stirring process in the physical model, and determine that the flow field of the resin stirring has reached a stable state.

[0051] Step 300: Add a tracer to the bottom of the stable flow field of resin stirring to replace resin precipitation by marking the unit region, and set its initial position and concentration, and then use the SIMPLE algorithm to perform transient simulation.

[0052] Step 400: Set up multiple monitoring points in the transient simulation flow field. When the concentration of resin precipitate replaced by tracer shown by multiple monitoring points meets the 95% mixing principle, obtain the time when the resin precipitate is uniformly mixed.

[0053] Methods for dividing the fluid domain in a physical model into rotating and stationary domains based on the structure and dimensional parameters of the agitator installed in the mixing tank include:

[0054] By performing Boolean operations on the physical model, the interface between the stirrer and the space inside the mixing tank is extracted, and the remaining area obtained by suppressing the solid model of the stirrer in the physical model is the fluid domain.

[0055] The rotational domain is determined based on the grid division of the agitators in the same layer, and the rest of the fluid domain is set as the stationary domain.

[0056] In the process of suppressing the remaining region obtained by the stirrer entity model in the physical model as a fluid domain, the top surface of the fluid domain is defined as the free liquid surface.

[0057] The stirrer includes a rotating shaft and stirring paddles. Multiple stirring paddles are evenly distributed along the same circumference of the rotating shaft to form a stirring body. Multiple stirring bodies are provided on the rotating shaft.

[0058] Before the mesh division of the stirring impeller of the agitator determines the rotation domain, the number of distributions of the stirring impeller of the agitator is increased, and a boundary layer is inserted between the stirring impeller and the fluid domain.

[0059] Specific methods for achieving stability by simulating and determining the flow field during resin stirring include:

[0060] By setting monitoring points in the flow field and determining that the velocity at the monitoring points no longer increases with the number of iterations;

[0061] Obtain the vertical velocity distribution cloud map of the model to obtain the maximum stirring velocity of the flow field; calculate the blade tip velocity using the formula, and make it the same as the maximum stirring velocity of the flow field.

[0062] The structure and dimensional parameters of the agitator include, but are not limited to: the distance between two adjacent agitator bodies, the diameter of the rotating shaft, the diameter of the rotating domain formed by the agitator bodies, and the distance from the bottom agitator body to the bottom of the mixing tank.

[0063] The diameter of the rotating zone is set to the diameter of the stirring body plus 10 mm.

[0064] The fluid inflow boundary conditions and fluid outflow boundary conditions of the rotating domain are constructed, and the stationary domain and the rotating domain exchange data through the fluid inflow boundary conditions and fluid outflow boundary conditions.

[0065] To more clearly illustrate the above implementation methods, this implementation method provides a specific example:

[0066] The agitator is selected based on the viscosity range and flow state of the medium. Taking a certain Huntsman fast-curing resin as an example, its viscosity at room temperature is 6.5-9.0 Pas; the flow state is divided into axial flow, radial flow, and mixed flow. Considering the size of this mixing tank, a mixed flow agitator is selected, typically a six-blade open turbine agitator.

[0067] Based on existing equipment, determine the structure of the mixing tank as follows: Figure 1 As shown, the tank has an inner diameter of 400 mm, an outer diameter of 460 mm, a height of 900 mm, a liquid height of 600 mm, a blade angle of 45°, a distance of 1 h from the bottom of the lower agitator to the bottom of the tank, a distance of 2 h between the two agitators, an agitator diameter of 1 d, and a rotation zone diameter of 2 d (Note: The agitator diameter and the rotation zone diameter should be equal, but considering the inconvenience of mesh division, the rotation zone diameter is increased by 10 mm). The density of the resin in the tank is 11503 g / kgm, the viscosity is 7 Pas, and the agitator rotates at a constant speed of 80 rpm around the Y-axis.

[0068] Regardless of whether a pressure-based or density-based solver is used, fluid flow requires solving the momentum equation, continuity equation, and turbulence equation, followed by solving the energy equation and composition equation depending on the operational settings. The difference lies in the solution method: pressure-based separate solvers solve the momentum equations in the UVW direction separately, and then solve the continuity equations. This leads to the need to ignore other conditions when calculating some equations, resulting in faster solution speeds. Pressure-based coupled solvers solve the momentum and continuity equations in a single system, and then solve the energy and composition equations separately. Density-based coupled solvers solve all equations together, and finally solve the turbulence equations separately, offering the highest accuracy but less convergence and longer solution time. Regardless of the solver used, both mass and momentum conservation equations must be satisfied.

[0069] For dividing the fluid domain into physical models:

[0070] The purpose of this numerical simulation is to examine the mixing uniformity of the stirrer. Since there is no calculation of the subsequent fluid's impact on the structural stress, Boolean operations were used to extract the interfaces. The solid model of the stirrer was then suppressed, and the rest was considered a fluid domain. The fluid domain was divided into a rotating domain and a stationary domain. An interface was created between interface1-out and interface1-in, and another between interface2-out and interface2-in. Data exchange between the rotating and stationary domains occurred through their internal interfaces. Figure 2 As shown.

[0071] Mesh generation: Before mesh generation, the size is adjusted, the blades in the two moving domains are densified, a boundary layer is inserted between the blades and the moving domain, and tetrahedral meshes are used for generation.

[0072] Boundary condition settings:

[0073] The multiphase flow model uses the VOF model; the viscous model uses the K-epsilon turbulence model; the walls are treated using standard wall functions; the Coupled algorithm is used in steady state, and the SIMPLE algorithm is used by default in transient state. The model calculation uses the multiple reference frame method to rotate the rotating domain. When setting the rotation direction, it is necessary to determine the rotation direction based on the blades, and the right-hand rule is used to specify whether the rotational speed is positive or negative. The top surface of the fluid domain is a free surface, set as a symmetric boundary condition (Note: Under this model condition, numerical simulations show that the stabilized liquid does not generate vortices, indicating minimal disturbance). The rotating axis wall is a moving wall; relative to the adjacent stationary region, the rotating axis is moving, with a rotational speed of -80 / minr. The blade walls are also moving walls; relative to the adjacent rotating region, the blades are relatively stationary. The remaining walls are stationary walls.

[0074] Research methods for mixing processes:

[0075] Using a VOF model, a steady-state simulation of resin stirring was first performed. After the flow field stabilized, component transport was initiated, precipitation areas were marked, tracers were added, and the component transport equation was calculated. Six equidistant monitoring points were set to monitor the tracer concentration, and a transient simulation was conducted. The convergence of the steady-state simulation was determined by the velocity field and the flow velocity at a specific point. When the flow velocity fluctuated little over time and remained almost a straight line, steady-state convergence was considered achieved, and then a transient tracer simulation was performed. The time required for homogenization is a crucial criterion for evaluating stirring uniformity, expressed as the 95% mixing time. The 95% mixing time refers to the time required for the modifier to reach a stable concentration within ±5% of the error after being added to the mixing tank.

[0076] Detailed simulation prediction and analysis process:

[0077] Taking the following model dimensions as an example, the velocity field and convergence are analyzed; the analysis method for other models is the same. Model dimensions: the stirrer diameter 1d is 200mm, the rotation domain diameter 2d is 210mm, 1h is 150mm, 2h is 300mm, and other dimensions remain unchanged. A monitoring point A (0, 500, 150) is selected near the upper surface. The velocity distribution cloud map of the plumb surface of this model is shown below. Figure 3 As shown, the velocity at monitoring point A changes with the number of iterations as follows: Figure 4 As shown;

[0078] Depend on Figure 3 The maximum stirring speed is 0.879 m / s, which occurs at the tip of the impeller. The tip speed can be calculated using the formula v = πnd, where n is the rotational speed in m / s and d is the diameter of the rotational domain in meters.

[0079] Substituting the model dimensions and parameters into the formula, the blade tip is calculated; Figure 4 It can be seen that the velocity at monitoring point A remains essentially unchanged with increasing iteration count. Based on these two points, the steady-state result calculation is considered accurate.

[0080] Initial resin position calibration and monitoring:

[0081] After the steady-state calculation is complete, switch to the transient calculation. On this basis, enable component delivery and add a tracer to ensure it mixes only with the resin. Here, the tracer is used instead of resin precipitation, therefore its viscosity and density are the same as the resin. Define the initial position and concentration of the tracer using cell region marking. Based on experience, precipitates generally accumulate at the bottom of the stirred tank. Assuming the precipitate height at the bottom of the stirred tank is 100mm and its mass-volume fraction is 1, use the patch function to add the precipitate to the bottom of the stirred tank, such as... Figure 5 As shown, the red area represents resin precipitation, and the blue area represents resin.

[0082] In this embodiment, simulation can also be performed by controlling single-factor variables, wherein the single factors include, but are not limited to: the distance between two adjacent stirring bodies, the diameter of the rotating shaft, the diameter of the rotating domain formed by the stirring bodies, and the distance from the bottom stirring body to the bottom of the mixing tank.

[0083] The above embodiments are merely exemplary embodiments of this application and are not intended to limit this application. The scope of protection of this application is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to this application within its substance and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of this application.

Claims

1. A simulation method for predicting resin stirring time in HP-RTM process, characterized in that, Includes the following steps: Step 100: Construct a physical model of the resin mixing tank, and divide the fluid domain in the physical model into a rotational domain and a stationary domain according to the structure and size parameters of the agitator set in the mixing tank. Step 200: Use the VOF model to perform a steady-state simulation of the resin stirring process in the physical model, and determine that the flow field of the resin stirring has reached a stable state. Step 300: Add a tracer to the bottom of the stable flow field of resin stirring to replace resin precipitation by marking the unit region, and set its initial position and concentration, and then use the SIMPLE algorithm to perform transient simulation. Step 400: Set up multiple monitoring points in the transient simulation flow field. When the concentration of resin precipitate replaced by tracer shown by multiple monitoring points meets the 95% mixing principle, obtain the time when the resin precipitate is uniformly mixed.

2. The method for predicting and simulating resin stirring time in HP-RTM process according to claim 1, characterized in that, Methods for dividing the fluid domain in a physical model into rotating and stationary domains based on the structure and dimensional parameters of the agitator installed in the mixing tank include: By performing Boolean operations on the physical model, the interface between the stirrer and the space inside the mixing tank is extracted, and the remaining area obtained by suppressing the solid model of the stirrer in the physical model is the fluid domain. The rotational domain is determined based on the grid division of the agitators in the same layer, and the rest of the fluid domain is set as the stationary domain.

3. The method for predicting and simulating resin stirring time in HP-RTM process according to claim 2, characterized in that, In the process of suppressing the remaining region obtained by the stirrer entity model in the physical model as a fluid domain, the top surface of the fluid domain is defined as the free liquid surface.

4. The method for predicting and simulating the resin stirring time in the HP-RTM process according to claim 3, characterized in that, The stirrer includes a rotating shaft and stirring paddles. Multiple stirring paddles are evenly distributed along the same circumference of the rotating shaft to form a stirring body. Multiple stirring bodies are provided on the rotating shaft. Before the mesh division of the stirring impeller of the agitator determines the rotation domain, the number of distributions of the stirring impeller of the agitator is increased, and a boundary layer is inserted between the stirring impeller and the fluid domain.

5. The method for predicting and simulating the resin stirring time in the HP-RTM process according to claim 4, characterized in that, Specific methods for achieving stability by simulating and determining the flow field during resin stirring include: By setting monitoring points in the flow field and determining that the velocity at the monitoring points no longer increases with the number of iterations; Obtain the vertical velocity distribution cloud map of the model to obtain the maximum stirring velocity of the flow field; calculate the blade tip velocity using the formula, and make it the same as the maximum stirring velocity of the flow field.

6. The method for predicting and simulating the resin stirring time in the HP-RTM process according to claim 5, characterized in that, Experimental parameters are determined by the structural dimensions of the stirrer and the relative structural dimensions between the stirrer and the mixing tank. These experimental parameters include, but are not limited to: the distance between two adjacent stirring bodies, the diameter of the stirring body, the diameter of the rotational domain formed by the stirring bodies, and the distance from the bottom stirring body to the bottom of the mixing tank.

7. The method for predicting and simulating the resin stirring time in the HP-RTM process according to claim 6, characterized in that, The data on the mixing time for each experimental parameter under different parameter values ​​were analyzed by controlling for a single variable. Select at least three sets of orthogonal experiments from each experimental parameter to determine the optimal combination of experimental parameters.

8. The method for predicting and simulating resin stirring time in HP-RTM process according to claim 6, characterized in that, The diameter of the rotating domain is set to the diameter of the stirring body plus 10 mm.

9. The method for predicting and simulating resin stirring time in HP-RTM process according to claim 6, characterized in that, The fluid inflow boundary conditions and fluid outflow boundary conditions of the rotating domain are constructed, and the stationary domain and the rotating domain exchange data through the fluid inflow boundary conditions and fluid outflow boundary conditions.

Images