Method for assisting filter material design based on filtering simulation technology

Abstract

The invention discloses a method for assisting filter material design based on a filtering simulation technology. The method comprises the following steps: determining a filtering performance target value and setting corresponding filter material design parameters; constructing a three-dimensional geometric model reflecting random arrangement of fibers according to the design parameters; carrying out flow field-particle coupling simulation on the model to obtain simulation predicted values of the filtering efficiency and the pressure drop; comparing the simulation result with a target value, and if not, adjusting the design parameters and iteratively simulating until the requirements are met; and finally, according to the optimized design scheme, carrying out physical sample trial-manufacture and verification. According to the method, performance verification and structure optimization are preposed to a computer simulation stage by constructing a design closed loop with priority of a virtual prototype, so that the trial-manufacture frequency of a physical sample is greatly reduced, the research and development cost is remarkably reduced, the development period is shortened, and the design scientificity and the product performance are improved.

Description

Technical Field

[0001] This invention relates to the field of dust removal filter media technology, and specifically to a method for assisting filter media design based on filtration simulation technology. Background Technology

[0002] In the design and development of nonwoven filter media (such as industrial dust collector bags and air filtration materials), traditional R&D models heavily rely on a physical iterative cycle of "design-prototype-testing." Specifically, engineers first determine key design parameters such as fiber material, fineness, mixing ratio, and needle punching density and depth based on working conditions and requirements and experience. Then, small-batch sample production is carried out on the production line. Finally, the physical samples are tested for performance such as filtration efficiency and pressure drop. If the test results do not meet the target requirements, the design parameters need to be adjusted, and the above prototyping and testing process needs to be repeated until a qualified product is obtained.

[0003] This R&D model, which relies on physical samples, has significant drawbacks: First, each sample production requires a large amount of raw materials, energy, and production line time, resulting in high economic costs for each iteration; second, the entire cycle, from parameter adjustment and production scheduling to performance testing, is usually measured in weeks or even months, severely slowing down the speed at which new products respond to market demands; and finally, because the performance of the design cannot be predicted before manufacturing, the R&D process is highly unpredictable and involves a great deal of trial and error, hindering the rapid exploration of products towards better performance. Summary of the Invention

[0004] In view of the technical problems existing in the prior art, the purpose of this invention is to provide a method for assisting filter media design based on filtration simulation technology, so as to save the cost of filter media design.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows: A method for assisting filter media design based on filtration simulation technology includes the following steps: Step 1: Determine the target value M for filtration performance, and set at least one set of filter media design parameters S based on the target value M. The set of design parameters S includes at least fiber parameters, filter media parameters, particle parameters, and flow field parameters. Step 2: Based on the design parameter set S, construct a three-dimensional geometric model that reflects the random arrangement of fibers inside the filter media; Step 3: Perform flow field-particle coupled simulation on the three-dimensional geometric model to obtain the simulation prediction value F, including filtration efficiency and pressure drop; The flow field and pressure field are obtained by performing flow field calculations on the three-dimensional geometric model in computational fluid dynamics software. In the discrete element method software, particle properties and contact models are set to simulate the motion of particles in the velocity field. Through a bidirectional coupling interface, flow field data and particle data are exchanged to realize the simulation prediction value F of the interaction between the flow field and particle motion. Step 4: Compare the simulated predicted value F with the target value M to determine whether F meets the requirements of M. If satisfied, the current design parameter set S is locked as the recommended design scheme, and step S5 is executed or the design parameter set S is modified and adjusted, and then step S2 is returned. Based on the adjusted design parameter set S, the three-dimensional geometric model is reconstructed and re-simulated and optimized. If not satisfied, adjust the design parameter set S and return to step S2, reconstruct the three-dimensional geometric model based on the adjusted design parameter set S and resimulate; Step 5: Select at least one of the recommended design schemes to conduct physical sample production and performance verification.

[0006] In step 1, The fiber parameters include: material, diameter and length, curvature, Young's modulus, Poisson's ratio, and density; The filter media parameters include hydroentangled / needle-punched parameters: needle-punching density and depth, areal density, and thickness. The particle parameters include: particle diameter, dust content, density, Young's modulus, Poisson's ratio, coefficient of restitution between particles and between particles and fibers, static friction coefficient between particles and between particles and fibers, rolling friction coefficient between particles and between particles and fibers, and surface energy parameters between particles and between particles and fibers.

[0007] In step S2, the construction of the three-dimensional geometric model is specifically as follows: a single fiber model is established based on the fiber diameter, length, and curvature. Then, based on the size and morphological parameters of the single fiber, it is randomly arranged within the set simulation domain, and a three-dimensional geometric model representing the microstructure of the filter material is generated according to the areal density and thickness constraints of the filter material.

[0008] The design parameter set S also includes coating parameters: film material, pore size / pore size distribution, thickness, and basis weight.

[0009] In step 2, when the design parameter set S includes membrane parameters, constructing the three-dimensional geometric model further includes: randomly generating a mesh structure within a set simulation domain according to the membrane parameters to obtain the geometric model of the membrane, and combining the geometric model of the membrane with the dust-facing surface of the filter media to generate a three-dimensional geometric model of the membrane filter media.

[0010] In step S1, multiple sets of different initial design parameters S are set to form multiple candidate design schemes; in steps S3 and S4, the multiple candidate design schemes are simulated and evaluated in parallel or sequentially to select one or more recommended design schemes that meet the target value M.

[0011] In step 3, the flow field-particle coupling simulation includes: The flow field of the three-dimensional geometric model is calculated in computational fluid dynamics software to obtain the velocity field and pressure field; particle properties and contact model are set in discrete element method software to simulate the motion of particles in the velocity field; flow field data and particle data are exchanged through a two-way coupling interface to realize the interaction calculation between the flow field and particle motion, so as to obtain the simulation prediction value F.

[0012] In step 1, multiple sets of different initial design parameters S are set to form multiple candidate design schemes; in steps 3 and 4, the multiple candidate design schemes are simulated and evaluated in parallel or sequentially to select one or more recommended design schemes that meet the target value M.

[0013] In step 4, if the simulation prediction value F satisfies the target value M, before or after locking in the recommended design scheme, the following steps are also included: based on the analysis of the simulation results, the design parameter set S is adjusted in a targeted manner to generate a new parameter set S', and the process returns to step 2 to rebuild the model and re-simulate in order to further optimize the design scheme.

[0014] By adopting the above scheme, this invention shifts the main focus of performance verification and structural optimization to the computer simulation stage, establishing a virtual iterative closed loop of "design-simulation-evaluation". A single design verification can be completed in a few hours or days, with only a very small number of physical samples verified after the simulation passes. This fundamentally reduces the number of physical sample trials, thereby significantly reducing the material and time costs of R&D and improving development efficiency by an order of magnitude.

[0015] Furthermore, this invention systematically introduces and sets parameters such as the coefficient of restitution, static friction coefficient, rolling friction coefficient, and surface energy between particles and fibers in the simulation, constructing a high-fidelity CFD-DEM coupled model based on the intrinsic properties of the materials. These parameters enable the simulation to realistically reproduce microscopic physical behaviors such as collision energy loss, sliding adhesion, rolling resistance, and short-range adhesion, thereby accurately predicting filtration efficiency and pressure drop, and intuitively revealing the flow field structure and particle trapping mechanism. Thus, design optimization can shift from empirical speculation to precise control based on physical mechanisms, allowing for targeted adjustments to fiber parameters and structure to quickly obtain design solutions with superior performance. Attached Figure Description

[0016] Figure 1 Flowchart of existing filter media design methods; Figure 2 This is a flowchart of the filter media design method of the present invention; Figure 3 This is a detailed flowchart of the present invention; Figure 4This is the geometric model of Scheme 1 in the embodiments of the present invention; Figure 5 This is the geometric model of Scheme 2 in the embodiments of the present invention; Figure 6 This is a velocity streamline diagram of Scheme 1 of the present invention; Figure 7 This is a velocity streamline diagram of Scheme 2 of the present invention; Figure 8 This is the optimized velocity streamline diagram of Scheme 2 of the present invention. Detailed Implementation

[0017] like Figure 2-3 As shown, this invention discloses a method for assisting filter media design based on filtration simulation technology, which includes the following steps: Step 1: Determine the target filtration performance value M, design the filter media scheme, and determine at least one set of filter media design parameters S.

[0018] The target value M for filtration performance refers to the quantitative design target based on the final application scenario of the filter media (such as power plant flue gas dust removal, chemical tail gas treatment) and operating conditions (such as temperature, inlet dust concentration, gas composition) and emission standards. It mainly includes the target filtration efficiency (such as ≥99.9%) and the target resistance pressure drop range.

[0019] The design parameter set S includes: fiber parameters, filter media parameters, membrane parameters, particle parameters, and flow field parameters.

[0020] Fiber parameters include: material (common filter media materials include polyphenylene sulfide (PPS), polytetrafluoroethylene (PTFE), imported polyimide (P84), etc.), diameter and length, curvature (degree of fiber bending), Young's modulus (a physical quantity that measures the elastic deformation ability of fiber materials), Poisson's ratio (a parameter that measures the elastic deformation of materials), and density. Filter media parameters include spunlace / needle-punched parameters: needle-punching density and needle-punching depth, areal density (mass of filter media per unit area (g / m³)). 2 ( ) and thickness. Needle punching density and needle punching depth are core reinforcement technologies in nonwoven fabric production, and these two parameters determine the microstructure and macroscopic properties of the final filter material. The coating parameters on the filter media surface include: membrane material (commonly PTFE membrane), pore size / pore size distribution, thickness, and basis weight (mass of membrane per unit area (g / m²)). 2 Pore ​​size / pore size distribution is a core parameter among membrane coating parameters. The more concentrated the pore size distribution, the more stable the filtration performance. In practical applications, if the filtration accuracy requirements of the filter media are not high, a membrane structure may not be required on the surface of the filter media.

[0021] The particle parameters are: particle diameter, dust proportion (the respective proportion of particles with different diameters), density, Young's modulus, Poisson's ratio, coefficient of restitution between particles, coefficient of restitution between particle and fiber, coefficient of static friction between particles, coefficient of static friction between particle and fiber, coefficient of rolling friction between particles, coefficient of rolling friction between particle and fiber, surface energy parameter between particles, and surface energy parameter between particle and fiber.

[0022] Among them, the coefficient of restitution between particles and the coefficient of restitution between particle and fiber reflect the degree of recoverable kinetic energy during the collision process. This coefficient is the ratio of the relative velocity after collision to the relative velocity before collision, and is usually based on the elastic and damping characteristics of the fiber and the particle itself. Both coefficients of restitution can be obtained by testing the velocity ratio before and after collision through simple inclined plane collision or falling ball experiment or by referring to the literature:

[0023] Among them, is the coefficient of restitution between particles or the coefficient of restitution between particle and fiber, is the normal relative velocity, and the subscripts "before" and "after" represent before and after collision respectively; e = 1: perfectly elastic collision (no energy loss); 0 < e < 1: partially elastic collision (actual materials); e = 0: perfectly inelastic collision (sticking together after collision).

[0024] The coefficient of static friction between particles and the coefficient of static friction between particle and fiber are the ratio of the maximum static friction force between two contact surfaces to the normal force, which determines whether relative sliding occurs. Both coefficients of static friction can be directly measured through an inclined experiment or obtained by referring to the literature:

[0025] Among them, is the coefficient of static friction between particles or the coefficient of static friction between particle and fiber, is the maximum static friction force between two contact surfaces, is the normal force.

[0026] The coefficient of rolling friction between particles and the coefficient of rolling friction between particle and fiber are the ratio of the torque resisting rolling to the normal force, which characterizes the plastic deformation or viscous effect of the contact surface. Both coefficients of rolling friction can be inversely calibrated through a packing angle experiment or obtained by referring to the literature;

[0027] Among them, is the coefficient of rolling friction between particles or the coefficient of rolling friction between particle and fiber, is the torque resisting rolling, is the normal force.

[0028] The flow field parameters include: inlet velocity, outlet pressure, physical model, and wall settings. The physical model calculates the Reynolds number (Re) using the internal flow field velocity, the hydraulic diameter of the flow field cross-section, fluid density, and fluid viscosity. A Reynolds number greater than 2300 indicates turbulent flow, less than 2300 indicates laminar flow, and less than 1 indicates creeping flow. The physical model required for the flow field simulation is selected based on the Reynolds number.

[0029] Step 2: Based on the design parameter set S, construct a geometric model that reflects the random distribution structure of fibers inside the three-dimensional filter media or membrane filter media.

[0030] Based on the design parameter set S, a geometric model of the filter media was created using 3D modeling software (SolidWorks). Specifically, a single fiber model was created based on the fiber diameter, length, and curvature. Then, based on the size and morphological parameters of the single fiber, they were randomly arranged within a defined simulation domain. A 3D geometric model representing the microstructure of the filter media was generated according to the areal density and thickness constraints. Symmetrical walls were set in the simulation software to reduce computational load and simplify the model. If membrane coating is required, a mesh structure was randomly generated within the defined simulation domain based on the membrane's pore size, thickness, and basis weight to obtain the membrane's geometric model. Finally, the membrane was combined with the dust-facing surface of the filter media to generate the geometric model of the membrane-coated filter media.

[0031] Step 3: Perform coupled simulation of the filtration performance of the three-dimensional filter media geometric model to obtain the velocity streamline diagram and simulation prediction value F of the geometric model. Import the filter media geometric model into the fluid dynamics software (Fluent software) and the discrete element method software (EDEM) respectively.

[0032] In fluid dynamics software, the flow field is set up by: calibrating the model size, setting the fiber material, selecting the physical model, the flow field solution method, and the fiber wall and flow field wall types.

[0033] In the discrete element method (DEM) software, parameters were set as follows: fiber and particle density, Young's modulus, Poisson's ratio, coefficient of restitution between particles and between particles and fibers, static friction coefficient between particles and between particles and fibers, rolling friction coefficient between particles and between particles and fibers, and surface energy between particles and between particles and fibers. The adhesion behavior between particles and between particles and fibers in the simulation was characterized using the Hertz-Mindlin with JKR adhesion contact model, whose key parameter is the material surface energy (γ, unit J / m²).

[0034] The flow field of the three-dimensional geometric model is calculated in computational fluid dynamics software to obtain the velocity and pressure fields. Particle properties and contact models are set in discrete element method (DEM) software to simulate particle motion in the velocity field. Through a bidirectional coupling interface, flow field data and particle data are exchanged to calculate the interaction between the flow field and particle motion, thereby obtaining velocity streamline diagrams and simulation prediction values ​​F. Essentially, this bidirectional coupling interface is an iterative solution process that performs real-time transfer of momentum and mass source terms between the Lagrange multimeter (DEM) and the Eulerian multimeter (CFD) framework through spatial mapping (particle-to-mesh) and spatial interpolation (mesh-to-particle).

[0035] The core of the simulation solution process is an iterative convergence process, the purpose of which is to obtain a stable physical field distribution that satisfies the conservation of mass, momentum, and energy. The iterative solution does not directly calculate the target performance parameters (filtration efficiency and pressure drop), but rather aims to obtain convergent solutions for the fundamental flow field variables throughout the simulation domain, primarily including the velocity field, pressure field, and particle field. The solver starts from an initial assumed flow field and iterates and corrects it repeatedly until the residuals of all governing equations are below the set convergence criterion, resulting in a realistic and stable flow field.

[0036] After the flow field converges, the drag pressure drop and filtration efficiency are calculated. The drag pressure drop is directly derived from the converged pressure field. It is calculated by subtracting the average pressure values ​​at a certain cross-section of the filter media inlet (upstream) and outlet (downstream) regions. The filtration efficiency is obtained through the converged velocity field and the particle model used. During the simulation, particles are released at the inlet and move in the velocity field. The filtration efficiency is calculated by statistically analyzing the ratio of the number of particles captured on the filter media surface to the total number of particles released at the inlet.

[0037] The velocity streamline diagram is a direct visual representation of the aforementioned convergent velocity field. Streamlines are woven together using post-processing software based on the direction and magnitude of velocity vectors at various points in space. This visually demonstrates the overall direction of airflow within the filter, the uniformity of velocity distribution, and flow characteristics such as eddies or flow around the filter. Through the velocity streamline diagram of the filter media's geometric model, the flow field structure can be clearly and intuitively analyzed, revealing the influence mechanism of the filter media's internal geometry on fluid flow and potential particulate matter capture behavior (such as inertial collisions, streamline interception, and diffusion).

[0038] Step 4: Compare the simulation prediction value F with the target value M, and choose whether to iteratively optimize based on the results.

[0039] The simulation prediction value F obtained in step 3 is compared with the target value M set in step 1. If F meets the requirements of M, the current design parameter set S is deemed feasible, and the current design parameter set S is locked as the recommended design scheme. Optionally, to further optimize the scheme, the design parameter set S is adjusted in a targeted manner based on the analysis of the simulation results, generating a new parameter set S'. Then, return to step 2, reconstruct the model based on S', and resimulate.

[0040] If F does not meet the requirements of M, then based on the analysis of simulation results (such as velocity streamline diagrams), the design parameter set S is adjusted in a targeted manner (e.g., increasing areal density to improve efficiency, changing the fiber mixing ratio to optimize resistance), generating a new parameter set S'. Subsequently, return to step 2, rebuild the model based on S' and re-simulate, forming a virtual iterative closed loop of "design-simulation-evaluation-redesign" until a recommended design scheme that meets the objective M is obtained.

[0041] Step 5: Produce and verify physical samples based on the recommended design scheme that meets the target value M.

[0042] Select at least one of the recommended design schemes finalized in step 4 and deliver it to the production line for physical sample prototyping. Subsequently, conduct standard performance tests on the physical samples to obtain measured filtration efficiency and measured pressure drop. Compare the measured results with the simulation prediction value F to verify the reliability of the simulation model and ultimately confirm that the product performance meets the standards.

[0043] In the actual filter media design stage (step 1), multiple sets of different design parameters S are set, forming multiple candidate design schemes. In this case, steps 3 and 4 require simulation and evaluation of multiple design schemes. In step 4, the scheme with an F value that meets the M is selected as the recommended design scheme. For design schemes with an F value that does not meet the M, parameters can be adjusted until a satisfactory design scheme is obtained, or no adjustment can be made. Design schemes with an F value that meets the M can also be further optimized by adjusting parameters.

[0044] To better illustrate the technical solution of the present invention, an embodiment will be described below.

[0045] Step 1: Determine the target filtration performance value M, design the filter media scheme, and determine two sets of filter media design parameters S.

[0046] The target filtration performance value M is as follows: The specific operating conditions are: flue gas temperature 160℃, flue gas composition 21% oxygen, 14% water, and 300 mg / Nm³ of nitrogen oxides. 3 Sulfur dioxide 50 mg / Nm 3 The dust concentration at the inlet is 10 g / Nm³.3 The required export concentration is ≤50mg / Nm³. 3 Filtration efficiency ≥99.9%.

[0047] In this embodiment, based on the target filtration performance value M, polyphenylene sulfide (PPS) is selected as the filter material, and two schemes are proposed as follows: Option 1: PPS fiber diameter is 2D, PPS fiber length is 51mm; PPS filter media thickness is tentatively set at 1.5mm. The areal density of this PPS filter media is designed to be 500g / m², and the needle-punching density is designed to be 380 needles / cm². 2 The needle penetration depth is 10 mm; the particle diameter is 1-10 μm, and the proportion of particles of each diameter satisfies the Gaussian distribution; the flow field is designed with an inlet velocity of 0.6 m / min and an outlet pressure of 0 Pa.

[0048] Option 2: The PPS fibers are a blend of 1.5D and 2D, with 1.5D accounting for 50% and 2D accounting for 50%. The PPS fiber length is 51mm. The PPS filter media thickness is tentatively set at 1.5mm. Based on requirements, the areal density of this PPS filter media is designed to be 500g / m², and the needle-punching density is designed to be 380 needles / cm². 2 The needle penetration depth is 10 mm; the particle diameter is 1-10 μm, and the proportion of particles of each diameter satisfies the Gaussian distribution; the flow field is designed with an inlet velocity of 0.6 m / min and an outlet pressure of 0 Pa.

[0049] Step 2: Based on the design parameter set S, construct a geometric model that reflects the three-dimensional filter media.

[0050] To simplify the fiber model in the two schemes above, the fibers are set as cylinders, and the fiber curvature is ignored. Based on the design scheme, the simulation domain for the filter media is determined to be 250×250×2000μm. A three-dimensional random geometric model of the filter media is generated within this 250×250×2000μm region using 3D modeling software. The generated geometric model is shown below. Figure 4 and Figure 5 As shown. In this embodiment, the filter media is not coated with a membrane.

[0051] Step 3: Perform a coupled simulation of the filtering performance on the three-dimensional geometric model to obtain the velocity streamline diagram and simulation prediction value F of the geometric model.

[0052] Aside from the differences in geometric models, Scheme 1 and Scheme 2 share the same basic parameter settings in both fluid dynamics and discrete element method (DEM) software. In flow field simulation, the boundary conditions of the filter media model need to be set, including: inlet velocity, outlet pressure, fiber surface, filter media wall, physical model, and solution method. Taking the illustrative case as an example, the inlet velocity is set to 0.6 m / min, the outlet pressure to 0 Pa, the fiber and surface to be set as wall, the physical model to laminar flow, and the solution method to be the first-order SIMPLE upwind approach. In the particle simulation software, the properties of particles and fibers are set, such as Young's modulus, Poisson's ratio, coefficient of restitution, static friction coefficient, rolling friction coefficient, and surface energy. Taking an illustrative example, the fiber's Young's modulus is set to 3 GPa and Poisson's ratio to 0.4, while the particle's Young's modulus is set to 100 MPa and Poisson's ratio to 0.25. Between particles: coefficient of restitution 0.5, static friction coefficient 0.2, rolling friction coefficient 0.1, and surface energy 0.01 J / m²; between particles and fibers: coefficient of restitution 0.5, static friction coefficient 0.2, rolling friction coefficient 0.1, and surface energy 0.01 J / m². The flow field and particle simulation software are connected via a two-way coupling interface for calculation, obtaining the velocity streamline diagram and the simulated predicted value F. Step 4: Compare the simulated predicted value F with the target value M, and select whether to iteratively optimize based on the results.

[0053] Specifically, through coupled calculations, the interception of the flow field by the filter media, the pressure difference before and after the filter media, and the filtration efficiency are obtained. The F-value obtained after post-processing the simulation results of Scheme 1 (pressure difference before and after the filter media is 15.8 Pa and filtration efficiency is 99.25%) does not meet the target value M and will not be modified further. The F-value obtained after simulation results of Scheme 2 (pressure difference before and after the filter media is 28.3 Pa and filtration efficiency is 99.99%) meets the target value M. Figure 6 and Figure 7 As can be seen, when using the same number and distribution of streamlines for post-treatment, the number of streamlines successfully generated and penetrating the filter media in Scheme 2 is significantly less than that in Scheme 1. This indicates that the geometry of Scheme 2 creates stronger flow resistance to the fluid, resulting in more flow stagnation zones or more tortuous flow channels, thereby enhancing the "retention" or "interception" effect on the fluid.

[0054] Simulation techniques can be used to further optimize the model. In this embodiment, the parameters of Scheme 2 are modified based on the velocity streamline diagram: the surface density is changed to 600 g / m², while other parameters remain unchanged, and simulation is then performed. Figure 7 and Figure 8 It can be seen that when using the same number and distribution of streamlines in the post-processing software for post-processing, the number of streamlines generated and penetrating the filter media in the optimized scheme 2 is less than that in the optimized scheme 2. At the same time, the pressure difference across the filter media increases to 57 Pa, and the filtration efficiency increases to 99.9935%.

[0055] Step 5: Produce and verify physical samples according to the recommended design scheme.

[0056] The design parameter set S' of the optimized scheme two, which was finally locked in step 4 and verified virtually, was delivered to the production line for the trial production of physical samples. Subsequently, standard performance tests were performed on the physical samples to obtain the measured filtration efficiency and measured pressure drop.

[0057] The filtration performance of the sample from optimized scheme two was tested, and the pressure difference was 61 Pa, with a filtration efficiency of 99.9930%. The simulation results are close to the filtration performance of the sample, indicating that the filter media meets the requirements.

[0058] In summary, the present invention has the following technical effects: First, this invention, by constructing a design closed loop prioritizing "virtual prototypes," shifts the main focus of performance verification and optimization from expensive physical production lines to a computer simulation environment. This effectively avoids the enormous costs of raw materials, energy consumption, and equipment wear and tear incurred in traditional processes that require multiple small-batch sample productions to verify each design hypothesis. Only a very small number of final verification samples need to be produced after simulation verification is successful, significantly reducing the marginal cost of R&D.

[0059] Secondly, the traditional "design-production-test" physical iteration cycle is often measured in weeks or months. This invention, through efficient digital simulation iteration, compresses the single design verification cycle to the order of hours or days. Designers can complete parallel evaluation and multiple rounds of optimization of multiple design schemes in a very short time, increasing the speed of new product development from conceptual design to performance finalization by an order of magnitude, and rapidly responding to market demands.

[0060] Third, traditional design relies on experience, resulting in unclear optimization directions. This invention, through high-fidelity CFD-DEM coupled simulation, can intuitively reveal the microscopic flow details and particle trapping mechanisms within the filter media (such as the regional distribution of inertial impaction, interception, and diffusion deposition). This transforms optimization from "empirical guesswork" to "mechanism-based precise control," enabling targeted structural improvements to address performance shortcomings (such as excessively high local flow velocities leading to penetration, or low efficiency in specific particle size ranges). This allows for the design of filter media structures with superior performance that are difficult to achieve with traditional experience.

[0061] The above description is merely an embodiment of the present invention and does not constitute any limitation on the technical scope of the present invention. Therefore, any minor modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention shall still fall within the scope of the technical solution of the present invention.

Claims

1. A method for filter media design based on filtration simulation technology, characterized in that, Includes the following steps: Step 1: Determine the target value M for filtration performance, and set at least one set of filter media design parameters S based on the target value M. The set of design parameters S includes at least fiber parameters, filter media parameters, particle parameters, and flow field parameters. Step 2: Based on the design parameter set S, construct a three-dimensional geometric model that reflects the random arrangement of fibers inside the filter media; Step 3: Perform flow field-particle coupled simulation on the three-dimensional geometric model to obtain the simulation prediction value F, including filtration efficiency and pressure drop; The flow field and pressure field are obtained by performing flow field calculations on the three-dimensional geometric model in computational fluid dynamics software. In the discrete element method software, particle properties and contact models are set to simulate the motion of particles in the velocity field. Through a bidirectional coupling interface, flow field data and particle data are exchanged to realize the simulation prediction value F of the interaction between the flow field and particle motion. Step 4: Compare the simulated predicted value F with the target value M to determine whether F meets the requirements of M. If satisfied, the current design parameter set S is locked as the recommended design scheme, and step S5 is executed or the design parameter set S is modified and adjusted, and then step S2 is returned. Based on the adjusted design parameter set S, the three-dimensional geometric model is reconstructed and re-simulated and optimized. If not satisfied, adjust the design parameter set S and return to step S2, reconstruct the three-dimensional geometric model based on the adjusted design parameter set S and resimulate; Step 5: Select at least one of the recommended design schemes to conduct physical sample production and performance verification.

2. The method for filter media design based on filtration simulation technology according to claim 1, characterized in that, In step 1, The fiber parameters include: material, diameter and length, curvature, Young's modulus, Poisson's ratio, and density; The filter media parameters include hydroentangled / needle-punched parameters: needle-punching density and depth, areal density, and thickness. The particle parameters include: particle diameter, dust content, density, Young's modulus, Poisson's ratio, coefficient of restitution between particles and between particles and fibers, static friction coefficient between particles and between particles and fibers, rolling friction coefficient between particles and between particles and fibers, and surface energy parameters between particles and between particles and fibers.

3. The method for filter media design based on filtration simulation technology according to claim 1, characterized in that, In step S2, the construction of the three-dimensional geometric model is specifically as follows: a single fiber model is established based on the fiber diameter, length, and curvature. Then, based on the size and morphological parameters of the single fiber, it is randomly arranged within the set simulation domain, and a three-dimensional geometric model representing the microstructure of the filter material is generated according to the areal density and thickness constraints of the filter material.

4. The method for filter media design based on filtration simulation technology according to claim 1, characterized in that, The design parameter set S also includes coating parameters: film material, pore size / pore size distribution, thickness, and basis weight.

5. The method for filter media design based on filtration simulation technology according to claim 4, characterized in that, In step 2, when the design parameter set S includes membrane parameters, constructing the three-dimensional geometric model further includes: randomly generating a mesh structure within a set simulation domain according to the membrane parameters to obtain the geometric model of the membrane, and combining the geometric model of the membrane with the dust-facing surface of the filter media to generate a three-dimensional geometric model of the membrane filter media.

6. The method for filter media design based on filtration simulation technology according to claim 1, characterized in that, In step S1, multiple sets of different initial design parameters S are set to form multiple candidate design schemes; in steps S3 and S4, the multiple candidate design schemes are simulated and evaluated in parallel or sequentially to select one or more recommended design schemes that meet the target value M.

7. The method for filter media design based on filtration simulation technology according to claim 1, characterized in that, In step 3, the flow field-particle coupling simulation includes: The flow field of the three-dimensional geometric model is calculated in computational fluid dynamics software to obtain the velocity field and pressure field; particle properties and contact model are set in discrete element method software to simulate the motion of particles in the velocity field; flow field data and particle data are exchanged through a two-way coupling interface to realize the interaction calculation between the flow field and particle motion, so as to obtain the simulation prediction value F.

8. The method for filter media design based on filtration simulation technology according to claim 1, characterized in that, In step 1, multiple sets of different initial design parameters S are set to form multiple candidate design schemes; In steps 3 and 4, the multiple candidate design schemes are simulated and evaluated in parallel or sequentially to select one or more recommended design schemes that meet the target value M.

9. The method for filter media design based on filtration simulation technology according to claim 1, characterized in that, In step 4, if the simulation prediction value F satisfies the target value M, before or after locking in the recommended design scheme, the following steps are also included: based on the analysis of the simulation results, the design parameter set S is adjusted in a targeted manner to generate a new parameter set S', and the process returns to step 2 to rebuild the model and re-simulate in order to further optimize the design scheme.

Images