Method for predicting flow blockage of multi-scale particles in pump

Abstract

The invention discloses a method for predicting flow blockage of multi-scale particles in a pump. According to the method, pump body unit flow channels are reconstructed for a pump body grid, a complete pump flow field is mapped, an expansion domain is constructed around a CFD unit, the space relation between particles and the expansion domain is recognized, the intersection volume is calculated, and therefore parameters such as local porosity, the specific surface area and the slippage speed are obtained. A structural blockage coefficient is established based on a Kozeny-Carman model, a dynamic blockage coefficient is formed in combination with a dimensionless slip velocity, and finally, a total blockage index is obtained to be used for judging the blockage risk. The method can accurately identify the particle blocking position and degree in the pump, and is suitable for operation analysis of various solid-containing pumps.

Description

Technical Field

[0001] This invention relates to the field of fluid machinery and multiphase flow analysis, and in particular to a method for predicting blockage in multi-scale particulate flow within pumps. This method can be used for identifying blockage risks and assessing operational status in centrifugal pumps, mixed-flow pumps, mixed-flow pumps, and other rotating fluid machinery containing complex media such as solids, sand, and particles. Background Technology

[0002] During the transport of particulate media within a pump, various forms of blockage frequently occur within the impeller channel, including particle accumulation, channel narrowing, localized stagnation, and dynamic imbalance, due to the complex particle size distribution, significant concentration variations, and highly three-dimensional flow patterns. Blockage not only reduces the pump's flow rate and head performance but can also lead to accelerated localized wear in the flow channel, increased vibration, and in severe cases, even pump failure. Therefore, accurately identifying the location and severity of blockages in multi-scale particle flow within the pump is a critical issue in the design, operation control, and engineering applications of multiphase pumps.

[0003] Existing blockage assessment methods often rely on single physical quantities, such as local volume fraction, velocity decay, eddy current intensity, or pressure distribution. However, in solid-liquid two-phase transport conditions, particle accumulation is the result of both structural changes and dynamic characteristics, and relying solely on a single physical quantity cannot fully reflect the essence of blockage formation. Therefore, it is necessary to propose a blockage prediction method that can simultaneously consider flow structure characteristics and particle dynamic behavior, and is applicable to multi-scale particle transport conditions, in order to accurately identify the location, degree, and risk of particle blockage within the pump. Summary of the Invention

[0004] The purpose of this application is to provide a method for predicting blockage in multi-scale particle flow within a pump, which considers both structural and dynamic blockage in multi-scale particle flow within the pump, thereby achieving accurate determination of the location and risk of blockage.

[0005] According to an embodiment of this application, a method for predicting multi-scale particulate flow blockage in a pump is provided, comprising: S1: Reconstruct the CFD mesh of the pump unit flow channel; S2: Map the fluid velocity field obtained from the three-dimensional flow calculation of the complete mixed-transport pump to the CFD mesh of the pump unit flow channel to obtain the pump unit flow channel CFD mesh with flow field velocity information. S3: Based on the CFD mesh of the pump body unit flow channel, construct an extended domain with each mesh as the center, and calculate the geometric volume of the extended domain; S4: Traverse all multi-scale particles in the pump in the EDEM software and read the spatial position, velocity, and particle size of the multi-scale particles in the pump; S5: Iterate through the CFD mesh of the pump unit flow channel to identify the relative spatial relationship between multi-scale particles and extended domains; S6: Based on the spatial position, velocity, particle size, geometric volume of the extended domain, and the relative spatial relationship, calculate the local porosity, local specific surface area, and slip velocity; S7: Determine whether all multi-scale particles have been traversed. If not, repeat steps S5–S6. S8: Calculate the structural blockage coefficient based on the local porosity and local specific surface area, calculate the dynamic blockage coefficient based on the sliding velocity, and combine the structural blockage coefficient and the dynamic blockage coefficient to obtain the total blockage coefficient. S9: Generate a visual cloud map based on the total clogging coefficient, and determine the clogging risk of multi-scale particles in the pump.

[0006] Furthermore, the CFD mesh of the pump body unit flow channel is obtained by reconstructing the complete pump body mesh, including splitting the original mesh, removing adjacent flow channel meshes, and retaining only the area between the two blades.

[0007] Furthermore, the fluid velocity field mapping is transferred to the CFD mesh of the pump unit flow channel through mesh matching or nearest node interpolation.

[0008] Furthermore, the extended domain is spherical in shape, and its radius is three times or more the radius of the largest particle.

[0009] Furthermore, the relative spatial relationship between the multi-scale particles and the extended domain is as follows: When the distance between the particle center and the center of the extended domain d satisfy When the particle is completely within the extended domain, it is determined that the particle is completely within the extended domain. when When a particle intersects with a portion of the extended domain, the intersection volume is calculated using spherical segment geometry. when When the particle is completely outside the extended domain, it is determined that the particle is completely outside the extended domain. in r p Where is the particle radius, R e The radius of the extended domain.

[0010] Furthermore, when it is determined that the particle is completely within the extended domain, the volume of the intersection between the particle and the extended domain is the particle volume; when it is determined that the particle is completely outside the extended domain, the volume of the intersection between the particle and the extended domain is 0; when it is determined that the spatial relationship between the particle and the extended domain is partially intersecting, the calculation method for the volume of the intersection between the particle and the extended domain is as follows: ; in Vin For the intersecting volume, R e To extend the domain radius, r p Where is the particle radius, h e and h p The spherical cap heights on the extended domain side and the particle side are respectively calculated using the following formula: ; ; in, x The distance from the boundary plane to the center of the extended domain. d To extend the distance between the center of the domain and the center of the particle; .

[0011] Furthermore, the structural blockage coefficient is obtained by logarithmizing and normalizing the permeability calculated using the Kozeny–Carman permeability model, and the formula is as follows: ; ; in, B K The structural congestion coefficient. K For penetration rate, Porosity S This refers to the specific surface area of ​​the particles. c This is a Kozeny constant.

[0012] Furthermore, the formula for calculating the dynamic blockage coefficient is: ; ; in, B s The kinetic clogging coefficient, u slip The slip velocity between the particle and the fluid. u f For fluid velocity, u p This represents the particle velocity.

[0013] Furthermore, the total blockage coefficient is obtained by multiplying the structural blockage coefficient and the dynamic blockage coefficient.

[0014] Furthermore, a visual cloud map is generated based on the total clogging coefficient, and the clogging risk of multi-scale particles within the pump is determined, including: A visual cloud map is generated based on the total congestion coefficient. The congestion risk of a local area is judged based on the local values ​​of the cloud map. The higher the total congestion coefficient, the higher the congestion risk of the corresponding local area.

[0015] The technical solutions provided by the embodiments of this application may include the following beneficial effects: This invention achieves accurate identification of multi-scale particle clogging behavior within pumps by constructing a pump body unit flow channel mesh, an extended domain statistical region, and a dual-criteria system of structural and dynamic clogging coefficients within a CFD-DEM coupled framework. Compared with existing single-physical-quantity discrimination methods, this invention offers advantages such as high computational accuracy, wide applicability, strong physical mechanism integrity, and high reliability in clogging identification. This method is suitable for complex solid-liquid two-phase mixed-phase transport conditions where multiple particle sizes and scales coexist, and can be widely applied to clogging prediction, structural optimization, and operational diagnosis of various fluid machinery equipment such as centrifugal pumps, mixed-flow pumps, and mixed-transport pumps.

[0016] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and do not limit this application. Attached Figure Description

[0017] Figure 1 This is a flowchart illustrating a method for predicting blockages in multi-scale particulate flow within a pump, according to an exemplary embodiment.

[0018] Figure 2 This is a schematic diagram of pump unit flow channel mesh reconstruction and flow field mapping according to an exemplary embodiment.

[0019] Figure 3 This is a schematic diagram illustrating the construction of an extended domain of a CFD mesh cell according to an exemplary embodiment.

[0020] Figure 4 This is a schematic diagram illustrating three relative positional relationships between particles and extended domains according to an exemplary embodiment.

[0021] Figure 5 The calculation results of the present invention for predicting blockage of multi-scale particulate flow in a pump are shown according to an exemplary embodiment. Detailed Implementation

[0022] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numbers in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this application as detailed in the appended claims.

[0023] The terminology used in this application is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. The singular forms “a,” “the,” and “the” used in this application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0024] The technical solution of the present invention will be described in detail below with reference to the accompanying drawings and embodiments.

[0025] Figure 1 This is a flowchart illustrating a method for predicting multi-scale particulate flow blockage within a pump according to an exemplary embodiment, such as... Figure 1 As shown, the method may include the following steps: S1: Reconstruct the CFD mesh of the pump unit flow channel; Specifically, a 3D geometric model of the pump body's single flow channel was first constructed directly using the 3D modeling software SolidWorks. Based on the impeller structural parameters of the target pump, the space between two adjacent blades was used as a single flow channel. The complete impeller model was imported, and the impeller was rotated and divided using the modeling software's cutting function at a cutting angle of 360° / 4 (4 being the number of blades). The remaining three flow channels were deleted, retaining only one representative flow channel. The retained single flow channel includes key structural features such as the blade suction surface, pressure surface, the gap region between the front and rear shrouds, and the impeller inlet throat to outlet edge. Appropriate lengths of inlet and outlet straight pipe sections were added before and after the flow channel according to the needs of the CFD solution to avoid boundary conditions interfering with the internal flow field. After completing the single flow channel solid model, it was imported into mesh generation software for CFD calculation mesh generation. An unstructured tetrahedral mesh was used, and local mesh refinement was applied at the impeller inlet, the near-wall region of the blades, the trailing edge of the blades, and the potential particle retention area. Reconstructing the CFD mesh of the pump unit flow channel can avoid the problem of adjacent flow channels being too close together in the original mesh, which leads to the miscalculation of particles from other flow channels in the extended domain statistics. This improves the accuracy of the neighboring particle statistics, ensuring that the blockage assessment only reflects the true situation of a single flow channel and is not affected by other areas of the pump body, thereby improving the spatial accuracy of blockage prediction.

[0026] S2: Map the fluid velocity field obtained from the three-dimensional flow calculation of the complete mixed-transport pump to the CFD mesh of the pump unit flow channel to obtain the pump unit flow channel CFD mesh with flow field velocity information. Specifically, after completing the geometric modeling and mesh generation of the single-channel flow, the fluid velocity field obtained from the three-dimensional flow field calculation of the complete mixed-transfer pump is mapped onto the aforementioned single-channel CFD mesh. Figure 2This diagram illustrates the reconstructing of the flow channel mesh and flow field mapping for a pump unit according to an exemplary embodiment. First, a three-dimensional numerical simulation is performed on the complete pump model to obtain the global velocity field under steady-state or quasi-steady-state conditions. Then, the complete pump CFD calculation file is imported into the flow field mapping module. By identifying the spatial relationships corresponding to the individual flow channel meshes, and using a node matching mapping method, the velocity points calculated from the complete pump are assigned to the individual flow channel meshes. Through this operation, the velocity distribution, pressure gradient, and shear characteristics in each individual flow channel are ensured to remain consistent with the actual flow state of the complete pump, eliminating the need for further CFD solving for each individual flow channel and significantly reducing computational costs. This step achieves accurate transfer of the macroscopic flow field of the complete pump to the local individual flow channels, enabling subsequent local blockage identification to be based on the actual operating flow field, ensuring the accuracy and physical consistency of the blockage prediction results.

[0027] S3: Based on the CFD mesh of the pump body unit flow channel, construct an extended domain with each mesh as the center, and calculate the geometric volume of the extended domain; Specifically, after successfully mapping the complete pump body flow field to a single-channel CFD mesh, a local extended domain is constructed for each CFD computational unit within the single channel, and its corresponding extended domain geometric volume is calculated. Figure 3 This is a schematic diagram illustrating the construction of an extended domain for a CFD grid cell according to an exemplary embodiment. The geometric center of each CFD cell serves as the center of the extended domain, and a spherical extended domain is constructed around it. The radius of this extended domain is determined based on the maximum particle size of the multi-scale particles within the pump, and is set to three times or more the radius of the maximum particle (three times is used in this example) to ensure that the extended domain can cover large particles and surrounding small particles, achieving sufficient statistical analysis of the local particle spatial distribution. The spherical structure of the extended domain maintains uniformity in all directions in three-dimensional space, avoiding statistical biases caused by asymmetric geometries such as cuboids and prisms, while also facilitating subsequent calculations of the intersection volume between particles and the extended domain. The volume of the extended domain is used for subsequent calculations of local porosity, specific surface area, and kinetic quantities. By constructing extended domains for each CFD cell, a local statistical space covering the entire single flow channel can be formed, enabling clogging characteristics to be discriminated at a fine scale using CFD cells, thereby improving the method's resolution and accuracy at different spatial locations.

[0028] S4: Traverse all multi-scale particles in the pump in the EDEM software and read the spatial position, velocity, and particle size of the multi-scale particles in the pump; Specifically, by calling the data export interface of the EDEM software, the three-dimensional spatial coordinates (x, y, z) and velocity vector (v) of each particle at the current calculation moment are extracted one by one. x , v y , v zThe data includes particle size and position. During the traversal, the state data of all particles is structured and stored to form a particle information table containing spatial location, velocity, and particle size. Particles located outside the pump body flow channel are removed to avoid unnecessary computational burden.

[0029] S5: Iterate through the CFD mesh of the pump unit flow channel to identify the relative spatial relationship between multi-scale particles and extended domains; Specifically, Figure 4 This is a schematic diagram illustrating three relative positional relationships between a particle and its extended domain according to an exemplary embodiment. The spatial relationships between the particle and the extended domain include: Obtain the spatial coordinates of all particles from step S4, and calculate the Euclidean distance between the particle centroid and the center of the extended domain sphere. d， When the distance between the particle center and the center of the extended domain d satisfy When the particle is completely within the extended domain, it is determined that the particle is completely within the extended domain. when When the particle intersects with the extended domain, the intersection volume will be calculated using the spherical cap stacking method. when When the particle is completely outside the extended domain, it is determined that the calculation of the extended domain is not included. in r p Where is the particle radius, R e The radius of the extended domain.

[0030] S6: Based on the spatial position, velocity, particle size, geometric volume of the extended domain, and the relative spatial relationship, calculate the local porosity, local specific surface area, and slip velocity; Specifically, when a particle is determined to be completely within the extended domain, the volume of the intersection between the particle and the extended domain is the particle volume; when a particle is determined to be completely outside the extended domain, the volume of the intersection between the particle and the extended domain is 0; when the spatial relationship between the particle and the extended domain is determined to be partially intersecting, the volume of the intersection between the particle and the extended domain is calculated as follows: ; in V in For the intersecting volume, R e To extend the domain radius, r p Where is the particle radius, h e and h p The spherical cap heights on the extended domain side and the particle side are respectively calculated using the following formula: ; ; in, x The distance from the boundary plane to the center of the extended domain. d To extend the distance between the center of the domain and the center of the particle; .

[0031] This step constructs a microscale "particle-fluid-space" coupling mapping by extending the statistical properties of multiple particles within the domain, enabling each CFD grid to possess the physical characteristics of three-dimensional structural blockage and flow disturbance. This method takes into account the realistic differences in particle scale, velocity coupling, and spatial distribution, significantly improving the accuracy and resolution of blockage prediction.

[0032] S7: Determine whether all multi-scale particles have been traversed. If not, repeat steps S5–S6. Specifically, a main loop counter is set up to record the total number of particles processed and compare it with the total number of particles in real time. When unprocessed particles are detected, it means that there are still particles that may have spatial overlap with the extended domain that has not yet been analyzed. Steps S5 and S6 need to be repeated to ensure a comprehensive evaluation of the spatial interaction status of all particles with all extended domain meshes. S8: Calculate the structural blockage coefficient based on the local porosity and local specific surface area, calculate the dynamic blockage coefficient based on the sliding velocity, and combine the structural blockage coefficient and the dynamic blockage coefficient to obtain the total blockage coefficient. Specifically, structural congestion coefficient B K The permeability K calculated from the Kozeny–Carman permeability model is obtained by logarithmic transformation and normalization, and its calculation formula is as follows:

[0033]

[0034] in, B K The structural congestion coefficient. K For penetration rate, Porosity S This refers to the specific surface area of ​​the particles. c It is the Kozeny constant, which characterizes the "geometric blockage" tendency caused by the narrowing of fluid channels and reduction of porosity due to local particle accumulation, reflecting the contribution of static structure to flow resistance.

[0035] Dynamic blockage coefficient B s The calculation formula is:

[0036]

[0037] in, B s The kinetic clogging coefficient, u slip The slip velocity between the particle and the fluid. u f For fluid velocity, u p For particle velocity, this coefficient characterizes the inconsistency between the particle and fluid motion trends, reflecting the "motion blockage" trend caused by dynamic disturbances and particle retention.

[0038] Total congestion coefficient B t Based on structural congestion coefficient B K With dynamic blockage coefficient B s The product is determined as follows:

[0039] The total blockage coefficient, as a comprehensive indicator, can fully reflect the level of blockage risk in local areas of the flow channel due to the coupling effect of multi-scale particle movement and distribution.

[0040] Based on the three types of blockage coefficients, a blockage identification and prediction system with physical meaning and numerical stability is constructed from the perspectives of static space occupation, dynamic velocity difference and their coupling. It is applicable to the internal analysis of mixed-transport pumps under different flow rates, particle size distributions and concentration fields, and has good generalization ability and engineering applicability.

[0041] S9: Generate a visual cloud map based on the total clogging coefficient and determine the clogging risk of multi-scale particles in the pump; Specifically, a visual cloud map is generated based on the total congestion coefficient, and the congestion risk of a local area is judged based on the local value of the cloud map. The higher the total congestion coefficient, the higher the congestion risk of the corresponding local area.

[0042] Obtain the total blockage coefficient corresponding to each CFD grid. B t Then, the entire pump unit flow channel area is further visualized, such as... Figure 5 As shown, an appropriate color mapping gradient can be set according to the magnitude of the value, with low-congestion areas designated as blue (indicating unobstructed flow), medium-congestion as yellow (indicating localized clustering), and high-congestion areas as red (indicating significant congestion risk). Through volumetric rendering or cross-sectional display, a cloud map with a three-dimensional sense of hierarchy can be generated to reveal the spatial distribution characteristics of congestion risk.

[0043] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only.

[0044] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. A method for predicting blockage caused by multi-scale particulate flow in a pump, characterized in that, include: S1: Reconstruct the CFD mesh of the pump unit flow channel; S2: Map the fluid velocity field obtained from the three-dimensional flow calculation of the complete mixed-transport pump to the CFD mesh of the pump unit flow channel to obtain the pump unit flow channel CFD mesh with flow field velocity information. S3: Based on the CFD mesh of the pump body unit flow channel, construct an extended domain with each mesh as the center, and calculate the geometric volume of the extended domain; S4: Traverse all multi-scale particles in the pump in the EDEM software and read the spatial position, velocity, and particle size of the multi-scale particles in the pump; S5: Iterate through the CFD mesh of the pump unit flow channel to identify the relative spatial relationship between multi-scale particles and extended domains; S6: Based on the spatial position, velocity, particle size, geometric volume of the extended domain, and the relative spatial relationship, calculate the local porosity, local specific surface area, and slip velocity; S7: Determine whether all multi-scale particles have been traversed. If not, repeat steps S5–S6. S8: Calculate the structural blockage coefficient based on the local porosity and local specific surface area, calculate the dynamic blockage coefficient based on the sliding velocity, and combine the structural blockage coefficient and the dynamic blockage coefficient to obtain the total blockage coefficient. S9: Generate a visual cloud map based on the total clogging coefficient, and determine the clogging risk of multi-scale particles in the pump.

2. The method for predicting multi-scale particulate flow blockage in a pump according to claim 1, characterized in that, The CFD mesh of the pump unit flow channel is obtained by reconstructing the complete pump body mesh, including splitting the original mesh, removing adjacent flow channel meshes, and retaining only the area between the two blades.

3. The method for predicting multi-scale particulate flow blockage in a pump according to claim 1, characterized in that, The fluid velocity field mapping is transferred to the CFD mesh of the pump unit flow channel through mesh matching or nearest node interpolation.

4. The method for predicting multi-scale particulate flow blockage in a pump according to claim 1, characterized in that, The extended domain is spherical in shape, and its radius is three times or more the radius of the largest particle.

5. The method for predicting multi-scale particulate flow blockage in a pump according to claim 1, characterized in that, The relative spatial relationship between the multi-scale particles and the extended domain is as follows: When the distance between the particle center and the center of the extended domain d satisfy When the particle is completely within the extended domain, it is determined that the particle is completely within the extended domain. when When a particle intersects with a portion of the extended domain, the intersection volume is calculated using spherical segment geometry. when When the particle is completely outside the extended domain, it is determined that the particle is completely outside the extended domain. in r p Where is the particle radius, R e The radius of the extended domain.

6. The method for predicting multi-scale particulate flow blockage in a pump according to claim 5, characterized in that, When a particle is determined to be completely within the extended domain, the volume of the intersection between the particle and the extended domain is equal to the particle's volume. When a particle is determined to be completely outside the extended domain, the volume of the intersection between the particle and the extended domain is 0. When the spatial relationship between the particle and the extended domain is determined to be partially intersecting, the volume of the intersection between the particle and the extended domain is calculated as follows: ； in V in For the intersecting volume, R e To extend the domain radius, r p Where is the particle radius, h e and h p The spherical cap heights on the extended domain side and the particle side are respectively calculated using the following formula: ； ； in, x The distance from the boundary plane to the center of the extended domain. d To extend the distance between the center of the domain and the center of the particle; 。 7. The method for predicting multi-scale particulate flow blockage in a pump according to claim 1, characterized in that, The structural blockage coefficient is obtained by logarithmizing and normalizing the permeability calculated using the Kozeny–Carman permeability model. The formula is as follows: ； ； in, B K The structural congestion coefficient. K For penetration rate, Porosity S This refers to the specific surface area of ​​the particles. c This is a Kozeny constant.

8. The method for predicting multi-scale particulate flow blockage in a pump according to claim 1, characterized in that, The formula for calculating the dynamic blockage coefficient is: ； ； in, B s The kinetic clogging coefficient, u slip The slip velocity between the particle and the fluid. u f For fluid velocity, u p This represents the particle velocity.

9. The method for predicting multi-scale particulate flow blockage in a pump according to claim 1, characterized in that, The total blockage coefficient is obtained by multiplying the structural blockage coefficient and the dynamic blockage coefficient.

10. The method for predicting multi-scale particulate flow blockage in a pump according to claim 1, characterized in that, A visual cloud map is generated based on the total clogging coefficient, and the clogging risk of multi-scale particles within the pump is determined, including: A visual cloud map is generated based on the total congestion coefficient. The congestion risk of a local area is judged based on the local values ​​of the cloud map. The higher the total congestion coefficient, the higher the congestion risk of the corresponding local area.

Images