A dedusting pipe network optimization design method based on fluid simulation

Abstract

The application discloses a dust removal pipe network optimization design method based on fluid simulation and belongs to the technical field of fluid simulation optimization of dust removal pipe networks. The method obtains dust removal pipe network geometric parameters and working condition parameters, generates pipe network topology data set with flow direction marks according to pipe sections, nodes, elbows, reducers, converging fittings and air suction hoods; a gas-solid two-phase simulation domain is established according to dust particle size distribution and gas phase flow boundary, pressure field, velocity field and particle concentration field are solved, and pipe section flow field data set is obtained; pipe section deposition coefficients and along-the-way dissipation coefficients are input into a pipe diameter and elbow curvature joint optimization process, total pressure drop constraints and minimum conveying speed constraints are screened, and pipe diameter adjustment amount, elbow curvature adjustment amount and converging angle adjustment amount are output; the topology data set is updated and repeated solving is performed, and pipe network design parameter set is obtained. The method jointly optimizes pipe section deposition coefficients and along-the-way dissipation coefficients, reduces deposition risks and reduces conveying energy consumption.

Description

Technical Field

[0001] This invention relates to the field of fluid simulation optimization technology for dust removal pipeline networks, specifically a method for optimizing the design of dust removal pipeline networks based on fluid simulation. Background Technology

[0002] Current technologies for optimizing dust collection pipeline networks primarily rely on one-dimensional resistance calculations, empirical design manuals, or single-phase airflow numerical simulations. One-dimensional resistance calculations treat pipe sections, elbows, diameter changes, and manifolds as local resistance elements, determine the pipe diameter based on the design wind speed, and summarize the network pressure drop using the equivalent resistance method to select the fan. Empirical design manuals control particle settling through minimum carry-over wind speeds and recommended elbow curvature radii, but lack quantitative descriptions of particle distribution within the network and wall deposition conditions. Some gas-solid two-phase simulation methods can obtain pressure, velocity, and particle concentration fields, but these are mostly used for verifying existing pipeline designs and rarely converted into segment-level parameters for parameter adjustment. Existing technical solutions suffer from the following drawbacks: Deposition risk assessment remains at the level of average wind speed or overall concentration, failing to identify high-deposition pipe sections near bends, diameter changes, and manifolds based on wall shear, particle size grouping, and local pressure drop ratios; pipe diameter, bend curvature, and manifold angle are typically adjusted independently based on experience, without simultaneously incorporating deposition sensitivity and friction loss into the joint optimization process. This leads to situations where the adjusted pipe network exhibits high local resistance, deposition in low-velocity sections, or wasted fan energy; the optimization results lack gas-solid two-phase simulation feedback, and the particle deposition state after changes in pipe diameter and bend curvature is not re-solved and verified, easily causing deviations between design parameters and actual gas-solid flow characteristics. These deficiencies make it difficult to coordinate deposition control and energy consumption control in dust removal pipe network design, and local pipe sections are prone to becoming sources of deposition blockage or points of abnormal resistance. The issues that need to be addressed are the lack of pipe section-level deposition coefficients and friction loss coefficients, and the problem of jointly optimizing and iteratively verifying pipe diameter, bend curvature, and manifold angle under multiple constraints. Summary of the Invention

[0003] To address the issues of missing pipe segment-level deposition coefficients and friction loss coefficients, and the inability to jointly optimize and iteratively verify pipe diameter, elbow curvature, and confluence angle in existing technologies, a dust removal pipeline optimization design method based on fluid simulation is proposed. This method establishes a gas-solid two-phase simulation domain using a pipeline topology dataset with flow direction indicators, solves for the pressure, velocity, and particle concentration fields of each pipe segment, and generates the pipe segment deposition coefficient and friction loss coefficient. These coefficients are then input into the joint optimization process of pipe diameter and elbow curvature. Under total pressure drop constraints and minimum delivery speed constraints, the method outputs adjustments for pipe diameter, elbow curvature, and confluence angle. By updating the topology dataset and repeating the solution process, a pipeline design parameter set consistent with the actual gas-solid two-phase flow state is obtained.

[0004] To achieve the above objectives, the present invention provides the following technical solution: The present invention provides a dust removal pipeline network optimization design method based on fluid simulation, comprising: acquiring the geometric parameters and operating parameters of the dust removal pipeline network; generating a pipeline network topology dataset with flow direction indicators according to pipe segments, nodes, elbows, diameter changes, manifolds, and suction hoods; establishing a gas-solid two-phase simulation domain based on the dust particle size distribution and gas phase flow boundary in the pipeline network topology dataset with flow direction indicators; solving the pressure field, velocity field, and particle concentration field in the gas-solid two-phase simulation domain to obtain a pipe segment-by-pipe flow field dataset; and based on the pipe segment-by-pipe flow field data... Collected wall shear data, particle concentration data, and local pressure drop data are used to generate pipe segment deposition coefficients and friction loss coefficients. These coefficients are then input into a joint optimization process for pipe diameter and elbow curvature, using total pressure drop constraints and minimum delivery speed constraints for filtering. The outputs are pipe diameter adjustment, elbow curvature adjustment, and confluence angle adjustment. The pipe network topology dataset with flow direction indicators is updated according to these adjustments, and the pressure field, velocity field, and particle concentration field solutions are repeated to obtain the pipe network design parameter set.

[0005] As a technical solution of this invention, when generating a pipeline topology dataset with flow direction identifiers, endpoint snapping is performed on the centerlines of pipe segments in the geometric parameters and operating parameters of the dust removal pipeline network to obtain a set of node coordinates; pipe segment numbers, upstream node numbers, and downstream node numbers are written according to the connection order in the set of node coordinates to obtain a directed connection table; elbow angles, cross-sectional ratios of diameter changes, and inlet angles of manifolds in the directed connection table are written into the corresponding pipe segment attribute fields to obtain a pipe segment attribute table; connectivity verification is performed on the pipe segment attribute table, and isolated pipe segment numbers and reverse connection numbers are written into a topology queue to be corrected; the directed connection table is updated according to the topology queue to be corrected to generate the pipeline topology dataset with flow direction identifiers. By using endpoint snapping and connectivity verification, isolated pipe segments and reverse connections in the topology data are reduced, improving the accuracy of the pipeline network model.

[0006] Preferably, when performing endpoint snapping on the pipe segment centerline, the first and last endpoints of the pipe segment centerline are extracted, and endpoint cluster numbers are generated according to coordinate tolerances. First and last endpoints belonging to the same endpoint cluster number are merged into common nodes, obtaining common node coordinates. Branch nodes in the common node coordinates are written with hood and valve identifiers to obtain a node attribute table. A set of node coordinates is generated based on the node attribute table and the pipe segment centerline. This allows for consideration of hood and valve information during the topology generation stage, improving the correspondence between the pipe network model and the actual system.

[0007] When establishing and solving the gas-solid two-phase simulation domain, pipe segment length, pipe diameter, elbow curvature, and node elevation are extracted from the pipe network topology dataset with flow direction identifiers to form fluid domain grid control parameters. The dust particle size distribution is mapped to inlet particle phase grouping, and the gas phase flow boundary is mapped to inlet velocity boundary and outlet pressure boundary. Pipe segment grids and locally refined grids are generated according to the fluid domain grid control parameters, and the inlet particle phase grouping, inlet velocity boundary, and outlet pressure boundary are written into the pipe segment grid. The gas phase pressure-velocity coupling field and particle phase concentration transport field are alternately solved on the pipe segment grid to obtain node pressure data, pipe segment velocity data, and particle deposition flux data. The node pressure data, pipe segment velocity data, and particle deposition flux data are encapsulated according to pipe segment number to obtain the pipe segment-by-pipe flow field dataset.

[0008] Preferably, when alternately solving the gas phase pressure-velocity coupled field and the particle phase concentration transport field on the pipe segment grid, the gas phase momentum equation is solved using an implicit pressure scheme to obtain initial velocity and initial pressure data; the initial velocity data is substituted into the particle phase transport process to update the particle concentration data; the source terms in the gas phase momentum equation are corrected based on the particle concentration data to obtain corrected pressure data; the corrected pressure data and the initial velocity data are alternately updated to obtain convergence node pressure data and convergence pipe segment velocity data; the wall deposition flux is obtained based on the convergence node pressure data and convergence pipe segment velocity data, generating particle deposition flux data. This alternating solution method can reflect the reverse influence of particles on the gas phase flow field, making the pressure and velocity fields closer to the actual operating state of the dust removal pipe network.

[0009] When generating the pipe segment deposition coefficient and friction loss coefficient, the wall shear data in the pipe segment flow field dataset is divided into shear intervals according to the circumferential position of the pipe segment to obtain a set of low-level shear intervals; the particle size groups in the particle concentration data are superimposed with the set of low-level shear intervals to obtain group deposition weights; the local pressure drop ratio is obtained based on the elbow pressure drop and diameter change pressure drop in the local pressure drop data; the group deposition weights and the local pressure drop ratios are merged according to the pipe segment number to generate the pipe segment deposition coefficient; the friction loss in the pipe segment flow field dataset is correlated with the pipe segment length, pipe diameter, and velocity data to generate the friction loss coefficient.

[0010] Preferably, when generating the pipe segment deposition coefficient by fusing group deposition weights and local pressure drop percentages according to pipe segment numbers, the weights of the largest and smallest grain size groups are extracted from the group deposition weights according to the pipe segment numbers; the weights of the largest and smallest grain size groups are written into the same pipe segment record; the same pipe segment record is sorted by weight to obtain a deposition sensitivity ranking value; the deposition sensitivity ranking values ​​are concatenated according to the pipe segment numbers to generate the pipe segment deposition coefficient. This method can highlight the deposition risk of different grain sizes in low-shear regions and identify easily deposited pipe segments by combining the local pressure drop percentage.

[0011] When inputting the pipe segment deposition coefficient and friction loss coefficient into the joint optimization process of pipe diameter and elbow curvature, the pipe segment numbers are sorted according to the pipe segment deposition coefficient to obtain a sequence of high-risk deposition pipe segments. From the sequence of high-risk deposition pipe segments, pipe segments are selected to establish discrete combination sets of candidate values ​​for pipe diameter, elbow curvature, and confluence angle. The friction loss coefficient is written into each combination in the discrete combination set to obtain combined energy consumption data. The combined energy consumption data is subjected to total pressure drop summation and minimum delivery speed verification, and combinations that meet the total pressure drop constraint and minimum delivery speed constraint are retained. From the retained combinations, the combination with the smallest total pressure drop and the largest coverage of the high-risk deposition pipe segment sequence is selected, and the pipe diameter adjustment, elbow curvature adjustment, and confluence angle adjustment are output.

[0012] Preferably, when selecting the combination with the smallest total pressure drop and the largest number of high-risk depositional pipe segment sequences covered from the retained combinations, the total pressure drop data and the number of high-risk depositional pipe segment sequences covered for each of the retained combinations are read. The number of high-risk depositional pipe segment sequences covered is then used to generate a coverage quantity based on the number of retained combinations. The retained combinations are arranged in ascending order of total pressure drop data and descending order of coverage quantity to obtain a combination sorting table. The first combination is selected from the sorting table, and the candidate values ​​for pipe diameter, elbow curvature, and confluence angle in the first combination are used as the pipe diameter adjustment, elbow curvature adjustment, and confluence angle adjustment, respectively. Through this optimization process, high-risk depositional pipe segments can be prioritized for coverage while meeting pressure loss and minimum transport speed requirements, reducing pipeline blockage risk and controlling energy consumption.

[0013] When the pipeline design parameter set is obtained, the pipe diameter adjustment is written into the pipe segment diameter field of the pipeline topology dataset with flow direction identification, the elbow curvature adjustment is written into the elbow curvature field, and the confluence angle adjustment is written into the manifold inlet angle field, resulting in an updated topology dataset. The fluid domain mesh control parameters are regenerated on the updated topology dataset, and the pressure field, velocity field, and particle concentration field are solved on the regenerated fluid domain mesh control parameters, resulting in an updated flow field dataset. The corresponding pipe segment data in the updated flow field dataset is compared with the pipe segment flow field dataset to obtain the non-converged pipe segment number. Based on the non-converged pipe segment number, the pipe diameter adjustment, elbow curvature adjustment, and confluence angle adjustment are written back to form an iterative adjustment record. When the non-converged pipe segment number in the iterative adjustment record is empty, the updated topology dataset and the updated flow field dataset are merged to obtain the pipeline design parameter set. Through iterative updates and convergence judgment, the design results gradually approach the optimal pipeline parameters, avoiding deviations from actual operating conditions in a single optimization.

[0014] The technical effects and advantages provided by the present invention in the above technical solution are as follows:

[0015] Based on wall shear data, particle concentration data, and local pressure drop data in the pipe segment flow field dataset, a pipe segment deposition coefficient and a friction loss coefficient are generated. Wall shear data is divided into shear intervals according to the circumferential position of the pipe segment, forming a set of low-shear intervals. Particle size groups from the particle concentration data are superimposed on this set of low-shear intervals to obtain group deposition weights. Elbow pressure drop and diameter change pressure drop form the local pressure drop proportion. The group deposition weights and local pressure drop proportions are merged according to the pipe segment number to generate the pipe segment deposition coefficient. Friction loss is correlated with pipe segment length, diameter, and velocity data to generate the friction loss coefficient. Deposition risk is refined from the overall average wind speed assessment of the pipeline network to each pipe segment. The pipe segment deposition coefficient reflects the sensitivity of particles of different sizes to accumulation in the low-shear region of the wall and the contribution of local resistance at elbows and diameter changes to deposition. The friction loss coefficient characterizes the differences in energy loss during the transport process of each pipe segment. Compared to the conventional approach of uniformly increasing wind speed or enlarging pipe diameter across the entire network, this method focuses adjustments on pipe sections with high deposition and dissipation coefficients. Low-risk sections do not require additional resistance increases or airflow changes, reducing ineffective modifications and energy consumption. The deposition coefficient and friction loss coefficient of each pipe section are input into a joint optimization process using pipe diameter and elbow curvature. A sequence of high-risk deposition sections is obtained by sorting the deposition coefficients. Discrete sets of candidate pipe diameter, elbow curvature, and confluence angle values ​​are established. For each combination, the friction loss coefficient is written to obtain combined energy consumption data. After summarizing the total pressure drop and verifying the minimum transport speed, combinations satisfying both constraints are retained. From these retained combinations, the combination with the smallest total pressure drop and the largest coverage of high-risk deposition sections is selected, outputting the pipe diameter, elbow curvature, and confluence angle adjustments. Instead of taking empirical values ​​for pipe diameter, elbow curvature, and confluence angle separately, their coupled effects on pressure drop, carry-over velocity, and deposition distribution are considered within the same discrete combination space. After updating the pipe network topology dataset with flow direction indicators, the fluid domain mesh control parameters are regenerated, and the gas-solid two-phase solution is repeated. The updated flow field dataset is compared with the previous round of pipe-segment flow field dataset. The adjustment amount is written back for the non-converged pipe segment numbers to form an iterative adjustment record. When the non-converged pipe segment number is empty, the pipe network design parameter set is output. Each round of adjustment is verified by resolving the problem. Changes in particle deposition caused by changes in pipe diameter and elbow curvature can be captured. Non-converged pipe segments continue to participate in the next round of adjustment to avoid the design parameters deviating from the actual gas-solid two-phase flow state due to a single calculation output. This results in dust removal pipe network parameters that meet the total pressure drop limit and minimum delivery speed requirements and have a reasonable deposition risk distribution. Attached Figure Description

[0016] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in this invention. For those skilled in the art, other drawings can be obtained based on these drawings.

[0017] Figure 1 This is a flowchart of a dust removal pipeline network optimization design method based on fluid simulation;

[0018] Figure 2 This is a flowchart of the iterative convergence process for pipeline design parameters. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] See Figure 1 This invention provides a method for optimizing the design of dust removal pipeline networks based on fluid simulation, comprising: acquiring the geometric and operating parameters of the dust removal pipeline network; generating a pipeline network topology dataset with flow direction indicators according to pipe segments, nodes, elbows, diameter changes, manifolds, and suction hoods; establishing a gas-solid two-phase simulation domain based on the dust particle size distribution and gas phase flow boundary in the pipeline network topology dataset with flow direction indicators; solving the pressure field, velocity field, and particle concentration field in the gas-solid two-phase simulation domain to obtain a pipeline segment-by-pipe flow field dataset; and determining the wall shear rate based on the pipeline segment flow field dataset. Data, particle concentration data, and local pressure drop data are used to generate pipe segment deposition coefficient and friction loss coefficient. The pipe segment deposition coefficient and friction loss coefficient are input into the joint optimization process of pipe diameter and elbow curvature. The total pressure drop constraint and minimum delivery speed constraint are used for screening, and the pipe diameter adjustment, elbow curvature adjustment, and confluence angle adjustment are output. The pipe network topology dataset with flow direction label is updated according to the pipe diameter adjustment, elbow curvature adjustment, and confluence angle adjustment, and the pressure field, velocity field, and particle concentration field are solved repeatedly to obtain the pipe network design parameter set.

[0021] In specific implementation, obtaining the geometric and operating parameters of the dust removal pipeline network includes: reading the geometric data of the centerline of each pipe segment, the cross-sectional dimensions of the pipe segment, the bending radius of the elbow, the cross-sectional dimensions of the variable diameter section, the inlet angle of the manifold, the position data of the suction hood, as well as the gas flow rate, dust particle size distribution, dust density, gas density, gas viscosity, and outlet pressure data from the three-dimensional model of the dust removal pipeline network. The above data are stored according to the pipe segment number and the node number to form a dataset of geometric and operating parameters of the dust removal pipeline network.

[0022] When performing endpoint snapping on the centerlines of pipe segments in the geometric and operating parameters of the dust removal pipeline network to obtain the set of node coordinates, the first and last endpoints of all pipe segment centerlines are extracted to obtain the set of endpoint coordinates of the pipe segment centerlines. The process of generating endpoint cluster numbers according to coordinate tolerance is as follows: for any first endpoint and any last endpoint of the pipe segment centerline, the Euclidean distance between the first and last endpoints of the pipe segment centerline is calculated. The formula for calculating the Euclidean distance is:

[0023]

[0024] in, Indicates the beginning and end points of the pipe segment centerline End point of the pipe section centerline The Euclidean distance between them; Indicates the beginning and end points of the pipe segment centerline The X coordinate value, Indicates the beginning and end points of the pipe segment centerline The Y-coordinate value, Indicates the beginning and end points of the pipe segment centerline The Z-coordinate value; Indicates the end point of the pipe segment centerline The X coordinate value, Indicates the end point of the pipe segment centerline The Y-coordinate value, Indicates the end point of the pipe segment centerline The Z-coordinate value. Coordinate tolerance is used. express, The value is set to one-thousandth of the minimum centerline length of all pipe segments in the dust collector network's geometric parameters. This setting is based on the following: the minimum centerline length reflects the shortest connection dimension in the network; this one-thousandth coordinate tolerance prevents incorrect merging of different pipe segment endpoints when addressing minute gaps caused by modeling errors, while ensuring that no necessary connection pairs are missed. When the calculated... Less than or equal to At that time, the starting point of the pipe section centerline will be... End point of the pipe section centerline Clusters are assigned to the same endpoint cluster number; when the calculated Greater than At that time, the starting and ending points of the pipe section centerline are not included. End point of the pipe section centerline Pipeline centerlines belonging to the same endpoint cluster number are grouped into common nodes. The starting and ending points of the pipe centerlines within the same endpoint cluster number are merged into a common node. The coordinates of the common node are the arithmetic mean of the coordinates of all endpoints within the same endpoint cluster number. Branch nodes in the common node coordinates are assigned hood and valve identifiers to obtain a node attribute table. Specifically, when a common node connects to 3 or more pipe centerlines, it is marked as a branch node; when a common node is located at a hood, it is assigned a hood identifier; and when a common node is associated with a valve control device, it is assigned a valve identifier. A node coordinate set is generated based on the node attribute table and the pipe centerlines. Each record in the node coordinate set contains a node number, node X coordinate, node Y coordinate, node Z coordinate, hood identifier, and valve identifier.

[0025] When creating a directed connection table by writing the pipe segment number, upstream node number, and downstream node number according to the connection order in the node coordinate set, read the two endpoints corresponding to the pipe segment centerline, take the common node corresponding to the first endpoint of the pipe segment centerline as the upstream node, and take the common node corresponding to the last endpoint of the pipe segment centerline as the downstream node; write the pipe segment number, upstream node number, and downstream node number according to the airflow direction. If the airflow direction is not specified in the original data, the upstream and downstream relationship of each pipe segment is determined by taking the suction hood in the dust removal pipeline as the upstream and the fan inlet as the downstream, along the airflow direction.

[0026] The elbow angle, reducer section ratio, and manifold inlet angle from the directed connection table are written into the corresponding pipe segment attribute fields to obtain the pipe segment attribute table. The elbow angle is calculated from the bend in the pipe segment's centerline, the reducer section ratio is the ratio of the cross-sectional areas at both ends of the reducer segment, and the manifold inlet angle is the angle between the centerlines of the two inlet pipe segments at the manifold. The pipe segment attribute table includes the pipe segment number, upstream node number, downstream node number, elbow angle, reducer section ratio, manifold inlet angle, pipe segment length, and pipe diameter.

[0027] When performing connectivity checks on the pipe segment attribute table and writing the isolated pipe segment number and reverse connection number into the topology queue to be corrected, it checks whether the upstream node number and downstream node number of each pipe segment both exist in the node coordinate set. If the upstream node number or downstream node number of a pipe segment does not exist in the node coordinate set, then the pipe segment number is marked as an isolated pipe segment number and written into the topology queue to be corrected. It checks whether there is a flow direction conflict between two pipe segments. If the downstream node number of the first pipe segment is the same as the downstream node number of the second pipe segment and the upstream node number of the first pipe segment is different from the upstream node number of the second pipe segment, then it is determined that the second pipe segment has a reverse connection, and the second pipe segment number is written into the topology queue to be corrected as a reverse connection number.

[0028] When updating the directed connection table based on the topology queue to be corrected and generating a pipeline topology dataset with flow direction identifiers, for isolated pipe segment numbers, the nearest common node to the endpoint of the isolated pipe segment is found again, and the isolated pipe segment is connected to the nearest common node. The upstream or downstream node number of the isolated pipe segment is then updated. For reverse connection numbers, the upstream and downstream node numbers of the pipe segment corresponding to the reverse connection number are swapped to ensure that the flow direction is consistent with that of adjacent pipe segments. After the update is completed, the directed connection table, the node coordinate set, and the pipe segment attribute table are merged to generate a pipeline topology dataset with flow direction identifiers. The pipeline topology dataset with flow direction identifiers includes a node coordinate set, a directed connection table, and a pipe segment attribute table. The pipe segment number, upstream node number, and downstream node number in the directed connection table together constitute the flow direction identifier.

[0029] In practical implementation, based on the dust particle size distribution and gas phase flow boundary in the pipeline topology dataset with flow direction indicators, a gas-solid two-phase simulation domain is established. When solving the pressure field, velocity field, and particle concentration field in the gas-solid two-phase simulation domain to obtain the flow field dataset for each pipe segment, the pipe segment length, pipe diameter, elbow curvature, and node elevation are extracted first. The pipeline topology dataset with flow direction indicators includes a set of node coordinates, a directed connection table, and a pipe segment attribute table. Each node record in the node coordinate set includes the node number, node X coordinate, node Y coordinate, node Z coordinate, hood identifier, and valve identifier. Each pipe segment record in the pipe segment attribute table includes the pipe segment number, upstream node number, downstream node number, elbow angle, cross-sectional area ratio, inlet angle of the manifold, pipe segment length, and pipe diameter. When extracting pipe segment length, the pipe segment length field in the pipe segment attribute table is read directly; when extracting pipe diameter, the pipe diameter field in the pipe segment attribute table is read directly; when extracting elbow curvature, the elbow curvature is recorded as the reciprocal of the elbow bending radius based on the elbow angle in the pipe segment attribute table and the elbow bending radius in the pipe segment centerline geometry data; when extracting node elevation, the node Z coordinate in the node coordinate set is used as the node elevation.

[0030] When generating the control parameters for the fluid domain mesh, pipe segment length, pipe diameter, elbow curvature, and node elevation are combined according to the pipe segment number to generate the pipe segment mesh baseline size and the locally refined mesh size. The pipe segment mesh baseline size is set to one-tenth of the pipe diameter, and the locally refined mesh size is calculated according to the following formula:

[0031]

[0032] in, The feature size represents the locally refined mesh. Indicates the diameter of the pipe section; Indicates the radius of curvature of the elbow; This represents the grid density adjustment factor, with a value ranging from 2 to 8, set according to the bend angle. When the bend angle is greater than or equal to 90 degrees... Set to 2; when the bend angle is less than 90 degrees, The setting is 4. The rationale is that the larger the bend angle, the stronger the flow separation and secondary flow within the bend, requiring a smaller local mesh refinement feature size. Using a smaller mesh density adjustment factor can reduce the amount of [unspecified component] in the denominator of the formula. Item, thus obtaining a smaller When the bend angle is small, the flow separation is relatively weak, so a larger bend angle is used. The size of the local densified mesh feature can be appropriately increased to reduce the number of meshes.

[0033] When mapping dust particle size distribution to inlet particle phase grouping, dust particle size distribution data is read. This data includes particle size intervals and the corresponding mass fraction for each interval. The particle size intervals are sorted from smallest to largest, and the mass fractions are accumulated. The particle sizes corresponding to cumulative mass fractions of 10%, 30%, 50%, 70%, and 90% are respectively used as the representative particle sizes for the 1st, 2nd, 3rd, 4th, and 5th inlet particle phase groups. The mass flow rate ratio of each inlet particle phase group is the mass fraction within the corresponding particle size interval. The representative particle size of each inlet particle phase and the mass flow rate ratio of each inlet particle phase group are then written into the inlet particle phase grouping.

[0034] When mapping the gas phase flow boundary to the inlet velocity boundary and the outlet pressure boundary, the gas phase volumetric flow rate or gas phase mass flow rate, gas phase density, and inlet cross-sectional area are read from the gas phase flow boundary. When the gas phase volumetric flow rate is known, the inlet velocity value in the inlet velocity boundary is the gas phase volumetric flow rate divided by the inlet cross-sectional area; when the gas phase mass flow rate is known, the gas phase mass flow rate is divided by the gas phase density to obtain the gas phase volumetric flow rate, and then divided by the inlet cross-sectional area to obtain the inlet velocity value in the inlet velocity boundary. The outlet pressure boundary is set according to the pressure measurement value at the outlet of the dust collector pipeline or the design negative pressure value at the fan inlet, and the outlet static pressure value in the outlet pressure boundary is written into the outlet pressure boundary.

[0035] When generating pipe segment meshes and locally refined meshes according to the fluid domain mesh control parameters, a hexahedral mesh is generated for each pipe segment according to the pipe segment reference size, and in the elbow area, the characteristic size of the locally refined mesh is used. The localized mesh for elbows is densified. In the diameter-changing region, the localized mesh for diameter-changing is densified by one-tenth of the smaller value of the pipe diameter before and after the diameter change. In the manifold region, the localized mesh for manifold is densified by one-tenth of the manifold inlet pipe diameter. The inlet particle phase group is written into the inlet boundary element of the pipe segment mesh, the inlet velocity boundary is written into the inlet boundary element of the pipe segment mesh, and the outlet pressure boundary is written into the outlet boundary element of the pipe segment mesh.

[0036] When alternately solving the gas-phase pressure-velocity coupled field and the particle-phase concentration-transport field on the pipe segment mesh, the gas-phase momentum equation is solved using a pressure implicit scheme on the pipe segment mesh to obtain initial velocity and pressure data. The pressure implicit scheme employs the SIMPLE algorithm. The gas-phase momentum equation includes the gas-phase continuity equation and the gas-phase momentum conservation equation, with the momentum exchange source term between the gas and particle phases considered in the gas-phase momentum conservation equation. In the initial solution, the force source term of the particle-to-gas phase interaction is set to zero.

[0037] Substituting the initial velocity data into the particulate phase transport process, when updating the particle concentration data, for each group of inlet particulate phases, the particulate phase concentration transport equation is solved on the pipe segment grid. This equation includes a particulate phase refluxing term, a particulate phase diffusion term, and a particulate phase deposition source term. The particulate phase deposition source term is calculated based on the wall shear velocity and the wall shear velocity in the initial velocity data. When the wall shear velocity in the initial velocity data is less than the critical shear velocity of that group of inlet particulate phases, deposition occurs near the wall, and the deposited mass is deducted from the particulate phase concentration. When the wall shear velocity in the initial velocity data is greater than or equal to the critical shear velocity of that group of inlet particulate phases, no deposition occurs. The particle concentration data for each group of inlet particulate phases is obtained through this process.

[0038] To obtain corrected pressure data, the source terms in the gas-phase momentum equation are corrected based on particle concentration data. Then, the particle concentration data is substituted into the gas-phase momentum equation to exchange source terms between the gas and particle phase momentum, and the gas-phase momentum equation is solved again to obtain corrected pressure and velocity data. This process is repeated, substituting the corrected velocity data into the particle phase transport process to update the particle concentration data, and then using the new particle concentration data to correct the source terms in the gas-phase momentum equation. This continues until the change in nodal pressure data between two consecutive solutions is less than the nodal pressure data convergence threshold, and the change in pipe segment velocity data is less than the pipe segment velocity data convergence threshold. At this point, the alternating updates stop, resulting in converged nodal pressure data and converged pipe segment velocity data. The nodal pressure data convergence threshold is set to 0.01% of the average value of all nodal pressure data, and the pipe segment velocity data convergence threshold is set to 0.01% of the average value of all pipe segment velocity data.

[0039] The wall deposition flux is obtained based on the convergence node pressure data and the convergence tube segment velocity data. When generating particle deposition flux data, for the inner wall of each tube segment, the velocity value of the first layer of grid near the wall in the convergence tube segment velocity data is read. The wall shear velocity is calculated based on the velocity value of the first layer of grid near the wall in the convergence tube segment velocity data. The wall deposition flux of each group of inlet particle phases is calculated by combining the critical shear velocity of each group of inlet particle phases. The wall deposition flux of each group of inlet particle phases in each tube segment is added together to obtain the particle deposition flux data of each tube segment.

[0040] When the node pressure data, pipe segment velocity data, and particle deposition flux data are encapsulated by pipe segment number to obtain a pipe segment-by-pipe flow field dataset, each pipe segment record in the pipe segment flow field dataset contains the pipe segment number, node pressure data, pipe segment velocity data, and particle deposition flux data. The node pressure data contains the node pressure value corresponding to the upstream node number and the downstream node number of the pipe segment in the pipe segment record. The pipe segment velocity data contains the average velocity of each section in the pipe segment corresponding to the pipe segment number and the velocity value of the first layer of grid near the wall. The particle deposition flux data contains the wall deposition flux value of each group of inlet particle phases in the pipe segment corresponding to the pipe segment number in the pipe segment record.

[0041] In practical implementation, when generating the pipe segment deposition coefficient and friction loss coefficient based on the wall shear data, particle concentration data, and local pressure drop data in the pipe segment flow field dataset, the pipe segment number, nodal pressure data, pipe segment velocity data, and particle deposition flux data contained in each pipe segment record in the pipe segment flow field dataset are first read. The nodal pressure data includes the upstream node pressure value corresponding to the upstream node number and the downstream node pressure value corresponding to the downstream node number. The pipe segment velocity data includes the average velocity of each section within the pipe segment and the velocity value of the first layer of grid near the wall. The particle deposition flux data includes the wall deposition flux value of each group of inlet particle phases within the pipe segment. The wall shear data is calculated from the velocity value of the first layer of grid near the wall in the pipe segment velocity data. The wall shear data is the tangential component of the sum of gas molecule viscous shear stress and turbulent shear stress at the wall.

[0042] When dividing the wall shear data in the pipe segment flow field dataset into shear intervals according to the circumferential position of the pipe segment to obtain the low-level shear interval set, the wall shear data of each pipe segment is divided into 36 equal-angle intervals on the circumference of the pipe segment cross-section, with each interval having an angle range of 10 degrees. The arithmetic mean of the wall shear data within an interval is taken to obtain the interval wall shear average. The arithmetic mean of the interval wall shear averages of all 36 intervals within a pipe segment is taken to obtain the pipe segment wall shear average. If the interval wall shear average is less than 70% of the pipe segment wall shear average, the interval is marked as a low-level shear interval, and the pipe segment number, interval start angle, interval end angle, and interval wall shear average are written into the low-level shear interval set. Each low-level shear interval in the low-level shear interval set corresponds to a pipe segment number and a circumferential angle range.

[0043] When grouping particle size from particle concentration data with the low-shear zone set to obtain the grouped deposition weights, the particle size grouping in the particle concentration data is the inlet particle facies grouping. The inlet particle facies grouping includes representative particle sizes of group 1, group 2, group 3, group 4, and group 5, with each group of representative particle sizes corresponding to a set of particle mass flow rates. For each low-shear zone in the low-shear zone set, the particle concentration values ​​corresponding to each group of representative particle sizes within that low-shear zone are extracted from the particle concentration data, and the wall deposition flux values ​​corresponding to each group of representative particle sizes are extracted from the particle deposition flux data. The particle concentration value of the same group of representative particle sizes is multiplied by the wall deposition flux value to obtain the deposition contribution value of that group of representative particle sizes. The deposition contribution values ​​of all 5 groups of representative particle sizes within a low-shear zone are summed to obtain the total deposition contribution value of that low-shear zone. The depositional contribution value of a group of representative grain sizes is divided by the total depositional contribution value of the low-shear section to obtain the group depositional weight of that group of representative grain sizes in the low-shear section. The group depositional weight is a dimensionless value, ranging from 0 to 1. The pipe segment number, the start angle and end angle of the low-shear section, the representative grain sizes of each group, and the corresponding group depositional weight are written into the group depositional weight dataset.

[0044] To determine the proportion of local pressure drop based on elbow and diameter change pressure drops in the local pressure drop data, the elbow angle, diameter change section ratio, and manifold inlet angle are read from the pipe segment attribute table. The upstream and downstream node pressure values ​​for each pipe segment are read from the node pressure data in the pipe segment flow field dataset. Friction pressure drop is the difference between the upstream and downstream node pressure values ​​for the same pipe segment. Elbow pressure drop is the difference between the average total pressure downstream of the elbow location and the average total pressure upstream of the elbow location; the elbow location is determined by the elbow angle in the pipe segment attribute table. Diameter change pressure drop is the difference between the average total pressure downstream of the diameter change location and the average total pressure upstream of the diameter change location; the diameter change location is determined by the diameter change section ratio in the pipe segment attribute table. For the same pipe segment, the elbow pressure drop and diameter change pressure drop are added to obtain the total local pressure drop, and the friction pressure drop and the total local pressure drop are added to obtain the total pressure drop for that pipe segment. The local pressure drop percentage is the result of dividing the total local pressure drop by the total pressure drop of the pipe section. The local pressure drop percentage is a dimensionless value, ranging from 0 to 1.

[0045] When generating the pipe segment deposition coefficient by merging the group deposition weights and local pressure drop percentages according to pipe segment numbers, the weights of the largest and smallest grain size groups are extracted from the group deposition weights according to the pipe segment number. The largest grain size group weight is the maximum value of the group deposition weight corresponding to the 5th representative grain size among all low-shear sections under the same pipe segment number. The smallest grain size group weight is the maximum value of the group deposition weight corresponding to the 1st representative grain size among all low-shear sections under the same pipe segment number. The largest grain size group weight, the smallest grain size group weight, and the local pressure drop percentage are written into the same pipe segment record. Weight sorting is performed on the same pipe segment record to obtain the deposition sensitivity sorting value. The specific weight sorting method is as follows: the largest grain size group weight in the same pipe segment record is multiplied by the sorting weight value of 0.4, the smallest grain size group weight is multiplied by the sorting weight value of 0.3, and the local pressure drop percentage is multiplied by the sorting weight value of 0.3. The three products are added together to obtain the deposition sensitivity sorting value. The ranking weights of 0.4, 0.3, and 0.3 are set as follows: the particles corresponding to the largest particle size group have greater inertia and the most significant deposition response in the low-shear zone, thus having the greatest impact on the pipe segment deposition coefficient; therefore, a weight of 0.4 is chosen. The particles corresponding to the smallest particle size group are easily affected by turbulent diffusion, but their wall deposition response is relatively weaker than that of the largest particle size group; therefore, a weight of 0.3 is chosen. The proportion of local pressure drop reflects the influence of geometric local resistance on gas phase flow and is on the same order of magnitude as the deposition weight; therefore, a weight of 0.3 is chosen. The deposition sensitivity ranking values ​​are concatenated according to the pipe segment number to generate the pipe segment deposition coefficient. The concatenation method is as follows: the numerical part of the pipe segment number is connected to the numerical part of the deposition sensitivity ranking value with a hyphen to form the pipe segment deposition coefficient. The pipe segment deposition coefficient is in string format, with the pipe segment number before the hyphen and the deposition sensitivity ranking value after the hyphen.

[0046] When generating the friction loss coefficient by correlating the friction loss with the pipe segment flow field dataset and the pipe segment length, diameter, and velocity data, the friction loss is calculated as the difference between the upstream and downstream pressure values ​​of the same pipe segment. The pipe segment length and diameter are read from the pipe segment attribute table, and the velocity data is read from the pipe segment velocity data in the pipe segment flow field dataset, using the average velocity of each cross-section within the pipe segment. The gas phase density is read from the geometric and operating parameters of the dust removal pipeline network. The friction loss coefficient is calculated using the following formula:

[0047]

[0048] in, This represents the friction loss coefficient, which is a dimensionless value. This represents the friction pressure drop in the flow field dataset for each pipe segment, which is the difference between the pressure value of the upstream node and the pressure value of the downstream node in the same pipe segment, and the unit is Pascal. This indicates the pipe segment diameter, read from the pipe diameter field in the pipe segment attribute table, in meters. This represents the gas phase density, which is read from the geometric parameters and operating parameters of the dust removal pipeline network, and the unit is kilograms per cubic meter. This indicates the pipe segment length, read from the pipe segment length field in the pipe segment attribute table, in meters. This represents the average gas velocity within the pipe segment, read from the pipe segment velocity data in the pipe segment flow field dataset, in meters per second. The calculated friction loss coefficient value for each pipe segment is written into the friction loss coefficient table, where each record contains the pipe segment number and the friction loss coefficient value.

[0049] In practice, the pipe segment deposition coefficient and friction loss coefficient are input into the joint optimization process of pipe diameter and elbow curvature. Using total pressure drop constraints and minimum transport speed constraints for screening, the pipe segment numbers are sorted according to the pipe segment deposition coefficient to obtain a sequence of high-risk deposition segments. The pipe segment deposition coefficient is in string form, consisting of the numerical part of the pipe segment number connected to the deposition sensitivity ranking value by a hyphen. The sorting process is as follows: extract the deposition sensitivity ranking value after the hyphen in the deposition coefficient of each pipe segment, and sort the pipe segment numbers from largest to smallest according to the deposition sensitivity ranking value to obtain the sequence of high-risk deposition segments. In the sorting results, the pipe segment number with the largest deposition sensitivity ranking value is placed at the beginning of the high-risk deposition segment sequence.

[0050] When selecting pipe segments from the high-risk deposition pipe segment sequence and establishing discrete combination sets for candidate pipe diameter, elbow curvature, and confluence angle values, the first Q pipe segments are selected from the high-risk deposition pipe segment sequence, where Q is 30% of the total number of pipe segments in the high-risk deposition pipe segment sequence, rounded up. The basis for setting Q is: the pipe segments ranked higher are those with higher deposition risk, having the most direct impact on dust removal network optimization; selecting 30% of the total number of pipe segments in the high-risk deposition pipe segment sequence can cover the area with concentrated deposition risk, while maintaining a reasonable size for the discrete combination set. For each selected pipe segment, a set of candidate pipe diameter values, a set of candidate elbow curvature values, and a set of candidate confluence angle values ​​are established. The set of candidate pipe diameter values ​​is obtained by multiplying the current pipe diameter by a pipe diameter scaling factor, with scaling factors set to 0.8, 0.9, 1.0, 1.1, and 1.2. The set of candidate elbow curvature values ​​is obtained by multiplying the current elbow curvature by an elbow curvature scaling factor, with scaling factors set to 0.9, 1.0, and 1.1. The candidate set of confluence angle values ​​is obtained by multiplying the current confluence angle by a confluence angle scaling factor, which takes values ​​of 0.85, 0.90, 0.95, 1.00, and 1.05. The candidate values ​​for pipe diameter, elbow curvature, and confluence angle for each selected pipe segment are combined using a Cartesian product to obtain a discrete combination set. Each combination in the discrete combination set contains one candidate value for pipe diameter, one candidate value for elbow curvature, and one candidate value for confluence angle corresponding to each selected pipe segment.

[0051] When obtaining combined energy consumption data by writing the friction loss coefficient into each combination in the discrete combination set, the friction loss coefficient value is read from the friction loss coefficient table generated in the above embodiment according to the pipe segment number. When the candidate pipe diameter value of a pipe segment in a combination within the discrete combination set is different from the original pipe diameter of the pipe segment, the friction loss coefficient of that pipe segment in the combination needs to be corrected based on the original friction loss coefficient value. The original friction loss coefficient is read from the friction loss coefficient table, and the original pipe diameter is read from the pipe segment attribute table. The correction method is: the friction loss coefficient of the pipe segment in the combination is equal to the original friction loss coefficient multiplied by the reciprocal of the ratio of the candidate pipe diameter value to the original pipe diameter raised to the power of 0.25. The corrected friction loss coefficient is written into the corresponding pipe segment in the discrete combination set to obtain the combined energy consumption data. The combined energy consumption data includes the friction loss coefficient, local resistance coefficient, candidate pipe diameter value, candidate elbow curvature value, and candidate confluence angle value of each pipe segment in each combination.

[0052] The combined energy consumption data is summarized for total pressure drop and checked for minimum transmission speed. When combinations that meet the total pressure drop and minimum transmission speed constraints are retained, the total pressure drop is calculated for each combination in the discrete combination set. The total pressure drop is calculated according to the following formula:

[0053]

[0054] in, This represents the total pressure drop of the m-th combination in the discrete combination set, in Pascals. It represents the friction loss coefficient corresponding to the nth pipe segment in the mth combination of the discrete combination set, and is a dimensionless value; This represents the gas phase density, which is read from the geometric parameters and operating parameters of the dust removal pipeline network, and the unit is kilograms per cubic meter. This represents the length of the nth pipe segment, read from the pipe segment attribute table, in meters. This represents the gas volumetric flow rate of the nth pipe segment, which is read from the geometric parameters and operating parameters of the dust removal pipeline network, and the unit is cubic meters per second. This represents the candidate pipe diameter value for the nth pipe segment within the mth combination in the discrete combination set, in meters. It represents the local resistance coefficient of the nth pipe segment within the mth combination in the discrete combination set. It is determined by the candidate values ​​of the bend curvature and the confluence angle of the pipe segment and is a dimensionless value. This represents the constant value of pi, with a value of 3.1415926. This represents the total number of pipe segments involved in the m-th combination in the discrete combination set; the superscript m indicates the combination number in the discrete combination set, and the superscripts 5 and 4 represent the power exponents, respectively. For each combination, the total pressure drop data is obtained, and the total pressure drop data is a fraction of that combination's total pressure drop. The total pressure drop constraint is: the combined total pressure drop data must be less than or equal to the allowable upper limit of the total pressure drop of the dust removal pipeline network. The upper limit of the total pressure drop is read from the geometric parameters and operating parameters of the dust removal pipeline network. The minimum conveying speed verification process is as follows: for each pipe section in the combination, use the gas phase volumetric flow rate of that pipe section. Divide by the cross-sectional area of ​​the pipe segment in the assembly to obtain the average gas velocity of the pipe segment. The cross-sectional area is equal to Multiply by the square of the candidate pipe diameter value and then divide by 4; when the average gas velocity of each pipe segment in the combination is greater than or equal to the minimum conveying velocity of that pipe segment, the combination satisfies the minimum conveying velocity constraint; the minimum conveying velocity is read from the geometric parameters and operating parameters of the dust removal pipeline network. The combination that simultaneously satisfies both the total pressure drop constraint and the minimum conveying velocity constraint is retained, resulting in the retained combination.

[0055] From the retained combinations, select the combination with the smallest total pressure drop and the largest number of high-risk depositional pipe segment sequences covered. When outputting pipe diameter adjustment, elbow curvature adjustment, and confluence angle adjustment, read the total pressure drop data and the number of high-risk depositional pipe segment sequences covered for each of the retained combinations. The total pressure drop data is the total pressure drop of the combination. The method for calculating the coverage number of high-risk pipe segment sequences is as follows: Check how many pipe segments in each selected combination are located within the high-risk pipe segment sequence; the number of pipe segments within the high-risk sequence is the coverage number of the high-risk pipe segment sequence. The coverage number of the high-risk pipe segment sequence is used to generate coverage quantities for each retained combination, with one coverage quantity corresponding to each retained combination. The retained combinations are sorted in ascending order of total pressure drop and descending order of coverage quantity to obtain a combination sorting table. During sorting, combinations are first sorted from largest to smallest coverage quantity; for retained combinations with the same coverage quantity, they are then sorted from smallest to largest total pressure drop. The first combination is selected from the combination sorting table; this first combination has the largest coverage quantity and the smallest total pressure drop. The candidate pipe diameter values, elbow curvature values, and confluence angle values ​​in the first combination are used as pipe diameter adjustment amounts, elbow curvature adjustment amounts, and confluence angle adjustment amounts, respectively. The pipe diameter adjustment amount, elbow curvature adjustment amount, and confluence angle adjustment amount are then output.

[0056] In specific implementation, please refer to Figure 2When obtaining the pipeline design parameter set, the pipe diameter adjustment amount output in the above embodiment is written into the pipe segment diameter field of the pipeline topology dataset with flow direction identification. The pipe diameter adjustment amount includes the pipe diameter candidate value in the first and second combinations. The pipe segment diameter field is located in the pipe segment attribute table of the pipeline topology dataset with flow direction identification. Each pipe segment record in the pipe segment attribute table includes the pipe segment number, upstream node number, downstream node number, elbow angle, diameter reduction ratio, manifold inlet angle, pipe segment length, and pipe diameter. During writing, the corresponding pipe segment record in the pipe segment attribute table is located according to the pipe segment number corresponding to the pipe diameter adjustment amount, and the pipe diameter candidate value in the pipe diameter adjustment amount replaces the pipe diameter field value in the corresponding pipe segment record. The elbow curvature adjustment amount output in the above embodiment is written into the elbow curvature field of the pipeline topology dataset with flow direction identification. The elbow curvature adjustment amount includes the elbow curvature candidate value in the first and second combinations. The elbow curvature field is located in the corresponding pipe segment record in the pipe segment attribute table, and the elbow curvature field is represented by the reciprocal of the elbow curvature radius. During the writing process, the candidate values ​​of the elbow curvature in the elbow curvature adjustment amount replace the elbow curvature field values ​​in the corresponding pipe segment record. The merging angle adjustment amount output in the above embodiment is written into the merging inlet angle field of the pipe network topology dataset with flow direction identification. The merging angle adjustment amount includes the merging angle candidate values ​​in the first and second combinations. The merging inlet angle field is located in the pipe segment attribute table. During the writing process, the candidate values ​​of the merging angle in the merging angle adjustment amount replace the merging inlet angle field values ​​in the corresponding pipe segment record. After completing the above field updates, the updated topology dataset is obtained.

[0057] When regenerating the fluid domain mesh control parameters for the updated topology dataset, the updated pipe segment length, pipe diameter, elbow curvature, and node elevation are extracted from the pipe segment attribute table of the updated topology dataset. The pipe segment length field remains unchanged, the pipe diameter field uses candidate pipe diameter values ​​from the pipe diameter adjustment amount, the elbow curvature field uses candidate elbow curvature values ​​from the elbow curvature adjustment amount, and the node elevation is extracted from the node coordinate set of the updated topology dataset using the node Z coordinate. The updated pipe segment length, updated pipe diameter, updated elbow curvature, and node elevation are combined according to the pipe segment number to generate the updated pipe segment mesh base size and the updated local refinement mesh size. The updated pipe segment mesh base size is set to one-tenth of the updated pipe diameter. The updated local refinement mesh size is calculated according to the local refinement mesh feature size formula in the above embodiment. During calculation, the pipe segment diameter is the updated pipe diameter, the elbow curvature radius is the elbow curvature radius corresponding to the updated elbow curvature adjustment amount, and the mesh density adjustment factor is set in the same way as in the above embodiment. The updated fluid domain mesh control parameters are obtained.

[0058] When solving for the pressure field, velocity field, and particle concentration field using the regenerated fluid domain mesh control parameters to obtain the updated flow field dataset, the process of establishing the gas-solid two-phase simulation domain in the previous embodiment is followed. The updated pipe segment mesh and the updated locally refined mesh are regenerated using the updated fluid domain mesh control parameters. The original inlet particle phase grouping, the original inlet velocity boundary, and the original outlet pressure boundary are written into the updated pipe segment mesh. The gas phase pressure-velocity coupling field and the particle phase concentration transport field are alternately solved on the updated pipe segment mesh to obtain updated node pressure data, updated pipe segment velocity data, and updated particle deposition flux data. The updated node pressure data, updated pipe segment velocity data, and updated particle deposition flux data are encapsulated according to pipe segment number to obtain the updated flow field dataset.

[0059] By comparing the corresponding pipe segment data in the updated flow field dataset and the pipe segment-by-pipe flow field dataset, when the non-converged pipe segment number is obtained, the pipe segment-by-pipe flow field dataset is the flow field dataset obtained in the first solution of the above embodiment. For each pipe segment number, the relative deviation between the pipe segment velocity data in the updated flow field dataset and the pipe segment velocity data in the pipe segment-by-pipe flow field dataset for the s-th pipe segment is calculated. The relative deviation is calculated using the following formula:

[0060]

[0061] in, This represents the relative deviation of the s-th pipe segment, and is a dimensionless value. This represents the average gas phase velocity within the s-th pipe segment in the updated flow field dataset, in meters per second. This represents the average gas phase velocity within the s-th pipe segment in the pipe-by-pipe flow field dataset, in meters per second. Compared to the non-convergence threshold, the non-convergence threshold is set to 0.01, based on the following criteria: when the relative deviation of the average gas velocity within the s-th pipe segment exceeds 1%, it indicates that the flow field in the s-th pipe segment is continuously changing due to updates in pipe diameter, bend curvature, or confluence angle, and the s-th pipe segment has not yet entered a convergence state; when the relative deviation does not exceed 1%, the change in the flow field of the s-th pipe segment is within an acceptable range. If the s-th pipe segment... If the value is greater than 0.01, the pipe segment number of the s-th pipe segment is written into the set of non-converged pipe segment numbers. Each element in the set of non-converged pipe segment numbers is the pipe segment number of the non-converged pipe segment.

[0062] When writing back the pipe diameter adjustment, elbow curvature adjustment, and manifold angle adjustment based on the non-converged pipe segment number to form an iterative adjustment record, if the non-converged pipe segment number set is not empty, for each non-converged pipe segment in the non-converged pipe segment number set, read the current pipe diameter, elbow curvature, and manifold inlet angle of the non-converged pipe segment from the pipe segment attribute table of the currently updated topology dataset. For each non-converged pipe segment, based on the current pipe diameter value, the pipe diameter write-back value for the next iteration is generated according to the pipe diameter adjustment direction. If the average gas velocity within the non-converged pipe segment in the updated flow field dataset is lower than the minimum delivery velocity of the non-converged pipe segment, the pipe diameter of the non-converged pipe segment is reduced by one level from the current value. If the average gas velocity within the non-converged pipe segment in the updated flow field dataset is higher than the minimum delivery velocity of the non-converged pipe segment and the pressure drop of the non-converged pipe segment is higher than the allowable pressure drop of the non-converged pipe segment, the pipe diameter of the non-converged pipe segment is increased by one level from the current value. The scaling factor corresponding to increasing or decreasing the pipe diameter by one level is 0.9 or 1.1. The scaling factor of 0.9 and 1.1 is set based on the fact that adjusting the pipe diameter by 10% based on the previous round can produce a identifiable change in velocity and pressure drop within the pipe segment, while avoiding excessive adjustment that would cause a jump in the flow field solution results of the next round. For each non-converging pipe segment, based on the current value of the elbow curvature, the elbow curvature write-back value for the next iteration is generated according to the elbow curvature adjustment direction. If the local pressure drop of the elbow corresponding to the non-converging pipe segment in the updated flow field dataset exceeds the allowable local pressure drop of the non-converging pipe segment, the elbow curvature of the non-converging pipe segment is multiplied by 1.1 based on the following: increasing the elbow curvature is equivalent to decreasing the elbow curvature radius, which increases the local resistance of the elbow, but here it is used to redistribute the local pressure drop so that the flow field gradually approaches convergence in subsequent iterations. For each non-converging pipe segment, based on the current value of the manifold inlet angle, the next iteration's manifold angle write-back value is generated according to the manifold angle adjustment direction. If the local pressure drop of the manifold corresponding to the non-converging pipe segment in the updated flow field dataset exceeds the allowable local pressure drop of the non-converging pipe segment, the manifold inlet angle is multiplied by 0.95 based on the current value. The rationale is that reducing the manifold inlet angle can reduce airflow impact and local pressure drop at the confluence. The above pipe diameter write-back value, elbow curvature write-back value, and manifold angle write-back value are used as the new pipe diameter, elbow curvature, and manifold inlet angle field values ​​for the current updated topology dataset, forming the pipe diameter adjustment amount, elbow curvature adjustment amount, and manifold angle adjustment amount for the next iteration. The set of non-converging pipe segment numbers for this iteration is recorded to form an iteration adjustment record. The iteration adjustment record includes the iteration round number, the set of non-converging pipe segment numbers, the pipe diameter write-back value, the elbow curvature write-back value, and the manifold angle write-back value.

[0063] When the non-converged pipe segment number in the iterative adjustment record is empty, the updated topology dataset and the updated flow field dataset are merged to obtain the pipeline design parameter set. If the set of non-converged pipe segment numbers is empty, it means that the corresponding pipe segment data in both the updated flow field dataset and the pipe segment-by-pipe flow field dataset meet the convergence condition, i.e., the relative deviation of all pipe segments is zero. If all values ​​are less than or equal to the non-convergence threshold of 0.01, iteration stops. When merging the updated topology dataset and the updated flow field dataset, the pipe segment number is used as the association key. The pipe segment length, pipe diameter, elbow curvature, and manifold inlet angle fields from the pipe segment attribute table of the updated topology dataset are merged with the updated node pressure data, updated pipe segment velocity data, and updated particle deposition flux data from the updated flow field dataset to form the network design parameter set. The network design parameter set includes the pipe segment number, upstream node number, downstream node number, pipe segment length, pipe diameter, elbow curvature, manifold inlet angle, updated node pressure data, updated pipe segment velocity data, and updated particle deposition flux data. As long as the set of non-converged pipe segment numbers in the iteration adjustment record is not empty, the iterative process of regenerating the fluid domain mesh control parameters, solving the updated flow field dataset, and comparing convergence and write-back continues until the set of non-converged pipe segment numbers is empty.

[0064] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A method for optimizing the design of dust removal pipeline networks based on fluid simulation, characterized in that, include: Obtain the geometric and operating parameters of the dust removal pipeline network, and generate a pipeline network topology dataset with flow direction labels according to pipe segments, nodes, elbows, diameter changes, manifolds and suction hoods; Based on the dust particle size distribution and gas phase flow boundary in the pipeline topology dataset with flow direction identification, a gas-solid two-phase simulation domain is established. The pressure field, velocity field and particle concentration field are solved in the gas-solid two-phase simulation domain to obtain the flow field dataset for each pipe segment. Based on the wall shear data, particle concentration data, and local pressure drop data in the pipe-by-pipe flow field dataset, the pipe-by-pipe deposition coefficient and friction loss coefficient are generated. The pipe section deposition coefficient and the friction loss coefficient are input into the joint optimization process of pipe diameter and elbow curvature. The total pressure drop constraint and the minimum delivery speed constraint are used for screening, and the pipe diameter adjustment amount, elbow curvature adjustment amount and confluence angle adjustment amount are output. The pipeline topology dataset with flow direction identifier is updated according to the pipe diameter adjustment, the elbow curvature adjustment, and the confluence angle adjustment, and the pressure field, velocity field, and particle concentration field are solved repeatedly to obtain the pipeline design parameter set.

2. The dust removal pipeline network optimization design method based on fluid simulation according to claim 1, characterized in that, The process of acquiring the geometric and operating parameters of the dust removal pipeline network, and generating a pipeline network topology dataset with flow direction indicators according to pipe sections, nodes, elbows, diameter changes, manifolds, and suction hoods, includes: By performing endpoint snapping on the centerline of the pipe segment in the geometric parameters and operating parameters of the dust removal pipeline network, a set of node coordinates is obtained; Write the pipe segment number, upstream node number and downstream node number according to the connection order in the node coordinate set to obtain a directed connection table; Write the elbow angle, diameter ratio and manifold inlet angle from the directed connection table into the corresponding pipe segment attribute field to obtain the pipe segment attribute table. Perform connectivity verification on the pipe segment attribute table and write the isolated pipe segment number and reverse connection number into the topology queue to be corrected; The directed connection table is updated according to the topology queue to be corrected, and the pipeline topology dataset with flow direction identifier is generated.

3. The dust removal pipeline network optimization design method based on fluid simulation according to claim 1, characterized in that, The process involves establishing a gas-solid two-phase simulation domain based on the dust particle size distribution and gas phase flow boundary in the pipeline topology dataset with flow direction indicators. The pressure field, velocity field, and particle concentration field are then solved within this simulation domain to obtain a pipe-segment flow field dataset, including: Pipe segment length, pipe diameter, elbow curvature, and node elevation are extracted from the pipeline topology dataset with flow direction identifiers to form fluid domain grid control parameters; The dust particle size distribution is mapped to inlet particle phase grouping, and the gas phase flow boundary is mapped to inlet velocity boundary and outlet pressure boundary; The pipe segment mesh and the local refinement mesh are generated according to the fluid domain mesh control parameters, and the inlet particle phase grouping, the inlet velocity boundary and the outlet pressure boundary are written into the pipe segment mesh; The gas phase pressure-velocity coupled field and the particle phase concentration transport field are alternately solved on the pipe segment grid to obtain nodal pressure data, pipe segment velocity data and particle deposition flux data; The node pressure data, the pipe segment velocity data, and the particle deposition flux data are packaged according to the pipe segment number to obtain the pipe segment flow field dataset.

4. The dust removal pipeline network optimization design method based on fluid simulation according to claim 1, characterized in that, The process of generating the pipe segment deposition coefficient and friction loss coefficient based on the wall shear data, particle concentration data, and local pressure drop data in the pipe segment flow field dataset includes: The wall shear data in the pipe segment flow field dataset is divided into shear intervals according to the circumferential position of the pipe segment to obtain a set of low-position shear segments. The particle size groups in the particle concentration data are superimposed with the low-position shear segment set to obtain the grouped deposition weights. The proportion of local pressure drop is obtained based on the elbow pressure drop and reducer pressure drop in the local pressure drop data. The grouping deposition weights and the local pressure drop ratios are merged according to the pipe segment number to generate the pipe segment deposition coefficient; The friction loss coefficient is generated by correlating the friction loss in the pipe segment flow field dataset with the pipe segment length, pipe diameter, and velocity data.

5. The dust removal pipeline network optimization design method based on fluid simulation according to claim 1, characterized in that, The process of inputting the pipe segment deposition coefficient and the friction loss coefficient into the joint optimization process of pipe diameter and elbow curvature, and filtering by total pressure drop constraints and minimum delivery speed constraints, outputs pipe diameter adjustment, elbow curvature adjustment, and confluence angle adjustment, including: The pipe segments are numbered according to their deposition coefficients to obtain a sequence of high-risk deposition pipe segments. Pipe segments are selected from the sequence of high-risk deposition pipe segments, and discrete combination sets of candidate values ​​for pipe diameter, elbow curvature, and confluence angle are established. Write the friction loss coefficient into each combination in the discrete combination set to obtain the combined energy consumption data; Perform total pressure drop summation and minimum transmission speed verification on the combined energy consumption data, and retain the combinations that satisfy the total pressure drop constraint and the minimum transmission speed constraint; Select the combination with the smallest total pressure drop and the largest number of high-risk pipe segment sequences covered from the retained combinations, and output the pipe diameter adjustment, the elbow curvature adjustment, and the confluence angle adjustment.

6. The dust removal pipeline network optimization design method based on fluid simulation according to claim 1, characterized in that, The obtained pipeline design parameter set includes: Write the pipe diameter adjustment amount into the pipe segment diameter field of the pipe network topology dataset with flow direction identification, write the elbow curvature adjustment amount into the elbow curvature field, and write the confluence angle adjustment amount into the confluence inlet angle field to obtain the updated topology dataset. The fluid domain mesh control parameters are regenerated on the updated topology dataset, and the pressure field, velocity field, and particle concentration field are solved on the regenerated fluid domain mesh control parameters to obtain the updated flow field dataset. Compare the updated flow field dataset with the corresponding pipe segment data in the pipe segment flow field dataset to obtain the unconverged pipe segment number; Based on the non-converged pipe segment number, the pipe diameter adjustment, the elbow curvature adjustment, and the confluence angle adjustment are written back to form an iterative adjustment record; When the number of the non-converged pipe segment in the iterative adjustment record is empty, the updated topology dataset and the updated flow field dataset are merged to obtain the set of pipe network design parameters.

7. The dust removal pipeline network optimization design method based on fluid simulation according to claim 2, characterized in that, The process of performing endpoint snapping on the centerline of the pipe segment in the geometric parameters and operating parameters of the dust removal pipeline network to obtain a set of node coordinates includes: Extract the first and last endpoints of the pipe segment centerline from the geometric parameters and operating parameters of the dust removal pipeline network, and generate endpoint cluster numbers according to coordinate tolerance; Merge the first and last endpoints of clusters belonging to the same endpoint cluster number into a common node and obtain the coordinates of the common node. Write the suction hood identifier and valve identifier into the branch nodes in the common node coordinates to obtain the node attribute table; The node coordinate set is generated based on the node attribute table and the pipe segment centerline.

8. The dust removal pipeline network optimization design method based on fluid simulation according to claim 3, characterized in that, The gas phase pressure-velocity coupled field and particle phase concentration-transport field are alternately solved on the pipe segment grid to obtain nodal pressure data, pipe segment velocity data, and particle deposition flux data, including: The gas phase momentum equation is solved on the pipe segment grid using a pressure implicit scheme to obtain initial velocity and initial pressure data; Substitute the initial velocity data into the particulate phase transport process to update the particle concentration data; Based on the particle concentration data, the source term in the gas phase momentum equation is corrected to obtain the corrected pressure data; The corrected pressure data and the initial velocity data are updated alternately to obtain the convergence node pressure data and the convergence segment velocity data. The wall deposition flux is obtained based on the convergence node pressure data and the convergence tube velocity data, and the particle deposition flux data is generated.

9. The dust removal pipeline network optimization design method based on fluid simulation according to claim 4, characterized in that, The step of fusing the grouped deposition weights and the local pressure drop ratios according to pipe segment numbers to generate the pipe segment deposition coefficient includes: Extract the weight of the largest and smallest grain size group from the grouped deposition weights according to the pipe segment number; The weight of the largest particle size group, the weight of the smallest particle size group, and the proportion of local pressure drop are written into the same pipe section record; Weighted sorting is performed on the records of the same pipe segment to obtain deposition sensitivity sorting values; The deposition sensitivity ranking values ​​are concatenated according to the pipe segment number to generate the pipe segment deposition coefficient.

10. The dust removal pipeline network optimization design method based on fluid simulation according to claim 5, characterized in that, The step of selecting the combination with the smallest total pressure drop and the largest coverage of high-risk deposition pipe segment sequences from the retained combinations, and outputting the pipe diameter adjustment, the elbow curvature adjustment, and the confluence angle adjustment includes: Read the total pressure drop data corresponding to each combination in the retained combinations and the coverage number of the high-risk deposition pipe segment sequence; The coverage number of the high-risk pipe segment sequence is generated by retaining the combination of coverage numbers; The combinations are sorted in ascending order of total pressure drop data and descending order of coverage quantity to obtain a combination sorting table; Select the first combination from the combination sorting table, and use the candidate values ​​of pipe diameter, elbow curvature and confluence angle in the first combination as the pipe diameter adjustment amount, the elbow curvature adjustment amount and the confluence angle adjustment amount, respectively.

Images