Adaptive simulation prediction method, system, medium and product for gas-solid two-phase flow

Abstract

The application discloses a kind of gas-solid two-phase flow self-adapting simulation prediction method, system, medium and product, belong to the field of digital simulation of fluid dynamics, the method is: obtaining industrial equipment geometry, particle and fluid data, constructs gas-solid two-phase flow CMPPIC model;Wherein, continuous phase is with Euler method, discrete phase is with Lagrange particle bag representation;Perform time advance iteration to termination condition, output result, wherein each round of iteration includes mapping particle bag data to Euler cell, obtaining particle statistical information and marking fixed state particle according to rules;Freeze fixed cell particle volume fraction, calculate porous medium parameter, reconstruct its continuous phase momentum conservation equation;Based on reconstruction and unreconstructed equation, update flow field, while updating the particle motion of unmarked fixed state cell. By implementing the application, the problem that calculation efficiency and physical precision are difficult to consider in simulating fixed particle structure in the prior art can be solved.

Description

Technical Field

[0001] This invention belongs to the field of digital simulation of fluid dynamics, and relates to an adaptive simulation prediction method, system, medium and product for gas-solid two-phase flow. Background Technology

[0002] In chemical, energy and environmental engineering, gas-solid two-phase flow is widely used in fixed-bed reactors, particulate filters, sedimentation layers and clogging structures. Its flow behavior is significantly affected by the particle packing morphology and pore structure, and numerical simulation methods are needed for prediction and optimization.

[0003] Existing techniques typically employ two approaches to simulate such problems: one is to directly use the CMPPIC method and force the particle velocity to zero, while still preserving the complete Lagrangian particle tracking, mapping, and time progression process; the other is to use traditional porous media models, introducing empirical Darcy or Forchheimer drag terms into the fluid control equations. The former introduces a large amount of unnecessary computational overhead when the particles are actually stationary, while the latter, due to its reliance on empirical parameters, struggles to accurately reflect particle size distribution, local porosity, and complex packing structures, resulting in insufficient physical accuracy. Furthermore, the two approaches are disconnected and cannot adaptively handle the evolution of particles from motion to stationary state within the same framework. Summary of the Invention

[0004] This application provides an adaptive simulation prediction method, system, medium, and product for gas-solid two-phase flow, which can solve the problem that existing technologies cannot simultaneously achieve both computational efficiency and physical accuracy when simulating fixed particle structures.

[0005] To achieve the above objectives, in a first aspect, the present invention provides an adaptive simulation and prediction method for gas-solid two-phase flow, comprising: Obtain the geometric structure data, particulate property data, and fluid boundary condition data of the industrial equipment to be simulated, and construct the corresponding gas-solid two-phase flow CMPPIC model; wherein, the continuous phase is represented by the Euler method in the gas-solid two-phase flow CMPPIC model; the discrete phase is represented by Lagrange particle packs in the gas-solid two-phase flow CMPPIC model. Based on the gas-solid two-phase flow CMPPIC model, a time-progressive iterative process is executed until a preset termination condition is reached, and numerical simulation results are output. Each round of time-progression iteration includes: The current Lagrange particle pack data is mapped to each Euler computing unit to obtain the particle statistics of each Euler computing unit. Combined with the preset statistical rules, it is determined and marked whether the particles in each Euler computing unit are in a fixed state. The particle volume fraction data corresponding to the Euler computing unit marked as fixed is frozen, and the corresponding porous medium parameters are calculated based on the frozen particle volume fraction data. Then, the momentum conservation equation of the continuous phase in the corresponding Euler computing unit is reconstructed. Based on the reconstructed momentum conservation equation and the remaining unreconstructed momentum conservation equation, the flow field update operation for the continuous phase is performed, and the particle phase motion update operation is performed for the Euler computational cells that are not marked as fixed states.

[0006] Compared with existing technologies, the embodiments of this application have the following beneficial effects: They acquire various data of the industrial equipment to be simulated and construct corresponding gas-solid two-phase flow CMPPIC models, starting from real engineering physics scenarios to ensure the physical fidelity of the modeling and the feasibility of the initial conditions; by mapping the current Lagrangian particle pack data to each Eulerian computational unit in each time-progression iteration to obtain particle statistical information, and combining this with preset statistical rules to judge and mark fixed states, adaptive identification of the macroscopic motion behavior of particles is achieved, avoiding the indiscriminate execution of high-cost Lagrangian calculations for all regions; by freezing the particle volume fraction data of the Eulerian computational units marked as fixed states, and calculating the porous medium parameters based on this frozen data to reconstruct the continuous phase momentum conservation equation of the corresponding unit, the local drag effect that originally relied on particle tracking is transformed into a continuous phase based on statistical information. The source term description preserves the spatial non-uniformity of particle packing structures while eliminating redundant calculations. By uniformly performing continuous phase flow field updates based on reconstructed and unreconstructed momentum conservation equations, and only performing particle phase motion updates on unlabeled elements, coupled solutions for the fixed and moving regions are achieved within the same time step, ensuring physical consistency and numerical stability across the entire flow field. The synergistic effect of these features eliminates the need for Lagrangian position / velocity time advance in fixed particle regions, significantly reducing computational load. Furthermore, the use of particle volume fractions inherent in the CMPPIC framework to construct a porous medium model with a clear physical origin avoids the inaccuracy of traditional empirical drag coefficients. Thus, computational efficiency and physical accuracy are simultaneously improved within a single numerical framework, ultimately solving the problem of balancing computational efficiency and physical accuracy in simulating fixed particle structures using existing techniques.

[0007] In some embodiments of the first aspect of this application, mapping the current Lagrange particle pack data to each Euler computing unit to obtain particle statistics information for each Euler computing unit includes: For each Lagrange particle bag, its volume fraction contribution weight to neighboring Euler computing units is calculated based on its location data, equivalent mass data, and particle density data. For each Euler computing unit, the volume fraction contribution weights of all Lagrange particle bags it receives are summed to obtain the particle volume fraction data of that Euler computing unit. For each Lagrange particle pack, velocity vector contribution is assigned to each neighboring cell based on its velocity vector data and the volume fraction contribution weight to the neighboring Euler computational cells; for each Euler computational cell, the average velocity vector data of the particles in the Euler computational cell is obtained by weighted averaging based on all the velocity vector contributions it receives. Based on the position data of each Lagrange particle bag in the current time step and the previous few time steps, calculate its displacement vector; for each Euler calculation unit, perform statistical averaging based on the particle bag displacement vector mapped to that unit to obtain the particle displacement change data of that unit. Obtain the current particle stress data of each Euler calculation unit, and use the particle volume fraction data, particle average velocity vector data, particle displacement change data, and particle stress data as particle statistical information.

[0008] Compared with existing technologies, the above embodiments have the following beneficial effects: By calculating the volume fraction contribution weight of each Lagrange particle pack to neighboring Eulerian computational units based on its position data, equivalent mass data, and particle density data, and accumulating all contribution weights for each unit to obtain particle volume fraction data, a conservative mapping from particle phase to Eulerian field is achieved, ensuring local mass conservation; by allocating velocity vector contributions based on particle velocity vector data and its volume fraction contribution weights, and performing a weighted average for each unit to obtain particle average velocity vector data, the macroscopic motion trend of particles at the unit scale is accurately characterized; by calculating displacement vectors based on the position data of particle packs at the current and previous time steps, and statistically averaging the displacement vectors mapped to units to obtain particle displacement change data, the temporal evolution characteristics of particle motion are effectively captured; by acquiring the current particle stress data of each Eulerian computational unit, and using particle volume fraction data, particle average velocity vector data, particle displacement change data, and particle stress data together as particle statistical information, a multi-dimensional and highly complete input basis is provided for subsequent state discrimination, enhancing the robustness and applicability of fixed state identification.

[0009] In some embodiments of the first aspect of this application, the step of determining and marking whether the particles within each Euler computational unit are in a fixed state, in accordance with preset statistical rules, includes: If any of the following conditions are met within a certain Euler calculation unit, the particles within that Euler calculation unit are determined to be in a fixed state, and fixed region marker data is generated: the average particle velocity data is lower than a preset velocity threshold, the particle volume fraction data is greater than a preset packing volume fraction threshold, the particle displacement change data is lower than a preset displacement threshold, and the difference between the fluid drag force and particle stress on the particles is less than a preset equilibrium tolerance.

[0010] Compared with existing technologies, the above embodiments have the following beneficial effects: By setting a fixed state when any of the following conditions are met: the average velocity data of particles in a certain Euler calculation unit is lower than a preset velocity threshold, the volume fraction data of particles is greater than a preset packing volume fraction threshold, the change in particle displacement is lower than a preset displacement threshold, or the difference between the fluid drag force and the particle stress is less than a preset equilibrium tolerance, the discrimination mechanism can cover various physical "fixed" situations such as low-speed stillness, high packing density, displacement stagnation, and force balance, avoiding the risk of misjudgment under complex working conditions by a single criterion; at the same time, by adopting the logic of triggering a flag when "any condition is met", users are given the freedom to flexibly configure the combination of criteria, adapting to the different needs of different engineering scenarios for the definition of "fixed", and improving the versatility and engineering adaptability of the method.

[0011] In some embodiments of the first aspect of this application, calculating the corresponding porous media parameters based on the frozen particle volume fraction data includes: Based on the particle volume fraction of each Euler calculation unit, the corresponding equivalent porosity data is calculated, and combined with the corresponding particle specific surface area data, the permeability data is calculated. Based on the permeability data and the fluid viscosity and density data in the current gas-solid two-phase flow CMPPIC model, the Darcy linear drag coefficient and the Fochheimer nonlinear drag coefficient are calculated.

[0012] Compared with existing technologies, the above embodiments have the following advantages: By calculating the equivalent porosity data based on the particle volume fraction of each Euler computing unit and combining it with the particle specific surface area data to calculate the permeability data, a direct mapping relationship from the microscopic packing state of particles to macroscopic transport characteristics is established; furthermore, by combining the permeability data, fluid viscosity data, and fluid density data to calculate the Darcy linear drag coefficient data and the Fochheimer nonlinear drag coefficient data, the porous media parameters not only reflect the viscous dissipation under low-speed laminar flow, but also characterize the inertial additional drag in high-speed flow, thereby constructing a full-condition drag model applicable to a wide Reynolds number range, which is significantly better than the traditional porous media method that relies only on a single empirical coefficient.

[0013] In some embodiments of the first aspect of this application, the reconstruction of the momentum conservation equations corresponding to the continuous phase within the Euler computational unit includes: Multiply the Darcy linear drag coefficient data with the fluid velocity data in the current gas-solid two-phase flow CMPPIC model to obtain the linear drag component data; Multiply the modulus of the Fochheimer nonlinear drag coefficient data, fluid density data, and fluid velocity data to obtain the nonlinear drag component data. The linear drag component data and the nonlinear drag component data are superimposed to construct a porous medium drag source term. This porous medium drag source term is then introduced as a source term into the momentum conservation equation of the continuous phase within the corresponding Euler computational unit to complete the reconstruction.

[0014] Compared with existing technologies, the above embodiments have the following beneficial effects: by multiplying Darcy linear drag coefficient data with fluid velocity data to obtain linear drag component data, the momentum dissipation intensity in viscous-dominated flow is quantified; by multiplying the modulus of Fochheimer nonlinear drag coefficient data, fluid density data, and fluid velocity data to obtain nonlinear drag component data, the additional pressure drop caused by vortex shedding and inertial effects at high flow velocities is accurately described; by superimposing linear and nonlinear drag component data to construct a porous media drag source term, and introducing it as a source term into the momentum conservation equation of the continuous phase in the corresponding Euler computational unit to complete the reconstruction, the fluid control equation of the fixed particle region can continuously and smoothly reflect the complex flow drag characteristics, while maintaining physical compatibility with the original CMPPIC phase coupling mechanism, avoiding numerical discontinuities or oscillations caused by model switching.

[0015] In some embodiments of the first aspect of this application, when determining and marking whether the particles in each Euler computational unit are in a fixed state in each round, if the Euler computational unit previously marked as fixed no longer meets the fixed condition in the current determination, its fixed mark is canceled and its particle phase motion update operation is resumed.

[0016] Compared with existing technologies, the above embodiments have the following beneficial effects: by re-evaluating whether the Euler computational units previously marked as fixed in each round of judgment still meet the fixed conditions, and canceling the fixed marking and resuming the particle phase motion update operation when the conditions are no longer met, the computational units can automatically switch from the porous medium model back to the Lagrange particle model when the particle state undergoes dynamic evolution (such as deposition unblocking, disturbance resuspension), realizing bidirectional adaptive conversion between fixed and moving states; this mechanism breaks through the limitation of static model fragmentation in traditional methods, enabling the numerical framework to continuously describe the entire process of particles from fixed to moving, significantly enhancing the applicability and predictive ability of the method in unsteady, evolutionary engineering problems.

[0017] Secondly, the present invention also provides an adaptive simulation and prediction system for gas-solid two-phase flow, comprising: a data acquisition module and an iterative output module; The data acquisition module is used to acquire the geometric structure data, particulate property data, and fluid boundary condition data of the industrial equipment to be simulated, and to construct the corresponding gas-solid two-phase flow CMPPIC model. The continuous phase is represented by the Euler method in the gas-solid two-phase flow CMPPIC model, and the discrete phase is represented by Lagrange particle packs in the gas-solid two-phase flow CMPPIC model. The iterative output module is used to execute a time-progressive iterative process based on the gas-solid two-phase flow CMPPIC model until a preset termination condition is reached, and output the numerical simulation results. Each round of time-progression iteration includes: The current Lagrange particle pack data is mapped to each Euler computing unit to obtain the particle statistics of each Euler computing unit. Combined with the preset statistical rules, it is determined and marked whether the particles in each Euler computing unit are in a fixed state. The particle volume fraction data corresponding to the Euler computing unit marked as fixed is frozen, and the corresponding porous medium parameters are calculated based on the frozen particle volume fraction data. Then, the momentum conservation equation of the continuous phase in the corresponding Euler computing unit is reconstructed. Based on the reconstructed momentum conservation equation and the remaining unreconstructed momentum conservation equation, the flow field update operation for the continuous phase is performed, and the particle phase motion update operation is performed for the Euler computational cells that are not marked as fixed states.

[0018] Compared with existing technologies, the above embodiments of this application have the following beneficial effects: They acquire various data of the industrial equipment to be simulated and construct corresponding gas-solid two-phase flow CMPPIC models, starting from real engineering physics scenarios to ensure the physical fidelity of the model and the feasibility of the initial conditions; by mapping the current Lagrangian particle pack data to each Eulerian computational unit in each round of time-progression iteration to obtain particle statistical information, and combining this with preset statistical rules to judge and mark fixed states, adaptive identification of the macroscopic motion behavior of particles is achieved, avoiding the indiscriminate execution of high-cost Lagrangian calculations for all regions; by freezing the particle volume fraction data of the Eulerian computational units marked as fixed states, and calculating the porous medium parameters based on this frozen data to reconstruct the continuous phase momentum conservation equation of the corresponding unit, the local drag effect that originally relied on particle tracking is transformed into a continuous phase based on statistical information. The source term description preserves the spatial non-uniformity of particle packing structures while eliminating redundant calculations. By uniformly performing continuous phase flow field updates based on reconstructed and unreconstructed momentum conservation equations, and only performing particle phase motion updates on unlabeled elements, coupled solutions for the fixed and moving regions are achieved within the same time step, ensuring physical consistency and numerical stability across the entire flow field. The synergistic effect of these features eliminates the need for Lagrangian position / velocity time advance in fixed particle regions, significantly reducing computational load. Furthermore, the use of particle volume fractions inherent in the CMPPIC framework to construct a porous medium model with a clear physical source avoids the inaccuracy of traditional empirical drag coefficients. Thus, computational efficiency and physical accuracy are simultaneously improved within a single numerical framework, ultimately solving the problem of balancing computational efficiency and physical accuracy in simulating fixed particle structures using existing techniques.

[0019] In some embodiments of the second aspect of this application, the iterative output module includes: a first calculation unit, a second calculation unit, a third calculation unit, and a statistical unit; The first calculation unit is used to calculate the volume fraction contribution weight of each Lagrange particle bag to the adjacent Euler calculation unit based on its location data, equivalent mass data and particle density data; and to accumulate the volume fraction contribution weights of all Lagrange particle bags received by each Euler calculation unit to obtain the particle volume fraction data of the Euler calculation unit. The second calculation unit is used to allocate velocity vector contributions to each neighboring unit for each Lagrange particle pack based on its velocity vector data and the volume fraction contribution weight to the neighboring Euler calculation units; and to obtain the average particle velocity vector data of the Euler calculation unit by performing a weighted average of all the velocity vector contributions it receives for each Euler calculation unit. The third calculation unit is used to calculate the displacement vector of each Lagrange particle bag based on the position data of each Lagrange particle bag in the current time step and the previous several time steps; for each Euler calculation unit, the particle displacement change data of the unit is obtained by statistical averaging based on the particle bag displacement vector mapped to the unit. The statistical unit is used to acquire the current particle stress data of each Euler calculation unit, and to use the particle volume fraction data, particle average velocity vector data, particle displacement change data and particle stress data as particle statistical information.

[0020] Compared with existing technologies, the above embodiments have the following beneficial effects: By calculating the volume fraction contribution weight of each Lagrange particle pack to neighboring Eulerian computational units based on its position data, equivalent mass data, and particle density data, and accumulating all contribution weights for each unit to obtain particle volume fraction data, a conservative mapping from particle phase to Eulerian field is achieved, ensuring local mass conservation; by allocating velocity vector contributions based on particle velocity vector data and its volume fraction contribution weights, and performing a weighted average for each unit to obtain particle average velocity vector data, the macroscopic motion trend of particles at the unit scale is accurately characterized; by calculating displacement vectors based on the position data of particle packs at the current and previous time steps, and statistically averaging the displacement vectors mapped to units to obtain particle displacement change data, the temporal evolution characteristics of particle motion are effectively captured; by acquiring the current particle stress data of each Eulerian computational unit, and using particle volume fraction data, particle average velocity vector data, particle displacement change data, and particle stress data together as particle statistical information, a multi-dimensional and highly complete input basis is provided for subsequent state discrimination, enhancing the robustness and applicability of fixed state identification.

[0021] Thirdly, the present invention also provides a computer program product, including a computer program or instructions, which, when executed, implement any one of the adaptive simulation and prediction methods for gas-solid two-phase flow of the present invention.

[0022] Fourthly, embodiments of this application also provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements any one of the adaptive simulation and prediction methods for gas-solid two-phase flow of the present invention. Attached Figure Description

[0023] Figure 1 This is a flowchart illustrating an adaptive simulation prediction method for gas-solid two-phase flow provided in some embodiments of the present invention.

[0024] Figure 2 This is a schematic diagram of the structure of an adaptive simulation and prediction system for gas-solid two-phase flow provided in some embodiments of the present invention.

[0025] Figure 3 This is a velocity cloud map provided in some embodiments of the present invention.

[0026] Figure 4 This is a pressure cloud map provided in some embodiments of the present invention.

[0027] Figure 5 This is a velocity curve provided in some embodiments of the present invention.

[0028] Figure 6 This is a pressure curve provided in some embodiments of the present invention. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] Example 1: Please refer to Figure 1 To address the problem of balancing computational efficiency and physical accuracy in simulating fixed particle structures using existing technologies, an embodiment of the present invention provides an adaptive simulation and prediction method for gas-solid two-phase flow, comprising steps S1 to S2: Step S1: Obtain the geometric structure data, particulate property data, and fluid boundary condition data of the industrial equipment to be simulated, and construct the corresponding gas-solid two-phase flow CMPPIC model; wherein, the continuous phase is represented by the Euler method in the gas-solid two-phase flow CMPPIC model; the discrete phase is represented by Lagrange particle packs in the gas-solid two-phase flow CMPPIC model. Step S2: Based on the gas-solid two-phase flow CMPPIC model, execute the time-progression iterative process until the preset termination condition is reached, and output the numerical simulation results; Each round of time-progression iteration includes: The current Lagrange particle pack data is mapped to each Euler computing unit to obtain the particle statistics of each Euler computing unit. Combined with the preset statistical rules, it is determined and marked whether the particles in each Euler computing unit are in a fixed state. The particle volume fraction data corresponding to the Euler computing unit marked as fixed is frozen, and the corresponding porous medium parameters are calculated based on the frozen particle volume fraction data. Then, the momentum conservation equation of the continuous phase in the corresponding Euler computing unit is reconstructed. Based on the reconstructed momentum conservation equation and the remaining unreconstructed momentum conservation equation, the flow field update operation for the continuous phase is performed, and the particle phase motion update operation is performed for the Euler computational cells that are not marked as fixed states.

[0031] In practical implementation, the Euler-Lagrange method is used to establish the model. The continuous phase (fluid phase) is discretized into several computational units within the computational domain using the Euler method, and the mass conservation equation, momentum conservation equation, and energy conservation equation for the continuous phase are solved. The discrete phase (particle phase) is described using the CMPPIC method, which represents a large number of actual physical particles as equivalent Lagrange parcels. Each parcel contains information on particle position, velocity, equivalent mass, and volume fraction contribution. The initial state of the particles can be set as a movable state according to the engineering problem. The particle phase and the continuous phase are coupled through interphase force terms, providing a unified computational framework for subsequent particle state discrimination and equivalent modeling.

[0032] Next, because directly applying CMPPIC in existing technologies is computationally inefficient, using CMPPIC's Lagrangian particle tracking, parcel-cell mapping, and time progression processes when particles are actually in a fixed state introduces a large amount of redundant computation, significantly increasing computational costs without substantially improving the accuracy of the results. Therefore, in this scheme, within each time step, the Lagrangian particle information is projected onto the Eulerian computational cells through a particle-mesh mapping process to obtain particle statistical information within each computational cell, including but not limited to: 1) Volume fraction of particles within a unit.

[0033] 2) Particle equivalent density distribution.

[0034] 3) Statistical momentum and velocity field of particles.

[0035] 4) The amount of momentum exchanged between particles and the continuous phase.

[0036] The particle statistics are stored in a continuous field format to characterize the spatial distribution of particles and serve as the basis for determining whether particles are in a fixed state and for constructing an equivalent model of porous media, as detailed below: Furthermore, in step S2, the Lagrange particle pack data mapping can be implemented through the following preferred embodiments, including steps S21-S24: S21: For each Lagrange particle bag, calculate its volume fraction contribution weight to neighboring Euler computing units based on its location data, equivalent mass data, and particle density data; for each Euler computing unit, sum the volume fraction contribution weights of all Lagrange particle bags it receives to obtain the particle volume fraction data of that Euler computing unit. S22: For each Lagrange particle pack, allocate velocity vector contribution to each neighboring cell based on its velocity vector data and the volume fraction contribution weight to the neighboring Euler computational cells; for each Euler computational cell, perform a weighted average based on all the velocity vector contributions it receives to obtain the average velocity vector data of the particles in that Euler computational cell. S23: Calculate the displacement vector of each Lagrange particle bag based on its position data in the current time step and the previous few time steps; for each Euler calculation unit, perform a statistical average based on the particle bag displacement vector mapped to that unit to obtain the particle displacement change data of that unit. S24: Obtain the current particle stress data of each Euler calculation unit, and use the particle volume fraction data, particle average velocity vector data, particle displacement change data and particle stress data as particle statistical information.

[0037] In this preferred embodiment, the volume fraction contribution weight of each Lagrange particle pack to neighboring Eulerian computational units is calculated based on its position data, equivalent mass data, and particle density data. The volume fraction data of the particles is obtained by summing all contribution weights across all units, thus achieving a conservative mapping from the particle phase to the Eulerian field and ensuring local mass conservation. The average velocity vector data of the particles is obtained by allocating velocity vector contributions based on particle velocity vector data and their volume fraction contribution weights, and by performing a weighted average across all units, accurately depicting the macroscopic motion trend of particles at the unit scale. The displacement vector is calculated based on the position data of the particle pack at the current and previous time steps, and the displacement vector mapped to the units is statistically averaged to obtain particle displacement change data, effectively capturing the temporal evolution characteristics of particle motion. Finally, the current particle stress data of each Eulerian computational unit is obtained, and the particle volume fraction data, average particle velocity vector data, particle displacement change data, and particle stress data are used together as particle statistical information, providing a multi-dimensional and highly complete input basis for subsequent state discrimination, enhancing the robustness and applicability of fixed-state identification.

[0038] Based on the particle statistics obtained in the preceding steps, the particle state within each computational unit is determined to ascertain whether the particles are in a fixed or approximately fixed state. This determination can be achieved using one or more of the following criteria: 1) The average velocity of particles within the unit is lower than the preset velocity threshold.

[0039] 2) The volume fraction of particles within the unit reaches or is close to the maximum packing volume fraction.

[0040] 3) The displacement of particles within a unit is lower than the preset displacement threshold over multiple consecutive time steps.

[0041] 4) The fluid force and particle stress on the particles reach an approximate equilibrium state.

[0042] When a computational unit meets the fixed criterion, it is marked as a fixed particle region; otherwise, it remains a movable particle region. For computational units marked as fixed particle regions, the Lagrangian motion calculations of particle packs within that region are stopped, including particle position updates and velocity time progression; the volume fraction distribution and statistical information of particles within that region are preserved to maintain stability over time; fixed particles are treated as solid skeletal structures that do not participate in macroscopic motion, and their effects on the fluid are no longer reflected through the particle momentum equation. This effectively eliminates unnecessary Lagrangian calculations within fixed particle regions, creating conditions for subsequent equivalent modeling of porous media. Specifically: Furthermore, in step S2, determining whether it is a fixed state can be achieved through the following preferred embodiment, including step S23: S23: If any of the following conditions are met in a certain Euler calculation unit, the particles in the Euler calculation unit are determined to be in a fixed state, and fixed area marker data is generated: the average velocity data of the particles is lower than the preset velocity threshold, the volume fraction data of the particles is greater than the preset packing volume fraction threshold, the change in particle displacement data is lower than the preset displacement threshold, and the difference between the fluid drag force and the particle stress is less than the preset equilibrium tolerance.

[0043] In this preferred embodiment, a fixed state is determined when any of the following conditions are met: the average velocity of particles within a certain Euler calculation unit is lower than a preset velocity threshold, the volume fraction of particles is greater than a preset packing volume fraction threshold, the change in particle displacement is lower than a preset displacement threshold, or the difference between the fluid drag force and the stress on the particles is less than a preset equilibrium tolerance. This allows the discrimination mechanism to cover various physically meaningful "fixed" situations, such as low-speed stillness, high packing density, displacement stagnation, and force balance, avoiding the risk of misjudgment under complex working conditions by a single criterion. At the same time, by adopting the logic of triggering a flag when "any condition is met", users are given the freedom to flexibly configure the combination of criteria, adapting to the different needs of "fixed" definitions in different engineering scenarios, and improving the versatility and engineering adaptability of the method.

[0044] Within a fixed particle region, based on frozen particle statistics, equivalent parameters of the porous medium are constructed, including: 1) Equivalent porosity, which is directly calculated from the particle volume fraction.

[0045] 2) Equivalent drag coefficient, used to describe the hindering effect of fixed particle structure on fluid momentum.

[0046] 3) Nonlinear drag correction term, used to characterize the inertial effect under high flow velocity or high Reynolds number conditions.

[0047] 4) Construct anisotropic drag parameters when needed to reflect the directional characteristics of particle packing.

[0048] The porous media parameters are not determined by external experience, but are directly derived from particle statistics within the CMPPIC framework, thus ensuring physical consistency. Specifically: Furthermore, the calculation of porous medium parameters can be implemented through the following preferred embodiments, including steps S24-S25: S24: Calculate the corresponding equivalent porosity data based on the particle volume fraction of each Euler calculation unit, and calculate the permeability data by combining the corresponding particle specific surface area data. S25: Based on the permeability data and the fluid viscosity and density data in the current gas-solid two-phase flow CMPPIC model, calculate the Darcy linear drag coefficient data and the Fochheimer nonlinear drag coefficient data.

[0049] In this preferred embodiment, by calculating the equivalent porosity data based on the particle volume fraction of each Euler computational unit and combining it with the particle specific surface area data to calculate the permeability data, a direct mapping relationship from the microscopic packing state of particles to macroscopic transport characteristics is established. Furthermore, by combining the permeability data, fluid viscosity data, and fluid density data to calculate the Darcy linear drag coefficient data and the Fochheimer nonlinear drag coefficient data, the porous medium parameters not only reflect the viscous dissipation under low-speed laminar flow but also characterize the inertial additional drag under high-speed flow. This constructs a full-condition drag model applicable to a wide Reynolds number range, which is significantly better than the traditional porous medium method that relies solely on a single empirical coefficient.

[0050] After obtaining the porous medium parameters, the continuous phase control equations are modified within the fixed particle region. A porous medium resistance source term is introduced into the continuous phase momentum equation, constraining the fluid flow behavior within this region to the fixed particle structure. The resistance source term maintains physical consistency with the interphase momentum exchange term in the original CMPPIC, but no longer relies on Lagrangian particle motion calculations, thus achieving a continuous description of the influence of the fixed particle structure. Specifically: Furthermore, the reconstructed momentum conservation equation can be implemented through the following preferred embodiments, including steps S26-S28: S26: Multiply the Darcy linear drag coefficient data with the fluid velocity data in the current gas-solid two-phase flow CMPPIC model to obtain the linear drag component data; S27: Multiply the modulus of the Fochheimer nonlinear drag coefficient data, fluid density data, and fluid velocity data to obtain the nonlinear drag component data; S28: Superimpose the linear resistance component data with the nonlinear resistance component data to construct a porous medium resistance source term, and introduce the porous medium resistance source term as a source term into the momentum conservation equation of the continuous phase in the corresponding Euler calculation unit to complete the reconstruction.

[0051] In this preferred embodiment, the linear drag component data is obtained by multiplying the Darcy linear drag coefficient data with the fluid velocity data, which quantifies the momentum dissipation intensity in viscous-dominated flow. The nonlinear drag component data is obtained by multiplying the modulus of the Fochheimer nonlinear drag coefficient data, fluid density data, and fluid velocity data, which accurately describes the additional pressure drop caused by vortex shedding and inertial effects at high flow velocities. The source term of porous medium drag is constructed by superimposing the linear and nonlinear drag component data, and it is introduced as the source term into the momentum conservation equation of the continuous phase in the corresponding Euler computational unit to complete the reconstruction. This allows the fluid control equation of the fixed particle region to continuously and smoothly reflect the complex flow drag characteristics, while maintaining physical compatibility with the original CMPPIC phase coupling mechanism, avoiding numerical discontinuities or oscillations caused by model switching.

[0052] After reconstructing the equations, throughout the entire numerical computation process: 1) For the region of movable particles, continue to use the CMPPIC method to solve for the Lagrange particle motion.

[0053] 2) For the fixed particle region, the effect of particles on the fluid is described by the porous medium equivalent model.

[0054] 3) After each or several time steps, the particle state determination process is re-executed. When the fixed conditions are no longer met, the computational unit is allowed to transform from a fixed particle region back into a movable particle region, realizing dynamic switching of particle states, thereby continuously describing complex evolution processes such as deposition, consolidation, or unblocking. Specifically: Furthermore, step S2 also includes step S29: S29: In each round of judging and marking whether the particles in each Euler computational unit are in a fixed state, if the Euler computational unit previously marked as fixed no longer meets the fixed condition in the current judgment, its fixed mark is canceled and its particle phase motion update operation is restored.

[0055] In this preferred embodiment, by re-evaluating whether the Euler computational units previously marked as fixed still meet the fixed conditions in each round of judgment, and canceling the fixed mark and resuming the particle phase motion update operation when the conditions are no longer met, the computational units can automatically switch from the porous medium model back to the Lagrange particle model when the particle state undergoes dynamic evolution (such as deposition unblocking, disturbance resuspension), realizing bidirectional adaptive conversion between fixed and moving states. This mechanism breaks through the limitation of static model fragmentation in traditional methods, enabling the numerical framework to continuously describe the entire process of particles from fixed to moving, significantly enhancing the applicability and predictive ability of the method in unsteady, evolutionary engineering problems.

[0056] To verify the effectiveness and accuracy of the CMPPIC-based equivalent modeling method for fixed-particle porous media proposed in this invention in describing the influence of fixed-particle structures on fluid flow, the classic Flow through a porous plug problem is selected as a verification example. This example is widely used to test the predictive ability of porous media models for fluid velocity decay characteristics, pressure loss characteristics, and flow field continuity, and has clear physical significance and a good basis for comparison.

[0057] A localized region within the computational domain is designated as a porous plug region, while the remaining region is a free-flowing region. The porous plug region consists of a high-volume-fraction, spatially stable particle structure. These particles do not undergo macroscopic movement during the computation, conforming to the physical characteristics of a fixed particle structure. In this invention, the porous plug region is initially described using the CMPPIC framework. Its fixed state is identified through particle volume fraction and statistical information, and it is adaptively transformed into an equivalent porous medium model during the computation. Continuous phase fluid flows through this porous plug region under pressure differential, resulting in typical flow hindrance. Some experimental results can be found by referring to... Figure 3 A velocity cloud map is shown. Figure 4 A pressure cloud map is shown. Figure 5 The speed curve shown is as follows: Figure 6 The diagram shows a pressure curve. The method of this invention can accurately describe the velocity decay behavior of fluid passing through a porous plug. Its velocity distribution trend is consistent with the traditional CMPPIC method and the classical porous media model, indicating that the porous media equivalent model constructed by this method is physically reasonable. The calculated pressure drop before and after the porous plug is consistent with the results of the comparative method, correctly reflecting the influence of the fixed particle structure on fluid momentum dissipation. At the boundary between the free flow region and the porous plug region, the fluid velocity and pressure field transition smoothly, without obvious numerical oscillations or discontinuities, indicating that the model switching process has good numerical stability.

[0058] In summary, compared with the prior art, the above embodiments of this application have the following beneficial effects: They acquire various data of the industrial equipment to be simulated and construct corresponding gas-solid two-phase flow CMPPIC models, starting from real engineering physics scenarios, ensuring the physical fidelity of the model and the feasibility of the initial conditions; by mapping the current Lagrangian particle pack data to each Eulerian computational unit in each round of time-progression iteration to obtain particle statistical information, and combining this with preset statistical rules to judge and mark fixed states, adaptive identification of the macroscopic motion behavior of particles is achieved, avoiding the indiscriminate execution of high-cost Lagrangian calculations for all regions; by freezing the particle volume fraction data of the Eulerian computational units marked as fixed states, and calculating the porous medium parameters based on this frozen data to reconstruct the continuous phase momentum conservation equation of the corresponding unit, the local drag effect originally dependent on particle tracking is transformed into one based on statistical information. The continuous source term description preserves the spatial non-uniformity of particle packing structures while eliminating redundant calculations. By uniformly performing continuous phase flow field updates based on reconstructed and unreconstructed momentum conservation equations, and only performing particle phase motion updates on unlabeled elements, coupled solutions for fixed and moving regions are achieved within the same time step, ensuring physical consistency and numerical stability across the entire flow field. The synergistic effect of these features eliminates the need for Lagrangian position / velocity time advance in fixed particle regions, significantly reducing computational load. Furthermore, the use of particle volume fractions inherent in the CMPPIC framework to construct a porous medium model with a clear physical origin avoids the inaccuracy of traditional empirical drag coefficients. Thus, computational efficiency and physical accuracy are simultaneously improved within a single numerical framework, ultimately solving the problem of balancing computational efficiency and physical accuracy in simulating fixed particle structures using existing techniques.

[0059] Example 2: Please refer to Figure 2 Based on the same inventive concept, the present invention discloses an adaptive simulation and prediction system for gas-solid two-phase flow, comprising: a data acquisition module M1 and an iterative output module M2; The data acquisition module M1 is used to acquire the geometric structure data, particulate property data and fluid boundary condition data of the industrial equipment to be simulated, and to construct the corresponding gas-solid two-phase flow CMPPIC model. The continuous phase is represented by the Euler method in the gas-solid two-phase flow CMPPIC model, and the discrete phase is represented by Lagrange particle packs in the gas-solid two-phase flow CMPPIC model. The iterative output module M2 is used to execute a time-progressive iterative process based on the gas-solid two-phase flow CMPPIC model until a preset termination condition is reached, and output the numerical simulation results. Each round of time-progression iteration includes: The current Lagrange particle pack data is mapped to each Euler computing unit to obtain the particle statistics of each Euler computing unit. Combined with the preset statistical rules, it is determined and marked whether the particles in each Euler computing unit are in a fixed state. The particle volume fraction data corresponding to the Euler computing unit marked as fixed is frozen, and the corresponding porous medium parameters are calculated based on the frozen particle volume fraction data. Then, the momentum conservation equation of the continuous phase in the corresponding Euler computing unit is reconstructed. Based on the reconstructed momentum conservation equation and the remaining unreconstructed momentum conservation equation, the flow field update operation for the continuous phase is performed, and the particle phase motion update operation is performed for the Euler computational cells that are not marked as fixed states.

[0060] In summary, compared with the prior art, the embodiments of this application have the following beneficial effects: They acquire various data of the industrial equipment to be simulated and construct corresponding gas-solid two-phase flow CMPPIC models, starting from real engineering physics scenarios to ensure the physical fidelity of the modeling and the feasibility of the initial conditions; by mapping the current Lagrangian particle pack data to each Eulerian computational unit in each round of time-progression iteration to obtain particle statistical information, and combining this with preset statistical rules to judge and mark fixed states, adaptive identification of the macroscopic motion behavior of particles is achieved, avoiding the indiscriminate execution of high-cost Lagrangian calculations for all regions; by freezing the particle volume fraction data of the Eulerian computational units marked as fixed states, and calculating the porous medium parameters based on this frozen data to reconstruct the continuous phase momentum conservation equation of the corresponding unit, the local drag effect originally dependent on particle tracking is transformed into a continuous phase based on statistical information. The continuous source term description preserves the spatial non-uniformity of the particle packing structure while eliminating redundant calculations. By uniformly performing continuous phase flow field updates based on reconstructed and unreconstructed momentum conservation equations, and performing particle phase motion updates only on unlabeled elements, coupled solutions for the fixed and moving regions are achieved within the same time step, ensuring physical consistency and numerical stability across the entire flow field. The synergistic effect of these features eliminates the need for Lagrangian position / velocity time advance in the fixed particle region, significantly reducing the computational load. Furthermore, the use of particle volume fractions inherent in the CMPPIC framework to construct a porous medium model with a clear physical origin avoids the inaccuracy of traditional empirical drag coefficients. Thus, computational efficiency and physical accuracy are simultaneously improved within a single numerical framework, ultimately solving the problem of balancing computational efficiency and physical accuracy in simulating fixed particle structures using existing techniques.

[0061] Furthermore, the iterative output module M2 includes: a first calculation unit, a second calculation unit, a third calculation unit, and a statistical unit; The first calculation unit is used to calculate the volume fraction contribution weight of each Lagrange particle bag to the adjacent Euler calculation unit based on its location data, equivalent mass data and particle density data; and to accumulate the volume fraction contribution weights of all Lagrange particle bags received by each Euler calculation unit to obtain the particle volume fraction data of the Euler calculation unit. The second calculation unit is used to allocate velocity vector contributions to each neighboring unit for each Lagrange particle pack based on its velocity vector data and the volume fraction contribution weight to the neighboring Euler calculation units; and to obtain the average particle velocity vector data of the Euler calculation unit by performing a weighted average of all the velocity vector contributions it receives for each Euler calculation unit. The third calculation unit is used to calculate the displacement vector of each Lagrange particle bag based on the position data of each Lagrange particle bag in the current time step and the previous several time steps; for each Euler calculation unit, the particle displacement change data of the unit is obtained by statistical averaging based on the particle bag displacement vector mapped to the unit. The statistical unit is used to acquire the current particle stress data of each Euler calculation unit, and to use the particle volume fraction data, particle average velocity vector data, particle displacement change data and particle stress data as particle statistical information.

[0062] In this preferred embodiment, the volume fraction contribution weight of each Lagrange particle pack to neighboring Eulerian computational units is calculated based on its position data, equivalent mass data, and particle density data. The volume fraction data of the particles is obtained by summing all contribution weights across all units, thus achieving a conservative mapping from the particle phase to the Eulerian field and ensuring local mass conservation. The average velocity vector data of the particles is obtained by allocating velocity vector contributions based on particle velocity vector data and their volume fraction contribution weights, and by performing a weighted average across all units, accurately depicting the macroscopic motion trend of particles at the unit scale. The displacement vector is calculated based on the position data of the particle pack at the current and previous time steps, and the displacement vector mapped to the units is statistically averaged to obtain particle displacement change data, effectively capturing the temporal evolution characteristics of particle motion. Finally, the current particle stress data of each Eulerian computational unit is obtained, and the particle volume fraction data, average particle velocity vector data, particle displacement change data, and particle stress data are used together as particle statistical information, providing a multi-dimensional and highly complete input basis for subsequent state discrimination, enhancing the robustness and applicability of fixed-state identification.

[0063] Furthermore, the iterative output module M2 also includes: a determination unit; The determination unit is used to determine that the particles in a certain Euler calculation unit are in a fixed state if any of the following conditions are met, and to generate fixed area marking data: the average velocity data of the particles is lower than a preset velocity threshold, the volume fraction data of the particles is greater than a preset packing volume fraction threshold, the change in particle displacement data is lower than a preset displacement threshold, and the difference between the fluid drag force and the particle stress is less than a preset equilibrium tolerance.

[0064] In this preferred embodiment, a fixed state is determined when any of the following conditions are met: the average velocity of particles within a certain Euler calculation unit is lower than a preset velocity threshold, the volume fraction of particles is greater than a preset packing volume fraction threshold, the change in particle displacement is lower than a preset displacement threshold, or the difference between the fluid drag force and the stress on the particles is less than a preset equilibrium tolerance. This allows the discrimination mechanism to cover various physically meaningful "fixed" situations, such as low-speed stillness, high packing density, displacement stagnation, and force balance, avoiding the risk of misjudgment under complex working conditions by a single criterion. At the same time, by adopting the logic of triggering a flag when "any condition is met", users are given the freedom to flexibly configure the combination of criteria, adapting to the different needs of "fixed" definitions in different engineering scenarios, and improving the versatility and engineering adaptability of the method.

[0065] Furthermore, the iterative output module M2 also includes: a fourth calculation unit and a fifth calculation unit; The fourth calculation unit is used to calculate the corresponding equivalent porosity data based on the particle volume fraction of each Euler calculation unit, and to calculate the permeability data in combination with the corresponding particle specific surface area data. The fifth calculation unit is used to calculate the Darcy linear drag coefficient and the Fochheimer nonlinear drag coefficient based on the permeability data and the fluid viscosity and fluid density data in the current gas-solid two-phase flow CMPPIC model.

[0066] In this preferred embodiment, by calculating the equivalent porosity data based on the particle volume fraction of each Euler computational unit and combining it with the particle specific surface area data to calculate the permeability data, a direct mapping relationship from the microscopic packing state of particles to macroscopic transport characteristics is established. Furthermore, by combining the permeability data, fluid viscosity data, and fluid density data to calculate the Darcy linear drag coefficient data and the Fochheimer nonlinear drag coefficient data, the porous medium parameters not only reflect the viscous dissipation under low-speed laminar flow but also characterize the inertial additional drag under high-speed flow. This constructs a full-condition drag model applicable to a wide Reynolds number range, which is significantly better than the traditional porous medium method that relies solely on a single empirical coefficient.

[0067] Furthermore, the iterative output module M2 also includes: a first component calculation unit, a second component calculation unit, and a reconstruction unit; The first component calculation unit is used to multiply the Darcy linear drag coefficient data with the fluid velocity data in the current gas-solid two-phase flow CMPPIC model to obtain the linear drag component data. The second component calculation unit is used to multiply the modulus of the Fochheimer nonlinear drag coefficient data, fluid density data, and fluid velocity data to obtain nonlinear drag component data; The reconstruction unit is used to superimpose the linear resistance component data and the nonlinear resistance component data to construct a porous medium resistance source term, and to introduce the porous medium resistance source term as a source term into the momentum conservation equation of the continuous phase in the corresponding Euler calculation unit to complete the reconstruction.

[0068] In this preferred embodiment, the linear drag component data is obtained by multiplying the Darcy linear drag coefficient data with the fluid velocity data, which quantifies the momentum dissipation intensity in viscous-dominated flow. The nonlinear drag component data is obtained by multiplying the modulus of the Fochheimer nonlinear drag coefficient data, fluid density data, and fluid velocity data, which accurately describes the additional pressure drop caused by vortex shedding and inertial effects at high flow velocities. The source term of porous medium drag is constructed by superimposing the linear and nonlinear drag component data, and it is introduced as the source term into the momentum conservation equation of the continuous phase in the corresponding Euler computational unit to complete the reconstruction. This allows the fluid control equation of the fixed particle region to continuously and smoothly reflect the complex flow drag characteristics, while maintaining physical compatibility with the original CMPPIC phase coupling mechanism, avoiding numerical discontinuities or oscillations caused by model switching.

[0069] Furthermore, the iterative output module M2 also includes: a recovery unit; The recovery unit is used to, in each round of judging and marking whether the particles in each Euler computational unit are in a fixed state, cancel the fixed mark and restore the particle phase motion update operation if the Euler computational unit previously marked as fixed no longer meets the fixed condition in the current judgment.

[0070] In this preferred embodiment, by re-evaluating whether the Euler computational units previously marked as fixed still meet the fixed conditions in each round of judgment, and canceling the fixed mark and resuming the particle phase motion update operation when the conditions are no longer met, the computational units can automatically switch from the porous medium model back to the Lagrange particle model when the particle state undergoes dynamic evolution (such as deposition unblocking, disturbance resuspension), realizing bidirectional adaptive conversion between fixed and moving states. This mechanism breaks through the limitation of static model fragmentation in traditional methods, enabling the numerical framework to continuously describe the entire process of particles from fixed to moving, significantly enhancing the applicability and predictive ability of the method in unsteady, evolutionary engineering problems.

[0071] Example 3: This invention also provides a computer program product, including a computer program or instructions, capable of running on a computing device or stored in any available medium. When the computer program product is run on at least one computing device, it causes the at least one computing device to execute any of the adaptive simulation and prediction methods for gas-solid two-phase flow according to this invention.

[0072] Example 4: This invention also provides a computer-readable storage medium storing at least one executable instruction that, when executed on an adaptive simulation prediction system for gas-solid two-phase flow, causes the system to perform an adaptive simulation prediction method for gas-solid two-phase flow according to any of the above-described method embodiments.

[0073] Numerous specific details are set forth in the specification provided herein. However, it will be understood that embodiments of this application may be practiced without these specific details. Similarly, for the purpose of simplification and aiding understanding of one or more aspects of the invention, in the above description of exemplary embodiments of this application, various features of the embodiments are sometimes grouped together in a single embodiment, figure, or description thereof. The claims, which follow the detailed description, are hereby expressly incorporated into that detailed description, wherein each claim itself is a separate embodiment of this application.

[0074] Those skilled in the art will understand that the modules in the system of the embodiments can be adaptively changed and placed in one or more systems different from that embodiment. Modules, units, or components in the embodiments can be combined into a single module, unit, or component, and further, they can be divided into multiple sub-modules, sub-units, or sub-components, except that at least some of such features and / or processes or units are mutually exclusive.

Claims

1. An adaptive simulation and prediction method for gas-solid two-phase flow, characterized in that, include: Obtain the geometric structure data, particulate property data, and fluid boundary condition data of the industrial equipment to be simulated, and construct the corresponding gas-solid two-phase flow CMPPIC model; wherein, the continuous phase is represented by the Euler method in the gas-solid two-phase flow CMPPIC model; the discrete phase is represented by Lagrange particle packs in the gas-solid two-phase flow CMPPIC model. Based on the gas-solid two-phase flow CMPPIC model, a time-progressive iterative process is executed until a preset termination condition is reached, and numerical simulation results are output. Each round of time-progression iteration includes: The current Lagrange particle pack data is mapped to each Euler computing unit to obtain the particle statistics of each Euler computing unit. Combined with the preset statistical rules, it is determined and marked whether the particles in each Euler computing unit are in a fixed state. The particle volume fraction data corresponding to the Euler computing unit marked as fixed is frozen, and the corresponding porous medium parameters are calculated based on the frozen particle volume fraction data. Then, the momentum conservation equation of the continuous phase in the corresponding Euler computing unit is reconstructed. Based on the reconstructed momentum conservation equation and the remaining unreconstructed momentum conservation equation, the flow field update operation for the continuous phase is performed, and the particle phase motion update operation is performed for the Euler computational cells that are not marked as fixed states.

2. The adaptive simulation and prediction method for gas-solid two-phase flow as described in claim 1, characterized in that, The process of mapping the current Lagrange particle pack data to each Euler computing unit to obtain particle statistics information for each Euler computing unit includes: For each Lagrange particle bag, its volume fraction contribution weight to neighboring Euler computing units is calculated based on its location data, equivalent mass data, and particle density data. For each Euler computing unit, the volume fraction contribution weights of all Lagrange particle bags it receives are summed to obtain the particle volume fraction data of that Euler computing unit. For each Lagrange particle pack, velocity vector contribution is assigned to each neighboring cell based on its velocity vector data and the volume fraction contribution weight to the neighboring Euler computational cells; for each Euler computational cell, the average velocity vector data of the particles in the Euler computational cell is obtained by weighted averaging based on all the velocity vector contributions it receives. Based on the position data of each Lagrange particle bag in the current time step and the previous few time steps, calculate its displacement vector; for each Euler calculation unit, perform statistical averaging based on the particle bag displacement vector mapped to that unit to obtain the particle displacement change data of that unit. Obtain the current particle stress data of each Euler calculation unit, and use the particle volume fraction data, particle average velocity vector data, particle displacement change data, and particle stress data as particle statistical information.

3. The adaptive simulation and prediction method for gas-solid two-phase flow as described in claim 2, characterized in that, The step of determining and marking whether the particles in each Euler computational unit are in a fixed state, based on preset statistical rules, includes: If any of the following conditions are met within a certain Euler calculation unit, the particles within that Euler calculation unit are determined to be in a fixed state, and fixed region marker data is generated: the average particle velocity data is lower than a preset velocity threshold, the particle volume fraction data is greater than a preset packing volume fraction threshold, the particle displacement change data is lower than a preset displacement threshold, and the difference between the fluid drag force and particle stress on the particles is less than a preset equilibrium tolerance.

4. The adaptive simulation and prediction method for gas-solid two-phase flow as described in claim 1, characterized in that, The step of calculating the corresponding porous media parameters based on the frozen particle volume fraction data includes: Based on the particle volume fraction of each Euler calculation unit, the corresponding equivalent porosity data is calculated, and combined with the corresponding particle specific surface area data, the permeability data is calculated. Based on the permeability data and the fluid viscosity and density data in the current gas-solid two-phase flow CMPPIC model, the Darcy linear drag coefficient and the Fochheimer nonlinear drag coefficient are calculated.

5. The adaptive simulation and prediction method for gas-solid two-phase flow as described in claim 4, characterized in that, The reconstructed momentum conservation equations for the continuous phase within the corresponding Euler computational unit include: Multiply the Darcy linear drag coefficient data with the fluid velocity data in the current gas-solid two-phase flow CMPPIC model to obtain the linear drag component data; Multiply the modulus of the Fochheimer nonlinear drag coefficient data, fluid density data, and fluid velocity data to obtain the nonlinear drag component data. The linear drag component data and the nonlinear drag component data are superimposed to construct a porous medium drag source term. This porous medium drag source term is then introduced as a source term into the momentum conservation equation of the continuous phase within the corresponding Euler computational unit to complete the reconstruction.

6. The adaptive simulation and prediction method for gas-solid two-phase flow as described in claim 1, characterized in that, In each round of judging and marking whether the particles in each Euler computational unit are in a fixed state, if the Euler computational unit previously marked as fixed no longer meets the fixed condition in the current judgment, its fixed mark is canceled and its particle phase motion update operation is restored.

7. An adaptive simulation and prediction system for gas-solid two-phase flow, characterized in that, include: Data acquisition module and iterative output module; The data acquisition module is used to acquire the geometric structure data, particulate property data, and fluid boundary condition data of the industrial equipment to be simulated, and to construct the corresponding gas-solid two-phase flow CMPPIC model. The continuous phase is represented by the Euler method in the gas-solid two-phase flow CMPPIC model, and the discrete phase is represented by Lagrange particle packs in the gas-solid two-phase flow CMPPIC model. The iterative output module is used to execute a time-progressive iterative process based on the gas-solid two-phase flow CMPPIC model until a preset termination condition is reached, and output the numerical simulation results. Each round of time-progression iteration includes: The current Lagrange particle pack data is mapped to each Euler computing unit to obtain the particle statistics of each Euler computing unit. Combined with the preset statistical rules, it is determined and marked whether the particles in each Euler computing unit are in a fixed state. The particle volume fraction data corresponding to the Euler computing unit marked as fixed is frozen, and the corresponding porous medium parameters are calculated based on the frozen particle volume fraction data. Then, the momentum conservation equation of the continuous phase in the corresponding Euler computing unit is reconstructed. Based on the reconstructed momentum conservation equation and the remaining unreconstructed momentum conservation equation, the flow field update operation for the continuous phase is performed, and the particle phase motion update operation is performed for the Euler computational cells that are not marked as fixed states.

8. The adaptive simulation and prediction system for gas-solid two-phase flow as described in claim 7, characterized in that, The iterative output module includes: a first calculation unit, a second calculation unit, a third calculation unit, and a statistical unit; The first calculation unit is used to calculate the volume fraction contribution weight of each Lagrange particle bag to the adjacent Euler calculation unit based on its location data, equivalent mass data and particle density data; and to accumulate the volume fraction contribution weights of all Lagrange particle bags received by each Euler calculation unit to obtain the particle volume fraction data of the Euler calculation unit. The second calculation unit is used to allocate velocity vector contributions to each neighboring unit for each Lagrange particle pack based on its velocity vector data and the volume fraction contribution weight to the neighboring Euler calculation units; and to obtain the average particle velocity vector data of the Euler calculation unit by performing a weighted average of all the velocity vector contributions it receives for each Euler calculation unit. The third calculation unit is used to calculate the displacement vector of each Lagrange particle bag based on the position data of each Lagrange particle bag in the current time step and the previous several time steps; for each Euler calculation unit, the particle displacement change data of the unit is obtained by statistical averaging based on the particle bag displacement vector mapped to the unit. The statistical unit is used to acquire the current particle stress data of each Euler calculation unit, and to use the particle volume fraction data, particle average velocity vector data, particle displacement change data and particle stress data as particle statistical information.

9. A computer program product, comprising a computer program or instructions, characterized in that, When the computer program or instructions are executed, they implement an adaptive simulation prediction method for gas-solid two-phase flow as described in any one of claims 1-6.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements an adaptive simulation prediction method for gas-solid two-phase flow as described in any one of claims 1-6.

Images