Digital twin modeling method for cement grinding process

Abstract

The invention provides a digital twin modeling method for a cement grinding process, which comprises the following steps of: simplifying cement grinding whole-process equipment, and setting a motion boundary of the equipment; and then connection among equipment is realized through a pipe section, a chute and a scraper, so that a virtual model of the whole process is established. Simulating a particle movement process and a crushing process based on a particle simulation technology; the method comprises the following steps: acquiring flow field information by coupling a computational fluid mechanics means and a particle simulation technology, solving by a computer to obtain a qualified product, and finally realizing result data visualization of the whole process of cement particle ore grinding. According to the digital twinborn modeling method, through full-process simulation, the problem that full-process collaborative optimization cannot be achieved for single module calculation in the past is solved, and meanwhile, based on comprehensive application of the particle simulation technology, the problems that simulation precision is poor and time is short in the past are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of digital modeling, and particularly to a digital twin modeling method for the cement grinding process. Background Art

[0002] In the process of mineral processing industry production, grinding, as an important link in mineral processing, is to fully dissociate useful minerals and gangue. The traditional grinding process involves multiple devices such as a feeding device, a roller press, a ball mill, a chute, a powder separator, etc. The process is long and has strong coupling, with problems such as non-linearity, multi-variables, strong hysteresis, and low intelligent level. And during the grinding process, there is a large amount of energy consumption and resource waste. Deeply understanding the working principle and operation law of the grinding process is of great significance for energy conservation and emission reduction in grinding.

[0003] CN109785378A discloses an on-line ore particle size detection device based on atlas image algorithm analysis technology: including an ore particle size analyzer and an industrial control computer; the ore particle size analyzer is connected to the industrial control computer through a dedicated local area network; the industrial control computer is connected to the Internet, and multiple clients are also connected in the local area network; the ore particle size analyzer is installed above the ore conveyor belt to take pictures of the ore on the conveyor belt and continuously and real-time collect the ore particle size images on the conveyor belt; an ore particle size analysis system is installed in the industrial control computer, including: an image preprocessing module, an image segmentation module, and an ore particle size calibration and analysis module. By on-line collecting the ore images on the conveyor belt, performing multi-level image algorithm processing, analyzing the ore particle size accurately, continuously and in real time, automatically statistically outputting the analysis results, and cooperating with optimizing the ore dressing process flow, improving the dissociation degree of the ore and the production efficiency.

[0004] CN115283123A discloses a grinding method, device, processor and electronic device for ore materials. The method includes: determining various minerals in the material to be ground and the occurrence state of each mineral; determining at least one target mineral from the various minerals according to the occurrence state; calculating the dissociation difficulty of each crystal plane of each target mineral, and determining the minimum dissociation difficulty from the dissociation difficulties of each crystal plane, obtaining the crystal plane corresponding to the minimum dissociation difficulty to get the first target crystal plane; determining the second target crystal plane from the adjacent crystal planes of the first target crystal plane, and calculating the included angle between the first target crystal plane and the second target crystal plane to get the target included angle; determining the target grinding medium with the target size and target ratio according to the target included angle, and grinding the material to be ground with the target grinding medium.

[0005] However, the above methods still cannot accurately predict the grinding result and have not solved the problems of multi-scale and multi-physical field coupling. Summary of the Invention

[0006] In view of the problems existing in the prior art, the present invention provides a digital twin modeling method for the cement grinding process. By simplifying the equipment in the entire cement grinding process, setting the motion boundaries of the equipment, and connecting the equipment through pipe sections, chutes, and scrapers, a virtual model of the entire process is established. Through the coupling of computational fluid dynamics means and particle simulation technology, the result data visualization of the entire cement particle grinding process is finally realized.

[0007] To achieve this purpose, the present invention adopts the following technical solutions:

[0008] The present invention provides a digital twin modeling method for the cement grinding process. The digital twin modeling method includes the following steps:

[0009] Step 1: Simplify the geometric structure of the equipment in the entire cement grinding process to obtain the characteristic dimensions, and establish a digital twin virtual model of the cement particle grinding process;

[0010] Step 2: According to the characteristic dimensions described in Step 1, set the particulate matter properties and fluid matter properties parameters in the entire cement processing process, and set the motion boundaries of the equipment in the entire cement grinding process;

[0011] Step 3: According to the working sequence of the equipment in the grinding process, use pipe sections, chutes, and scrapers to establish the connection relationship between the equipment in the entire cement grinding process to ensure the flow of cement materials between each equipment;

[0012] Step 4: Based on the interaction between cement particles and between cement particles and equipment, use particle simulation technology to describe the motion process and crushing process at the cement particle scale;

[0013] Step 5: Establish the fluid kinematic equation in the cement grinding process. Through the coupling of computational fluid dynamics means and particle simulation technology, obtain the flow field information inside the equipment during the cement processing, as well as the separation and mixing characteristics of cement particles under the action of the flow field, so as to obtain qualified products;

[0014] Step 6: Solve the coupling process of cement particles and cement particle fluid on the computer to realize the result data visualization of the entire cement particle grinding process, and further perform data monitoring and data analysis, so as to establish a digital twin model of the cement processing process.

[0015] The digital twin modeling method for the cement grinding process of the present invention first simplifies the equipment in the whole cement grinding process and sets the motion boundaries of the equipment. Then, the connection between the equipment is realized through pipe sections, chutes and scrapers, so as to establish a virtual model of the whole process. Based on the particle simulation technology, the particle movement process and the crushing process are simulated; by coupling the computational fluid dynamics method with the particle simulation technology, the flow field information is obtained, and the qualified products are obtained. Through computer solution, the result data visualization of the whole process of cement particle grinding is finally realized.

[0016] The present invention solves the problem that the previous simulation of a single module cannot achieve the collaborative optimization of the whole process through the whole process simulation. At the same time, based on the comprehensive application of the particle simulation technology, the problems of poor simulation accuracy and short time in the past are overcome.

[0017] The digital twin modeling method for the cement grinding process of the present invention establishes a digital twin model for the grinding process, maps the actual operation state of the grinding equipment in the digital virtual world, and realizes the prediction of the whole grinding process and the optimization of operation parameters through the information interaction between the virtual body and the entity. It solves the problems of multi-scale and multi-physical field coupling, more accurately predicts the results of cement grinding, and thus improves the production efficiency of the industry.

[0018] Preferably, the equipment in the whole cement grinding process described in step one includes a roller press, a ball mill and a powder separator.

[0019] Preferably, the simplified roller press includes a fixed roller and a moving roller.

[0020] Preferably, the simplified ball mill includes a cylinder body and lifting bars.

[0021] Preferably, the simplified powder separator includes a feed inlet, an air outlet pipe, a fine powder discharge pipe and a coarse powder discharge pipe.

[0022] The reason why the present invention simplifies the geometric structure of the equipment in the whole cement grinding process is that the actual geometric structure of the cement grinding equipment is too complex. Therefore, on the premise of ensuring the same geometric dimensions and operation effects, some component details are ignored, and the actual structure is simplified into a virtual entity. Among them, power devices such as gears and transmission shafts have little influence on the simulation process and can be reflected by the rotational speed under the dynamic boundary conditions.

[0023] Preferably, the characteristic dimensions include: the diameter D of the roller shaft roll , the length L of the roller shaft roll , the roll gap spacing d roll ; the diameter D of the ball mill cylinder body ball , the length L of the cylinder body ball , the upper bottom a of the lifting bar, the lower bottom b of the lifting bar, the height h of the lifting bar, the number n of the lifting bars; the diameter D of the feed inlet of the powder separator in , the diameter D of the coarse powder discharge pipe coarse , the diameter D of the fine powder discharge pipefine 、Outlet duct diameter D wind 。

[0024] Preferably, the particulate physical parameters described in step two include particle size, density ρ p , Young's modulus E, elastic coefficient k, and Poisson's ratio ν.

[0025] Preferably, the fluid physical parameters include the density ρ of the fluid f , viscosity μ f , initial velocity u f0 。

[0026] In the present invention, the moving boundary refers to the rotating components, rolling components, and hitting components in the cement grinding equipment.

[0027] Preferably, the boundary conditions of the moving boundary include the roller shaft rotation speed ω roll , ball mill rotation speed ω ball , classifier cage rotation speed ω cage , and the axis direction and rotation direction of the roller shaft, the axis direction and rotation direction of the ball mill, and the axis direction and rotation direction of the classifier cage.

[0028] Preferably, the establishment of the connection relationship between the full-process equipment of cement grinding by using pipe sections, chutes, and scrapers in step three includes the transportation of the rough crushing product of the roller press to the ball mill through pipe sections and scrapers, and the transportation of the grinding product of the ball mill to the classifier through the chute.

[0029] In the present invention, the roller press performs rough crushing on the cement material, and the rough crushing product enters the ball mill through the conveying device. The ball mill grinds it, and the grinding product is transported to the classifier through the chute, and the classifier separates the qualified cement product.

[0030] Preferably, the description of the movement process of the cement particle size by using the particle simulation technology in step four specifically includes: Cement exists in the form of particles throughout the grinding process. Under the Lagrangian system, the translational equation and rotational equation of the cement particles are established. Combining the particulate physical parameters, based on the resultant force acting on the cement particles on the time scale, the position, velocity, and rotational speed information of the cement particles at the current moment are obtained through Newton's second law. Based on the position coordinates of the cement particles, the contacts between the cement particles are retrieved to describe the interaction process between cement particle - cement particle and cement particle - wall surface and the movement of the cement particles within the updated time step.

[0031] Preferably, the translational equation is:

[0032]

[0033] Preferably, the rotational equation is:

[0034]

[0035] where m p represents the mass of the cement particle, u p represents the velocity vector of the cement particle, t is time, F n and F t respectively represent the normal component force and the tangential component force when the cement particles are in contact with each other, F drag represents the drag force exerted by the fluid on the cement particle, g represents the acceleration due to gravity; I p represents the moment of inertia of the cement particle; ω p represents the rotational speed of the cement particle; R represents a vector pointing from the centroid of the cement particle to the collision contact point, R is the magnitude of the vector, μ r represents the rolling friction coefficient of the cement particle.

[0036] Preferably, the use of the particle simulation technology to describe the crushing process of the cement particle size in step four specifically includes: the cement changes from large particles to small particles and then to powdery particles in the grinding process, and "size effect" will occur with the change of the particle size. At the same time, based on the crushing criterion, the particle simulation technology is used to judge whether the cement particle breaks when it is collided, and the particle size distribution result of the product after crushing. At the same time, it is coupled with the simulation of the cement particle movement process to realize the real-time update of the position, velocity and particle size information of the crushed product during the continuous crushing process of equipment such as roller presses and ball mills.

[0037] Preferably, the particle simulation technology includes establishing a bonded particle model, establishing a particle replacement model, combining the finite discrete element method and the peridynamics method.

[0038] In the present invention, the bonded particle model combines several small balls into a particle cluster through bonding bonds, and depicts the crushing process through the fracture of the bonds; the particle replacement method replaces a large particle with a group of sub-particles when a certain crushing criterion is met; the combined finite discrete element method describes the deformation and crushing of particles based on the finite element method, and calculates the contact and movement of particles through the discrete element method; peridynamics discretizes a single particle into a series of interacting material points, and depicts the initiation of cracks and particle crushing through the fracture of the bonds between the material points.

[0039] Preferably, the crushing criterion includes a force crushing criterion, a stress crushing criterion, a bond length crushing criterion and a crushing probability model.

[0040] Preferably, the fluid kinematic equations in step five include a continuity equation and a momentum equation:

[0041] Preferably, the continuity equation is:

[0042]

[0043] Preferably, the momentum equation is as follows:

[0044]

[0045] where ε f represents the fluid volume fraction in the grid, ρ f represents the fluid density, t represents time, u f represents the fluid velocity, p represents pressure, S represents the source term, τ f represents the viscous stress, and g represents the acceleration due to gravity.

[0046] Preferably, the flow field information in step five includes the flow field distribution characteristics and the separation result of cement particles.

[0047] Preferably, the steps for obtaining the flow field distribution characteristics include: in the Eulerian system, considering the fluid volume fraction and the source term, establishing the control equation of the fluid, introducing the turbulence model, mapping the positions of the particles to the grid of computational fluid dynamics at the same time, calculating the velocity field and concentration field in a single grid, and obtaining the flow field distribution characteristics.

[0048] Preferably, the separation result of the cement particles includes: calculating the drag force exerted by the fluid on the particles based on the relative slip velocity of the fluid particles, updating the resultant force on the particles, and characterizing the selection result of the fluid for the qualified grinding products in the classifier.

[0049] Preferably, the solution of the coupling process of cement particles and cement particle fluid in step six on the computer includes: solving the coupling process of particle fluid on the computer through heterogeneous parallelism; the specific parallel mode includes: establishing a data shared memory block to realize data interaction between the fluid and the particles. When the GPU finishes solving the particles, copy the positions, types, and velocities of the cement particles to the shared memory. The shared memory transmits the particle information to the CFD and updates the porosity. Similarly, the CFD transmits the flow field information to the GPU through the shared memory to update the particle drag force.

[0050] Preferably, the actual solution times of the CFD and the DEM are not synchronized, and an atomic lock is established for the data receiving process, and when one party finishes the calculation, it receives the calculation result of the other party.

[0051] The present invention uses high-performance parallel computing to improve the computing power and solution speed of the digital twin model. The number of particles involved in the whole process of cement grinding is huge, and the number of particles increases sharply with the progress of crushing. One GPU thread is adopted to update the motion information of each particle in steps four and five; the CPU solves the fluid control equation, while the GPU is used to solve the large-scale particle motion, and the information interaction between the CPU and the GPU is realized by establishing a shared memory.

[0052] In the present invention, CPU is the central processing unit, GPU is the graphics processing unit, CFD is computational fluid dynamics, and DEM is the discrete element model.

[0053] Preferably, the method for visualizing the result data of the entire cement particle grinding process in step six includes: with the aid of post-processing visualization software or through virtual image technology, virtually demonstrating the connection relationship between the equipment structure and the process and the model calculation result data on the interface.

[0054] Preferably, the post-processing visualization software includes Paraview and Tecplot.

[0055] In the present invention, based on multi-source heterogeneous data such as data from actual production records and the solution results of the digital twin model, through three-dimensional visualization technology, the real-time operating status of the overall process and the flow, crushing, classification, etc. of the cement material in the equipment are mapped to a virtual environment, realizing data visualization of the entire cement grinding process, and problems generated in the grinding process can be discovered in a timely manner, and the operating conditions can be optimized.

[0056] As a preferred technical solution of the present invention, the digital twin modeling method includes the following steps:

[0057] Step 1: Simplify the geometric structure of the equipment for the entire cement grinding process to obtain characteristic dimensions, and establish a digital twin virtual model for the cement particle grinding process;

[0058] The equipment for the entire cement grinding process includes a roller press, a ball mill, and a powder separator; the simplified roller press includes a fixed roller and a moving roller; the simplified ball mill includes a cylinder body and lifting bars; the simplified powder separator includes a feed inlet, an air outlet pipe, a fine powder discharge pipe, and a coarse powder discharge pipe; the characteristic dimensions include: the roller shaft diameter D roll , the roller shaft length L roll , the roll gap spacing d roll ; the ball mill cylinder diameter D ball , the cylinder body length L ball , the upper bottom a of the lifting bar, the lower bottom b of the lifting bar, the height h of the lifting bar, the number n of lifting bars; the feed inlet diameter D of the powder separator in , the coarse powder discharge pipe diameter D coarse , the fine powder discharge pipe diameter D fine , the air outlet pipe diameter D wind ;

[0059] Step 2: According to the characteristic dimensions described in step 1, set the particulate matter properties parameters and fluid matter properties parameters during the cement full-process processing, and set the motion boundaries of the equipment for the entire cement grinding process.

[0060] The particulate matter properties parameters include particle size grade, density ρ p, Young's modulus E, elastic coefficient k, Poisson's ratio ν; the fluid physical property parameters include the density ρ of the fluid f , viscosity μ f , initial velocity u f0 ; the boundary conditions of the moving boundary include the roller speed ω roll , the ball mill speed ω ball , the cage speed ω of the classifier cage , and the axis direction and rotation direction of the roller, the axis direction and rotation direction of the ball mill, and the axis direction and rotation direction of the classifier cage;

[0061] Step Three: According to the working sequence of the equipment in the grinding process, the conveying between the roller press coarse crushing product and the ball mill is realized through the pipe section and the scraper, and the grinding product of the ball mill is conveyed to the classifier through the chute to ensure the circulation of the cement material between each equipment;

[0062] Step Four: Based on the interaction between cement particles and between cement particles and equipment, the particle simulation technology is used to describe the movement process and crushing process of cement particles at the particle scale;

[0063] The use of particle simulation technology to describe the movement process of cement particles at the particle scale specifically includes: Cement exists in the form of particles throughout the grinding process. Under the Lagrangian system, the translational equation and rotational equation of cement particles are established. Combining the particle physical property parameters, based on the resultant force on the cement particles on the time scale, the position, velocity, and rotation speed information of the cement particles at the current moment is obtained through Newton's second law. Based on the position coordinates of the cement particles, the contacts between cement particles are retrieved to describe the interaction process between cement particle-cement particle and cement particle-wall surface and update the movement of the cement particles at the next time step;

[0064] The translational equation is:

[0065]

[0066] The rotational equation is:

[0067]

[0068] Where m p represents the mass of the cement particle, u p represents the velocity vector of the cement particle, t is time, F n and F t respectively represent the normal component force and tangential component force when the cement particles are in contact with each other, F drag represents the drag force exerted by the fluid on the cement particle, g represents the acceleration due to gravity; I p represents the moment of inertia of the cement particle; ω prepresents the rotational speed of cement particles; R represents a vector pointing from the centroid of the cement particle to the collision contact point, R is the magnitude of the vector, and μ r represents the rolling friction coefficient of cement particles;

[0069] The use of particle simulation technology to describe the crushing process of cement particle size specifically includes: Cement changes from large particles to small particles and then to powdery particles in the grinding process. And with the change of particle size, there will be a "size effect". At the same time, based on the crushing criterion, through particle simulation technology, it is judged whether the cement particles break when they are collided, as well as the particle size distribution result of the products after crushing. At the same time, it is coupled with the simulated cement particle movement process to realize the real-time update of the position, speed, and particle size information of the crushed products during the continuous crushing process of equipment such as roller presses and ball mills;

[0070] The particle simulation technology includes establishing a bonded particle model, establishing a particle replacement model, combining the finite discrete element method and the peridynamics method; the crushing criteria include the force crushing criterion, the stress crushing criterion, the bond length crushing criterion, and the crushing probability model;

[0071] Step Five: Establish the fluid kinematic equation in the cement grinding process, couple it with particle simulation technology through computational fluid dynamics means, obtain the flow field information inside the equipment during the cement processing process, as well as the separation and mixing characteristics of cement particles under the action of the flow field, so as to obtain qualified products;

[0072] The fluid kinematic equation includes the continuity equation and the momentum equation:

[0073] The continuity equation is:

[0074]

[0075] The momentum equation is:

[0076]

[0077] Among them, ε f represents the fluid volume fraction in the grid, ρ f represents the fluid density, t represents time, u f represents the fluid velocity, p represents the pressure, S represents the source term, τ f represents the viscous stress, and g represents the acceleration due to gravity;

[0078] The flow field information includes the flow field distribution characteristics and the separation results of cement particles:

[0079] The steps for obtaining the flow field distribution characteristics include: in the Eulerian system, considering the fluid volume fraction and source terms, establishing the control equations of the fluid, introducing a turbulence model, mapping the positions of the particles to the computational fluid dynamics (CFD) grid, calculating the velocity field and concentration field within a single grid, and obtaining the flow field distribution characteristics;

[0080] The separation results of the cement particles include: calculating the drag force exerted by the fluid on the particles based on the relative slip velocity between the fluid and the particles, updating the resultant force on the particles, and characterizing the selection result of the qualified grinding products by the fluid in the classifier;

[0081] Step 6: Solve the coupling process of the cement particles and the cement particle-fluid on a computer, visualize the result data of the entire cement particle grinding process, further perform data monitoring and data analysis, and thus establish a digital twin model of the cement processing process;

[0082] Solve the coupling process of the particle-fluid on a computer through heterogeneous parallel computing. The specific parallel method includes: establishing a data shared memory block to achieve data interaction between the fluid and the particles. When the GPU completes the solution of the particles, copy the positions, types, and velocities of the cement particles to the shared memory. The shared memory transmits the particle information to the CFD to update the porosity. Similarly, the CFD transmits the flow field information to the GPU through the shared memory to update the particle drag force. The actual solution times of the CFD and the DEM are not synchronized. An atomic lock is established for the data receiving process, and when one party completes the calculation, it receives the calculation results of the other party;

[0083] The method for visualizing the result data of the entire cement particle grinding process includes: virtually demonstrating the connection relationship of the equipment structure and process and the model calculation result data on the interface with the aid of post-processing visualization software or through virtual image technology.

[0084] Compared with the prior art, the present invention has at least the following beneficial effects:

[0085] The digital twin modeling method for the cement grinding process provided by the present invention couples computational fluid dynamics means with particle simulation technology, realizes the visualization of the result data of the entire cement particle grinding process, realizes the collaborative optimization of the entire process, has high simulation accuracy and a long simulation time, and has broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0086] Figure 1 is a flowchart of the digital twin modeling method for the cement grinding process in a specific embodiment of the present invention.

[0087] Figure 2 is the simulation result of the particle crushing in the roller press based on the bonded particle model in a specific embodiment of the present invention.

[0088] Figure 3It is the simulation result of particle breakage in a ball mill based on the population balance method in the specific embodiment of the present invention.

[0089] Figure 4 It is the simulation result of selecting qualified products in a powder separator based on the CFD-DEM coupling in the specific embodiment of the present invention.

[0090] Figure 5 It is the visualization result of the full-process simulation result based on Paraview rendering in the specific embodiment of the present invention. Specific embodiment

[0091] The technical solution of the present invention will be further described below in conjunction with the drawings and through specific embodiments.

[0092] The present invention will be described in further detail below. However, the following examples are only simple examples of the present invention and do not represent or limit the scope of the patent protection of the present invention. The scope of protection of the present invention shall be subject to the claims.

[0093] As a specific embodiment of the present invention, a digital twin modeling method for the cement grinding process is provided, and its flowchart is as Figure 1 shown.

[0094] The digital twin modeling method includes the following steps:

[0095] Step 1: Simplify the geometric structure of the equipment in the entire cement grinding process to obtain characteristic dimensions, and establish a digital twin virtual model for the cement particle grinding process;

[0096] The equipment in the entire cement grinding process includes a roller press, a ball mill, and a powder separator; the simplified roller press includes a fixed roller and a moving roller; the simplified ball mill includes a cylinder and lifting bars; the simplified powder separator includes a feed inlet, an air outlet pipe, a fine powder discharge pipe, and a coarse powder discharge pipe; the characteristic dimensions include: the roller shaft diameter D roll = 0.4 m, the roller shaft length L roll = 0.4 m, the roll gap spacing d roll = 0.04 m; the ball mill cylinder diameter D ball = 0.5 m, the cylinder length L ball = 0.6 m, the upper bottom a of the lifting bar = 0.08 m, the lower bottom b of the lifting bar = 0.08 m, the height h of the lifting bar = 0.08 m, the number of lifting bars n = 8; the feed inlet diameter D in = 0.28 m of the powder separator, the diameter D coarse = 0.30 m of the coarse powder discharge pipe, the diameter D fine = 0.1 m of the fine powder discharge pipe, the diameter D wind = 0.3 m of the air outlet pipe;

[0097] Step 2: According to the characteristic dimensions described in Step 1, set the particulate physical properties and fluid physical properties in the whole process of cement processing, and set the motion boundaries of the equipment in the whole process of cement grinding;

[0098] The particulate physical properties include the feed particle size distribution (-10 + 5mm accounts for 23%, -5 + 2mm accounts for 24%, -2 + 1mm accounts for 22%, -1 + 0.5mm accounts for 16%, -0.5 + 0.008mm accounts for 15%, and the total particle mass is 1398.5g), density ρ p = 2800 kg / m 3 、Young's modulus E = 3.5×10 8 Pa, elastic coefficient k = 1000 N·m, Poisson's ratio ν = 0.33;

[0099] The fluid physical properties include the density ρ f = 1.29 kg / m 3 、viscosity μ f = 0.000018 Pa·s, initial velocity u f0 = 12 m / s;

[0100] The boundary conditions of the motion boundary include: the rotational speed ω roll = 25 rpm of the roller press, the rotational speed ω ball = 70 rpm of the ball mill, the rotational speed ω cage = 10 rpm of the classifier rotor, and the rotational axis direction and rotational direction of the roller shaft, the rotational axis direction and rotational direction of the ball mill, and the rotational axis direction and rotational direction of the classifier rotor;

[0101] Step 3: According to the working sequence of the equipment in the grinding process, use pipe sections and scrapers to convey the rough crushing products of the roller press to the ball mill, and use a chute to convey the pulverized products of the ball mill to the classifier to ensure the flow of cement materials between various equipment;

[0102] Step 4: Based on the interaction between cement particles and between cement particles and equipment, use particle simulation technology to describe the motion process and crushing process at the scale of cement particles;

[0103] The use of particle simulation technology to describe the motion process at the scale of cement particles specifically includes: Cement exists in the form of particles in the whole process of grinding. In the Lagrangian system, establish the translational equation and rotational equation of cement particles. Combining the particulate physical properties, based on the resultant force on the cement particles on the time scale, obtain the position, velocity, and rotational speed information of the cement particles at the current moment through Newton's second law. Based on the position coordinates of the cement particles, retrieve the contacts between cement particles to describe the interaction process of cement particle - cement particle and cement particle - wall surface and update the motion of cement particles in the next time step;

[0104] The translational equation is as follows:

[0105]

[0106] The rotational equation is as follows:

[0107]

[0108] Where m p represents the mass of cement particles, u p represents the velocity vector of cement particles, t is time, F n and F t respectively represent the normal component force and tangential component force when cement particles are in contact with each other, F drag represents the drag force exerted by the fluid on the cement particles, g represents the acceleration due to gravity; I p represents the moment of inertia of the cement particles; ω p represents the rotational speed of the cement particles; R represents a vector pointing from the centroid of the cement particle to the collision contact point, R is the magnitude of the vector, μ r represents the rolling friction coefficient of the cement particles;

[0109] The description of the crushing process of cement particle size by using particle simulation technology specifically includes: in the grinding process, cement changes from large particles to small particles and then to powdery particles, and "size effect" will occur with the change of particle size. At the same time, based on the crushing criterion, it is judged whether the cement particles are broken when they are collided by using particle simulation technology, as well as the particle size distribution result of the products after crushing. At the same time, it is coupled with the simulation of the movement process of cement particles to realize the real-time update of the position, velocity and particle size information of the crushed products in the continuous crushing process of equipment such as roller presses and ball mills;

[0110] For the crushing process of cement particles, the bonded particle model is used to describe the crushing process of particles under the action of roller pressure in a roller press. As Figure 2 shown, the bonded particle model combines several small balls into a particle cluster through bonding bonds, and depicts the crushing process through the fracture of the bonds;

[0111] In a ball mill, the population balance method is used to perform long-time simulation of particle impact crushing. As Figure 3 shown, the population balance model establishes differential equations based on the material conservation at the process scale, including the crushing rate and the crushing distribution function, and introduces the minimum crushing energy of particles. When the collision energy of particles exceeds this value, the particles are broken, and as the collision energy and collision frequency increase, the crushing rate increases. The change of particle size grade with time can be predicted based on this method;

[0112] Step 5: Establish the kinematic equation of fluid in the cement grinding process. By coupling computational fluid dynamics and particle simulation technology, obtain the flow field information inside the equipment during the cement processing, as well as the separation and mixing characteristics of cement particles under the action of the flow field, so as to obtain qualified products;

[0113] The kinematic equation of the fluid includes the continuity equation and the momentum equation:

[0114] The continuity equation is:

[0115]

[0116] The momentum equation is:

[0117]

[0118] Among them, ε f represents the fluid volume fraction in the grid, ρ f represents the fluid density, t represents time, u f represents the fluid velocity, p represents pressure, S represents the source term, τ f represents the viscous stress, and g represents the acceleration due to gravity;

[0119] The flow field information includes the flow field distribution characteristics and the separation results of cement particles:

[0120] The steps to obtain the flow field distribution characteristics include: in the Eulerian system, considering the fluid volume fraction and the source term, establish the control equation of the fluid, introduce the turbulence model, and at the same time map the positions of the particles to the grids of computational fluid dynamics, calculate the velocity field and concentration field in a single grid, and obtain the flow field distribution characteristics;

[0121] The separation results of the cement particles include: based on the relative slip velocity of the fluid particles, calculate the drag force exerted by the fluid on the particles, update the resultant force on the particles, and characterize the selection result of the fluid on the qualified grinding products in the classifier;

[0122] Step 6: Solve the coupling process of cement particles and cement particle fluid on the computer, realize the visualization of the result data of the whole process of cement particle grinding, and further conduct data monitoring and data analysis, so as to establish a digital twin model of the cement processing process;

[0123] Solve the coupled process of particulate fluid on a computer through heterogeneous parallelism. The specific parallel methods include: establishing a data shared memory block to achieve data interaction between the fluid and particles. After the GPU completes the solution of the particles, copy the positions, types, and velocities of the cement particles to the shared memory. The shared memory transmits the particle information to the CFD to update the porosity. Similarly, the CFD transmits the flow field information to the GPU through the shared memory to update the particle drag force. The actual solution consumption times of the CFD and DEM are asynchronous. An atomic lock is established for the data reception process, and when one party completes the calculation, it receives the calculation results of the other party; Figure 4 Shows the calculation results based on the CFD-DEM method in the classifier;

[0124] The method for realizing the visualization of the result data of the entire process of cement particle grinding includes: with the help of post-processing visualization software or through virtual image technology, virtually demonstrate the connection relationship of the equipment structure and process and the model calculation result data on the interface; Figure 5 Is the visualization result of the full-process simulation result based on Paraview rendering.

[0125] The applicant declares that the above is only the specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Those skilled in the art should understand that any changes or substitutions that can be easily thought of by any person skilled in the art within the technical scope disclosed by the present invention fall within the protection scope and the disclosure scope of the present invention.

Claims

1. A digital twin modeling method for the cement grinding process, characterized in that, The digital twin modeling method includes the following steps: Step 1: Simplify the geometric structure of the full-process equipment for cement grinding to obtain characteristic dimensions, and establish a digital twin virtual model for the cement particle grinding process. Step 2: According to the characteristic dimensions described in Step 1, set the particulate physical properties and fluid physical properties parameters in the full-process cement processing, and set the motion boundaries of the full-process equipment for cement grinding. Step 3: According to the working sequence of the equipment in the grinding process, use pipe sections, chutes and scrapers to establish the connection relationship between the full-process equipment for cement grinding, ensuring the flow of cement materials between each equipment. Step 4: Based on the interaction between cement particles and between cement particles and equipment, use particle simulation technology to describe the motion process and crushing process at the scale of cement particles. Step 5: Establish the fluid kinematic equation in the cement grinding process, couple it with the particle simulation technology through computational fluid dynamics means, obtain the flow field information inside the equipment during the cement processing, as well as the separation and mixing characteristics of cement particles under the action of the flow field, so as to obtain qualified products. Step 6: Solve the coupling process of cement particles and cement particle fluid on the computer, realize the visualization of the result data of the full process of cement particle grinding, and further conduct data monitoring and data analysis, so as to establish a digital twin model for the cement processing process.

2. The digital twin modeling method according to claim 1, characterized in that, The full-process equipment for cement grinding described in Step 1 includes a roller press, a ball mill and a powder separator. Preferably, the simplified roller press includes a fixed roller and a moving roller. Preferably, the simplified ball mill includes a cylinder body and lifting bars. Preferably, the simplified powder separator includes a feed inlet, an air outlet pipe, a fine powder discharge pipe and a coarse powder discharge pipe. Preferably, the characteristic dimensions include: the diameter D of the roller shaft roll , the length L of the roller shaft roll , the gap spacing d roll ; the diameter D of the ball mill cylinder ball , the length L of the cylinder ball , the upper base a of the lifting bar, the lower base b of the lifting bar, the height h of the lifting bar, the number n of the lifting bars; the diameter D of the inlet of the classifier in , the diameter D of the coarse powder discharge pipe coarse , the diameter D of the fine powder discharge pipe fine , the diameter D of the air outlet pipe wind .

3. The digital twin modeling method according to claim 1 or 2, characterized in that, The particulate matter properties described in Step 2 include particle size, density ρ p , Young's modulus E, elastic coefficient k, and Poisson's ratio ν; Preferably, the fluid physical property parameters include the density ρ of the fluid f , viscosity μ f , initial velocity u f0 ; Preferably, the boundary conditions of the movement boundary include the roller speed ω roll , the ball mill speed ω ball , the cage speed ω cage of the classifier, as well as the axis direction and rotation direction of the roller, the axis direction and rotation direction of the ball mill, and the axis direction and rotation direction of the classifier cage.

4. The digital twin modeling method according to any one of claims 1 to 3, characterized in that, The connection relationship between the full-process equipment for cement grinding established by using pipe sections, chutes and scrapers in Step 3 includes transporting the coarse crushing product of the roller press to the ball mill through pipe sections and scrapers, and transporting the grinding product of the ball mill to the powder separator through a chute.

5. The digital twin modeling method according to any one of claims 1 to 4, characterized in that, The specific description of the motion process at the scale of cement particles by using particle simulation technology in Step 4 specifically includes: Cement exists in the form of particles throughout the grinding process. Under the Lagrangian system, establish the translational equation and rotational equation of cement particles. Combining with the particulate physical properties parameters, based on the resultant force received by cement particles on the time scale, obtain the position, velocity and rotational speed information of cement particles at the current moment through Newton's second law. Based on the position coordinates of cement particles, retrieve the contact between cement particles to describe the interaction process between cement particle-cement particle and cement particle-wall surface and update the motion of cement particles within the time step. Preferably, the translational equation is: Preferably, the rotational equation is: where m p represents the mass of the cement particle, u p represents the velocity vector of the cement particle, t is time, F n and F t respectively represent the normal component force and the tangential component force when the cement particles are in contact with each other, F drag represents the drag force exerted by the fluid on the cement particle, g represents the acceleration due to gravity; I p represents the moment of inertia of the cement particle; ω p represents the rotational speed of the cement particle; R represents a vector pointing from the centroid of the cement particle to the collision contact point, R is the magnitude of the vector, μ r represents the rolling friction coefficient of the cement particle.

6. The digital twin modeling method according to any one of claims 1 to 5, characterized in that, The specific description of the crushing process at the scale of cement particles by using particle simulation technology in Step 4 specifically includes: Cement changes from large particles to small particles and then to powdery particles in the grinding process, and there will be a "size effect" with the change of particle size. Based on the crushing criterion, use particle simulation technology to judge whether cement particles break when they are collided, and the particle size distribution result of the broken products. At the same time, couple it with the simulation process of cement particle motion to realize the real-time update of the position, velocity and particle size information of the broken products during the continuous crushing process of equipment such as roller presses and ball mills. Preferably, the particle simulation technology includes establishing a bonded particle model, establishing a particle replacement model, combining the finite discrete element method and the peridynamics method; Preferably, the crushing criteria include the crushing criterion of force, the crushing criterion of stress, the crushing criterion of bond length, and the crushing probability model.

7. The digital twin modeling method according to any one of claims 1 to 6, characterized in that, The fluid kinematic equation in Step Five includes the continuity equation and the momentum equation: Preferably, the continuity equation is: Preferably, the momentum equation is: where, ε f represents the fluid volume fraction in the grid, ρ f represents the fluid density, t represents time, u f represents the fluid velocity, p represents pressure, S represents the source term, τ f represents the viscous stress, and g represents the acceleration due to gravity.

8. The digital twin modeling method according to any one of claims 1 to 7, characterized in that, The flow field information in Step Five includes the flow field distribution characteristics and the separation result of cement particles; Preferably, the steps for obtaining the flow field distribution characteristics include: in the Eulerian system, considering the fluid volume fraction and the source term, establishing the control equation of the fluid, introducing the turbulence model, and at the same time mapping the positions of the particles to the grid of computational fluid dynamics, calculating the velocity field and the concentration field in a single grid, and obtaining the flow field distribution characteristics; Preferably, the separation result of the cement particles includes: based on the relative slip velocity of the fluid particles, calculating the drag force exerted by the fluid on the particles, updating the resultant force on the particles, and characterizing the selection of qualified products by the action of the air flow in the classifier.

9. The digital twin modeling method according to any one of claims 1 to 8, characterized in that, The solution of the coupling process of cement particles and cement particle fluid in Step Six on the computer includes: solving the coupling process of particle fluid on the computer through heterogeneous parallelism; the specific parallelism method includes: establishing a data shared memory block to realize data interaction between the fluid and the particles. When the GPU completes the solution of the particles, copy the positions, types, and velocities of the cement particles to the shared memory, and the shared memory transmits the particle information to the CFD to update the porosity. Similarly, the CFD transmits the flow field information to the GPU through the shared memory to update the particle drag force; Preferably, the actual solution times of the CFD and the DEM are out of sync, and an atomic lock is established for the data receiving process, and when one party completes the calculation, it receives the calculation result of the other party.

10. The digital twin modeling method according to any one of claims 1 to 9, characterized in that, The method for realizing the visualization of the result data of the full process of cement particle grinding in Step Six includes: virtual demonstration of the connection relationship of the equipment structure and the process and the model calculation result data on the interface by means of post-processing visualization software or through virtual image technology.

Images