Raymond mill operation grading prediction and evaluation method

Abstract

A Raymond mill operation grading prediction and evaluation method aims at solving the problem that related processing modes capable of correspondingly evaluating grading efficiency and whole machine pressure loss of a Raymond mill do not exist at present, the prediction and evaluation method in the invention is based on an inlet wind speed D, a main shaft rotating speed E and a grader rotating speed F, and through combination of a Box-Behnken method, a DPM model and an MRF multi-coordinate reference system, the grading efficiency and the whole machine pressure loss of the Raymond mill can be evaluated, and the grading efficiency and the whole machine pressure loss of the Raymond mill can be evaluated. And performing response surface simulation analysis by using the R-R distributed discrete phase, and calculating to obtain the maximum value of the grading efficiency M and the minimum value of the complete machine pressure loss N, and completing the contrastive analysis process with the actually acquired data according to the maximum value of the grading efficiency M and the minimum value of the complete machine pressure loss N.

Description

Technical Field

[0001] The invention belongs to the technical field of mechanical engineering, and in particular relates to a prediction and evaluation method for operation classification of a Raymond mill. Background Art

[0002] As a kind of mining equipment, Raymond mill mainly grinds ore raw materials into fine powder through mechanical processing and rolling. During the working process, the internal flow field is mainly gas-solid two-phase flow, usually presenting a complex fluid-particle multiphase flow state, covering the vortex flow generated between the grinding roller and the grinding ring, the contraction flow phenomenon in the gravity classification area and the strong vortex flow in the centrifugal separation area. Its internal structure is complex, requiring multi-point monitoring and a long product life cycle. During the working process of these particle and fluid coupling systems, the internal fluid will exert drag, lift or pressure on the particles, causing the particles to move. This process not only involves collisions between particles and particles and between particles and solid walls, but also the movement of particles will in turn affect the flow of fluids. This complex interaction makes it more complicated and difficult to fully and accurately understand the physical processes such as the movement, distribution and collision characteristics of particle groups in the fluid system. At present, there is a lack of relevant estimation methods that can accurately grasp the fluid-solid two-phase flow state in the powder system to improve the transportation efficiency of the equipment and the surface wear of the equipment. When selecting a classifier as a common classification equipment for Raymond mills, there are usually some problems, especially when coarse and fine classification of limestone powder. These problems are mainly concentrated on the mismatch between performance parameters such as air speed at the air inlet box, spindle speed and classifier speed. This phenomenon leads to low efficiency of Raymond Mill in classifying materials, serious coarseness, and more classification is required to meet the use requirements. There is currently no relevant systematic and quantitative evaluation and treatment method. Summary of the invention

[0003] In view of the fact that there is no evaluation method for both the classification efficiency and the pressure loss of the whole machine of the Raymond Mill in the prior art, a predictive evaluation method for the operation classification of the Raymond Mill is proposed.

[0004] A predictive evaluation method for Raymond Mill operation classification is disclosed. The predictive evaluation method is based on inlet wind speed D, main shaft speed E and classifier speed F. The Box-Behnken method, DPM model and MRF multi-coordinate reference system are combined to perform response surface simulation analysis with the discrete phase of RR distribution and calculate the maximum value of classification efficiency M and the minimum value of whole machine pressure loss N. The maximum value of classification efficiency M and the minimum value of whole machine pressure loss N are used to complete the comparison and analysis process with the actually obtained data.

[0005] As a preferred solution: with an inlet wind speed D of 25-33m / s, a main shaft speed E of 72-92r / min, and a classifier speed F of 308-338r / min as flow field boundary conditions, experiments are carried out through Ansys Fluent simulation software according to the Box-Behnken design method. The collection efficiency calculation method in the Box-Behnken design method is: after the Fluent steady-state simulation calculation is completed, the statistics of the particle information captured at the outlet of the whole machine are completed in the Sample of Discrete Phase under Reports, and then the mass flow rate of particles of each mesh size is obtained, and the collection efficiency is calculated accordingly. The collection efficiency is the initial preparation data of the classification efficiency M.

[0006] As a preferred solution: By combining the Box-Behnken method, the DPM model and the MRF multi-coordinate reference system, the response surface simulation analysis is performed with the discrete phase of the RR distribution and the process of calculating the pressure loss N of the whole machine is as follows:

[0007] Raymond mill flow field model construction and boundary condition setting:

[0008] The 3D model of the Raymond mill was constructed by using AutoCAD and UG software. Then the Spaceclaim software under ANSYS was used to simplify the whole machine model and extract the flow field model inside the Raymond mill to ensure that the flow field model was obtained correctly. Secondly, the meshing software under ANSYS was used to mesh the flow field of the Raymond mill. Finally, the drawn mesh file was imported into Fluent. The appropriate calculation model was used to set the boundary conditions and solution method for numerical simulation. The numerical simulation results were compared and analyzed with the actual data obtained from the Raymond mill to verify the calculation model, boundary conditions and solution conditions. Accuracy, steady-state and pressure solvers are used for solution. The turbulence model adopts the RNG model in the k-ε double equation, the DPM model and the MRF multi-coordinate reference system are used, the discrete phase adopts the RR distribution, the particles are launched perpendicular to the grinding roller surface with an initial velocity of zero, the hydraulic diameter at the air inlet is 0.485m, and the coupling solution method is used for solution. The default under-relaxation factor is used to perform statistics on the Box-Behnken test design and results, and the variance analysis is performed on the fitting model to form the standard table 1 for the variance analysis of the main engine pressure loss. According to the data in the standard table 1 for the variance analysis of the main engine pressure loss, the fitting regression equation of the main engine pressure loss N is established, which is as follows:

[0009] N=949.06-145.46D+38.81E-0.42F-2.70DE-0.02DF+0.01EF+4.91D 2 +0.48E 2 +1.63×10 -5 F 2 (1)

[0011] The Pr value of the model used in the regression analysis results is compared with the significant level standard value of 0.05. When the Pr value is less than the significant level standard value of 0.05, it means that the model is reliable; when the Pr value is greater than the significant level standard value of 0.05, it means that the model is unreliable; the lack of fit term is used as an indicator to evaluate the degree of fit between the model and the test data. Its value is 0.052. The lack of fit term is greater than 0.05, indicating that there is no lack of fit phenomenon, that is, there is good consistency between the model and the test data;

[0012] The correlation coefficient r of the model 2 It indicates the degree to which the model can explain the response value. According to the fitting regression equation of the main engine pressure loss Y, the factors affecting the comprehensive score are ranked in order of importance as follows: inlet wind speed D, main shaft speed E, and whole machine outlet pressure. An increase in inlet wind speed D will lead to an increase in pressure loss, while an increase in main shaft speed E will help reduce pressure loss.

[0013] As a preferred solution: statistics are conducted on the Box-Behnken test design and results, variance analysis is conducted on the fitting model, and the variance analysis standard table 2 of the graded efficiency M is formed. Based on the data of the variance analysis standard table 2 of the main engine pressure loss, a fitting regression equation of the graded efficiency M is established. The response surface fitting equation of the graded efficiency M is:

[0014] M=-31.42314-43.83017D+6.52373E+3.12442F+0.12819DE+0.038208DF-0.037308EF+0.35797D 2 +0.013450E 2 -2.09167e -3 F 2 ; (2)

[0015] The response surface fitting equation of the classification efficiency M was input into Origin 2022 to obtain the response surface of the classification efficiency M. When the spindle speed E is a fixed value, the classification efficiency M increases with the increase of the inlet wind speed D and the decrease of the classifier speed F. When the inlet wind speed D is constant, with the increase of the spindle speed E and the decrease of the classifier speed F, the collection efficiency shows a trend of first decreasing and then increasing. The optimal value of the spindle speed E and the optimal value of the classifier speed F are determined when the collection efficiency is close to 100%.

[0016] As a preferred solution, we start the optimization calculation with the goal of maximizing the classification efficiency M and minimizing the whole machine pressure loss N, and establish the optimization equation:

[0017]

[0018] According to the above response surface equation, it is calculated that the maximum value of the classification efficiency M and the minimum value of the whole machine pressure loss N are obtained by the inlet wind speed D, the main shaft speed E and the classifier speed F. By comparing and analyzing with the actually obtained data, the evaluation process of the actually obtained data is obtained. When the classification efficiency M in the actually obtained data is lower than 60% of the maximum value of the classification efficiency M compared with the maximum value of the classification efficiency M, and the whole machine pressure loss N in the actually obtained data is higher than 60% of the minimum value of the whole machine pressure loss N compared with the minimum value of the whole machine pressure loss N, then the actually obtained data is low-quality data, indicating that the corresponding Raymond mill is in a poor operating classification state and should be maintained and adjusted accordingly; when the classification efficiency M in the actually obtained data is 60-80% of the maximum value of the classification efficiency M compared with the maximum value of the classification efficiency M, and the whole machine pressure loss N in the actually obtained data is compared with the minimum value of the whole machine pressure loss N, the difference between the actual data of the whole machine pressure loss N and the minimum value of the whole machine pressure loss N is 10-40% of the minimum value of the whole machine pressure loss N, then the actually obtained data is high-quality data, indicating that the corresponding Raymond mill is in a good operating classification state.

[0019] Compared with the prior art, the present invention has the following beneficial effects:

[0020] The present invention realizes the in-depth regular optimization processing process of the classifier of the Raymond mill, optimizes the speed and pressure flow field inside the classifier, so as to achieve the goals of improving classification efficiency, reducing energy consumption and reducing the particle size distribution range. In the optimization and evaluation process, the present invention determines the coordination of key performance parameters combining the air intake box inlet wind speed, the spindle speed and the classifier speed. By adjusting these parameters, we can achieve better material classification effect and reduce the occurrence of coarse running. Through in-depth research and optimization, it is possible to conduct corresponding evaluations on the operation classification and the whole machine pressure loss of the existing Raymond mill, and provide favorable and reliable basis support for subsequent rectification, so that the Raymond mill can more efficiently cope with the coarse and fine classification of limestone powder, thereby helping to improve the overall production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 This is a schematic diagram of the force analysis of particles in the gravity classification area;

[0022] Figure 2a The response surface simulation diagram of DE factor in the response surface of fractional efficiency M;

[0023] Figure 2b The response surface simulation diagram of DF factor in the response surface of fractional efficiency M;

[0024] Figure 2c It is a response surface simulation diagram of EF factor in the response surface of fractional efficiency M;

[0025] Figure 3aIt is a schematic diagram of the response surface simulation of the interaction between inlet wind speed D and main shaft speed E;

[0026] Figure 3b It is a schematic diagram of response surface simulation of the interaction between inlet wind speed D and outlet pressure of the whole machine;

[0027] Figure 3c This is a schematic diagram of the response surface simulation of the interaction between the spindle speed E and the outlet pressure of the whole machine;

[0028] Figure 4 This is a schematic diagram of the wind field trajectory simulation during the test at 9 o'clock inside the Raymond machine;

[0029] Figure 5 This is a schematic diagram of the wind field trajectory simulation during the test at 14 o'clock inside the Raymond machine;

[0030] Figure 6 This is the velocity cloud diagram of the internal flow field during the test 2 inside the Raymond machine;

[0031] Figure 7 This is the velocity cloud diagram of the internal flow field inside the Raymond machine during the test at 7 o'clock;

[0032] Figure 8 This is the internal pressure cloud diagram of the Raymond machine during test 2;

[0033] Fig. 9 This is the internal pressure cloud diagram of the Raymond machine during test 7;

[0034] Fig.10 It is a comparative diagram of the effect of classifier rotation speed on the passing rate and classification rate of gravity classification area;

[0035] Fig.11 This is a schematic diagram of the retention time of finished particles;

[0036] Fig.12 This is the velocity cloud diagram of the coarse particle trajectory. DETAILED DESCRIPTION

[0037] In order to make the purpose, technical scheme and advantages of the present invention clearer, the present invention is described below by the specific embodiments shown in the accompanying drawings. However, it should be understood that these descriptions are only exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, the description of well-known structures and technologies is omitted to avoid unnecessary confusion of the concept of the present invention.

[0038] Specific implementation method 1: Combination Figures 1 to 12This embodiment is described. The prediction and evaluation method for the operation classification of the Raymond mill in this embodiment is based on the inlet wind speed D, the main shaft speed E, and the classifier speed F. By combining the Box-Behnken method, the DPM model and the MRF multi-coordinate reference system, the response surface simulation analysis is performed with the discrete phase of the RR distribution, and the maximum value of the classification efficiency M and the minimum value of the whole machine pressure loss N are calculated. The comparison and analysis process with the actually obtained data is completed based on the maximum value of the classification efficiency M and the minimum value of the whole machine pressure loss N, wherein the inlet wind speed D is the inlet wind speed D of the air inlet box.

[0039] Combination Figure 1 As shown, the force analysis process of the particles in the classifier in this embodiment is as follows:

[0040] During the working process of the classifier, the classification of material particles is mainly carried out under the combined action of three forces, including the spiral rising gas drag force F d , centrifugal inertia force F c and gravity G. The interaction of these forces determines the trajectory of the particles in the classifier, which in turn affects the classification effect. The velocity of the particles can be decomposed into the axial velocity (u p ) L , radial velocity (u p ) r and tangential velocity (u p ) t Three mutually perpendicular component velocities are used for analysis:

[0041] 1) Axial speed (u p ) L . Due to the interaction between the rising airflow and gravity, the particles in the Raymond Mill may move upward or downward in the axial direction because the two forces are in opposite directions. For some coarse particles, when the rising speed of the airflow is less than the settling speed of gravity, the particles move downward. When the rising speed of the airflow on the particles is greater than the settling speed of gravity, the particles move upward. Therefore, the smaller the particle diameter, the rising speed of the particles is almost close to the rising speed of the airflow.

[0042] 2) Radial velocity (u p ) r It is caused by the inertial centrifugal force generated by the material particles around the center of the classifier's rotating shaft.

[0043] 3) Tangential velocity (u p ) t It is caused by the circumferential speed of the material particles and airflow rotating around the center of the classifier shaft.

[0044] Based on the motion theory of particle centrifugal classification, when particles move in a circular motion with the airflow, the combined effect of the axial velocity and radial velocity causes the particles to take a spiral motion trajectory. This spiral motion can be upward or downward. Among them, fine particles will move in a spiral upward compared to coarse particles.

[0045] The particle force calculation method process is:

[0046] In the centrifugal field, the resistance of the particles in the medium is opposite to the centrifugal force, which can be expressed by the Tosks resistance formula:

[0047] F d =3πηdV r (5-1)

[0048] Where η is the medium viscosity, Pa·s; δ and ρ are the particle material and medium density, kg / m 3 ; V r Medium flow rate, m / s;

[0049] The centrifugal force on a particle with density δ is:

[0050]

[0051] Where r is the particle's radius of gyration, m; ω is the particle's gyration speed, r / min; V t is the tangential velocity of the particle, m / s;

[0052] If F d >F c , that is, the resistance of the particles is greater than the centrifugal force, the particles will move toward the wall of the classifier and be discharged and collected as coarse particles. d <F c When the centrifugal force on the particles is greater than the resistance, the particles will be discharged with the airflow and finally collected at the outlet. d =F c When the particle group theoretically moves in a circular motion, the final centrifugal sedimentation velocity and the classification particle size can be obtained as follows:

[0053]

[0054] Where: d r -Grading particle size, m; Vt-impeller average speed, m / s; Vr-medium flow rate, m / s; η-air viscosity; δ-material density, kg / m 3 ; r-mean radius of impeller, m; ρ-air flow density, kg / m 3 .

[0055] Specific implementation method two: This implementation method is a further limitation of specific implementation method one. In this implementation method, an inlet wind speed D of 25 to 33 m / s, a spindle speed E of 72 to 92 r / min, and a classifier speed F of 308 to 338 r / min are used as flow field boundary conditions. The experiment is carried out through Ansys Fluent simulation software according to the Box-Behnken design method. The collection efficiency calculation method in the Box-Behnken design method is: after the Fluent steady-state simulation calculation is completed, the statistics of the particle information captured at the outlet of the whole machine are completed in the Sample of Discrete Phase under Reports, and then the mass flow rate of particles of each mesh size is obtained, and the collection efficiency is calculated accordingly. The collection efficiency is the initial preparation data of the classification efficiency M.

[0056] In this implementation, the inlet wind speed D, the main shaft speed E, and the classifier speed F are selected as three factors, and the classification efficiency M and the whole machine pressure loss N are selected as the target optimization parameters. In order to deeply understand the comprehensive influence of the three factors on the whole machine performance, the inlet wind speed D is selected as 25, 29 and 33m / s, the main shaft speed E is selected as 72, 82 and 92r / min, and the classifier speed is selected as 308, 338 and 368r / min. The boundary conditions are set as shown in Table 1.

[0057] Table 1 Flow field boundary conditions of Raymond mill

[0058]

[0059] The steady-state and pressure solvers are used for solution, and the turbulence model adopts the RNG model in the k-ε double equation. Because it has better effect in handling flows with high strain rates and large streamline curvature, it can better simulate the internal flow field of the Raymond mill. The DPM model and MRF multi-coordinate reference system are used, and the discrete phase adopts RR distribution. The particles are launched perpendicular to the grinding roller surface with an initial velocity of zero. The hydraulic diameter at the air inlet is 0.711m, and the hydraulic diameter at the air inlet is 0.721m. The coupled solution method is used for solution, and the default under-relaxation factor is used.

[0060] Specific implementation method three: This implementation method is a further limitation of specific implementation method one or two. In this implementation method, Ansys Fluent simulation software is used to perform experimental design based on the Box-Behnken design method in Design-Expert10.0 software. The following Table 2 is the Box-Behnken experimental design and result statistics. The collection efficiency calculation method is as follows: After the Fluent steady-state simulation calculation is completed, the statistics of the particle information captured at the outlet of the whole machine are completed in the Sample of Discrete Phase under Reports, and then the mass flow rate of particles of each mesh size is obtained. For example, when the inlet wind speed D is 25m / s, the spindle speed E is 72r / min, and the classifier speed F is 338r / min, the collected 200-mesh mass flow rate is 3.3114, respectively, then the collection efficiency m is

[0061]

[0062] The calculation method of Raymond mill collection efficiency for other simulation conditions is the same as above. The obtained classification efficiency M and whole machine pressure loss N are input into the result statistical table to obtain Table 3, which is the variance analysis table of classification efficiency M and whole machine pressure loss N:

[0063] Table 2 Box-Behnken test design and result statistics

[0064]

[0065] Table 3 Variance analysis of classification efficiency M and whole machine pressure loss N

[0066]

[0067] In the above table, when Pr < 0.05, ※ indicates that the factor has a significant impact on the test index.

[0068] Specific implementation method 4: This implementation method is a further limitation of specific implementation methods 1, 2 or 3. Figure 3a , Figure 3b and Figure 3c As shown, in this embodiment, the Box-Behnken method, the DPM model and the MRF multi-coordinate reference system are combined to perform response surface simulation analysis with the discrete phase of the RR distribution and calculate the process of the whole machine pressure loss N as follows:

[0069] Raymond mill flow field model construction and boundary condition setting:

[0070] The 3D model of the Raymond Mill machine was constructed by using AutoCAD and UG software. Then the Spaceclaim software under ANSYS was used to simplify the whole machine model and extract the flow field model inside the Raymond Mill machine to ensure that the flow field model was obtained correctly. Secondly, the meshing software under ANSYS was used to mesh the flow field of the Raymond Mill machine. Finally, the drawn mesh file was imported into Fluent. The appropriate calculation model was used to set the boundary conditions and solution method for numerical simulation. The numerical simulation results were compared and analyzed with the actual data obtained from the Raymond Mill machine to verify the accuracy of the calculation model, boundary conditions and solution conditions. , the steady-state and pressure solvers are used for solving, the turbulence model adopts the RNG model in the k-ε double equation, the DPM model and the MRF multi-coordinate reference system are used, the discrete phase adopts the RR distribution, the particles are launched perpendicular to the grinding roller surface with an initial velocity of zero, the hydraulic diameter at the air inlet is 0.485m, and the coupling solution method is used for solving. The default under-relaxation factor is used to statistically analyze the Box-Behnken test design and results to form Table 4, and the variance analysis of the fitting model is performed to form the standard table 5 of the variance analysis of the main engine pressure loss. According to the data of the standard table 4 and table 5 of the variance analysis of the main engine pressure loss, the fitting regression equation of the main engine pressure loss N is established, which is as follows:

[0071] Table 4 Box-Behnken test results statistics

[0072]

[0073]

[0074] Table 5. Standard table for variance analysis of main engine pressure loss

[0075]

[0076]

[0077] The fitting regression equation of the main engine pressure loss N established based on the data in Table 4 and Table 5 of the main engine pressure loss variance analysis standard is:

[0078] N=949.06-145.46D+38.81E-0.42F-2.70DE-0.02DF+0.01EF+4.91D 2 +0.48E 2 +1.63×10 -5 F 2 (1)

[0080] The Pr value of the model used in the regression analysis results is 0.0309, which is less than 0.05, indicating that the model is statistically reliable. The lack of fit term is an indicator of the degree of fit between the model and the experimental data. Its value is 0.052 and is greater than 0.05, indicating that there is no lack of fit, that is, there is good consistency between the model and the experimental data. The correlation coefficient r of the model 2 It is 0.8511, indicating that the model can explain 85.11% of the response value changes. According to this model, the factors affecting the comprehensive score are ranked in order of importance: inlet wind speed D, spindle speed E, and whole machine outlet pressure F. Among them, the inlet wind speed D and spindle speed E have a significant impact on the model. The increase in inlet wind speed D will lead to an increase in pressure loss, while the increase in spindle speed E will help reduce pressure loss.

[0081] The final conclusion is: the Pr value of the model used is compared with the significant level standard value of 0.05. When the Pr value is less than the significant level standard value of 0.05, it means that the model is reliable; when the Pr value is greater than the significant level standard value of 0.05, it means that the model is unreliable; the lack of fit term is an indicator of the degree of fit between the model and the test data, and its value is 0.052. The lack of fit term is greater than 0.05, indicating that there is no lack of fit phenomenon, that is, there is good consistency between the model and the test data;

[0082] The correlation coefficient r of the model 2 It indicates the degree to which the model can explain the response value. According to the fitting regression equation of the main engine pressure loss Y, the factors affecting the comprehensive score are ranked in order of importance as follows: inlet wind speed D, main shaft speed E, and whole machine outlet pressure F. An increase in inlet wind speed D will lead to an increase in pressure loss, while an increase in main shaft speed E will help reduce pressure loss.

[0083] The main engine pressure loss analysis process in this embodiment is as follows:

[0084] The energy equation for a steady flow with energy input is described as:

[0085]

[0086] In the above formula, W shaft，net in is the unit mass shaft work input to the fluid; is the flow energy per unit mass; is the kinetic energy per unit mass; gz is the potential energy per unit mass of the fluid; e is the internal energy; q net in is the unit mass heat transferred to the fluid; intermolecular collisions are not considered in this paper, so the internal energy and the heat transferred to the fluid (u2-u1-q net in ) is negligible; P is absolute pressure, Pa; ρ is density, kg / m3; V is velocity, m / s; g is gravitational acceleration, m / s2; z represents spatial position, m;

[0087] According to the above analysis, formula 6 is reorganized, and the change law of the flow energy of gas at the inlet and outlet of the mill main engine is as follows formula 7:

[0088]

[0089] The difference between P2 and P1 in the above formula is the pressure loss of the Raymond mill.

[0090] When the fluid flows through the wall structure of the mill, the pressure is inevitably reduced due to the generation of structural resistance and the dissipation of flow energy by vortices. Combined with Formula 7, it can be seen that the flow energy of the pressure component is a component of the energy balance equation. The greater the pressure loss, the greater the flow energy loss of the mill. For the collection of finished products, the kinetic energy change caused by the speed change should remain stable, and the potential energy change should be consistent with the equipment height; therefore, the increase in pressure loss will lead to the need for greater shaft power input to maintain energy conservation, that is, to ensure a greater main engine air intake, spindle speed E and classifier speed F motor energy input. Therefore, pressure loss is an important component that can reflect the energy consumption of the mill.

[0091] Box-Behnken test results combined with response surface Figure 3a , Figure 3b and Figure 3c As shown in the figure, when the outlet pressure of the whole machine remains unchanged, with the increase of inlet wind speed D and the continuous decrease of spindle speed E, the pressure loss of the main engine also increases continuously. When the inlet wind speed D is 46m / s and the spindle speed E is 100r / min, the pressure loss of the main engine reaches a maximum value of 1660Pa. When the spindle speed E remains unchanged, with the continuous increase of inlet wind speed D and the continuous decrease of main engine outlet pressure, the pressure loss of the main engine also increases continuously, with a maximum value of about 1320Pa. When the inlet wind speed D remains unchanged, with the continuous decrease of spindle speed E and main engine outlet pressure, the pressure loss of the main engine increases slowly, with a maximum value of about 1260Pa.

[0092] Specific implementation method 5: This implementation method is a further limitation of specific implementation methods 1, 2, 3 or 4, combined with Table 4, Table 5, Figure 4 and Figure 5 As shown, the parameters corresponding to test 9 are 38m / s, 120r / min, 3500Pa, and the parameters corresponding to test 14 are 46m / s, 100r / min, 3500Pa. The two are used as comparative tests. The minimum and maximum values ​​of the corresponding host pressure loss results are 880Pa and 1660Pa respectively. Figure 4 and Figure 5The wind field traces in the grinding roller area are shown. Fluid experiments show that when the critical Reynolds number is exceeded, a series of complex changes will occur, which will lead to a sharp change in flow characteristics, and the flow will be in a disordered chaotic state, which is called turbulence. Turbulent flow with a vortex flow structure is a turbulent vortex. The wind field characteristics inside the Raymond mill show complex turbulent flow characteristics, and there are a large number of vortex areas. There are moving parts in the main engine and classifier area that rotate around the center of the structure and have a high speed, which will cause drastic changes in the flow field. Combined with Figure 4 and Figure 5 It can be seen that the wind field traces in the grinding roller area show an overall trend of circling upward, and local vortices caused by wall effects are generated on the lower side of the plum blossom frame and the top area of ​​the plum blossom frame. The main engine pressure loss of test 9 is the smallest, and the wind field traces emitted from the grinding roller have a spiral upward trend and enter the classifier area more concentratedly. However, the main engine pressure loss of test 14 is the largest. Specifically, the distribution of the wind field traces emitted from the grinding roller is relatively chaotic, and there are large-sized vortex areas. The large-sized vortices will continuously obtain energy from the mainstream, resulting in huge flow field energy losses, which is reflected in the significant increase in pressure loss in test 14. In order to further determine the influence of wind speed and rotation speed on pressure loss, a single-factor comparative study is conducted below.

[0093] Specific implementation method 6: This implementation method is a further limitation of specific implementation methods 1, 2, 3, 4 or 5. Figure 2a , 2b As shown in Figure 2c, in this embodiment, the Box-Behnken test design and results are statistically analyzed, and the variance analysis is performed on the fitting model to form the standard table 2 for variance analysis of the fractional efficiency M. The fitting regression equation of the fractional efficiency M is established based on the data of the standard table 2 for variance analysis of the main engine pressure loss. The response surface fitting equation of the fractional efficiency M is:

[0094] M=-31.42314-43.83017D+6.52373E+3.12442F+0.12819DE+0.038208DF-0.037308EF+0.35797D 2 +0.013450E 2 -2.09167e -3 F 2 ; (2)

[0096] The response surface fitting equation of the classification efficiency M was input into Origin 2022 to obtain the response surface of the classification efficiency M. When the spindle speed E is a fixed value, the classification efficiency M increases with the increase of the inlet wind speed D and the decrease of the classifier speed F. When the inlet wind speed D is constant, with the increase of the spindle speed E and the decrease of the classifier speed F, the collection efficiency shows a trend of first decreasing and then increasing. The optimal value of the spindle speed E and the optimal value of the classifier speed F are determined when the collection efficiency is close to 100%.

[0097] Specific implementation method 7: This implementation method is a further limitation of specific implementation methods 1, 2, 3, 4, 5 or 6. The optimization calculation is started with the goal of maximizing the classification efficiency M and minimizing the whole machine pressure loss N, and the optimization equation is established, specifically:

[0098]

[0099] According to the above response surface equation, it is calculated that when the inlet wind speed D is 23.21m / s, the spindle speed E is 70r / min and the classifier speed F is 320.70r / min, the target result is optimal, and the classification efficiency M is 99.95% and the pressure loss is 910.43Pa. This parameter combination is set in the simulation, and the simulation results are shown in Table VI. After optimization, the classification efficiency M is increased by 4.33%. Based on the fact that the pressure loss cannot be measured by the company's current detection, it can be seen from the simulation results that the pressure loss N of the whole machine after optimization is reduced by 140.99Pa compared with the actual relevant data.

[0100] Table 6 Response surface optimization results and simulation results

[0101]

[0102] In summary, the inlet wind speed D, the main shaft speed E and the classifier speed F have all decreased compared with the initial operating parameters. While improving the classification efficiency M of the Raymond mill, the energy consumption of the whole machine has also been reduced.

[0103] According to the above response surface equation, it is calculated that the maximum value of the classification efficiency M and the minimum value of the whole machine pressure loss N are obtained by the inlet wind speed D, the main shaft speed E and the classifier speed F. By comparing and analyzing with the actually obtained data, the evaluation process of the actually obtained data is obtained. When the classification efficiency M in the actually obtained data is lower than 60% of the maximum value of the classification efficiency M compared with the maximum value of the classification efficiency M, and the whole machine pressure loss N in the actually obtained data is higher than 60% of the minimum value of the whole machine pressure loss N compared with the minimum value of the whole machine pressure loss N, then the actually obtained data is low-quality data, indicating that the corresponding Raymond mill is in a poor operating classification state and should be maintained and adjusted accordingly; when the classification efficiency M in the actually obtained data is 60-80% of the maximum value of the classification efficiency M compared with the maximum value of the classification efficiency M, and the whole machine pressure loss N in the actually obtained data is compared with the minimum value of the whole machine pressure loss N, the difference between the actual data of the whole machine pressure loss N and the minimum value of the whole machine pressure loss N is 10-40% of the minimum value of the whole machine pressure loss N, then the actually obtained data is high-quality data, indicating that the corresponding Raymond mill is in a good operating classification state.

[0104] Specific implementation eight: This implementation is a further limitation of specific implementation one, two, three, four, five, six or seven. In this implementation, the influence of the classifier rotation speed F on the flow field velocity of the Raymond mill and the analysis process are as follows:

[0105] Combination Figure 6 and Figure 7 As shown, based on the analysis of the response surface, in order to further analyze the influence of the classifier speed F on the collection efficiency, Experiment 2 and Experiment 7 were selected as comparative experiments to analyze the changes in the internal flow field of the Raymond mill, as well as the pass rate and classification efficiency M in the gravity classification zone, to provide a reference for optimizing the classifier speed F. The operating parameters of Experiment 2 were 25m / s, 82r / min, and 308r / min, and the operating parameters of Experiment 7 were 25m / s, 82r / min, and 368r / min.

[0106] When other operating parameters remain unchanged, the classifier speed F reaches the maximum value of 368r / min. Figure 6 and Figure 7As shown, it can be observed that the local wind field formed on the periphery of the classifier has a relatively high speed. Especially at the edge of the frame position, the wind field speed is more significant, even as high as 157m / s. This local wind field enables the particles to obtain greater kinetic energy, accelerates the movement of the particles, and causes more coarse particles to be brought into the classifier, or causes the "coarse running" phenomenon. When the classifier speed F drops to 308r / min, the velocity flow field distribution inside the Raymond mill is more uniform. This uniform velocity distribution contributes to the uniform distribution and collection of particles, thereby improving the efficiency and accuracy of classification. A relatively low speed means a smoother wind field, and the particles will no longer obtain greater kinetic energy locally, making the particles more easily attracted to the collection area. Therefore, choosing an appropriate speed can effectively improve the efficiency and accuracy of the entire classification process.

[0107] Combination Figure 8 and Fig. 9 As shown in the figure, the influence of the classifier speed F on the flow field pressure of the Raymond mill in this embodiment and the analysis process are as follows: Test 2 and Test 7 are also used to compare and analyze the flow field pressure of the whole machine. When other operating parameters remain unchanged, if the speed of the classifier is adjusted to the maximum value of 368r / min, according to Figure 8 and Fig. 9 As shown, it can be observed that a local high-pressure area is formed on the periphery of the classifier, with a pressure as high as -25.6Pa. This area is almost consistent with the high-speed area of ​​the velocity field, which means that the particles with greater kinetic energy have intensified collisions with the classifier shell and the rotating cage. This collision causes increased energy consumption, resulting in a large pressure loss in this area, and thus the particle collection efficiency is reduced. When the classifier speed F is 308r / min, the pressure distribution inside the Raymond mill is more uniform, the pressure loss is smaller, and the particles can enter the classifier more smoothly and be collected. Therefore, by adjusting the speed of the classifier to 308r / min, the pressure distribution inside the Raymond mill can be uniform, the pressure loss can be reduced, thereby promoting the smooth flow of particles and improving the overall performance of the classifier and the particle collection efficiency.

[0108] In this embodiment, the influence of the classifier speed F on the Raymond mill classification efficiency M and the analysis process are as follows:

[0109] according to Figure 3a , 3bFrom the response surface analysis results shown in Figure 3c, it can be seen that the classifier speed F is a significant factor. When the inlet wind speed D is 25m / s, it meets the relevant requirements for the actual 50 mesh cutting particle size. The operating parameters of this test are inlet wind speed D of 25m / s and main spindle speed E of 82r / min. On this basis, experiment 2 with operating parameters of 25m / s, 82r / min, 308r / min and experiment 7 with operating parameters of 25m / s, 82r / min, 368r / min are selected as comparative experiments to study the relationship between gravity classification pass rate and collection efficiency at different classifier speeds F.

[0110] Depend on Fig.10 It can be seen that the finished product passing rate in the gravity classification area is 80.41%. When the classifier rotates and the speed is 308r / min, the Raymond mill shaft work increases, the kinetic energy provided to the particles increases, and more finished products are collected, so the collection efficiency of the whole machine increases to 97.9%, which is a high collection efficiency. However, when the classifier speed F increases to 368r / min, the collection efficiency of the whole machine decreases to 64.41%. This is because the local wind field speed is high and the wind field force is greater than the centrifugal force, so many coarse products are collected.

[0111] Specific implementation method 9: This implementation method is a further limitation of specific implementation methods 1, 2, 3, 4, 5, 6, 7 or 8. Fig.11 and Fig.12 It can be seen that in this embodiment, the finished particle ID is searched in the collected particle information file, named ID1 and ID2 respectively, and the particle code is entered in Fluent to obtain the combined Fig.11 Particle trajectories shown.

[0112] By analyzing the particle trajectory, it can be seen that the particles drill into the rotating cage along the outer shell of the classifier under the action of the classifier to complete the final classification process. In this process, the retention time of the finished particles with ID 1 and 2 is 0.534s and 1.46s. In particular, from the particle trajectory of ID2, it can be seen that when the diameter of the classifier increases, the particle's running trajectory is almost the same as Figure 1 The running trajectories of the particles in the simulation remain consistent, all spiraling up along the wall, indicating that the simulation can accurately reflect the physical processes in the real system and provide a reliable basis for further engineering analysis and optimization.

[0113] Refer to the selection method of finished product particle ID, combined with Fig.12As shown in the figure, two groups of particles with ID 3 and 4 are selected from the coarse particles and their movement under the action of the classifier is observed. It can be found that the speed of the particles increases under the action of the classifier, and then after spiraling up, the speed decreases to 0m / s at the wall of the classifier, and finally adheres to the wall. After the Raymond Mill has been working for a period of time, this phenomenon may cause dust to accumulate on the wall. The accumulated dust will affect the movement path of the particles, resulting in a decrease in the screening performance of the classifier, and even equipment failure or shutdown.

[0114] The present invention may also have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

Claims

1. A method for predicting and evaluating the operation classification of a Raymond Mill, characterized in that: The prediction and evaluation method is based on the inlet wind speed D, the main shaft speed E, and the classifier speed F. By combining the Box-Behnken method, the DPM model and the MRF multi-coordinate reference system, the response surface simulation analysis is carried out with the discrete phase of the RR distribution, and the maximum value of the classification efficiency M and the minimum value of the whole machine pressure loss N are calculated. The analysis process of comparing with the actual acquired data is completed based on the maximum value of the classification efficiency M and the minimum value of the whole machine pressure loss N.

2. The method for predicting and evaluating the operation classification of a Raymond mill according to claim 1, characterized in that: The inlet wind speed D of 25-33m / s, the main shaft speed E of 72-92r / min, and the classifier speed F of 308-338r / min were used as flow field boundary conditions through Ansys Fluent simulation software and the experiment was carried out according to the Box-Behnken design method. The collection efficiency calculation method in the Box-Behnken design method is: after the Fluent steady-state simulation calculation is completed, the statistics of the particle information captured at the outlet of the whole machine are completed in the Sample of Discrete Phase under Reports, and then the mass flow rate of particles of each mesh size is obtained, and the collection efficiency is calculated accordingly. The collection efficiency is the initial preparation data of the classification efficiency M.

3. The method for predicting and evaluating the operation classification of a Raymond mill according to claim 2, characterized in that: By combining the Box-Behnken method, the DPM model and the MRF multi-coordinate reference system, the response surface simulation analysis is performed with the discrete phase of the RR distribution and the process of calculating the pressure loss N of the whole machine is as follows: Raymond mill flow field model construction and boundary condition setting: The 3D model of the Raymond Mill machine was constructed by using AutoCAD and UG software. Then the Spaceclaim software under ANSYS was used to simplify the whole machine model and extract the flow field model inside the Raymond Mill machine to ensure that the flow field model was obtained correctly. Secondly, the meshing software under ANSYS was used to mesh the flow field of the Raymond Mill machine. Finally, the drawn mesh file was imported into Fluent, and the appropriate calculation model was used to set the boundary conditions and solution method for numerical simulation. The numerical simulation results were compared and analyzed with the actual data obtained from the Raymond mill to verify the accuracy of the calculation model, boundary conditions and solution conditions. The steady-state and pressure solvers were used for solution. The turbulence model adopted the RNG model in the k-ε double equation, the DPM model and the MRF multi-coordinate reference system were used, the discrete phase adopted the RR distribution, the particles were launched perpendicular to the grinding roller surface with an initial velocity of zero, the hydraulic diameter at the air inlet was 0.485m, and the coupling solution method was used for solution. The default under-relaxation factor was used to perform statistics on the Box-Behnken test design and results, and the variance analysis was performed on the fitting model to form the standard table 1 for the variance analysis of the main engine pressure loss. According to the data in the standard table 1 for the variance analysis of the main engine pressure loss, the fitting regression equation of the main engine pressure loss N was established, which is as follows: N=949.06-145.46D+38.81E-0.42F-2.70DE-0.02DF+0.01DF+4.91D 2 +0.48E 2 +1.63×10 -5 F 2 (1) The Pr value of the model used in the regression analysis results is compared with the significant level standard value of 0.

05. When the Pr value is less than the significant level standard value of 0.05, it means that the model is reliable; when the Pr value is greater than the significant level standard value of 0.05, it means that the model is unreliable; the lack of fit term is used as an indicator to evaluate the degree of fit between the model and the test data. Its value is 0.

052. The lack of fit term is greater than 0.05, indicating that there is no lack of fit phenomenon, that is, there is good consistency between the model and the test data; The correlation coefficient r of the model 2 It indicates the degree to which the model can explain the response value. According to the fitting regression equation of the main engine pressure loss Y, the factors affecting the comprehensive score are ranked in order of importance as follows: inlet wind speed D, main shaft speed E, and whole machine outlet pressure. An increase in inlet wind speed D will lead to an increase in pressure loss, and an increase in main shaft speed E will correspond to a decrease in pressure loss.

4. A method for predicting and evaluating the operation classification of a Raymond mill according to claim 2 or 3, characterized in that: The Box-Behnken test design and results were statistically analyzed, and the variance analysis of the fitting model was performed to form the standard table 2 for variance analysis of the fractional efficiency M. The fitting regression equation of the fractional efficiency M was established based on the data of the standard table 2 for variance analysis of the main engine pressure loss. The response surface fitting equation of the fractional efficiency M is: M=-31.42314-43.83017D+6.52373E+3.12442F+0.12819DE+0.038208DF-0.037308EF+0.35797D 2 +0.013450E 2 -2.09167e -3 F 2 ; (2) The response surface fitting equation of the classification efficiency M was input into Origin 2022 to obtain the response surface of the classification efficiency M. When the spindle speed E is a fixed value, the classification efficiency M increases with the increase of the inlet wind speed D and the decrease of the classifier speed F. When the inlet wind speed D is constant, with the increase of the main shaft speed E and the decrease of the classifier speed F, the collection efficiency shows a trend of first decreasing and then increasing. The optimal value of the main shaft speed E and the optimal value of the classifier speed F are determined when the collection efficiency is close to 100%.

5. The method for predicting and evaluating the operation classification of a Raymond mill according to claim 4, characterized in that: The optimization calculation was started with the goal of maximizing the classification efficiency M and minimizing the whole machine pressure loss N, and the optimization equation was established, which is as follows: According to the above response surface equation, it is calculated that the maximum value of the classification efficiency M and the minimum value of the whole machine pressure loss N are obtained by the inlet wind speed D, the main shaft speed E and the classifier speed F. By comparing and analyzing with the actually obtained data, the evaluation process of the actually obtained data is obtained. When the classification efficiency M in the actually obtained data is lower than 60% of the maximum value of the classification efficiency M compared with the maximum value of the classification efficiency M, and the whole machine pressure loss N in the actually obtained data is higher than 60% of the minimum value of the whole machine pressure loss N compared with the minimum value of the whole machine pressure loss N, then the actually obtained data is low-quality data, indicating that the corresponding Raymond mill is in a poor operating classification state and should be maintained and adjusted accordingly; when the classification efficiency M in the actually obtained data is 60-80% of the maximum value of the classification efficiency M compared with the maximum value of the classification efficiency M, and the whole machine pressure loss N in the actually obtained data is compared with the minimum value of the whole machine pressure loss N, the difference between the actual data of the whole machine pressure loss N and the minimum value of the whole machine pressure loss N is 10-40% of the minimum value of the whole machine pressure loss N, then the actually obtained data is high-quality data, indicating that the corresponding Raymond mill is in a good operating classification state.

Images