Ring roller milling performance optimization method based on pendulum mill

Abstract

The invention discloses a ring roller milling performance optimization method based on a pendulum mill, and belongs to the technical field of pendulum mill optimization. In order to solve the problem that a currently designed pendulum mill cannot ensure higher yield within a short time, the method comprises the following steps: firstly, carrying out a primary grinding simulation test and a total grinding simulation test on a pendulum mill to be optimized, and determining whether the optimization direction is to improve the yield of one time or multiple times according to the grinding frequency and the corresponding yield; the shape of the grinding roller is optimized by further combining the relation between the abrasion degree of the grinding roller and the grinding ring and the gap; then single-factor simulation analysis is carried out to determine a to-be-fixed single factor and an optimized value thereof, wherein the single-factor simulation analysis comprises scraper knife angle single-factor simulation analysis, feeding quantity single-factor simulation analysis and main shaft rotating speed single-factor simulation analysis; and taking the optimization value of the to-be-fixed single factor as a fixed factor, taking the other two factors as variables, carrying out analysis based on a response surface, and determining optimization results of the other two factors by taking the yield as a target.

Description

Technical Field

[0001] The invention belongs to the technical field of swing mill optimization and relates to a swing mill optimization method. Background Art

[0002] The mainframe of the swing mill is the main mechanical structure for material grinding. The mainframe of the swing mill mainly includes a plum blossom frame 1, a feed port 2, a central shaft 3, an air inlet 4, a suspension cover 5, a grinding ring 6, a grinding roller 7, a scraper 8, a base 9 and other parts. Figure 1 As shown. During operation, the transmission system is responsible for transferring energy to the central axis, thereby driving the plum blossom frame to rotate. As the plum blossom frame rotates, the grinding roller device moves around the central axis and swings outward under the action of centrifugal force. This movement enables the grinding roller to exert pressure on the grinding ring, thereby generating a grinding effect between the grinding roller and the grinding ring. This is also the main working area for the swing mill to process materials. The material falling from the inlet first enters the grinding roller and grinding ring area, is captured by the grinding roller, and is crushed and ground for the first time. The qualified powder produced is blown to the classification area by the wind field of the main air duct; and the material that does not reach the classification fineness falls down into the scraper area, and is thrown back to the grinding area by the scraper rotating with the central axis, and is ground twice or multiple times to form a finished product.

[0003] In actual working conditions, the grinding roller and grinding ring are in the zero gap state in the installed state, and the wear caused by the production process gradually disappears; after entering the stable working state, the material thrown by the scraper will form a dynamic material layer between the grinding roller and the grinding ring. The existence of the dynamic material layer will reduce the vibration of the equipment and have a beneficial effect on the grinding. The particles in the material layer contact each other and transmit the grinding force through friction, which is continuously reduced in the process to reach the final classification size. In addition, as the wear of the grinding roller and the grinding ring increases, a gradually increasing gap will be formed between the grinding roller and the grinding ring until the wear limit is reached and they are replaced.

[0004] The existing pendulum mills are all designed through forward research and development. In the forward research and development process, the basic dimensions are basically determined first based on the working conditions and environmental requirements of the equipment, and then further designs are carried out based on the basic dimensions, including the actual conditions of each component and its accessories. Therefore, the existing solutions are not designed based on the yield as the goal, and the yield is basically adjusted by adjusting the grinding time according to the actual yield requirements to meet the requirements. Therefore, the currently designed pendulum mills cannot guarantee a high yield in a short time, and the ring roller grinding performance lacks a standardized and detailed optimization method, and a standardized and systematic optimization process has not been formed. Summary of the invention

[0005] The present invention aims to solve the problem that the currently designed swing mill has no standard and systematic method for optimizing the ring roller grinding performance and cannot guarantee a high yield rate in a short time.

[0006] A method for optimizing the performance of a ring roller mill based on a swing mill comprises the following steps:

[0007] For the swing mill to be optimized, Rocky DEM is first used to conduct a single grinding simulation test and a total grinding simulation test; the single yield rate corresponding to each gap is obtained based on the single grinding simulation result analysis, and the total grinding yield rate corresponding to each gap is obtained based on the total grinding simulation result analysis, and the gap is the gap between the grinding roller and the grinding ring; the single yield rate of the single grinding simulation result and the total grinding simulation result are compared to obtain the yield rate corresponding to each gap, and for the purpose of improving efficiency and reducing energy consumption, the optimization direction is determined according to the number of grinding times and the corresponding yield rate, whether to improve the single yield rate or the multiple yield rate, and the shape of the grinding roller is further optimized in combination with the relationship between the wear degree of the grinding roller and the grinding ring and the gap;

[0008] Then, a single factor simulation analysis based on CFD-DEM is carried out, which includes single factor simulation analysis of scraper angle, single factor simulation analysis of feed amount and single factor simulation analysis of spindle speed. According to the results of single factor analysis, the quasi-fixed single factor and its optimized value are determined, and the optimized value of the quasi-fixed single factor is taken as the fixed factor, and the other two factors are variables. The response surface analysis based on CFD-DEM simulation is carried out, and the optimization results of the other two factors are determined with the yield as the target.

[0009] Furthermore, in the process of determining whether the optimization direction is to improve the yield once or the yield multiple times according to the number of grinding times and the corresponding yield rate, improving the yield once is the optimization direction.

[0010] Furthermore, when optimizing the shape of the grinding roller based on the relationship between the wear degree and the clearance of the grinding roller and the grinding ring, the shape of the grinding roller is optimized to a tire-shaped grinding roller, or the shape of the grinding roller is optimized to a truncated cone-shaped grinding roller.

[0011] Furthermore, the grinding roller and the grinding ring are made of high manganese steel.

[0012] Furthermore, the quasi-fixed single factor determined according to the results of the single factor analysis is the blade angle.

[0013] Furthermore, the optimized value of the blade angle, which is assumed to be a fixed single factor, is 35°.

[0014] Furthermore, based on the response surface analysis of CFD-DEM simulation, the optimization results of the other two factors were determined to be feed rate 12t / h and spindle speed 74.5rpm with the yield as the target.

[0015] Beneficial effects:

[0016] The present invention firstly performs a single factor analysis, and according to the characteristics of the single factor analysis, determines a fixed single factor, takes the other two factors as variables, performs analysis based on the response surface of CFD-DEM simulation, and determines the optimization results of the other two factors with the yield as the target. Therefore, the present invention can form an optimization process for improving the yield of the swing mill in the shortest time, which is beneficial to improving the performance specification of the ring roller grinding in the whole machine and the optimization processing quality of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a schematic diagram of the main structure of a certain type of swing mill.

[0018] Figure 2 This is the limit state diagram of grinding roller wear.

[0019] Figure 3 This is a curve of the collision frequency between particles and grinding roller.

[0020] Figure 4 This is the energy spectrum of the accumulated impact power between the material and the grinding ring.

[0021] Figure 5 This is the curve diagram of the total mass change of particles.

[0022] Figure 6 This is a graph of particle size ratio.

[0023] Figure 7 This is the energy spectrum of the accumulated impact power between the material and the grinding ring.

[0024] Figure 8 This is the change diagram of the total mass of particles.

[0025] Fig. 9 This is a graph of particle size ratio.

[0026] Fig.10 This is a cloud diagram of component wear.

[0027] Fig.11 This is the stress curve of the grinding roller.

[0028] Fig.12 This is the wear curve of grinding roller.

[0029] Fig.13 The wear cloud diagram of grinding rollers with different shapes.

[0030] Fig.14 is the particle size ratio of grinding rollers of different shapes.

[0031] Fig.15 This is the wear curve of grinding roller.

[0032] Fig.16 Schematic diagram of the blade angle.

[0033] Fig.17 This is the distribution ratio of each particle size range after 8s of single factor simulation of the blade angle.

[0034] Fig.18 This is the wear curve of grinding roller.

[0035] Fig.19 The simulation diagram of particle trajectory under different blade angles.

[0036] Fig. 20 This is a diagram of the grinding roller wear status under different scraper angles before and after simulation.

[0037] Fig.21 This is the distribution ratio of each particle size range after 8s of single factor simulation of feed amount.

[0038] Fig. 22 This is the energy spectrum of the accumulated impact power between the material and the grinding ring.

[0039] Fig.23 This is the distribution ratio diagram of each particle size range after 8s of single factor simulation of spindle speed.

[0040] Fig.24 This is the impact frequency diagram of particles and grinding ring.

[0041] Fig.25 This is the response surface diagram of grinding rate. DETAILED DESCRIPTION Specific implementation method one:

[0043] This embodiment is a method for optimizing the performance of ring roller grinding based on a pendulum mill, comprising the following steps:

[0044] According to the actual working conditions of the mill, the installed state is the state of zero clearance between the grinding roller and the grinding ring; when the swing mill is working stably, the material thrown by the scraper forms a dynamic material layer with flow characteristics between the grinding roller and the grinding ring, and the thickness of the material layer is 3mm, 5mm, and 8mm; the wear limit of the grinding roller is 25mm ( Figure 2 The gaps between the grinding roller and the grinding ring are set to 0, 3, 5, 8, and 25 mm in the simulation. The material property settings in the simulation test are shown in Table 1.

[0045] Table 1 Simulation model material parameters

[0046]

[0047] DEM software was used for simulation analysis. In the DEM software, the particle crushing model uses the Ab-T10 model, and the specific impact energy (e imp), which is the sum of the work done by all contact forces acting on the particle at a certain moment. The impact energy includes both elastic energy and dissipated energy in the normal direction. Depending on the normal contact model adopted (the normal contact model is suitable for simulating inviscid and dry granular materials), the dissipated energy is plastic or viscous. In addition, the elastic part of the tangentially transmitted energy is also considered in the calculation. In order to break the particle, the specific impact energy should be greater than the minimum crushing energy of the particle, e min , the minimum crushing energy is as follows:

[0048]

[0049] In the formula, e min,ref is the reference minimum specific energy value of the material, and the reference minimum specific energy value of limestone is set to 5J / kg; L ref is the reference particle size, which is set to 25.4 mm by default in Rocky DEM; L is the particle size, which is set to 25 mm.

[0050] Based on RockyDEM, a grinding simulation test is first conducted: the material falls into the main machine from the feed port, and first enters the grinding roller and grinding ring area to be captured for crushing and grinding, which is called primary grinding; then the material that does not meet the fineness standard and is not crushed will fall down into the scraper area and be thrown by the scraper. After another or multiple grindings, it finally reaches the finished product fineness, which is called scraper throwing grinding. The superposition effect of primary grinding and throwing grinding is called total grinding. After primary grinding, some materials reach the finished product fineness, which can effectively reduce the energy consumption of secondary or multiple throwing. In order to study the primary grinding efficiency of the above five gaps, the grinding roller is kept rotating at a fixed speed of 82rev / min in the simulation, and the scraper does not rotate (ignoring the effect of scraper throwing on the grinding efficiency). The feed particles are polygonal ore with a particle size of 25mm; the feed is 3s, the total mass is 10kg, and a total simulation test is carried out for 5s. Considering the simulation cost, the target finished product particle size is set to 5mm.

[0051] Then a total grinding simulation test was carried out: In order to study the total grinding effect and further reveal the superimposed influence of the scraper throwing action on the material grinding, the speed of the scraper and the grinding roller were set to 82rev / min, with the same setting parameters as the primary grinding, and the grinding simulation test was only extended to 10s to ensure that the material was fully ground.

[0052] Grinding roller and grinding ring are grinding working elements, which bear large alternating stress loads. Generally, the service life of grinding roller should be no less than 500h and the service life of grinding ring should be no less than 1000h. Both are easy-to-wear parts and need to be replaced regularly when they reach their service limit.

[0053] Rocky DEM applies Archard wear criterion, and the wear amount is as follows:

[0054] A*dh=B*dw

[0055] Where A is the surface area of ​​the structure, m 2 ; h is the wear depth, m; W is the shear work, J; B is the wear rate, m 3 / J.

[0056] Through the study of wear conditions, the grinding roller and grinding ring are made of wear-resistant high manganese steel, and the volume / shear work ratio is 5e×10 -7 ; The surface mesh of the grinding roller and grinding ring is encrypted and controlled with a mesh size of 5mm to extract accurate wear data.

[0057] Analysis of a milling simulation result:

[0058] Extract information about the frequency of collision between particles and a grinding roller, such as Figure 3 As shown in the figure, during the falling process of the material, the material first collides with the grinding roller and grinding ring area, and the grinding roller rotates continuously at high speed to ensure that the material falling from the feeding port can be captured in time and crushed and ground once. In the figure, as the gap between the grinding roller and the grinding ring decreases, the collision frequency increases significantly with time. The collision frequency of the 0mm gap model is as high as 1.4×10 5 Hz; the gap between the grinding roller and the grinding ring increases, and the collision frequency between the material and the grinding roller decreases successively; the collision frequency of the model with a gap of 25mm is only 6200 times / s at most. Observing the simulation results, since the material size is comparable to the gap size, part of the material falls directly without being captured by the grinding roller and the grinding ring.

[0059] Extract the energy spectrum of the collision between the material and the grinding ring (ignore the 25mm gap in the statistics). Among them, energy is used to characterize the ability of the physical system to do work, and power is a physical quantity that describes the speed of work. The crushing of a single material is mainly determined by the energy level of a single collision and the sum of the collision power. If the energy of a single collision is large, the material can be effectively destroyed, and the amount of material crushed is related to the total power of the energy level; if the energy of a single collision is small, the material cannot be crushed at once, but more small energy level collisions can cause repeated superposition of material damage, which will eventually produce effective crushing. Figure 4 As shown, when the energy is 1.0×10 -4 -4.4×10 -3 The impact power maintains a large value in the J range. The impact power shows a decreasing trend as the gap increases from 0mm to 8mm. The impact power of 3mm and 5mm is 19kW and 12kW respectively. When the gap increases to 8mm, the impact power decreases significantly to 0.65kW. The larger the impact power value, the faster the corresponding crushing and the better the effect. Energy greater than 4.4×10 -3 After the energy increases to 1 J, the corresponding impact power of each gap model begins to decrease; when the energy increases to 1 J, the corresponding cumulative power is basically zero, proving that there are fewer particles with high energy levels.

[0060] The change of the mass of the simulated particles after one grinding is shown in Figure 5 , after 3s of feeding, the particle quality of the five simulation tests showed a downward trend. After 4s, the particles were basically broken and the total quality tended to be stable. By default, Rocky software no longer counts particles smaller than the preset size after grinding. The finished product size was set to 5mm during simulation, so particles smaller than 5mm were not counted. Relative to the total feeding mass of 10kg, the yield rate K was introduced:

[0061]

[0062] Where m1 is the mass of particles with a particle size of 0-5 mm after crushing; M is the total mass of particles.

[0063] According to the calculation of the yield rate K, the first-time yield rate of the 0mm gap model is 77.2%, the first-time yield rate of the 3mm, 5mm and 8mm gap models are 53.1%, 21.4% and 6.1% respectively, and the first-time yield rate of the 25mm gap model is only 1.7%. With the increase of the gap between the grinding roller and the grinding ring, the first-time yield rate is greatly reduced, that is, the grinding efficiency of the equipment is reduced.

[0064] The particle size ratio is used to evaluate the particle size after grinding, and the particle size ratios of [0-5], [5-10], [10-15], [15-20], and [20-25] mm are extracted in turn, such as Figure 6 As shown. Except for the 25mm gap, the particle size proportions of the 0mm, 3mm, 5mm, and 8mm gap models in the interval size [5-15]mm increase successively, and the [15-25]mm interval is basically the same, indicating that more fine particles can be produced when the gap increases, but the change trend of the yield is the opposite. Overall, the increase in the gap between the grinding roller and the grinding ring leads to a significant decrease in the first-time yield, but the fine particles of [5-15]mm increase with the increase in the gap. These particles need to continue to be ground under the throwing of the scraper to finally reach the finished product fineness, so more energy is consumed. It can be concluded that the first-time grinding yield is an effective way for the swing mill to achieve high efficiency and low energy consumption.

[0065] Analysis of total grinding simulation results:

[0066] In the total grinding simulation, the blade rotates synchronously with the main shaft, and the combined effect of primary grinding and blade throwing can be obtained. The energy spectrum of the impact power of the material and the grinding ring is statistically analyzed. Figure 7 The impact power distribution under different energies is shown in Figure 2. Compared with the primary grinding, when the energy is 1.0×10 -4 -4.4×10 -3The J range is the larger impact power range, and the better the crushing effect. The impact power shows a decreasing trend as the gap increases. At this time, the impact power of 3mm is 20kW, which is 1.6kW higher than the first crushing. The impact power of 5mm increases by 0.9kW, and the impact power of 8mm also increases accordingly.

[0067] Particle mass changes Figure 8 Compared with the simulation of primary grinding, the quality of the model particles in each gap decreased more under the throwing of the scraper. The finished product crushing rate was calculated. Compared with the primary grinding, the 0mm, 3mm, 5mm, and 8mm gap models increased by 1.3%, 12.5%, 7.6%, and 0.7%, respectively, and the 25mm model did not improve. The improvement in the yield rate proves that the throwing of the scraper produces effective secondary and multiple grinding effects. The gap between the grinding roller and the grinding ring formed by the dynamic material layer of a certain thickness is conducive to capturing more materials during the throwing of the scraper, resulting in more efficient grinding. The proportion of particle size in the five groups of experiments is shown in Fig. 9 , observing the mass proportion of the fineness range, it is consistent with the law of primary grinding. When the gap increases, more fine particles with a particle size of [5-15] mm can be produced. When there is a gap (3mm, 5mm and 8mm gap model), the proportion of coarse particles with a particle size of [15-25] mm is very small.

[0068] Based on the above analysis, it is determined that the optimization direction is to improve the first-time yield. In fact, it is determined that the optimization direction is to have the highest yield with the least number of grinding times. Then, based on this optimization direction, the wear degree of the grinding roller and grinding ring is analyzed. Combined with the relationship between the wear degree of the grinding roller and grinding ring and the gap, the shape of the grinding roller is optimized.

[0069] In this embodiment, two improved structures of grinding rollers (tire-shaped grinding rollers and round table-shaped grinding rollers) are proposed by studying the wear conditions. The grinding rollers and grinding rings are made of wear-resistant high manganese steel, and the volume / shear work ratio is 5e×10 -7 ; The surface grid of the grinding roller and grinding ring is encrypted and controlled with a grid size of 5mm to extract accurate wear data.

[0070] In order to study the wear of the grinding roller and grinding ring, the wear cloud map of the simulated components is extracted, such as Fig.10 As shown in the figure, (a) is the grinding ring, (b) is the grinding roller, and the grinding roller and grinding ring are prone to wear and tear, which are more concentrated in the middle area. Figure 2 ), the simulation results are consistent with the actual wear of the grinding roller.

[0071] Extract the instantaneous stress of a grinding roller as follows Fig.11As shown. The spindle speed is 105rev / min, which corresponds to a period of 0.57s for one rotation of the grinding roller, which coincides with the time interval of the six peaks in the figure. When the grinding roller reaches the bottom of the feed port from the initial position, stress changes will occur (starting from 0.65s). When the grinding roller passes under the feed port, it can capture the falling material and grind it. The particle grinding and crushing requires a large stress. Since the scraper does not rotate, the grinding stress drops rapidly after the grinding roller leaves the position below the feed port, so there is a phenomenon of periodic stress changes. Since the continuous feeding time is 3s, the first four groups of peaks are averaged, and the corresponding average stresses of the 0mm, 3mm, 5mm, and 8mm gap models are 0.78MPa, 0.65MPa, 0.54MPa, and 0.41MPa, respectively. The smaller the gap, the greater the stress of the grinding roller, the higher the one-time grinding efficiency, and the greater the stress value, the greater the wear of the grinding roller.

[0072] Fig.12 The radial wear variation law of each group of test rollers. The wear of the roller increases in a step-by-step manner. After 3s of feeding, the growth trend of the wear of the roller slows down. After about 5s of particle crushing, the wear of the roller tends to be stable. The wear of the rollers with gaps of 0mm, 3mm, 5mm, and 8mm are 2.36μm, 2.35μm, 2.00μm, and 1.23μm, respectively. The size of the gap is inversely proportional to the wear of the roller. The smaller the gap, the greater the wear. In order to improve the wear condition of the roller, tire-shaped rollers and round table rollers are designed. Fig.13 Figure 2 shows different grinding roller structures and corresponding wear area cloud diagrams. Among them, the cylindrical grinding roller shown in (a) is the original grinding roller form of the mill, (b) is a tire-type grinding roller, and (c) is a frustum-type grinding roller, where the bottom of the grinding roller is a cylinder and the top of the cylinder forms a frustum structure.

[0073] The particle size ratios of three groups of grinding rollers with different shapes are as follows Fig.14 Compared with the cylindrical roller, the first-time yield of the frustum roller increased by 1.8%, and the average particle size after crushing decreased slightly. The first-time yield of the tire roller decreased by 1%, and the grinding effect of the frustum roller was relatively good.

[0074] Extract the wear variation rules of the grinding rollers in the radial direction of each shape as follows: Fig.15 As shown in the figure, the wear of the tire-shaped grinding roller and the frustum-shaped grinding roller are 2.84μm and 1.36μm respectively. Both the tire-shaped grinding roller and the frustum-shaped grinding roller can effectively reduce wear. According to the comprehensive grinding efficiency analysis, the frustum-shaped grinding roller has a better overall effect.

[0075] Then a single factor simulation analysis based on CFD-DEM is carried out:

[0076] (1) Single factor simulation analysis of blade angle:

[0077] In the single-factor simulation with the scraper angle as the variable, the spindle speed is set to 82 (rev / min), the feed rate is set to the actual feed rate of 15t / h in production (the particle mass in the 8s simulation is 33.3kg), and the scraper angle is set to 25°, 30°, 35°, and 40° according to the actual production experience of the enterprise. Fig.16 After the simulation is completed, the material particle parameters and roller wear parameters after crushing are derived from DEM, as shown in Table 2. Fig.16 shown.

[0078] The particle crushing process will significantly increase the calculation pressure during computer simulation. For this reason, the minimum size to which the particles may be crushed must be defined in the simulation configuration (the minimum crushing particle size set in this study is 5 mm) to ensure that all particles smaller than this size will no longer participate in further calculations during the simulation. The total mass of these excluded small particles is counted through DEM software and calculated as the mass of the final finished particles.

[0079] Table 2 Single factor analysis of blade angle

[0080]

[0081] Table 2 shows the single factor results of the scraper angle. When analyzing the effect of the scraper group angle change on the average particle size in the simulation results, it was observed that as the scraper group angle increased, the average particle size showed an increasing trend. The difference between each group was small, and the average particle size of the particles in the main engine was 10.35mm, 10.4mm, 10.52mm, and 10.96mm, respectively. The yield of the particles showed an upward trend and there was a large difference between the scraper group angles of 30° and 35°. The yield increased from 74.5% to 78.2%, with a significant optimization effect. In the range of 35° to 40°, the wear amount increased from 6.33μm to 7.4μm during the increase of the scraper group angle. Fig.17 To simulate the distribution of each particle size range after 8s, the particles in the [5-15] mm range account for the majority, and the 40° blade has the highest proportion in the [15-25] mm large particle size range. Considering the service life of the components and saving energy, although the 40° blade has the highest yield rate, the yield rate is not significantly improved compared to the 35° blade, and the wear of the grinding roller is greatly increased. The 35° blade is used in the single factor analysis of the blade.

[0082] The scraper is an important component used to send the material to the grinding area between the grinding roller and the grinding ring for grinding. The angle of the scraper, that is, the angle between the scraper and the horizontal plane, directly affects the force of the material on the grinding roller, thereby affecting the degree of wear of the grinding roller. Fig.18The wear distribution of the scraper at different angles shows that the wear of the grinding roller increases step by step with the increase of feed amount. When the scraper angle is 40°, the wear of the grinding roller is the largest, which is 7.4μm; when the scraper angle is 25°, the wear of the grinding roller is the smallest, which is 5.95μm. Fig.19 The particle movement trajectory and roller wear are shown in Figure 2 when the scraper angle is 25° and 40°. Fig.19 In (a), the force on the material at the lower part of the grinding roller is more concentrated, which reduces the vertical impact force on the grinding roller and relatively reduces wear. In addition, under the 25-degree blade, the material stays on the grinding roller for a short time, which is not conducive to the full grinding of the material, thereby reducing the grinding efficiency. Under the 40-degree blade, Fig.19 In (b), the force on the grinding roller is more uniform. The increase in the scraper angle means that the cutting force of the material is enhanced, and the impact force on the grinding roller is also increased, which leads to increased wear. The material can be sent to the grinding area in time by the scraper, and the force on the grinding roller is more dispersed. In addition, under the 40° scraper, the material stays on the grinding roller for a longer time, which is conducive to the full grinding of the material, thereby improving the grinding efficiency.

[0083] according to Fig.19 The particle trajectory under different scraper angles can be used to obtain the wear state of the grinding roller under each scraper angle. Fig. 20 As shown in the figure, (a) is the original grinding roller, (b) is the wear of the grinding roller under the 25° scraper, and (c) is the wear of the grinding roller under the 40° scraper. When the particles are ground once and then thrown by the scraper for secondary grinding, compared with the original grinding roller, the wear of the grinding roller under the 25° scraper angle is more uniform, without obvious unevenness; the upper part of the grinding roller under the 40° scraper angle is more seriously worn, because the scraper angle is too high, so the particles thrown for the secondary time are first ground by the upper part of the grinding roller, which leads to this phenomenon.

[0084] (2) Single factor simulation analysis of feed volume:

[0085] In the single-factor simulation with feed rate as the variable, the blade group angle was set to 30°, the spindle speed was set to 82 (rev / min), and the feed rate was set to 15, 20, and 25 t / h (the particle masses in the 8s simulation were 33.3 kg, 44.4 kg, and 55.5 kg, respectively) based on the company's actual production experience, and the 8s simulation was completed.

[0086] Table 3 Single factor analysis of feed amount

[0087]

[0088] The feed rate has a direct impact on the grinding efficiency of the pendulum mill. Under ideal conditions, the grinding efficiency of the pendulum mill is proportional to the feed rate. The larger the feed rate, the higher the grinding efficiency. Because, when the feed rate increases, the pendulum mill has more materials to grind, which can improve the grinding efficiency. However, this does not mean that the feed rate can be increased indefinitely. Because when the feed rate exceeds the carrying capacity of the pendulum mill, the equipment will be overloaded, resulting in a decrease in grinding efficiency. On the contrary, if the feed rate is too small, although high-quality products can be obtained, the production efficiency will be greatly reduced and cannot meet production needs. Therefore, reasonable control of the feed rate is the key to improving the grinding efficiency of the pendulum mill. As shown in the single factor analysis results of the blade angle in Table 3, with the increase of the feed rate, the yield rate shows a gradually decreasing trend, which is 74.5%, 66.4%, and 58.9% respectively, the wear of the grinding roller shows an upward trend, and the average particle size of the particles does not change much, which has little effect on it. Fig.21 To simulate the distribution proportion of each particle size range after 8s, particles in the [5-15] mm range account for the majority, and when the feed rate is 15t / h, the proportion in the [15-25] mm large particle size range is the highest, and the proportion in the [5-10] mm small particle size range is the lowest, at 53.69%.

[0089] In the simulation, the blade rotates synchronously with the main shaft, and the effect of particles in different feed amounts on the grinding ring can be obtained under the action of the blade. The energy spectrum of the impact power of the material and the grinding ring is statistically analyzed. Fig. 22 The impact power distribution under different energies. When the energy is 1.0×10 -4 -4.4×10 -3 The J range is the larger impact power range, and the better the crushing effect. The impact power increases with the increase of feed amount. At this time, the impact power of 20t / h feed amount is 35kW, which is 6kW higher than that of 15t / h feed amount. The maximum impact power of 25t / h feed amount is 40kW.

[0090] (3) Single factor simulation analysis of spindle speed:

[0091] The swing mill is mainly composed of a base, a bracket, a main shaft, a grinding roller, an adjustment device, etc. When the main shaft rotates at high speed, the grinding roller is driven to swing through the gear transmission, so that the material is squeezed and ground between the grinding roller and the grinding ring, thereby achieving the purpose of crushing. Therefore, the main shaft speed has an important influence on the grinding effect of the swing mill.

[0092] The spindle speed is one of the key parameters for the grinding effect of a pendulum mill. Generally speaking, the higher the spindle speed, the faster the grinding roller swings, the longer the grinding time of the material between the grinding roller and the grinding ring, and the better the grinding effect. However, too high a spindle speed will also cause the material to pass between the grinding roller and the grinding ring too quickly, which will not be fully ground and will reduce the grinding effect. Therefore, choosing the right spindle speed is crucial to improving the grinding effect of a pendulum mill.

[0093] The spindle speed also has a great influence on the energy consumption of the swing mill. Generally speaking, the higher the spindle speed, the faster the grinding roller swings, the greater the friction generated during the grinding process, and the higher the energy consumption. However, as the spindle speed increases, the grinding time of the material between the grinding roller and the grinding disc is shortened, and the energy consumption per unit time is relatively reduced. Therefore, under the premise of ensuring the grinding effect, appropriately increasing the spindle speed can reduce the energy consumption of the swing mill.

[0094] The spindle speed also has a certain impact on the service life of the pendulum mill. Too high a spindle speed will increase the wear between the grinding roller and the grinding disc, resulting in a shortened equipment life. In addition, too high a spindle speed will also increase the vibration and noise of the equipment, affecting the stable operation of the equipment. Therefore, when selecting the spindle speed, the wear and service life requirements of the equipment should be fully considered.

[0095] The spindle speed also has a certain influence on the stability of the pendulum mill. Too high a spindle speed will increase the vibration and noise of the equipment, affecting the stable operation of the equipment. In addition, too high a spindle speed may also cause the material to pass between the grinding roller and the grinding disc too quickly, which cannot be fully ground and reduces the grinding effect. Therefore, when selecting the spindle speed, the stability requirements of the equipment should be fully considered.

[0096] In the single-factor simulation with spindle speed as the variable, the blade group angle was set to 30°, the feed rate was set to the actual feed rate of 15t / h in production (the mass of the particles in the 8s simulation was 33.3kg), and the spindle speed was set to 62, 72, 82, 92, and 102 (rev / min) according to the range of the company's actual production experience to complete the 8s simulation.

[0097] Table 4 Single factor analysis of spindle speed

[0098]

[0099] Table 4 shows the single factor result of spindle speed. It can be found that as the spindle speed increases, the average particle size after the simulation is completed shows a downward trend. The average particle sizes of the particles in the main machine are 12.43mm, 11.16mm, 10.40mm, 10.35mm, and 10.26mm, respectively. The yield rate of particles shows a downward trend and there is a large difference between the groups with speeds of 62rev / min to 82rev / min. The yield rate increases from 74.47% to 77.09% and then drops to 74.5%, indicating that there is a maximum yield rate in this range. At the same time, with the increase of spindle speed, the wear of the grinding roller increases successively, from 3.37μm to 6.08μm. Fig.23 To simulate the distribution proportion of each particle size range after 8s, the particles in the [5-15] mm range also account for the majority, and the 62rev / min speed combination has the lowest proportion in this range. In addition, due to the low speed, the particles cannot be captured in time to be ground. Although the roller wear is the lowest, the crushing effect is not ideal, resulting in the 62rev / min speed combination having the highest proportion in the [15-25] mm large particle size range.

[0100] The impact frequency of particles on the grinding ring refers to the number of collisions between particles and the grinding ring per unit time. The impact frequency is closely related to the grinding efficiency, product fineness, energy consumption and other indicators of the swing mill. Generally speaking, the higher the impact frequency, the higher the material grinding efficiency and the finer the product fineness, but the energy consumption also increases accordingly. The impact frequency of particles on the grinding ring when the lowest, original model and highest spindle speeds are 62rev / min, 82rev / min and 102rev / min respectively is extracted. Fig.24 As shown in the figure, with the increase of the spindle speed, the collision frequency of particles and the grinding ring gradually increases. At a speed of 102rev / min, the collision frequency of particles on the grinding ring is as high as 2.1E+07 times / s. If the speed of the scraper is too high, the more times the material is thrown to the grinding ring, the higher the collision frequency. However, too high a speed will cause the material to stay on the grinding disc too short, which is not conducive to the crushing of the material. Therefore, at a speed of 102rev / min, the yield is the lowest.

[0101] (4) According to the characteristics of single factor analysis, the single factor machine optimization value that needs to be fixed is determined, and the other two factors are used as variables. The response surface analysis based on CFD-DEM simulation is performed, and the optimization results of the other two factors are determined with the yield as the target.

[0102] In order to explore the crushing efficiency of limestone powder, the main shaft speed and feed amount are used as optimization parameters, and the crushing efficiency of the swing mill is used as the objective function to conduct a response surface experiment. According to the parameter range determined by enterprise practice, the Box-Behnken response surface optimization method is used for scheme design. The blade installation angle is 35 degrees according to the single factor analysis of the blade angle. The blade angle is set to 35° in the response surface experimental design.

[0103] The experimental analysis was performed using Design-Expert13 software. The experimental factor level table is shown in Table 5.

[0104] Table 5 Experimental factor level table

[0105]

[0106] Crushing efficiency is an important evaluation index in the production of Raymond mill, and its level directly determines the competitiveness of Raymond mill in the ultrafine powder industry.

[0107] After simulation tests, the simulation results of each group are shown in Table 6.

[0108] Table 6 Simulation test and yield value

[0109]

[0110]

[0111] The yield rate obtained by analyzing the three groups of simulation tests is shown in Table 6. When the spindle speed is 62rpm, with the increase of feed amount, the yield rate shows a trend of increasing first and then decreasing, with the highest being 79.6%; when the spindle speed is 72rpm, the yield rate shows a downward trend, with the highest being 83.1%; when the spindle speed is 82rpm, the yield rate also shows a downward trend, with the highest being 82.6%. This shows that when the spindle speed is constant, the feed amount is too large, which has a deteriorating effect on the overall grinding, the mill is in an oversaturated state, and the yield rate decreases. When the feed amount is constant, the spindle speed is too high, which makes the particles stay in the grinding area too short, which also leads to a decrease in the yield rate.

[0112] Table 7 Results of variance analysis of yield rate

[0113]

[0114] ※Indicates that the factors have a significant impact on the experimental index (P<0.05)

[0115] From Table 7, we can see that the main factors affecting the yield rate η are A, B, AB, A 2 , B 2 Among them, A has a significant effect on the yield of materials (P<0.05).

[0116] Design Expert software was used to establish a response surface curve for the test results, such as Fig.25 When the spindle speed is constant, the yield rate increases first and then decreases with the decrease of the feed amount; when the feed amount is constant, the yield rate increases first and then decreases with the decrease of the spindle speed. When the spindle speed is in the range of 72rpm-80rpm, the yield rate is higher; the regression equation is fitted, and the fitting formula of the yield rate is as follows:

[0117] η=82.08-2.95A+0.8333B-2.0AB-2.22A 2 -3.67B 2

[0118] Using the Optimization module in the Design Expert software, with the maximum yield as the optimization goal, the best optimization parameter combination is obtained as shown in Table 8.

[0119] Table 8 Response surface recommendation results

[0120]

[0121] Based on the above optimization results and according to the original working conditions of the enterprise, while ensuring that the output is not reduced, the feed rate is 12t / h, the spindle speed is 74.5rpm, and the yield rate is 83.3% for verification. The optimal yield rate of the material is 82.96%, which is 0.41% different from the optimization result.

[0122] 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 optimizing the performance of a ring roller mill based on a pendulum mill, characterized in that: The following steps are involved: For the swing mill to be optimized, Rocky DEM is first used to carry out a primary grinding simulation test and a total grinding simulation test; Based on the analysis of the simulation results of one grinding, the yield rate of one grinding corresponding to each gap is obtained, and based on the analysis of the total grinding simulation results, the total grinding yield rate of each gap is obtained, and the gap is the gap between the grinding roller and the grinding ring; the yield rate of one grinding simulation result and the total grinding simulation result are compared to obtain the yield rate of each gap, and the optimization direction is to improve the yield rate of one grinding or multiple times according to the grinding times and the corresponding yield rate, and the shape of the grinding roller is further optimized in combination with the relationship between the wear degree of the grinding roller and the grinding ring and the gap; Then, a single factor simulation analysis based on CFD-DEM is carried out, which includes single factor simulation analysis of scraper angle, single factor simulation analysis of feed amount and single factor simulation analysis of spindle speed. According to the results of single factor analysis, the quasi-fixed single factor and its optimized value are determined, and the optimized value of the quasi-fixed single factor is taken as the fixed factor, and the other two factors are variables. The response surface analysis based on CFD-DEM simulation is carried out, and the optimization results of the other two factors are determined with the yield as the target.

2. The method for optimizing the performance of a ring roller mill based on a pendulum mill according to claim 1, characterized in that: In the process of determining whether the optimization direction is to increase the yield once or multiple times based on the number of grinding times and the corresponding yield rate, increasing the yield once is the optimization direction.

3. The method for optimizing the performance of a ring roller mill based on a pendulum mill according to claim 2, characterized in that: In combination with the relationship between the wear degree and the clearance of the grinding roller and the grinding ring, when optimizing the shape of the grinding roller, the shape of the grinding roller is optimized to a tire-shaped grinding roller.

4. The method for optimizing the performance of a ring roller mill based on a pendulum mill according to claim 2, characterized in that: In combination with the relationship between the wear degree and the clearance of the grinding roller and the grinding ring, when optimizing the shape of the grinding roller, the shape of the grinding roller is optimized to a frustum-shaped grinding roller.

5. The method for optimizing the performance of a ring roller mill based on a pendulum mill according to claim 1, characterized in that: Grinding roller and grinding ring are made of high manganese steel.

6. A method for optimizing roller milling performance based on a pendulum mill according to any one of claims 1 to 5, characterized in that: The quasi-fixed single factor determined according to the results of univariate analysis is the blade angle.

7. The method for optimizing the performance of a ring roller mill based on a pendulum mill according to claim 6, characterized in that: The optimal value of the blade angle, which is a quasi-fixed single factor, is 35°.

8. The method for optimizing the performance of a ring roller mill based on a pendulum mill according to claim 7, characterized in that: Based on the response surface analysis of CFD-DEM simulation, the optimization results of the other two factors were determined to be feed rate 12t / h and spindle speed 74.5rpm with the yield as the goal.

Images