Ball mill for producing aluminum silicate

Abstract

The application relates to the technical field of ball mills, and particularly discloses a ball mill for aluminum silicate production, which comprises a rack, an outer cylinder, a driving assembly, a discharge pipe, a feeding pipe and a control interaction unit. First and second bearing seats are fixedly connected to the top of the rack, the outer cylinder is rotationally connected to the rack through the first and second bearing seats, an inner cylinder is fixedly connected in the outer cylinder, a plurality of steel balls are arranged in the inner cylinder, and a counterweight adjusting assembly is arranged in the steel balls. In the application, the inertia and impact force of the steel balls can be changed through the counterweight adjusting assembly, so that the working state of the steel balls can be flexibly adjusted according to the properties and grinding requirements of different materials.

Description

Technical Field

[0001] This invention relates to the field of ball mill technology, and in particular to a ball mill for the production of aluminum silicate. Background Technology

[0002] Alumina silicate, an important inorganic non-metallic material widely used in refractory and insulation materials, relies on raw materials such as bauxite, kaolin, and quartz sand, which have high hardness. The performance of the product is closely related to the particle size distribution of the raw materials. For example, alumina silicate powder used in refractory fibers typically requires an average particle size of 5-10 μm or even finer. Furthermore, industrial production requires a continuous process of "crushing-grinding-mixing-forming-sintering," placing high demands on the capacity and particle size control of grinding equipment. Overflow ball mills, based on the principle of "cylinder rotation + steel ball media grinding," use the rotation of the cylinder to drive the steel balls to pulverize the material in all directions through throwing, grinding, and impact. Its unique discharge structure, without grids and relying on the slurry's own weight for overflow, allows fine powder to remain in the cylinder for a longer time, achieving more thorough grinding and a concentrated particle size distribution. This efficiently meets the fine grinding requirements and strict particle size requirements of alumina silicate, making it suitable for its industrial continuous production process. In existing technologies, the steel balls inside ball mills are solid, making it impossible to adjust their working state according to different media materials. Summary of the Invention

[0003] In view of the technical problem that the steel balls in the ball mill are all solid in the prior art, and it is impossible to adjust the working state of the steel balls according to different media materials, the present invention provides a ball mill for aluminum silicate production.

[0004] The technical solution adopted in this invention is: a ball mill for aluminum silicate production, the ball mill comprising: a frame, an outer cylinder, a drive assembly, a discharge pipe, a feed pipe, and a control interaction unit. A first bearing seat and a second bearing seat are fixedly connected to the top two sides of the frame, respectively. The outer cylinder is rotatably connected to the frame via the first and second bearing seats. An inner cylinder is fixedly connected inside the outer cylinder, and a plurality of steel balls are disposed within the inner cylinder. A counterweight adjustment assembly is disposed within each steel ball. The drive assembly is located on the frame and is used to drive the outer cylinder to rotate. The discharge pipe is fixedly connected to the outside of the first bearing seat, and a discharge port is provided on the side end of the outer cylinder corresponding to the position of the discharge pipe. The feed pipe is fixedly connected to the outside of the second bearing seat, and a feed port is provided on the side end of the outer cylinder corresponding to the position of the feed pipe. The control interaction unit monitors and adjusts the operating status of the ball mill and performs fault diagnosis.

[0005] In a further embodiment of the present invention, the drive assembly includes a first motor, a gearbox, a first gear, and a gear ring. The gear ring is fixedly connected to the outer cylinder. The first gear is rotatably connected to the frame. The gearbox is fixedly connected to the frame. A second gear and a third gear are rotatably connected inside the gearbox. The first motor is fixedly connected to the frame. The output end of the first motor is fixedly connected to the second gear. The third gear is coaxially fixedly connected to a synchronous shaft. The synchronous shaft is coaxially fixedly connected to the first gear. The first gear meshes with the gear ring.

[0006] A further configuration of the present invention includes: a first fixed frame fixedly connected to the outside of the discharge pipe; a second fixed frame fixedly connected to the outside of the feed pipe; a second motor fixedly connected to the outside of the first fixed frame; a third motor fixedly connected to the outside of the second fixed frame; a first spiral conveying rod fixedly connected to the output end of the second motor; a second spiral conveying rod fixedly connected to the output end of the third motor; a feed hopper provided above the feed pipe; the output end of the feed hopper fixedly connected to the feed pipe; inlets and outlets provided in the outer cylinder and inner cylinder; sealing caps provided in the inlets and outlets; the sealing caps fastening into the inlets and outlets; and long bolts passing through the inlets and outlets and the sealing caps.

[0007] A further configuration of the present invention includes a spiral blade on the outside of the steel ball, an inner ball inside the steel ball, a connecting rod fixedly connected to the outside of the inner ball, the connecting rod being fixedly connected to the steel ball, and multiple sets of sliding holes in the inner ball. The counterweight adjustment assembly includes a first threaded rod, a second threaded rod, and a third threaded rod rotatably connected inside the steel ball. A third bevel gear is fixedly connected to the bottom of the first threaded rod, a second bevel gear is fixedly connected to the outside of the second threaded rod, and a first bevel gear is fixedly connected to one end of the third threaded rod. The third bevel gear meshes with the second bevel gear, and the second bevel gear meshes with the first bevel gear. The second threaded rod is a bidirectional threaded rod with opposite thread groove directions. Counterweight balls are threadedly connected to the outside of the first, second, and third threaded rods. Two counterweight balls are provided outside the second threaded rod. Multiple sets of guide rods are also fixedly connected inside the steel ball, and the guide rods penetrate the counterweight balls.

[0008] A further embodiment of the present invention is that a fixing block is fixedly connected to the outside of the first threaded rod, the second threaded rod, and the third threaded rod corresponding to the outside of the steel ball. One of the fixing blocks is provided with a positioning groove. One end of the first threaded rod extends into the positioning groove and is fixedly connected to a rotating block. A positioning plate is provided outside the positioning groove. A positioning block is fixedly connected to the bottom of the positioning plate. A positioning bolt is threadedly connected to the positioning plate. The positioning bolt is threadedly connected to the fixing block. A hexagonal groove is provided at the position of the positioning block corresponding to the outside of the rotating block.

[0009] The present invention is further configured such that the control interaction unit includes a ball mill operation status monitoring module, a ball mill grinding process mathematical model establishment module, an objective function optimization module, a parameter optimization module based on intelligent algorithms, a real-time control and adjustment module, a fault diagnosis and early warning module, a ball mill operation data recording module, a human-machine interface module, and a system integration module.

[0010] A further embodiment of the present invention is that the ball mill operating status monitoring module includes a vibration sensor, a temperature sensor, and a speed sensor;

[0011] A vibration sensor is mounted on either the first or second bearing housing; a temperature sensor is mounted outside the first motor, the first or second bearing housing; and a speed sensor is located outside the frame and is used to measure the rotational speed of the outer cylinder. Let the vibration signal be represented in the time domain as... The temperature signal collected by the temperature sensor is The rotational speed collected by the speed sensor is ,in Representing time; preprocessing the raw signals acquired by the sensor, including filtering, noise reduction, and normalization, and introducing multi-scale data fusion technology; and integrating the vibration signals. and temperature signal Multi-scale decomposition is performed, and wavelet transform is used to obtain signal features at different scales.

[0012] For vibration signals ,go through Low-frequency components are obtained after layer wavelet decomposition. and high frequency components Feature extraction is performed on components at different scales to construct multi-scale feature vectors. : ;in, This represents the mean calculation function. This represents the function for calculating standard deviation.

[0013] Temperature signal The normalization process is performed using the following formula: ;in, This is the normalized temperature value. and These represent the minimum and maximum values ​​of the temperature signal during the monitoring period. A dynamic noise compensation algorithm is introduced to adaptively compensate for noise characteristics under different operating conditions. Based on the spectral analysis of the vibration signal, the noise energy proportion is calculated. ;in, For the frequency domain representation of the vibration signal, This refers to the frequency range where noise is mainly distributed. For the entire spectrum range; develop a spatiotemporal feature fusion network to simultaneously process spatially distributed multi-sensor data and time-series features, and construct a spatiotemporal feature tensor. ;in, Indicates the first One sensor in The feature vectors at each time step are used to extract spatiotemporal features through a spatiotemporal convolutional network;

[0014] .

[0015] A further provision of this invention is that the mathematical modeling module for the ball mill grinding process includes the introduction of a nonlinear dynamics model; assuming that the pulverization process of the material within the ball mill conforms to an improved second-order pulverization kinetic equation: ;in, For the material with a particle size of Particle content, For particle size The first-order grinding rate constant of the particles, The secondary grinding rate constant is used. A deep learning-based grinding efficiency prediction model is established, which uses a convolutional neural network to learn from a large amount of operating data. The input consists of sensor data and operating parameters after feature extraction, and the output is the predicted grinding efficiency value. : Construct a dynamic energy consumption prediction model to determine the actual output power of the first motor. The calculation is as follows:

[0016] ;in, For input power, The efficiency of the first motor. Transmission efficiency; First motor efficiency and transmission efficiency Calculate using the following formulas respectively:

[0017] ;

[0018] ;

[0019] in, The rated power of the first motor, , , This is an empirical coefficient. The optimal speed point;

[0020] Combining quantum computing with neural networks, the output of the quantum layer is calculated as follows: ;in, For quantum weight matrix, For parameterized quantum circuits, For the initial quantum state, For bias, The activation function combines the output of the quantum layer with that of the classical layer: ;in, This is a classic weight matrix. For bias, To construct the output function, a multi-scale grinding kinetic model is built, considering the differences in grinding behavior of materials with different particle sizes. The materials are divided into categories according to particle size. There are several intervals, and the crushing kinetic equation for each interval is: ;in, For the first Material content in each particle size range and Let be the crushing rate constant for this interval. For from the first From the interval to the first The conversion rate constants for each interval; where the objective function optimization module optimizes the original grinding efficiency. Maximize and consume energy Minimize the basis and introduce equipment wear. As the optimization objective, a three-objective optimization function is established: ;in, , and These are the weighting coefficients, and To maximize grinding efficiency, To minimize energy consumption, Maximum permissible equipment wear;

[0021] Equipment wear Based on the wear of the steel ball and inner cylinder wear thickness calculate:

[0022] ;in, and These are the weighting coefficients. This represents the maximum permissible wear of the steel ball. Maximum allowable wear thickness of the inner cylinder; energy consumption Power of the ball mill and running time The relevant calculation formula is as follows: ;

[0023] ball mill power Calculated using the following empirical formula:

[0024] ;in, Power coefficient; For load capacity; The density of the material; This represents the volume of the material inside the inner cylinder; Let be the angular velocity of the ball mill. ; Let the radius of the inner cylinder be denoted by ; a quality stability index is introduced. This measures the degree of fluctuation in product quality by calculating the standard deviation of the particle size distribution through real-time monitoring of the product's particle size distribution. : ;in, The target standard deviation is used; the quality stability index is added to the optimization objective function: ;in, This is the weighting coefficient for quality stability;

[0025] A dynamic weight adaptive optimization algorithm is proposed, which automatically adjusts the weights of the objective function based on the running state. The weight adjustment formula is as follows:

[0026] ;

[0027] ;

[0028] ;

[0029] ;

[0030] in, , , , For learning rate, , , , Let the partial derivatives of the objective function with respect to each weight be denoted; a carbon emission index is introduced. Establish the green optimization objective function: ;in, To achieve the minimum carbon emission target, This is the carbon emission weighting coefficient, carbon emissions Calculated as: ;in, For the first Energy consumption For the first The carbon emission factors of various energy sources; among them, the parameter optimization module based on intelligent algorithms uses an improved quantum particle swarm optimization algorithm to optimize the operating parameters of the ball mill;

[0031] In the quantum particle swarm optimization algorithm, quantum behavior and chaotic search strategies are introduced, and the position update formula for each particle in quantum space is:

[0032] ;in, The parameters of the quantum rotation gate are dynamically adjusted through chaotic mapping; Step size factor; For the quantum Levy flight path; For the first The optimal position of each individual particle; This is the globally optimal position for the entire particle swarm;

[0033] An adaptive learning rate optimization algorithm is proposed to dynamically adjust the learning parameters and learning rate of quantum particle swarm optimization. With the number of iterations Dynamic changes:

[0034] ;in, The initial learning rate, For the final learning rate, The decay period;

[0035] Introducing a population diversity maintenance mechanism, the average Euclidean distance between particles is calculated. : ;in, For the number of particles, and The particle position vector;

[0036] Develop a multi-agent cooperative optimization framework, leveraging communication and collaboration among agents: ;in, For the first The objective function of each agent. To control its input, As the control input for other agents, each agent uses a deep deterministic policy gradient algorithm for decision-making:

[0037] ;in, The optimal strategy is... For action value function, To address the state, a hybrid optimization method combining quantum genetic algorithm and particle swarm optimization is proposed. The update formula for the quantum chromosome is as follows: ;in, For the first The phase angle of each qubit. This is the phase adjustment amount. For frequency, Using the initial phase, chromosome evolution is achieved through quantum rotation gate operations:

[0038] The real-time control and adjustment module controls the ball mill's feed rate, rotation speed, and inner cylinder loading in real time based on the optimal combination of operating parameters.

[0039] An adaptive fuzzy PID control algorithm is introduced to adjust the rotational speed of the outer cylinder;

[0040] Let the actual rotational speed of the outer cylinder be... The target rotational speed is The variable frequency drive adjusts and controls the quantity. The calculation is as follows: First, based on the speed deviation... and rate of change of deviation Adjusting PID control parameters using fuzzy inference rules , and :

[0041] ;

[0042] ;

[0043] ;

[0044] in, , and These are the initial PID control parameters. , and The adjustment amount is determined based on the parameters obtained from fuzzy inference; then, the regulation control amount is calculated. ; For the control of the feeding speed, the adaptive fuzzy PID control algorithm is also adopted. Based on the real-time feedback of the grinding efficiency, the feeding speed of the second screw conveyor is adjusted to keep the feeding speed at the optimal value.

[0045] A multivariable collaborative control strategy is designed, considering the coupling effect between various control parameters, and a coupling coefficient matrix is ​​established. Calculate the coupling degree between parameters: ;in, For the first One output variable, For the first One control variable; dynamically adjust the control parameters based on the coupling coefficient matrix to improve system response speed and stability; develop a predictive control compensation algorithm to predict the system state at future moments; based on the current state and historical data, use a long short-term memory network to predict the future... Step system output ; ;in, For system output, To control the input, and Based on the historical data length, control parameters are adjusted in advance according to the prediction results.

[0046] A dynamic adaptive control architecture is proposed, which automatically switches control strategies based on the system's operating state, and the system operating state index S is defined as follows: ;

[0047] in, For the i-th state feature, For the corresponding weights, switch between different control strategies based on the state index S:

[0048] ;

[0049] in, For the i-th control strategy, For the corresponding state regions; develop a distributed cooperative control algorithm, assigning control tasks to multiple controllers, each controller responsible for local control, and exchanging information through a communication network: ;in, Let i be the control law for the i-th controller. This is a local error. For communication weights, This is the predicted output of the j-th controller.

[0050] A further provision of the present invention is that the fault diagnosis and early warning module establishes a fault diagnosis model based on transfer learning;

[0051] First, a deep neural network model is pre-trained on a large amount of general mechanical equipment fault data, and then fine-tuned for specific fault data of ball mills. The collected sensor data and extracted feature parameters are used as inputs, and the trained fault diagnosis model is used to determine whether the ball mill has a fault and the type of fault. An improved convolutional long short-term memory network is used as the main structure of the fault diagnosis model, which can simultaneously process the spatial and temporal features of time series data.

[0052] The model output is a failure probability vector. ,in This indicates that the ball mill has reached its first occurrence. The probability of a certain type of failure; when the maximum probability Exceeding the set threshold When necessary, a fault warning signal is issued, along with the fault type and cause.

[0053] Develop a fault evolution prediction model and predict fault development trends based on a Hidden Markov Model (HMM); train the HMM using historical fault data to obtain the state transition probability matrix. and observation probability matrix :

[0054] ;

[0055] ;

[0056] in, for The hidden state at all times for The system uses observations at specific times; calculates state sequences using the Viterbi algorithm to predict fault evolution paths and remaining lifetime; constructs a multi-dimensional fault feature fusion model that integrates time-domain, frequency-domain, and time-frequency-domain features; extracts the essential representations of features from each domain using an autoencoder; and then concatenates the features. ;in, , , These are time-domain, frequency-domain, and time-frequency-domain features, respectively. , , For the corresponding autoencoder;

[0057] Design a quantum entanglement-based fault diagnosis algorithm to improve the discriminative power of fault features by utilizing the principle of quantum entanglement; define quantum feature vectors. : ;in, For the first Each fault characteristic state The corresponding probability amplitude; fault diagnosis results are obtained through quantum measurement: ;in, For the first The eigenstates of various faults; the development of a federated learning fault diagnosis system to achieve knowledge sharing among multiple ball mills; local training of fault diagnosis models for each ball mill and uploading of model parameters: ;in, For the first Model parameters of the billiard mill Its sample size, By aggregating global models through federated learning, the accuracy of fault diagnosis can be improved.

[0058] The invention is further configured such that the ball mill operation data recording module records various operating parameters, sensor data, control commands, and fault information of the ball mill in real time; a reinforcement learning algorithm is introduced to analyze the recorded data, and the optimal control strategy is learned through continuous interaction with the environment; and a state space is defined. This refers to the set of operating parameters and sensor data for the ball mill, and the action space. For an adjustable set of operating parameters, the reward function Designed based on indicators such as grinding efficiency, energy consumption, and equipment wear;

[0059] Develop a knowledge graph construction system to transform operational data into structured knowledge; construct a ball mill knowledge graph through entity recognition, relation extraction, and knowledge fusion. ,in For a collection of entities, For a set of relationships; using knowledge graphs for fault diagnosis and decision support: ;in, For the set of possible decisions, For the observed anomalies, For knowledge graphs Make a decision For abnormal The explanatory probability;

[0060] An online learning mechanism is designed to enable the system to adapt to changes in operating conditions in real time; a concept drift detection algorithm is adopted to trigger model updates when a change in data distribution is detected. ;in, For characteristic function, and These represent the sizes of the old and new data windows, respectively. To determine the drift threshold; a digital twin-driven reinforcement learning framework was developed, utilizing the digital twin model to generate simulation data for policy training; the state transition function of the digital twin model is:

[0061] ;in, For state, For action, For random disturbances; optimize the control strategy using an Actor-Critic architecture: ;in, For the policy function, The action value function;

[0062] A causal inference analysis method is proposed to identify causal relationships between operating parameters; a causal graph is constructed. Calculate causal effects using Do-calculus: ;

[0063] in, As an intervention variable, For the outcome variable, To avoid obfuscating variables, the system integration module introduces digital twin technology during system operation to establish a digital twin model of the ball mill. By synchronizing the ball mill's operating data in real time, the digital twin model is updated. Virtual simulation experiments are then conducted using the digital twin model to test the impact of different control strategies and parameter adjustments on the ball mill's operating performance. Simultaneously, based on new material characteristics and production process requirements, and combined with the simulation results of the digital twin model, the fault diagnosis model is retrained and the operating parameters are updated.

[0064] The system calculates the expected benefits under different maintenance strategies. ;in, For system status, For maintenance actions, In order to take action The system is now in a state. The probability, For state Take action below The benefits;

[0065] Developing a multi-agent cooperative optimization framework: ;in, For the first The objective function of each agent. To control its input, Provide control input for other intelligent agents; design a quantum-safe communication protocol to ensure the data transmission security of the ball mill control system; and use quantum key distribution technology to generate encryption keys. ;in, and For both parties in the communication, For shared keys; utilizing the properties of quantum entanglement to detect potential eavesdropping: ;in, For conditional probability, It is an unconditional probability. To measure the number of times, Set a threshold; develop a blockchain-driven device management system to achieve full lifecycle management of devices; upload device information and operation records to the blockchain: ;in, It contains information such as the hash value and timestamp of the previous block. Includes equipment status, maintenance records, etc.; utilizes smart contracts to automatically execute maintenance plans: ;in, To uphold the contract, As a triggering condition, and To perform the action.

[0066] The beneficial effects of this invention are:

[0067] Compared with existing technologies, this invention, through the setting of the counterweight adjustment component, can change the inertia and impact force of the steel balls, enabling the ball mill to flexibly adjust the working state of the steel balls according to the properties of different materials and grinding requirements. For materials with high hardness, increasing the rotational inertia and impact force of the steel balls can more effectively crush the materials; while for materials with low hardness, it can make the steel balls more stable during the grinding process, improving the uniformity and fineness of grinding. Compared with traditional fixed-structure steel balls, this adjustable design greatly improves the versatility and adaptability of the ball mill, and can meet diverse production needs.

[0068] Second, compared with the prior art, the present invention significantly improves the control precision of the ball mill by setting up the control interaction unit, making the control of operating parameters more precise and the grinding process more stable and efficient; effectively reduces energy consumption and balances equipment wear, extending the service life of the equipment; greatly enhances the fault diagnosis capability and provides early warning of potential faults; realizes in-depth intelligent application, and provides strong support for the application of ball mills in complex scenarios. Attached Figure Description

[0069] Figure 1 This is a schematic diagram of the structure of the present invention;

[0070] Figure 2 This is a schematic diagram of the structure of the frame and outer cylinder in this invention;

[0071] Figure 3 This is a schematic diagram of the internal structure of the gearbox in this invention;

[0072] Figure 4 This is a schematic diagram of the outer cylinder in this invention;

[0073] Figure 5 This is a schematic diagram of the inner cylinder structure in this invention;

[0074] Figure 6 This is a schematic diagram of the steel ball structure in this invention;

[0075] Figure 7 This is a schematic diagram of the main structure of the steel ball in this invention;

[0076] Figure 8 This is a schematic diagram of the cross-sectional structure of the steel ball in this invention;

[0077] Figure 9 This is a schematic diagram of the rotating block in the steel ball of the present invention;

[0078] Figure 10 This is a schematic diagram of the positioning block in this invention.

[0079] The diagram is marked as follows:

[0080] 1. Frame; 2. Outer cylinder; 3. First bearing housing; 4. First gear; 5. Gear ring; 6. Sealing cover; 7. Long bolt; 8. Inlet / outlet; 9. First motor; 10. Gearbox; 11. Second gear; 12. Third gear; 13. Synchronous shaft; 14. Discharge pipe; 15. Discharge port; 16. First fixed frame; 17. Second motor; 18. First screw conveyor; 19. Second fixed frame; 20. Third motor; 21. Feed pipe; 22. Second screw conveyor; 2 3. Feed hopper; 24. Second bearing seat; 25. Inner cylinder; 26. Steel ball; 27. Spiral blade; 28. Fixing block; 29. ​​Positioning plate; 30. Positioning bolt; 31. Inner ball; 32. Connecting rod; 33. Sliding hole; 34. First threaded rod; 35. Positioning groove; 36. Rotating block; 37. Second threaded rod; 38. Third threaded rod; 39. Guide rod; 40. Counterweight ball; 41. First bevel gear; 42. Second bevel gear; 43. Third bevel gear; 44. Positioning block. Detailed Implementation

[0081] In the description of this invention, it should be noted that the terms "front", "up", "down", "left", "right", "vertical", "horizontal", etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0082] The following is in conjunction with the appendix Figure 1-10 The present invention will be further described below.

[0083] Example 1:

[0084] To address the problems existing in the background technology, this application proposes the following technical solution: a ball mill for aluminum silicate production, comprising: a frame 1, an outer cylinder 2, a drive assembly, a discharge pipe 14, a feed pipe 21, and a control interaction unit. A first bearing seat 3 and a second bearing seat 24 are fixedly connected to the top two sides of the frame 1, respectively. The outer cylinder 2 is rotatably connected to the frame 1 via the first bearing seat 3 and the second bearing seat 24. An inner cylinder 25 is fixedly connected inside the outer cylinder 2, and several steel balls 26 are disposed inside the inner cylinder 25. The drive assembly is located on the frame 1 and is used to drive the outer cylinder 2 to rotate. The discharge pipe 14 is fixedly connected to the outside of the first bearing seat 3. A discharge port 15 is provided at the side end of the outer cylinder 2 corresponding to the position of the discharge pipe 14. The feed pipe 21 is fixedly connected to the outside of the second bearing seat 24. A feed port is provided at the side end of the outer cylinder 2 corresponding to the position of the feed pipe 21. The control interaction unit monitors and adjusts the operating status of the ball mill and performs fault diagnosis.

[0085] In this embodiment, the drive assembly includes a first motor 9, a gearbox 10, a first gear 4, and a gear ring 5. The gear ring 5 is fixedly connected to the outer cylinder 2. The first gear 4 is rotatably connected to the frame 1. The gearbox 10 is fixedly connected to the frame 1. A second gear 11 and a third gear 12 are rotatably connected inside the gearbox 10. The first motor 9 is fixedly connected to the frame 1. The output end of the first motor 9 is fixedly connected to the second gear 11. The third gear is coaxially fixedly connected to a synchronous shaft 13, which is coaxially fixedly connected to the first gear 4. The first gear 4 meshes with the gear ring 5. The first motor 9, as the power source of the drive assembly, has the characteristics of high power and high efficiency, and can stably output strong power, providing a solid power foundation for the operation of the ball mill. It is fixedly connected to the frame 1 to ensure stability during operation and reduce power loss and equipment damage caused by vibration. The gearbox 10 optimizes and transmits the power output of the motor. The second gear 11 and the third gear 12 inside the gearbox 10 achieve multi-stage power transmission and speed change through precise meshing design. The second gear 11 is directly connected to the output end of the first motor 9, receiving the rotational power of the motor and transmitting it to the third gear 12. The third gear 12 then transmits the power to the first gear 4 through the synchronous shaft 13, which is coaxially fixedly connected.

[0086] The first gear 4 is rotatably connected to the frame 1 and meshes with the gear ring 5 fixed on the outer cylinder 2. This gear transmission method has the advantages of accurate transmission ratio, high power transmission, and high efficiency. During power transmission, the rotation of the first gear 4 can precisely drive the gear ring 5 to rotate, thereby causing the outer cylinder 2 and the inner cylinder 25 to rotate synchronously. Through the combination and design of different gears in the gearbox 10, the speed and torque of the first gear 4 can be adjusted to meet the speed and power requirements of the ball mill at different grinding stages. For example, in the initial stage of grinding, a larger torque is required to crush the material. At this time, the torque transmitted from the first gear 4 to the gear ring 5 can be increased by adjusting the transmission ratio of the gearbox 10. In the later stage of grinding, in order to make the material grind more finely, the speed of the outer cylinder 2 can be appropriately increased.

[0087] Furthermore, the enclosed design of gearbox 10 effectively prevents dust and impurities from entering, protecting the normal operation of the internal gears and extending their service life. The entire drive assembly is compact and rationally designed with a scientific layout. While ensuring efficient power transmission, it also reduces the operating noise of the equipment, creating a relatively quiet working environment for operators. Its stable operating performance also helps improve the grinding quality and production efficiency of the ball mill, ensuring the continuity and stability of the aluminum silicate production process.

[0088] In this embodiment, a first fixing frame 16 is fixedly connected to the outside of the discharge pipe 14, a second fixing frame 19 is fixedly connected to the outside of the feed pipe 21, a second motor 17 is fixedly connected to the outside of the first fixing frame 16, a third motor 20 is fixedly connected to the outside of the second fixing frame 19, a first spiral conveying rod 18 is fixedly connected to the output end of the second motor 17, a second spiral conveying rod 22 is fixedly connected to the output end of the third motor 20, a feed hopper 23 is provided above the feed pipe 21, the output end of the feed hopper 23 is fixedly connected to the feed pipe 21, an inlet and outlet 8 are provided in the outer cylinder 2 and the inner cylinder 25, a sealing cover 6 is provided in the inlet and outlet 8, the sealing cover 6 is fastened in the inlet and outlet 8, and a long bolt 7 is passed through the inlet and outlet 8 and the sealing cover 6; the first fixing frame 16 and the second fixing frame 19 are respectively fixed to the outside of the discharge pipe 14 and the feed pipe 21, providing a stable installation foundation for the second motor 17, the third motor 20 and the spiral conveying rod. They are made of high-strength materials, capable of withstanding the vibrations generated during motor operation and the forces exerted by the screw conveyor when transporting materials, ensuring the stability of the entire conveying system during operation without swaying or displacement. The second motor 17 and the third motor 20, as the power units for material conveying, have excellent speed regulation performance and stable torque output.

[0089] The second spiral conveyor 22, driven by the third motor 20, is connected to the feed pipe 21 and is responsible for conveying the material in the feed hopper 23 to the inner cylinder 25. The design of the feed hopper 23 increases the material storage space, facilitates centralized feeding by operators, reduces the frequency of feeding, and improves work efficiency. Driven by the motor, the second spiral conveyor 22, through the rotation of the spiral blades 27, pushes the material evenly and continuously into the inner cylinder 25, ensuring a timely and stable supply of material to meet the grinding requirements of the ball mill.

[0090] The first spiral conveyor 18, driven by the second motor 17, works in conjunction with the discharge pipe 14 to perform screening and reflow functions. When the ground slurry overflows from the discharge port 15 into the discharge pipe 14, it may contain some large, incompletely ground materials or steel balls 26. Driven by the second motor 17, the first spiral conveyor 18 rotates in the opposite direction, pushing these materials and steel balls 26 back from the discharge pipe 14 into the inner cylinder 25 for further grinding and processing. This design avoids material waste, improves grinding efficiency and finished product quality, and also ensures that the steel balls 26 are not lost during discharge, thus reducing production costs.

[0091] The inlet / outlet 8, along with the matching sealing cap 6 and long bolt 7, provided by the outer cylinder 2 and inner cylinder 25, facilitate the maintenance of the equipment and the addition and replacement of steel balls 26. When the equipment needs maintenance or the addition of steel balls 26, the operator only needs to unscrew the long bolt 7 and remove the sealing cap 6 to perform the corresponding operations through the inlet / outlet 8. The design of the long bolt 7 ensures a tight connection between the sealing cap 6 and the inlet / outlet 8, effectively preventing material leakage during ball mill operation and ensuring the equipment's sealing and safety. This design not only meets the needs of daily maintenance and operation but also ensures the stability and reliability of the ball mill during normal operation, making the entire production process more efficient and convenient.

[0092] In this embodiment, the steel ball 26 has a spiral blade 27 on its exterior and an inner ball 31 inside. A connecting rod 32 is fixedly connected to the exterior of the inner ball 31, and the connecting rod 32 is fixedly connected to the steel ball 26. The inner ball 31 has multiple sets of sliding holes 33. The counterweight adjustment assembly includes a first threaded rod 34, a second threaded rod 37, and a third threaded rod 38 rotatably connected inside the steel ball 26. A third bevel gear 43 is fixedly connected to the bottom of the first threaded rod 34, a second bevel gear 42 is fixedly connected to the exterior of the second threaded rod 37, and a first bevel gear 42 is fixedly connected to one end of the third threaded rod 38. Gear 41, third bevel gear 43 meshes with second bevel gear 42, and second bevel gear 42 meshes with first bevel gear 41. Second threaded rod 37 is a bidirectional threaded rod with opposite thread groove directions. Counterweight balls 40 are threadedly connected to the exterior of first threaded rod 34, second threaded rod 37, and third threaded rod 38. Two counterweight balls 40 are provided on the exterior of second threaded rod 37. Multiple sets of guide rods 39 are fixedly connected inside steel ball 26, passing through the counterweight balls 40. Spiral blades 27 on the exterior of steel ball 26 play a unique role during ball mill operation. When outer cylinder 2 and inner cylinder 25 rotate, steel ball 26 moves accordingly. Spiral blades 27 can change the contact method and trajectory between steel ball 26 and material. They can guide the material to flow between steel balls 26, making the material more evenly distributed within the grinding area of ​​steel ball 26, avoiding localized material accumulation or insufficient grinding. Meanwhile, the spiral blade 27, during the friction and stirring process with the material, can also assist the steel ball 26 in the initial crushing and mixing of the material, thereby improving the grinding efficiency of the ball mill.

[0093] In this embodiment, the inner ball 31 is fixedly connected to the steel ball 26 via the connecting rod 32, providing stable support for the internal adjustment mechanism. Multiple sets of sliding holes 33 in the inner ball 31 provide a track for the movement of the counterweight ball 40, ensuring that the counterweight ball 40 can slide smoothly inside the steel ball 26. The first threaded rod 34, the second threaded rod 37, and the third threaded rod 38 mesh with each other via bevel gears, forming a linked adjustment system. When one of the threaded rods is rotated, the other two threaded rods rotate synchronously under the transmission of the bevel gears, achieving simultaneous adjustment of the positions of multiple counterweight balls 40.

[0094] In this embodiment, the second threaded rod 37 is a bidirectional threaded rod, and its two externally mounted counterweight balls 40 can move to the sides or center respectively, cooperating with the counterweight balls 40 on the first threaded rod 34 and the third threaded rod 38, enabling more flexible adjustment of the center of gravity and mass distribution of the steel ball 26. The counterweight balls 40 are made of high-density metal material and have a large mass; by changing their positions within the steel ball 26, the inertia and impact force of the steel ball 26 can be significantly altered. The guide rod 39 provides guidance and limits for the movement of the counterweight balls 40, preventing them from shifting or jamming during movement, ensuring a smooth and stable adjustment process.

[0095] The adjustable counterweight steel ball 26 design allows the ball mill to flexibly adjust the working state of the steel balls 26 according to the properties of different materials and grinding requirements. For materials with higher hardness, moving the counterweight ball 40 to the outside of the steel ball 26 increases the rotational inertia and impact force of the steel ball 26, enabling more effective material crushing. For materials with lower hardness, moving the counterweight ball 40 towards the center of the steel ball 26 makes the steel ball 26 more stable during grinding, improving the uniformity and fineness of the grinding. Compared to the traditional fixed structure of the steel ball 26, this adjustable design greatly improves the versatility and adaptability of the ball mill, meeting diverse production needs, reducing the time and cost associated with changing different types of steel balls 26, and improving production efficiency.

[0096] In this embodiment, fixing blocks 28 are fixedly connected to the outside of the first threaded rod 34, the second threaded rod 37, and the third threaded rod 38 corresponding to the outside of the steel ball 26. One fixing block 28 is provided with a positioning groove 35. One end of the first threaded rod 34 extends into the positioning groove 35 and is fixedly connected to a rotating block 36. A positioning plate 29 is provided outside the positioning groove 35. A positioning block 44 is fixedly connected to the bottom of the positioning plate 29. A positioning bolt 30 is threadedly connected to the positioning plate 29 and is threadedly connected to the fixing block 28. A hexagonal groove is provided at the position of the positioning block 44 corresponding to the outside of the rotating block 36. The fixing blocks 28, positioning grooves 35, positioning plates 29, positioning bolts 30 and other components provided outside the steel ball 26 together constitute a reliable positioning and locking system to ensure that the adjusted position of the counterweight ball 40 will not change during the operation of the ball mill, thus ensuring the stable working state of the steel ball 26. The fixing blocks 28 are fixed to the outside of the first threaded rod 34, the second threaded rod 37 and the third threaded rod 38 respectively, providing additional support for the threaded rods, enhancing the stability of the threaded rods during rotation, and preventing the adjustment accuracy and reliability of the counterweight ball 40 from being affected by the shaking of the threaded rods.

[0097] One of the fixing blocks 28 has a positioning groove 35 that engages with a rotating block 36 at the end of the first threaded rod 34. The rotating block 36 can rotate freely within the positioning groove 35, allowing the operator to rotate the first threaded rod 34 by inserting a tool into the hexagonal groove on the outside of the rotating block 36, thereby adjusting the position of the counterweight ball 40. Once the counterweight ball 40 is in the appropriate position, the positioning plate 29 comes into play. The positioning block 44 at the bottom of the positioning plate 29 matches the shape of the hexagonal groove on the outside of the rotating block 36. The positioning plate 29 is installed on the fixing block 28, so that the positioning block 44 is embedded in the hexagonal groove. Then, by tightening the positioning bolt 30, the positioning plate 29 is firmly fixed to the fixing block 28. In this way, the positioning plate 29 and the positioning block 44 limit the rotation of the rotating block 36, preventing the first threaded rod 34 from continuing to rotate, thereby locking the position of the counterweight ball 40.

[0098] This positioning and locking design offers multiple advantages. Firstly, it ensures that even under strong vibration and impact during high-speed operation of the ball mill, the counterweight ball 40 will not shift due to the rotation of the threaded rod, ensuring that the steel ball 26 maintains its set center of gravity and mass distribution, thus maintaining a stable grinding effect. Secondly, it is simple and quick to operate. When the operator needs to adjust the position of the counterweight ball 40, they only need to loosen the positioning bolt 30, remove the positioning plate 29, and easily rotate the threaded rod for adjustment. After adjustment, reinstalling the positioning plate 29 and tightening the positioning bolt 30 quickly completes the positioning and locking process. Furthermore, this design is compact, occupies little space, and does not affect the normal movement of the steel ball 26 within the ball mill. This reliable positioning and locking system, combined with the adjustable counterweight structure inside the steel ball 26, ensures both the flexibility of steel ball 26 adjustment and the stability of its working state, providing a solid guarantee for the ball mill to efficiently and accurately grind materials of different properties. This further enhances the performance and reliability of the ball mill, meeting the demand for high-quality grinding equipment in the production of aluminum silicate.

[0099] The usage method of this embodiment is as follows: Material is added to the feed hopper 23. The third motor 20 drives the second threaded rod 37 to rotate, thereby conveying the material into the inner cylinder 25. Simultaneously, water or grinding fluid is added for synchronous grinding. After conveying, steel balls 26 are then conveyed into the inner cylinder 25 through the inlet / outlet 8. The inlet / outlet 8 is then sealed with the sealing cap 6, and the sealing cap 6 is fixed with long bolts 7. The first motor 9 is started, driving the second gear 11 to rotate. The second gear 11 drives the third gear 12 to rotate, and the third gear 12 drives... The rotation of the first gear 4 drives the gear ring 5 to rotate, thereby rotating the outer cylinder 2 and the inner cylinder 25. The material is ground by the impact of the moving steel balls 26. The ground slurry overflows into the discharge pipe 14 and is discharged. The second motor 17 drives the first spiral conveyor rod 18 to rotate in the opposite direction, pushing the overflowing steel balls 26 and large pieces of material back into the inner cylinder 25 for further processing. It should be noted that the counterweight ball 40 is made of a high-density metal material, such as lead or tungsten alloy, for the steel balls 26. Rotating the first threaded rod 34 with a tool changes the position of the counterweight ball 40 within the steel balls 26, thus adjusting the center of gravity and mass distribution of the steel balls 26. When grinding materials with higher hardness, the counterweight ball 40 can be moved to the outside of the steel balls 26 to increase the impact force during collision; while for materials with lower hardness, the counterweight ball 40 can be moved towards the center of the steel balls 26 to improve the grinding uniformity. This adjustable counterweight design allows the steel balls 26 to adapt to the grinding needs of materials with different properties, greatly improving the versatility and working efficiency of the ball mill.

[0100] Example 2:

[0101] The control interaction unit includes a ball mill operation status monitoring module, a ball mill grinding process mathematical model establishment module, an objective function optimization module, a parameter optimization module based on intelligent algorithms, a real-time control and adjustment module, a fault diagnosis and early warning module, a ball mill operation data recording module, a human-machine interface module, and a system integration module. The ball mill operation status monitoring module includes a vibration sensor, a temperature sensor, and a speed sensor. The speed sensor is a non-contact speed sensor, an existing product, used to measure the rotational speed of the outer cylinder 2. The vibration sensor is installed on the first bearing seat 3 or the second bearing seat 24. The temperature sensor is installed outside the first motor 9, the first bearing seat 3, or the second bearing seat 24. The speed sensor is located outside the frame 1 and is used to measure the rotational speed of the outer cylinder 2. Let the vibration signal be represented in the time domain as... The temperature signal collected by the temperature sensor is The rotational speed collected by the speed sensor is ,in Indicates time;

[0102] In this embodiment, a comprehensive operational status monitoring system is constructed by installing various sensors, such as vibration, temperature, and rotational speed sensors, on key parts of the ball mill. Compared with traditional monitoring methods, this system can collect key data during the ball mill's operation in real time and accurately. The collected data covers multiple dimensions of equipment operation, providing rich and accurate basic information for subsequent analysis and control. Based on this, a dynamic noise compensation algorithm and a spatiotemporal feature fusion network are introduced to further improve data quality and feature extraction capabilities. The dynamic noise compensation algorithm can adaptively suppress noise for different operating conditions, ensuring that the collected signals truly reflect the equipment's operating status. The spatiotemporal feature fusion network can simultaneously process the spatial distribution and time series features of multi-sensor data, effectively mining the potential correlations between data, thereby gaining a more comprehensive and in-depth understanding of the ball mill's operating status and laying a solid foundation for subsequent precise control and fault diagnosis.

[0103] The raw signals acquired by the sensors are preprocessed, including filtering, noise reduction, and normalization, and multi-scale data fusion technology is introduced; the vibration signals are then processed. and temperature signal Multi-scale decomposition is performed, and wavelet transform is used to obtain signal features at different scales; for vibration signals ,go through Low-frequency components are obtained after layer wavelet decomposition. and high frequency components Feature extraction is performed on components at different scales to construct multi-scale feature vectors. ; ;in, This represents the mean calculation function. The standard deviation calculation function is represented by this multi-scale feature extraction, which can capture more comprehensive details of the ball mill's operating status.

[0104] Temperature signal The normalization process is performed using the following formula: ;in, This is the normalized temperature value. and These represent the minimum and maximum values ​​of the temperature signal during the monitoring period, respectively.

[0105] Normalization helps to unify the scale of data from different sensors, facilitating subsequent analysis and processing. A dynamic noise compensation algorithm is introduced to adaptively compensate for noise characteristics under different operating conditions. Based on the spectral analysis of the vibration signal, the noise energy percentage is calculated. ; ;in, For the frequency domain representation of the vibration signal, This refers to the frequency range where noise is mainly distributed. It covers the entire frequency spectrum; the filtering parameters are dynamically adjusted based on the proportion of noise energy to improve signal quality.

[0106] Develop a spatiotemporal feature fusion network to simultaneously process spatially distributed multi-sensor data and time-series features, and construct a spatiotemporal feature tensor. ;in, Indicates the first One sensor in The feature vectors at each time step are used to extract spatiotemporal features through a spatiotemporal convolutional network; ;

[0107] By combining spatial and temporal convolution, it can capture the spatial correlation and temporal evolution characteristics between sensors;

[0108] The mathematical modeling module for the ball mill grinding process includes the introduction of a nonlinear dynamics model.

[0109] Assume that the material pulverization process inside the ball mill conforms to the improved two-stage pulverization kinetic equation: ;

[0110] in, For the material with a particle size of Particle content, For particle size The first-order grinding rate constant of the particles, These are the secondary grinding rate constants, and these two constants are related to the rotational speed of the ball mill. Load capacity The results are related to factors such as material properties and can be obtained by fitting experimental data with a nonlinear regression algorithm.

[0111] The above technical solution employs conventional processing methods such as filtering, noise reduction, and normalization, while also introducing multi-scale data fusion technology to decompose and extract features from vibration and temperature signals at multiple scales. This allows for a more comprehensive capture of detailed information about the ball mill's operating status. A dynamic noise compensation algorithm dynamically adjusts filtering parameters based on vibration signal spectrum analysis, effectively improving signal quality. The application of a spatiotemporal feature fusion network combines the spatial and temporal features of multi-sensor data, extracting key features through a spatiotemporal convolutional network. Compared to single data processing methods, this approach more accurately extracts effective features reflecting the ball mill's operating status, providing high-quality data support for subsequent model building and parameter optimization, and significantly improving the accuracy and effectiveness of data processing.

[0112] Grinding efficiency of ball mill With feed rate Rotation speed Based on parameters such as [missing information], a deep learning-based grinding efficiency prediction model is established. A convolutional neural network is used to learn from a large amount of operational data. The input consists of sensor data after feature extraction and operational parameters, and the output is the predicted grinding efficiency value. : ;

[0113] A dynamic energy consumption prediction model is constructed, comprehensively considering factors such as the efficiency of the first motor, transmission losses, and material characteristics, to determine the actual output power of the first motor. The calculation is as follows: ;in, For input power, The efficiency of the first motor. Transmission efficiency; First motor efficiency and transmission efficiency Calculate using the following formulas respectively:

[0114] ;

[0115] ;

[0116] in, The rated power of the first motor, , , This is an empirical coefficient. The optimal speed point;

[0117] Combining quantum computing with neural networks improves model prediction accuracy. The output of the quantum layer is calculated as follows: ;in, For quantum weight matrix, For parameterized quantum circuits, For the initial quantum state, For bias, The activation function combines the output of the quantum layer with that of the classical layer: ;

[0118] in, This is a classic weight matrix. For bias, To construct the output function, a multi-scale grinding kinetic model is built, considering the differences in grinding behavior of materials with different particle sizes. The materials are divided into categories according to particle size. There are several intervals, and the crushing kinetic equation for each interval is:

[0119] ;in, For the first Material content in each particle size range and Let be the crushing rate constant for this interval. For from the first From the interval to the first The conversion rate constant for each interval.

[0120] The above technical solutions fully consider the nonlinear characteristics of material crushing and adopt an improved two-stage crushing kinetic equation, which is closer to the actual grinding process than the traditional model. The grinding efficiency prediction model based on deep learning uses a convolutional neural network to learn from a large amount of operating data, which can accurately predict grinding efficiency. The innovative quantum hybrid neural network model integrates quantum computing and classical neural networks, which greatly improves the model's prediction accuracy. The multi-scale grinding kinetic model considers the differences in crushing behavior of materials with different particle sizes, making the model's description of the grinding process more detailed and accurate. The establishment of these models provides a strong theoretical basis and technical support for a deeper understanding of the ball mill grinding process and optimization of operating parameters, effectively improving the controllability of the grinding process and the accuracy of efficiency prediction.

[0121] Among them, the objective function optimization module improves the original grinding efficiency. Maximize and consume energy Minimize the basis and introduce equipment wear. As the optimization objective, a three-objective optimization function is established: ;in, , and These are the weighting coefficients, and Their values ​​are adjusted according to actual production needs to balance the relationship between grinding efficiency, energy consumption and equipment wear; To obtain the maximum grinding efficiency under the experimental conditions, To minimize energy consumption, Maximum permissible equipment wear;

[0122] Equipment wear Based on the wear of the steel ball and inner cylinder wear thickness calculate: ;in, and These are the weighting coefficients. This represents the maximum permissible wear of the steel ball. This represents the maximum permissible wear thickness of the inner cylinder. Energy consumption. Power of the ball mill and running time The relevant calculation formula is as follows: ;

[0123] ball mill power The following empirical formula can be used for calculation: ;in, The power coefficient was determined experimentally. Load capacity (unit: ); Density of the material (unit: ); Volume of material inside the inner cylinder (unit: ); Angular velocity of the ball mill (unit: ), ; The radius of the inner cylinder (unit: );

[0124] Introducing quality stability index This measures the degree of fluctuation in product quality by calculating the standard deviation of the particle size distribution through real-time monitoring of the product's particle size distribution. : ;in, To achieve the target standard deviation, the quality stability index is added to the optimization objective function: ;in, This is the weighting coefficient for quality stability.

[0125] A dynamic weight adaptive optimization algorithm is proposed, which automatically adjusts the weights of the objective function based on the running state. The weight adjustment formula is:

[0126] ;

[0127] ;

[0128] ;

[0129] ;

[0130] in, , , , For learning rate, , , , Let the partial derivatives of the objective function with respect to each weight be denoted; a carbon emission index is introduced. Establish a green optimization objective function: ;

[0131] in, To achieve the minimum carbon emission target, This is the carbon emission weighting coefficient, carbon emissions Calculated as: ;

[0132] in, For the first Energy consumption For the first Carbon emission factors of various energy sources.

[0133] In the above technical solution, the process of determining the objective function not only considers maximizing grinding efficiency and minimizing energy consumption, but also introduces equipment wear and quality stability indicators to establish a multi-objective optimization function. Compared with traditional single-objective optimization, this is more in line with actual production needs. The dynamic weight adaptive optimization algorithm can automatically adjust the objective function weights according to the operating status, making the optimization process more flexible and intelligent. The introduction of carbon emission indicators to construct a green optimization objective function responds to energy conservation and emission reduction requirements, promoting green production of ball mills. These innovations make the optimization of ball mill operating parameters more comprehensive and scientific, enabling a comprehensive balance between grinding efficiency, energy consumption, equipment wear, quality stability, and carbon emissions under different production needs, thereby improving the economic and social benefits of enterprises. Specifically, the parameter optimization module based on intelligent algorithms uses an improved quantum particle swarm optimization algorithm to optimize the operating parameters of the ball mill.

[0134] In the quantum particle swarm optimization algorithm, quantum behavior and chaotic search strategies are introduced to improve the algorithm's global search capability and convergence speed. The formula for updating the position of each particle in quantum space is: ;in, The parameters of the quantum rotation gate are dynamically adjusted through chaotic mapping; Step size factor; This is the quantum Levy flight path, used to enhance the algorithm's local search capability; For the first The optimal position of each individual particle; This represents the globally optimal position for the entire particle swarm. (Based on the feed rate) Rotation speed and loading capacity As optimization variables, under the constraints of ensuring the safe operation and process requirements of the ball mill, the objective function is searched using an improved quantum particle swarm optimization algorithm. To achieve the optimal combination of parameters.

[0135] An adaptive learning rate optimization algorithm is proposed to dynamically adjust the learning parameters and learning rate of quantum particle swarm optimization. With the number of iterations Dynamic changes: ;in, The initial learning rate, For the final learning rate, The decay period is used to balance global and local search capabilities through this adaptive adjustment.

[0136] Introducing a population diversity maintenance mechanism, the average Euclidean distance between particles is calculated. : ;

[0137] in, For the number of particles, and For particle position vectors, when diversity is below a threshold, a population perturbation operation is triggered to prevent the algorithm from getting trapped in local optima.

[0138] A multi-agent collaborative optimization framework was developed to collaboratively optimize the ball mill control system and the entire production line. Through communication and cooperation among the agents, global optimization was achieved. ;in, For the first The objective function of each agent. To control its input, As the control input for other agents, each agent uses the Deep Deterministic Policy Gradient (DDPG) algorithm to make decisions: ;in, The optimal strategy is... For action value function, For the state, a hybrid optimization method combining quantum genetic algorithm (QGA) and particle swarm optimization is proposed. The update formula for the quantum chromosome is: ;

[0139] in, For the first The phase angle of each qubit. This is the phase adjustment amount. For frequency, Using the initial phase, chromosome evolution is achieved through quantum rotation gate operations:

[0140] ;

[0141] The above technical solution employs an improved quantum particle swarm optimization algorithm, combining quantum behavior and chaotic search strategies. Compared to traditional optimization algorithms, it exhibits stronger global search and convergence capabilities. A newly developed multi-agent collaborative optimization framework enables collaborative optimization between the ball mill and the production line, improving overall production efficiency. A hybrid optimization method combining quantum genetic algorithm and particle swarm optimization integrates the advantages of both algorithms, further enhancing parameter optimization. The application of these intelligent algorithms can quickly and accurately find the optimal combination of ball mill operating parameters, not only improving the ball mill's own operating efficiency but also achieving collaborative operation with the production line, optimizing the entire production process, and enhancing enterprise production efficiency. Specifically, the real-time control and adjustment module controls the ball mill's feed rate, rotational speed, and inner cylinder loading in real time based on the obtained optimal operating parameter combination; an adaptive fuzzy PID control algorithm is introduced to adjust the outer cylinder rotational speed.

[0142] Let the actual rotational speed of the outer cylinder be... The target rotational speed is The variable frequency drive adjusts and controls the quantity. The calculation is as follows:

[0143] First, based on the speed deviation and rate of change of deviation Adjusting PID control parameters using fuzzy inference rules , and ; ; ; ;

[0144] in, , and These are the initial PID control parameters. , and This refers to the adjustment amount of the parameters obtained based on fuzzy inference.

[0145] Then, calculate the regulation control quantity. : ;

[0146] For the control of the feeding speed, an adaptive fuzzy PID control algorithm is also adopted. Based on the real-time feedback of grinding efficiency, the feeding speed of the second screw conveyor is adjusted to keep the feeding speed at the optimal value. A multi-variable collaborative control strategy is designed, considering the coupling effect between various control parameters, and a coupling coefficient matrix is ​​established. Calculate the coupling degree between parameters: ;

[0147] in, For the first One output variable, For the first With several control variables, the control parameters are dynamically adjusted based on the coupling coefficient matrix to improve system response speed and stability. A predictive control compensation algorithm is developed to predict the system state at future moments. Based on the current state and historical data, a Long Short-Term Memory (LSTM) network is used to predict the future. Step system output :

[0148] ;in, For system output, To control the input, and Based on the historical data length, control parameters are adjusted in advance according to the prediction results to reduce system lag;

[0149] A dynamic adaptive control architecture is proposed, which automatically switches control strategies based on the system's operating state, and the system operating state index S is defined as follows:

[0150] ;in, For the i-th state feature, For the corresponding weights, switch between different control strategies based on the state index S:

[0151] ;

[0152] in, For the i-th control strategy, For the corresponding state regions, a distributed cooperative control algorithm is developed, which assigns control tasks to multiple controllers. Each controller is responsible for local control and exchanges information through a communication network. ;

[0153] in, Let i be the control law for the i-th controller. This is a local error. For communication weights, This is the predicted output of the j-th controller;

[0154] In the above technical solution, an adaptive fuzzy PID control algorithm is introduced into the real-time control and regulation stage. This algorithm can dynamically adjust control parameters according to speed deviation and rate of change. Compared with traditional PID control, it has higher control accuracy and faster response. The designed multivariable cooperative control strategy considers the coupling effect between parameters to improve system stability. The predictive control compensation algorithm uses LSTM to predict the system state and adjust parameters in advance to reduce system lag. The dynamic adaptive control architecture can automatically switch control strategies according to the system operating state, making control more flexible. The distributed cooperative control algorithm distributes control tasks to multiple controllers to achieve efficient cooperative control. The comprehensive application of these control technologies ensures that the ball mill can maintain stable and efficient operation under various working conditions, significantly improving the dynamic performance and robustness of the system.

[0155] Among them, the fault diagnosis and early warning module establishes a fault diagnosis model based on transfer learning. First, a deep neural network model is pre-trained on a large amount of general mechanical equipment fault data. Then, it is fine-tuned for specific fault data of the ball mill. The sensor data collected in step one and the feature parameters extracted in step two are used as inputs. The trained fault diagnosis model is used to determine whether the ball mill has a fault and the type of fault.

[0156] An improved Convolutional Long Short-Term Memory (ConvLSTM) network is used as the main structure of the fault diagnosis model, which can simultaneously process the spatial and temporal features of time series data. The model output is a fault probability vector. ,in This indicates that the ball mill has reached its first occurrence. The probability of a certain type of failure, when the maximum probability Exceeding the set threshold When necessary, a fault warning signal is issued, along with the fault type and possible causes.

[0157] A fault evolution prediction model was developed, which predicts the fault development trend based on a Hidden Markov Model (HMM). The HMM was trained using historical fault data to obtain the state transition probability matrix. and observation probability matrix : ; ;in, for The hidden state at all times for The observations at each time point are used to calculate the most likely state sequence using the Viterbi algorithm to predict the failure evolution path and remaining lifetime.

[0158] A multi-dimensional fault feature fusion model is constructed, fusing features from the time domain, frequency domain, and time-frequency domain. An autoencoder (AE) is used to extract the essential representations of the features from each domain, followed by feature concatenation.

[0159] ;in, , , These are time-domain, frequency-domain, and time-frequency-domain features, respectively. , , This is the corresponding autoencoder.

[0160] A quantum entanglement-based fault diagnosis algorithm is designed to improve the discriminative power of fault features by utilizing the principle of quantum entanglement. A quantum feature vector is defined. : ;in, For the first Each fault characteristic state For the corresponding probability amplitude, the fault diagnosis result is obtained through quantum measurement: ;in, For the first The eigenstate of a fault.

[0161] Develop a federated learning fault diagnosis system to achieve knowledge sharing among multiple ball mills. Each ball mill trains a fault diagnosis model locally and uploads the model parameters. ;

[0162] in, For the first Model parameters of the billiard mill Its sample size, By aggregating global models through federated learning, the accuracy of fault diagnosis can be improved.

[0163] In the above technical solutions, the fault diagnosis model based on transfer learning is pre-trained using general equipment fault data and fine-tuned for ball mills, which improves the efficiency and accuracy of fault diagnosis. The improved ConvLSTM model effectively processes time series data features, and the quantum entanglement fault diagnosis algorithm uses quantum properties to improve the distinguishability of fault features. The federated learning fault diagnosis system realizes multi-machine knowledge sharing, further improving the diagnostic accuracy. The fault evolution prediction model can predict the fault development trend, and the multi-dimensional fault feature fusion model integrates features from various domains.

[0164] The ball mill operation data recording module records various operating parameters, sensor data, control commands, and fault information of the ball mill in real time. It introduces a reinforcement learning algorithm to analyze the recorded data, learning the optimal control strategy through continuous interaction with the environment and defining the state space. This refers to the set of operating parameters and sensor data for the ball mill, and the action space. For an adjustable set of operating parameters, the reward function Based on indicators such as grinding efficiency, energy consumption, and equipment wear, the control strategy is continuously optimized through reinforcement learning algorithms to ensure efficient and stable operation of the ball mill under different working conditions. A knowledge graph construction system is developed to transform operational data into structured knowledge, and a ball mill knowledge graph is constructed through entity recognition, relation extraction, and knowledge fusion. ,in For a collection of entities, For a set of relationships, knowledge graphs are used for fault diagnosis and decision support: ;

[0165] in, For the set of possible decisions, For the observed anomalies, For knowledge graphs Make a decision For abnormal To explain the probabilities, an online learning mechanism is designed to enable the system to adapt to changes in operating conditions in real time. A concept drift detection algorithm is adopted to trigger a model update when a change in data distribution is detected. ;in, For characteristic function, and These represent the sizes of the old and new data windows, respectively. Let be the drift threshold. A digital twin-driven reinforcement learning framework is developed, utilizing the digital twin model to generate simulation data for policy training. The state transition function of the digital twin model is: ;in, For state, For action, To address random disturbances, the control strategy is optimized using an Actor-Critic architecture: ;

[0166] in, For the policy function, For the action value function, a causal inference analysis method is proposed to identify the causal relationships between operating parameters and to construct a causal graph. Calculate causal effects using Do-calculus: ;in, As an intervention variable, For the outcome variable, To obfuscate variables, the control strategy is optimized based on causal relationships to improve system performance.

[0167] In the above technical solution, the operational data recording system established by the operational data recording and analysis steps records various data of the ball mill in real time, providing rich materials for data analysis. The introduction of reinforcement learning algorithms to learn the optimal control strategy improves the operating efficiency of the ball mill. The knowledge graph construction system transforms data into structured knowledge, supporting fault diagnosis and decision-making. The online learning mechanism enables the system to adapt to changes in operating conditions. The digital twin-driven reinforcement learning framework uses simulation data to train strategies, improving learning efficiency. The causal inference analysis method identifies causal relationships between parameters and optimizes control strategies. These features enable in-depth mining and utilization of ball mill operating data, providing strong support for continuously optimizing ball mill operating performance and improving production management.

[0168] The human-machine interface module allows operators to view the ball mill's operating status, parameters, and fault warnings in real time. Alternatively, virtual reality (VR) technology can be incorporated, enabling operators to immerse themselves in the ball mill's internal structure and operating status via VR devices, facilitating equipment inspection and maintenance. Furthermore, the human-machine interface integrates an intelligent voice assistant, supporting voice command input and operation prompts to enhance ease of use. The interface also features data query, report generation, and printing functions, aiding production management and data analysis. An augmented reality (AR) assisted maintenance system can be developed, displaying equipment maintenance information in real time through AR glasses. The system overlays virtual information onto the actual equipment, guiding operators in maintenance operations. ;in, To augment reality images, Images of real-world scenes. For virtual information, For transparency, Proj is the projection function. For camera parameters, For the coordinate transformation matrix, an emotion computing interaction module is designed. By analyzing the operator's voice tone, facial expressions, etc., the interface display and interaction methods are adjusted. ;in, For emotional states extracted from speech and facial expressions, For historical interactive records, For the emotion response function, Adjust the strategy for the interface.

[0169] Alternatively, a holographic projection interactive interface can be designed to display the ball mill's operating status in three-dimensional space using holographic projection technology. Operators can interact with the holographic interface through gestures and voice. ;in, For three-dimensional data models, The interface supports multi-user collaborative operation for observation point parameters, improving team collaboration efficiency.

[0170] In the above technical solutions, traditional VR and intelligent voice assistant technologies have improved the ease of operation in the human-computer interaction interface design stage.

[0171] The system integration module forms a complete ball mill control and management system. During system operation, digital twin technology is introduced to establish a digital twin model of the ball mill. By synchronizing the ball mill's operating data in real time, the digital twin model is updated and optimized. Virtual simulation experiments are conducted using the digital twin model to test the impact of different control strategies and parameter adjustments on the ball mill's operating performance, providing a reference for optimizing the actual system. At the same time, based on new material characteristics and production process requirements, and combined with the simulation results of the digital twin model, the fault diagnosis model is retrained and operating parameters are optimized; the control algorithm is adjusted and optimized to improve the system's control accuracy and response speed.

[0172] A smart maintenance decision-making system is constructed, which recommends the best maintenance strategy based on a digital twin model and reinforcement learning. The system calculates the expected benefits under different maintenance strategies. : ;in, For system status, For maintenance actions, In order to take action The system is now in a state. The probability, For state Take action below To maximize benefits, a multi-agent collaborative optimization framework was developed to collaboratively optimize the ball mill control system and the entire production line. Through communication and cooperation among agents, global optimization was achieved. ;

[0173] in, For the first The objective function of each agent. To control its input, This serves as the control input for other intelligent agents. A quantum-safe communication protocol is designed to ensure the security of data transmission in the ball mill control system. Quantum key distribution (QKD) technology is used to generate encryption keys. ;

[0174] in, and For both parties in the communication, For shared keys. Utilizing the properties of quantum entanglement to detect potential eavesdropping: ;in, For conditional probability, It is an unconditional probability. To measure the number of times, The threshold value is used.

[0175] Develop a blockchain-driven device management system to achieve full lifecycle management of devices and upload device information and operation records to the blockchain: ;in, It contains information such as the hash value and timestamp of the previous block. Includes equipment status, maintenance records, etc., and automatically executes maintenance plans using smart contracts: ;

[0176] in, To uphold the contract, As a triggering condition, and To execute actions, the system integration modules form a complete system. Digital twin technology enables virtual simulation and system optimization. The newly designed quantum-secure communication protocol uses quantum key distribution technology to ensure data transmission security and prevent information leakage and tampering. The blockchain-driven equipment management system realizes full lifecycle management of equipment and uses smart contracts to automatically execute maintenance plans, improving the transparency and automation level of equipment management. The application of these technologies not only improves the overall performance and reliability of the ball mill control system, but also provides innovative solutions for equipment management and data security, making the ball mill control and management system more intelligent, safe, and efficient.

[0177] In summary, this embodiment, in terms of operational status monitoring, employs a comprehensive monitoring system built with multiple sensors. Combined with dynamic noise compensation algorithms and a spatiotemporal feature fusion network, it not only accurately collects key data in real time but also effectively suppresses noise and uncovers potential data correlations, providing high-quality foundational information for subsequent analysis and control. Data preprocessing and feature extraction, along with multi-scale data fusion techniques, delve into signal details. The spatiotemporal feature fusion network integrates the spatiotemporal features of multi-sensor data, improving data processing accuracy and laying a solid foundation for model building and parameter optimization. Regarding the mathematical model construction for the grinding process, the improved two-stage grinding kinetic equation and the deep learning-based grinding efficiency prediction model better reflect actual working conditions. The introduction of a quantum hybrid neural network model and a multi-scale grinding kinetic model allows for the comprehensive consideration of multiple factors, including grinding efficiency, energy consumption, equipment wear, quality stability, and carbon emissions, when determining the objective function. A multi-objective optimization function is established, and the weights are flexibly adjusted using a dynamic weight adaptive optimization algorithm to achieve a balance between multiple objectives. Parameter optimization based on intelligent algorithms employs an improved quantum particle swarm optimization algorithm, combined with a multi-agent collaborative optimization framework and hybrid optimization methods, significantly enhancing global search and convergence capabilities. This enables collaborative operation between the ball mill and the production line. Real-time control and regulation utilize adaptive fuzzy PID control algorithms, multivariable collaborative control strategies, and predictive control compensation algorithms to improve control accuracy and response speed, reducing system lag. The application of dynamic adaptive control architecture and distributed collaborative control algorithms allows the system to flexibly adapt to different operating conditions, maintaining stable and efficient operation. In terms of fault diagnosis and early warning, a fault diagnosis model based on transfer learning, combined with quantum entanglement fault diagnosis algorithms and federated learning fault diagnosis systems, significantly improves the efficiency, accuracy, and feature discrimination of fault diagnosis. Fault evolution prediction models and multi-dimensional fault feature fusion models can predict fault trends in advance, constructing a comprehensive and accurate fault diagnosis and early warning system, reducing equipment failure risks and maintenance costs. Operational data recording and analysis, through the establishment of a recording system and the application of algorithms such as reinforcement learning, achieves in-depth data mining and utilization. A digital twin-driven reinforcement learning framework and causal inference analysis methods further optimize control strategies, continuously improving the ball mill's operating performance.

[0178] Although embodiments of the invention have been shown and described, the scope of the invention will be defined by the appended claims and their equivalents by those skilled in the art.

Claims

1. A ball mill for producing aluminum silicate, characterized in that, The ball mill includes: The frame (1) has a first bearing seat (3) and a second bearing seat (24) fixedly connected to the top two sides of the frame (1). The outer cylinder (2) is rotatably connected to the frame (1) via the first bearing seat (3) and the second bearing seat (24). The inner cylinder (25) is fixedly connected inside the outer cylinder (2). The inner cylinder (25) is provided with a number of steel balls (26). The steel balls (26) are provided with a counterweight adjustment component. A drive assembly located on the frame (1) for driving the outer cylinder (2) to rotate; The discharge pipe (14) is fixedly connected to the outside of the first bearing seat (3), and the side end of the outer cylinder (2) is provided with a discharge port (15) corresponding to the position of the discharge pipe (14). Feed pipe (21), the feed pipe (21) is fixedly connected to the outside of the second bearing seat (24), and the feed port is provided at the position of the feed pipe (21) corresponding to the side end of the outer cylinder (2); A control interaction unit monitors and adjusts the operating status of the ball mill and performs fault diagnosis. The steel ball (26) has a spiral blade (27) on its outside and an inner ball (31) inside. A connecting rod (32) is fixedly connected to the outside of the inner ball (31). The connecting rod (32) is fixedly connected to the steel ball (26). The inner ball (31) has multiple sets of sliding holes (33). The counterweight adjustment assembly includes a first threaded rod (34), a second threaded rod (37), and a third threaded rod (38) rotatably connected inside the steel ball (26). A third bevel gear (43) is fixedly connected to the bottom of the first threaded rod (34). A second bevel gear (42) is fixedly connected to the outside of the second threaded rod (37). The third threaded rod (38) is fixedly connected to the outside of the second threaded rod (37). One end of the threaded rod (38) is fixedly connected to a first bevel gear (41). The third bevel gear (43) meshes with the second bevel gear (42). The second bevel gear (42) meshes with the first bevel gear (41). The second threaded rod (37) is a bidirectional threaded rod with opposite thread groove directions. The first threaded rod (34), the second threaded rod (37), and the third threaded rod (38) are all threaded with counterweight balls (40). There are two counterweight balls (40) on the outside of the second threaded rod (37). Multiple sets of guide rods (39) are also fixedly connected inside the steel ball (26). The guide rods (39) pass through the counterweight balls (40).

2. The ball mill for producing aluminum silicate according to claim 1, characterized in that, The drive assembly includes a first motor (9), a gearbox (10), a first gear (4), and a gear ring (5). The gear ring (5) is fixedly connected to the outer cylinder (2). The first gear (4) is rotatably connected to the frame (1). The gearbox (10) is fixedly connected to the frame (1). A second gear (11) and a third gear (12) are rotatably connected inside the gearbox (10). The first motor (9) is fixedly connected to the frame (1). The output end of the first motor (9) is fixedly connected to the second gear (11). The third gear (12) is coaxially fixedly connected to a synchronous shaft (13). The synchronous shaft (13) is coaxially fixedly connected to the first gear (4). The first gear (4) meshes with the gear ring (5).

3. The ball mill for producing aluminum silicate according to claim 2, characterized in that, The discharge pipe (14) is fixedly connected to a first fixed frame (16), the feed pipe (21) is fixedly connected to a second fixed frame (19), the first fixed frame (16) is fixedly connected to a second motor (17), the second fixed frame (19) is fixedly connected to a third motor (20), the output end of the second motor (17) is fixedly connected to a first spiral conveying rod (18), the output end of the third motor (20) is fixedly connected to a second spiral conveying rod (22), a feed hopper (23) is provided above the feed pipe (21), the output end of the feed hopper (23) is fixedly connected to the feed pipe (21), the outer cylinder (2) and the inner cylinder (25) are provided with inlets and outlets (8), the inlets and outlets (8) are provided with sealing caps (6), the sealing caps (6) are fastened in the inlets and outlets (8), and long bolts (7) are provided through the inlets and outlets (8) and the sealing caps (6).

4. A ball mill for producing aluminum silicate according to claim 3, characterized in that, The steel ball (26) is externally connected to the first threaded rod (34), the second threaded rod (37) and the third threaded rod (38) with fixed blocks (28). One of the fixed blocks (28) is provided with a positioning groove (35). One end of the first threaded rod (34) extends into the positioning groove (35) and is fixedly connected to a rotating block (36). The positioning groove (35) is externally provided with a positioning plate (29). The bottom of the positioning plate (29) is fixedly connected with a positioning block (44). The positioning plate (29) is threadedly connected with a positioning bolt (30). The positioning bolt (30) is threadedly connected to the fixed block (28). The rotating block (36) is externally provided with a hexagonal groove at the position of the positioning block (44).

5. A ball mill for producing aluminum silicate according to claim 1, characterized in that, The control and interaction unit includes a ball mill operation status monitoring module, a ball mill grinding process mathematical model establishment module, an objective function optimization module, a parameter optimization module based on intelligent algorithms, a real-time control and adjustment module, a fault diagnosis and early warning module, a ball mill operation data recording module, a human-machine interface module, and a system integration module.

Images