Method and system for automatic regulation of the dynamic gap of a nanogrinder

Abstract

The application relates to the technical field of nanometer sand mill, and particularly discloses a dynamic gap automatic adjusting method and system of a nanometer sand mill. The method comprises the following steps: collecting a medium dust concentration sequence in real time, constructing a medium wear rate model and predicting an equivalent particle size attenuation curve through time sequence analysis; applying a short-time pulse excitation in a grinding starting stage, collecting response data, online identifying hardness coefficients, viscosity indexes and breaking energy consumption characteristics of the measured material, and constructing a dynamic matching model of the material and the grinding parameters; calculating a prospective pre-adjustment amount of the separator gap based on the attenuation curve and the matching model; collecting online particle size detection values to obtain real-time feedback adjustment amounts; fusing the prospective pre-adjustment amount and the real-time feedback adjustment amounts according to dynamic weights to generate a composite adjustment instruction; and driving an electromagnetic servo mechanism to adjust the gap to a target position. The application realizes a fundamental change from lagging feedback to prospective pre-compensation, and effectively solves the precision fluctuation problem caused by the coupling of medium wear and material change.

Description

Technical Field

[0001] This invention relates to the field of nano-sand mill technology, and in particular to a method and system for automatic adjustment of dynamic gap in nano-sand mills. Background Technology

[0002] In the field of nano-grinding mills, precise control of the separator gap directly determines the separation efficiency of the grinding media, the uniformity of the slurry output particle size, and the overall energy consumption level. With increasingly stringent requirements for nanoscale material particle size distribution in industries such as new energy materials, high-end coatings, and pharmaceuticals, dynamic gap adjustment technology has become a core element in the intelligent upgrading of grinding mills. Existing technologies employ mechanical telescopic structures to achieve adjustable gaps, but the adjustment logic relies on manual experience; other solutions introduce online particle size detection for feedback control, achieving automatic adjustment based on the output particle size. These technologies improve the adaptability of the equipment to some extent, but none fundamentally solve the control challenges arising from the coupling between continuous wear of the grinding media and dynamic changes in material properties during the grinding process.

[0003] Current technologies generally employ a post-feedback adjustment mode, meaning that gap compensation is only performed after the output particle size deviates from the target value. This delayed adjustment method has inherent drawbacks: as the grinding media gradually wears down, causing the equivalent particle size to decrease, the matching relationship between the separator gap and the media particle size shifts, and adjustment is only triggered when the output particle size exceeds the standard, by which time a large number of defective products have already been generated. Furthermore, different materials exhibit significant differences in hardness, viscosity, and crushing energy consumption characteristics, making it impossible to adapt to sudden changes in operating conditions when materials are switched using uniform adjustment parameters. The root cause of these problems lies in the lack of forward-looking prediction capabilities for media wear trends and online adaptive capabilities for material characteristics in current technologies. This results in gap adjustment always being in a passive response state, unable to pre-compensate before disturbances occur, and failing to meet the extreme requirements of process stability and product consistency in nanoscale grinding. In particular, when preparing core soft magnetic materials for energy-saving motors or transformers, the magnetic properties of the soft magnetic materials (such as permeability, coercivity, iron loss, etc.) are extremely sensitive to the uniformity and consistency of powder particle size. The lag adjustment mode of the existing technology causes particle size fluctuations to exceed the process window, which directly restricts the stable mass production of high-performance soft magnetic materials, and thus affects the improvement of the energy efficiency level of energy-saving motors and transformers.

[0004] Therefore, this invention proposes a method and system for automatic dynamic gap adjustment of nano-sand mills. Summary of the Invention

[0005] This invention provides a method and system for automatic dynamic gap adjustment of a nano-sand mill. By combining wear trend prediction with material characteristic identification, the invention overcomes the defects of particle size fluctuation caused by the lagging adjustment in the prior art and achieves precise adaptive control of the gap of the nano-sand mill.

[0006] This invention provides a method for automatic dynamic gap adjustment of a nano-sand mill, comprising: The concentration sequence of media dust discharged after grinding by the nano-sand mill is collected in real time. By performing time series analysis on the media dust concentration sequence, a media wear rate model is constructed, and the equivalent particle size decay curve of the grinding media in the future preset time window is predicted based on the media wear rate model. During the grinding start-up phase, a preset short-time pulse excitation is applied to the test material, and the response data of the test material under the preset short-time pulse excitation is collected. Based on the response data, the hardness coefficient, viscosity index and crushing energy consumption characteristics of the test material are identified online, and a dynamic matching model between the material and grinding parameters is constructed. Based on the equivalent particle size decay curve and dynamic matching model, the forward-looking pre-adjustment amount of the separator gap of the nano-sand mill is calculated at each moment within a future preset time window. The online particle size detection value at the discharge port is collected in real time, the real-time deviation between the online particle size detection value and the target particle size value is calculated, and the real-time feedback adjustment amount of the gap is obtained based on the real-time deviation. The forward-looking pre-adjustment amount and the real-time feedback adjustment amount are weighted and fused according to the preset dynamic weights to generate a composite adjustment command; The electromagnetic servo mechanism connected to the separator is driven to move according to the compound adjustment command, and the separator gap is adjusted to the target position.

[0007] Furthermore, the steps of collecting the real-time concentration sequence of media dust discharged after grinding by the nano-grinding mill, and constructing a media wear rate model by performing time series analysis on the media dust concentration sequence include: A laser dust concentration sensor is installed on the discharge pipe downstream of the nano-sand mill discharge port to continuously collect the instantaneous value of the media dust concentration discharged with the slurry after grinding at a preset sampling frequency. The instantaneous values ​​of media dust concentration collected continuously are arranged in order of collection time to generate a media dust concentration time series; Outlier removal and smoothing filtering are performed on the media dust concentration time series to obtain the preprocessed media dust concentration series; The cumulative integral of the pretreated media dust concentration sequence is calculated to obtain the cumulative mass of media dust discharged per unit time. The cumulative wear rate of the media at the current moment is calculated based on the ratio of the cumulative discharged mass to the initial total mass of the grinding media. Using grinding operation time as the independent variable and cumulative wear rate of the media as the dependent variable, an exponential decay regression model was used for curve fitting to obtain the media wear rate model.

[0008] Furthermore, the step of predicting the equivalent particle size decay curve of the grinding media within a future preset time window based on the media wear rate model includes: The wear rate constant is extracted from the medium wear rate model, and the predicted cumulative wear rate is calculated at each moment within a future preset time window; Based on the predicted cumulative wear rate and the initial particle size of the medium, the equivalent remaining particle size at each moment within the future preset time window is calculated according to the mapping relationship between particle size and wear rate. The equivalent remaining particle size at each time point within a future preset time window is connected in chronological order to generate an equivalent particle size decay curve.

[0009] Furthermore, the step of applying a preset short-duration pulse excitation to the test material during the grinding start-up phase and collecting the response data of the test material under the preset short-duration pulse excitation includes: During the initial period after the grinding starts, the stirrer is controlled to apply an excitation signal according to a preset pulse waveform, which includes a square wave pulse sequence or a triangular wave sweep pulse. The instantaneous power response curve of the agitator drive motor, the instantaneous pressure fluctuation curve in the grinding chamber, and the instantaneous flow fluctuation curve at the discharge port are collected simultaneously. The instantaneous power response curve, instantaneous pressure fluctuation curve, and instantaneous flow fluctuation curve are used as the response data of the tested material under a preset short-time pulse excitation.

[0010] Furthermore, based on the response data, the hardness coefficient, viscosity index, and crushing energy consumption characteristics of the tested material are identified online, and a dynamic matching model between the material and grinding parameters is constructed, including: Based on the pre-stored hardness-response mapping table, the peak power, power rise slope and power decay time constant extracted from the instantaneous power response curve during pulse excitation are mapped to the hardness coefficient of the material. The ratio of pressure pulsation amplitude to excitation frequency is extracted from the instantaneous pressure fluctuation curve and used as a characterization value of material viscosity. The characterization value of material viscosity is then normalized to obtain the viscosity index of the material. Extract the cumulative material throughput corresponding to the unit excitation energy input from the instantaneous flow fluctuation curve as the material crushing energy consumption efficiency value, and take the reciprocal of the crushing energy consumption efficiency value as the material crushing energy consumption characteristic. Based on hardness coefficient, viscosity index and crushing energy consumption characteristics, a dynamic matching model is constructed with material characteristics as input and recommended grinding process parameters as output.

[0011] Furthermore, based on the equivalent particle size decay curve and dynamic matching model, the step of calculating the forward-looking pre-adjustment amount of the separator gap at each time point within a future preset time window includes: Extract the equivalent remaining particle size at each time point within a future preset time window from the equivalent particle size decay curve; Obtain the optimal grinding gap coefficient corresponding to the current material characteristics from the dynamic matching model; Based on the equivalent residual particle size and the optimal grinding gap coefficient, the basic pre-adjustment amount is calculated at each moment within the future preset time window; Based on the decay rate of the equivalent particle size decay curve, the basic pre-adjustment amount at each moment within the future preset time window is dynamically corrected to obtain the forward-looking pre-adjustment amount at each moment within the future preset time window.

[0012] Furthermore, the step of weighting and fusing the forward-looking pre-adjustment amount and the real-time feedback adjustment amount according to a preset dynamic weight to generate a composite adjustment command includes: Based on the historical prediction error of the media wear rate model and the identification stability of the material dynamic matching model, the comprehensive confidence index of the media wear rate model and the material dynamic matching model at the current moment is calculated. Based on the fluctuation amplitude of the online particle size detection value and the measurement signal-to-noise ratio, the confidence index of the feedback adjustment at the current moment is calculated; Based on the comprehensive confidence index of the media wear rate model and the material dynamic matching model, as well as the confidence index of the feedback adjustment amount, dynamic weights are calculated so that the higher the comprehensive confidence of the media wear rate model and the material dynamic matching model, the greater the weight of the forward adjustment amount; and the higher the confidence of the feedback adjustment amount, the greater the weight of the real-time feedback adjustment amount. The composite adjustment amount is obtained by weighting and summing the forward-looking pre-adjustment amount and the real-time feedback adjustment amount according to the dynamic weights. The composite adjustment amount is converted into a displacement drive command for the electromagnetic servo mechanism, which is then used as the composite adjustment command.

[0013] Furthermore, the step of adjusting the separator gap to the target position by driving the electromagnetic servo mechanism connected to the separator according to the composite adjustment command includes: The composite adjustment command is converted into the target displacement value of the separator, and the target displacement value is sent to the servo driver of the electromagnetic servo mechanism; The servo driver generates a drive current, which drives the electromagnetic servo mechanism to move the separator. The feedback values ​​of the displacement sensors installed on the separator are collected in real time to form a closed-loop position control until the deviation between the actual displacement of the separator and the target displacement value is less than the preset positioning accuracy threshold.

[0014] Furthermore, it also includes collaborative protection steps for abnormal operating conditions: Real-time monitoring of the slurry temperature, grinding chamber pressure, and spindle motor current in the nano-sand mill. When the slurry temperature exceeds the preset temperature threshold, the grinding chamber pressure exceeds the preset pressure threshold, or the spindle motor current exceeds the preset current threshold, the abnormal protection mode is triggered. In abnormal protection mode, the weighted fusion process of forward pre-adjustment amount and real-time feedback adjustment amount is suspended, and the emergency command to increase the gap is forcibly executed to quickly increase the separator gap to the preset safe gap value. At the same time, a deceleration command is sent to the agitator drive motor to reduce the agitator linear speed to the preset safe linear speed. Once all monitoring parameters return to the normal range and remain within the preset time, the abnormal protection mode will exit, and the normal automatic gap adjustment process will resume.

[0015] This invention provides a dynamic gap automatic adjustment system for a nano-sand mill, comprising: The wear prediction module is used to collect the concentration sequence of media dust discharged after grinding by the nano-grinding mill in real time. By performing time series analysis on the media dust concentration sequence, a media wear rate model is constructed, and the equivalent particle size decay curve of the grinding media in the future preset time window is predicted based on the media wear rate model. The material identification module is used to apply a preset short-time pulse excitation to the material under test during the grinding start-up stage, collect the response data of the material under test under the preset short-time pulse excitation, and identify the hardness coefficient, viscosity index and crushing energy consumption characteristics of the material under test online based on the response data, and construct a dynamic matching model between the material and grinding parameters. The pre-adjustment calculation module is used to calculate the forward-looking pre-adjustment amount of the separator gap of the nano-sand mill at each moment within a future preset time window, based on the equivalent particle size decay curve and dynamic matching model. The feedback adjustment module is used to collect the online particle size detection value at the discharge port in real time, calculate the real-time deviation between the online particle size detection value and the target particle size value, and obtain the real-time feedback adjustment amount of the gap based on the real-time deviation. The instruction fusion module is used to weight and fuse the forward-looking pre-adjustment amount and the real-time feedback adjustment amount according to the preset dynamic weight to generate a composite adjustment instruction; The execution control module is used to drive the electromagnetic servo mechanism connected to the separator to adjust the separator gap to the target position according to the compound adjustment command.

[0016] The beneficial effects of this invention compared to existing technologies are as follows: It achieves a fundamental shift in the gap control of nano-grinding mills from hysteretic feedback to proactive pre-compensation. By predicting the wear trend of the grinding media in real time and identifying material characteristics online, it can proactively pre-adjust the separator gap before the equivalent particle size of the grinding media decays and material characteristics change, significantly improving the response speed and accuracy of particle size control. At the same time, by dynamically weighting the proactive pre-adjustment amount and the real-time feedback adjustment amount, it effectively solves the hysteresis of single feedback control and the uncertainty of single prediction model, greatly reducing the fluctuation range of output particle size and the defect rate, ensuring the high stability of the nano-grinding process and product consistency, and improving the equipment's adaptability to different materials and overall operating efficiency.

[0017] Other features and advantages of the invention will be set forth in the description which follows, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other advantages of the invention may be realized and obtained by means of the structures particularly pointed out in this application.

[0018] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0019] The accompanying drawings are provided to further illustrate the invention and form part of the specification. They are used in conjunction with embodiments of the invention to explain the invention and do not constitute a limitation thereof. In the drawings: Figure 1 This is an overall flowchart of the automatic dynamic gap adjustment method for nano-sand mill in an embodiment of the present invention; Figure 2 This is a flowchart of the dual-mode fusion control and execution feedback process in an embodiment of the present invention. Detailed Implementation

[0020] The preferred embodiments of the present invention will be described below with reference to the accompanying drawings. It should be understood that the preferred embodiments described herein are for illustration and explanation only and are not intended to limit the present invention.

[0021] like Figure 1 and Figure 2 As shown, this invention provides an embodiment of a method for automatically adjusting the dynamic gap of a nano-sand mill, comprising: The concentration sequence of media dust discharged after grinding by the nano-sand mill is collected in real time. By performing time series analysis on the media dust concentration sequence, a media wear rate model is constructed, and the equivalent particle size decay curve of the grinding media in the future preset time window is predicted based on the media wear rate model. During the grinding start-up phase, a preset short-time pulse excitation is applied to the test material, and the response data of the test material under the preset short-time pulse excitation is collected. Based on the response data, the hardness coefficient, viscosity index and crushing energy consumption characteristics of the test material are identified online, and a dynamic matching model between the material and grinding parameters is constructed. Based on the equivalent particle size decay curve and dynamic matching model, the forward-looking pre-adjustment amount of the separator gap of the nano-sand mill is calculated at each moment within a future preset time window. The online particle size detection value at the discharge port is collected in real time, the real-time deviation between the online particle size detection value and the target particle size value is calculated, and the real-time feedback adjustment amount of the gap is obtained based on the real-time deviation. The forward-looking pre-adjustment amount and the real-time feedback adjustment amount are weighted and fused according to the preset dynamic weights to generate a composite adjustment command; The electromagnetic servo mechanism connected to the separator is driven to move according to the compound adjustment command, and the separator gap is adjusted to the target position.

[0022] In this embodiment, the concentration sequence of media dust discharged after grinding by the nano-sand mill refers to the data set formed by arranging the concentration values ​​of the grinding media wear dust discharged with the slurry, which are continuously collected by the laser dust concentration sensor installed on the downstream discharge pipe at a sampling frequency of once per second, according to the order of collection time.

[0023] In this embodiment, the media wear rate model is obtained by calculating the ratio of the cumulative mass of media dust discharged per unit time to the initial total mass of the grinding media obtained by cumulative integral calculation, and by using the grinding operation time as the independent variable and the media cumulative wear rate as the dependent variable, and then using the exponential decay regression method to perform curve fitting to obtain the mathematical relationship. Its input is the grinding operation time, and the output is the media cumulative wear rate at the corresponding time.

[0024] In this embodiment, the grinding media refers to zirconia beads or other tiny spheres with a diameter of 0.1 to 0.3 mm that are filled in the grinding chamber of the nano-sand mill. They are used to grind material particles to the nanoscale through the shearing and impact forces generated by high-speed motion.

[0025] In this embodiment, the future preset time window refers to a fixed time interval extending backward from the current moment, with the duration set to thirty seconds to two minutes, used to predict the changing trend of the grinding media and material properties within this time period.

[0026] In this embodiment, the equivalent particle size decay curve refers to the curve formed by connecting the equivalent remaining particle size values ​​of the grinding medium at each time within a future preset time window based on the media wear rate model in chronological order, reflecting the trend of the media gradually wear down and become smaller as the grinding time increases.

[0027] In this embodiment, the grinding start-up phase refers to the initial 30-second period after each new batch of material is ground, during which material characteristics are identified to avoid interfering with normal production.

[0028] In this embodiment, the material being tested refers to the slurry-like material that is about to enter or has just entered the grinding chamber for grinding, and can be different types such as lithium iron phosphate, lithium cobalt oxide, and soft magnetic ferrite.

[0029] In this embodiment, the preset short-time pulse excitation refers to the excitation signal applied by controlling the stirrer according to a preset waveform during the grinding start-up stage. The waveform includes a square wave pulse sequence or a triangular wave sweep pulse, with a duration of 5 to 10 seconds, which is used to excite the material to produce a measurable response.

[0030] In this embodiment, the hardness coefficient of the tested material refers to the value determined by comparing three characteristic values ​​extracted from the instantaneous power response curve—power peak value, power rise slope, and power decay time constant—with a pre-stored hardness-response mapping table. The value ranges from 0 to 1, with a higher value indicating higher material hardness. The viscosity index of the tested material is the value obtained after normalizing the ratio of pressure pulsation amplitude to excitation frequency extracted from the instantaneous pressure fluctuation curve. The value ranges from 0 to 1, with a higher value indicating higher material viscosity. The crushing energy consumption characteristic of the tested material refers to the reciprocal of the cumulative material throughput corresponding to a unit excitation energy input extracted from the instantaneous flow fluctuation curve, reflecting the energy required to crush the material to the target particle size.

[0031] In this embodiment, constructing a dynamic matching model between materials and grinding parameters means taking the hardness coefficient, viscosity index, and crushing energy consumption characteristics obtained online as inputs, and outputting recommended values ​​of grinding process parameters such as the optimal grinding gap coefficient, the optimal agitator linear speed range, and the optimal media filling rate that match the current material characteristics through a pre-established lookup table relationship or linear regression method.

[0032] In this embodiment, the forward-looking pre-adjustment amount of the separator gap of the nano-sand mill calculated based on the equivalent particle size decay curve and the dynamic matching model at each time in the future preset time window refers to first extracting the equivalent remaining particle size at each time in the future from the equivalent particle size decay curve, then obtaining the optimal grinding gap coefficient corresponding to the current material from the dynamic matching model, subtracting the current actual gap value from the product of the equivalent remaining particle size and the optimal grinding gap coefficient to obtain the basic pre-adjustment amount, and then dynamically correcting the basic pre-adjustment amount according to the decay rate of the equivalent particle size decay curve to obtain the numerical sequence.

[0033] In this embodiment, the real-time acquisition of the online particle size detection value at the discharge port refers to the particle size distribution characteristic value of the solid particles in the discharged slurry at the current moment, which is continuously measured and recorded by a laser particle size analyzer installed on the discharge pipeline at a frequency of 5 to 10 times per second. Typically, D50 or D90 is taken as the representative particle size value.

[0034] In this embodiment, the real-time feedback adjustment amount of the gap obtained based on the real-time deviation refers to the real-time deviation obtained by subtracting the online particle size detection value from the preset target particle size value, inputting the real-time deviation into the proportional-integral-derivative controller, and outputting the separator gap adjustment amount after calculation to eliminate the current deviation.

[0035] In this embodiment, the preset dynamic weight refers to two weighting coefficients calculated in real time based on the comprehensive confidence index of the medium wear rate model and the material dynamic matching model, as well as the confidence index of the feedback adjustment amount. The sum of the two coefficients is always equal to 1, which is used to balance the proportion of forward adjustment amount and real-time feedback adjustment amount in the final composite command.

[0036] In this embodiment, the composite adjustment command refers to the final adjustment amount obtained by weighting and summing the forward-looking pre-adjustment amount and the real-time feedback adjustment amount according to dynamic weights, and then converting the adjustment amount into a displacement drive command that the electromagnetic servo mechanism can recognize and execute. The command format is the position control word of the standard industrial bus protocol.

[0037] In this embodiment, driving the electromagnetic servo mechanism connected to the separator to adjust the separator gap to the target position according to the composite adjustment command means sending the composite adjustment command to the servo driver of the electromagnetic servo mechanism. The servo driver generates a three-phase drive current according to the command to drive the linear motor to move the separator. At the same time, the displacement sensor provides real-time feedback on the actual position of the separator to form a closed-loop control, which is continuously corrected until the deviation between the actual position of the separator and the target position is less than the set positioning accuracy threshold.

[0038] Furthermore, the steps of collecting the real-time concentration sequence of media dust discharged after grinding by the nano-grinding mill, and constructing a media wear rate model by performing time series analysis on the media dust concentration sequence include: A laser dust concentration sensor is installed on the discharge pipe downstream of the nano-sand mill discharge port to continuously collect the instantaneous value of the media dust concentration discharged with the slurry after grinding at a preset sampling frequency. The instantaneous values ​​of media dust concentration collected continuously are arranged in order of collection time to generate a media dust concentration time series; Outlier removal and smoothing filtering are performed on the media dust concentration time series to obtain the preprocessed media dust concentration series; The cumulative integral of the pretreated media dust concentration sequence is calculated to obtain the cumulative mass of media dust discharged per unit time. The cumulative wear rate of the media at the current moment is calculated based on the ratio of the cumulative discharged mass to the initial total mass of the grinding media. Using grinding operation time as the independent variable and cumulative wear rate of the media as the dependent variable, an exponential decay regression model was used for curve fitting to obtain the media wear rate model.

[0039] In this embodiment, installing a laser dust concentration sensor on the discharge pipe downstream of the nano-sand mill discharge port means fixing the laser dust concentration sensor to the discharge pipe by means of flange or clamp. The sensor probe extends into the inside of the pipe and directly contacts the flowing slurry, continuously measuring and outputting the concentration value of the medium dust contained in the slurry at a sampling frequency of once per second.

[0040] In this embodiment, outlier removal and smoothing filtering preprocessing of the medium dust concentration time series refers to first using the Laida criterion to remove outlier data points that deviate from the mean by more than three standard deviations, and then using the five-point triple moving average method to smooth the removed series, thereby eliminating high-frequency random noise interference and obtaining a preprocessed series that better reflects the true trend of change.

[0041] In this embodiment, the cumulative mass of media dust discharged per unit time by performing cumulative integration on the pre-treated media dust concentration sequence refers to multiplying the pre-treated concentration sequence with the corresponding slurry flow rate sequence to obtain the instantaneous dust discharge rate, and then performing trapezoidal integration on the instantaneous dust discharge rate per unit time to obtain the total mass of media dust discharged from the grinding chamber per unit time.

[0042] In this embodiment, the initial total mass of the grinding media refers to the total mass of the newly loaded zirconia beads and other grinding media before the start of grinding, which is obtained by weighing during loading and serves as a benchmark value for calculating the wear rate of the media.

[0043] In this embodiment, calculating the cumulative wear rate of the grinding media at the current moment based on the ratio of the cumulative discharged mass to the initial total mass of the grinding media means dividing the cumulative discharged mass of the grinding media dust per unit time by the initial total mass of the grinding media to obtain the wear rate per unit time, and then summing the wear rates of all unit times from the start of grinding to the current moment to obtain the cumulative wear rate of the grinding media at the current moment.

[0044] In this embodiment, the media wear rate model is obtained by using an exponential decay regression model with grinding operation time as the independent variable and media cumulative wear rate as the dependent variable. This model uses grinding operation time as input data and media cumulative wear rate at the corresponding moment as output data. It uses an exponential decay function to perform nonlinear least squares fitting to obtain a regression equation that can describe the change law of media wear over time. This equation can predict the media cumulative wear rate at any grinding moment.

[0045] Furthermore, the step of predicting the equivalent particle size decay curve of the grinding media within a preset time window based on the media wear rate model includes: The wear rate constant is extracted from the medium wear rate model, and the predicted cumulative wear rate is calculated at each moment within a future preset time window; Based on the predicted cumulative wear rate and the initial particle size of the medium, the equivalent remaining particle size at each moment within the future preset time window is calculated according to the mapping relationship between particle size and wear rate. The equivalent remaining particle size at each time point within a future preset time window is connected in chronological order to generate an equivalent particle size decay curve.

[0046] In this embodiment, extracting the wear rate constant from the wear rate model refers to reading the decay rate parameter value in the exponential decay function expression from the wear rate model obtained by fitting through exponential decay regression. This parameter reflects the rate at which the grinding media wears as the grinding time increases.

[0047] In this embodiment, calculating the predicted cumulative wear rate at each moment within a future preset time window means substituting each moment within the future preset time window as an input quantity into the medium wear rate model, and outputting the predicted value of the medium cumulative wear rate corresponding to each moment through model calculation.

[0048] In this embodiment, the initial particle size of the media refers to the original diameter of the grinding media newly loaded into the grinding chamber before grinding begins. It is obtained by the nominal value provided by the supplier or by sampling measurement before use, and serves as a benchmark value for calculating the wear degree of the media.

[0049] In this embodiment, the particle size-wear rate mapping relationship refers to the correspondence between the cumulative wear rate of the medium and the remaining particle size of the medium, established through pre-experimental calibration. This is represented by a curve where the remaining particle size gradually decreases as the wear rate increases. The particle size-wear rate mapping relationship is calibrated through wear experiments on batch grinding media, and a power function fitting is used to obtain the mathematical expression: D equals the product of D0 and (1-W) raised to the power of k, where D is the equivalent remaining particle size, D0 is the initial particle size, W is the cumulative wear rate, and k is the wear correction coefficient, ranging from 0.8 to 1.2.

[0050] In this embodiment, calculating the equivalent remaining particle size at each time within a future preset time window based on the predicted cumulative wear rate and the initial particle size of the medium according to the particle size and wear rate mapping relationship means substituting the predicted cumulative wear rate at each future time into the particle size and wear rate mapping relationship to obtain the corresponding particle size attenuation ratio, and then multiplying the initial particle size of the medium by the attenuation ratio to calculate the equivalent remaining particle size value of the grinding medium at each future time.

[0051] In this embodiment, the equivalent residual particle size refers to the equivalent sphere diameter calculated after the grinding media has been ground and worn for a certain period of time, ignoring shape changes. It is used to characterize the current grinding capacity of the media and its matching relationship with the separator gap.

[0052] In this embodiment, connecting the equivalent remaining particle size at each time point within a future preset time window in chronological order to generate an equivalent particle size decay curve means plotting the calculated equivalent remaining particle size values ​​at each future time point in a coordinate system in chronological order, and connecting them sequentially with a smooth curve to form a curve reflecting the decay trend of the medium particle size over time.

[0053] Furthermore, the step of applying a preset short-duration pulse excitation to the test material during the grinding start-up phase and collecting the response data of the test material under the preset short-duration pulse excitation includes: During the initial period after the grinding starts, the stirrer is controlled to apply an excitation signal according to a preset pulse waveform, which includes a square wave pulse sequence or a triangular wave sweep pulse. The instantaneous power response curve of the agitator drive motor, the instantaneous pressure fluctuation curve in the grinding chamber, and the instantaneous flow fluctuation curve at the discharge port are collected simultaneously. The instantaneous power response curve, instantaneous pressure fluctuation curve, and instantaneous flow fluctuation curve are used as the response data of the tested material under a preset short-time pulse excitation.

[0054] In this embodiment, the preset pulse waveform refers to the excitation signal waveform parameters that are preset and stored in the controller, including the pulse amplitude, pulse width, pulse interval and number of pulses of the square wave pulse sequence, or the start frequency, end frequency, sweep time and amplitude of the triangular wave sweep pulse.

[0055] In this embodiment, controlling the stirrer to apply an excitation signal according to a preset pulse waveform means that the controller converts the preset pulse waveform into a speed or torque command and sends it to the stirrer drive motor, so that the stirrer generates a corresponding speed change or torque change according to the command, thereby applying mechanical excitation to the material being tested in the grinding chamber.

[0056] In this embodiment, the synchronous acquisition of the instantaneous power response curve of the stirrer drive motor, the instantaneous pressure fluctuation curve in the grinding chamber, and the instantaneous flow fluctuation curve at the discharge port means that while applying pulse excitation, the real-time power value of the stirrer drive motor, the real-time pressure value of the pressure sensor installed on the grinding chamber wall, and the real-time flow value of the flow sensor installed at the discharge port are recorded at a high sampling frequency of 1,000 times per second, respectively forming three response curves that change with time.

[0057] Furthermore, based on the response data, the hardness coefficient, viscosity index, and crushing energy consumption characteristics of the tested material are identified online, and a dynamic matching model between the material and grinding parameters is constructed, including: Based on the pre-stored hardness-response mapping table, the peak power, power rise slope and power decay time constant extracted from the instantaneous power response curve during pulse excitation are mapped to the hardness coefficient of the material. The ratio of pressure pulsation amplitude to excitation frequency is extracted from the instantaneous pressure fluctuation curve and used as a characterization value of material viscosity. The characterization value of material viscosity is then normalized to obtain the viscosity index of the material. Extract the cumulative material throughput corresponding to the unit excitation energy input from the instantaneous flow fluctuation curve as the material crushing energy consumption efficiency value, and take the reciprocal of the crushing energy consumption efficiency value as the material crushing energy consumption characteristic. Based on hardness coefficient, viscosity index and crushing energy consumption characteristics, a dynamic matching model is constructed with material characteristics as input and recommended grinding process parameters as output.

[0058] In this embodiment, the pre-stored hardness-response mapping table refers to a database table that is pre-established and stored in the controller through a large number of offline experiments. The table records the range of three characteristic values ​​of standard materials with different known hardness coefficients under the same pulse excitation conditions: power peak value, power rise slope, and power decay time constant, as well as the hardness coefficient calibration value corresponding to these characteristic value ranges.

[0059] In this embodiment, the power peak, power rise slope, and power decay time constant extracted from the instantaneous power response curve during the pulse excitation period refer to analyzing and processing the acquired instantaneous power response curve, finding the maximum power value reached during the pulse excitation period as the power peak, calculating the average slope of the curve from the start of excitation to the peak as the power rise slope, and calculating the time length of the curve from the peak to the stable value as the power decay time constant.

[0060] In this embodiment, mapping the peak power, power rise slope, and power decay time constant extracted from the instantaneous power response curve during the pulse excitation period to the hardness coefficient of the material based on the pre-stored hardness and response mapping table means comparing the three extracted feature values ​​with the feature value ranges of each standard material recorded in the hardness and response mapping table, finding the hardness coefficient calibration value corresponding to the range with the highest matching degree, and using this calibration value as the hardness coefficient of the current material being tested.

[0061] In this embodiment, extracting the ratio of pressure pulsation amplitude to excitation frequency from the instantaneous pressure fluctuation curve as a characterization value of material viscosity means performing a fast Fourier transform on the collected instantaneous pressure fluctuation curve, extracting the pressure amplitude corresponding to the frequency component with the same excitation frequency, and dividing the amplitude by the excitation frequency as the characterization value of material viscosity.

[0062] In this embodiment, normalizing the viscosity characterization value of the material to obtain the viscosity index of the material means dividing the current viscosity characterization value of the material by a preset maximum reference viscosity value to obtain a normalized value between 0 and 1 as the viscosity index of the material.

[0063] In this embodiment, extracting the cumulative material throughput corresponding to a unit excitation energy input from the instantaneous flow fluctuation curve as the material crushing energy efficiency value means dividing the total energy input to the agitator drive motor during the pulse excitation period by the cumulative material mass flowing out of the outlet during the excitation period, and obtaining the material mass that can be processed per unit of energy as the crushing energy efficiency value.

[0064] In this embodiment, taking the reciprocal of the crushing energy efficiency value as the crushing energy consumption characteristic of the material means dividing 1 by the crushing energy efficiency value to obtain the energy required to process a unit mass of material as the crushing energy consumption characteristic of the material. The larger the value, the more difficult the material is to crush.

[0065] In this embodiment, constructing a dynamic matching model based on hardness coefficient, viscosity index, and crushing energy consumption characteristics, with material characteristics as input and recommended grinding process parameters as output, refers to using multiple linear regression or table lookup interpolation methods to establish a mathematical relationship model with three parameters—hardness coefficient, viscosity index, and crushing energy consumption characteristics—as input variables, and grinding process parameters such as optimal grinding gap coefficient, optimal agitator linear speed, and optimal media filling rate as output variables.

[0066] In this embodiment, the recommended values ​​of grinding process parameters refer to the optimal combination of process parameters calculated and output by the dynamic matching model based on the current material characteristics, including the optimal matching coefficient between the separator gap and the media particle size, the recommended value of the agitator linear velocity, the recommended value of the grinding media filling rate, and the recommended value of the feed flow rate.

[0067] Furthermore, based on the equivalent particle size decay curve and dynamic matching model, the step of calculating the forward-looking pre-adjustment amount of the separator gap at each time point within a future preset time window includes: Extract the equivalent remaining particle size at each time point within a future preset time window from the equivalent particle size decay curve; Obtain the optimal grinding gap coefficient corresponding to the current material characteristics from the dynamic matching model; Based on the equivalent residual particle size and the optimal grinding gap coefficient, the basic pre-adjustment amount is calculated at each moment within the future preset time window; Based on the decay rate of the equivalent particle size decay curve, the basic pre-adjustment amount at each moment within the future preset time window is dynamically corrected to obtain the forward-looking pre-adjustment amount at each moment within the future preset time window.

[0068] In this embodiment, extracting the equivalent remaining particle size at each time point within a future preset time window from the equivalent particle size decay curve means substituting each time point within the future preset time window into the equivalent particle size decay curve and reading the equivalent remaining particle size value corresponding to each time point on the curve.

[0069] In this embodiment, the optimal grinding gap coefficient corresponding to the current material characteristics refers to the optimal matching relationship determined in advance through process experiments based on the hardness coefficient, viscosity index and crushing energy consumption characteristics of the current material being tested. This coefficient represents the ideal ratio between the separator gap and the equivalent residual particle size of the grinding media, and the value range is usually from 0.8 to 1.2.

[0070] In this embodiment, obtaining the optimal grinding gap coefficient corresponding to the current material characteristics from the dynamic matching model means substituting the currently identified hardness coefficient, viscosity index, and crushing energy consumption characteristics as inputs into the dynamic matching model. After calculation, the model outputs the optimal grinding gap coefficient value that matches the current material.

[0071] In this embodiment, the current material characteristics refer to the real-time values ​​of three parameters of the material being tested: hardness coefficient, viscosity index, and crushing energy consumption characteristics, obtained through online identification.

[0072] In this embodiment, calculating the basic pre-adjustment amount for each moment within a future preset time window based on the equivalent remaining particle size and the optimal grinding gap coefficient means multiplying the equivalent remaining particle size for each future moment by the optimal grinding gap coefficient to obtain the ideal gap value for each moment, and then subtracting the actual gap value of the separator at the current moment to obtain the basic pre-adjustment amount that needs to be adjusted at each moment.

[0073] In this embodiment, the decay rate of the equivalent particle size decay curve refers to the slope value of each point on the curve as a function of time obtained by taking the first derivative of the equivalent particle size decay curve. This value represents the rate at which the equivalent remaining particle size of the grinding media decreases with time.

[0074] In this embodiment, the forward-looking pre-adjustment amount for each moment within the future preset time window is obtained by dynamically correcting the basic pre-adjustment amount based on the decay rate of the equivalent particle size decay curve. This means multiplying the decay rate at each moment by a preset correction coefficient (the correction coefficient is determined according to the grinding media material; for zirconia media, it is 0.05-0.1, and for alumina media, it is 0.1-0.15, or it can be obtained through fitting of previous process experiments) to obtain the correction factor at that moment. Then, the basic pre-adjustment amount at each moment is multiplied by the corresponding correction factor to obtain the forward-looking pre-adjustment amount at each moment after dynamic adjustment.

[0075] Furthermore, the step of weighting and fusing the forward-looking pre-adjustment amount and the real-time feedback adjustment amount according to a preset dynamic weight to generate a composite adjustment command includes: Based on the historical prediction error of the media wear rate model and the identification stability of the material dynamic matching model, the comprehensive confidence index of the media wear rate model and the material dynamic matching model at the current moment is calculated. Based on the fluctuation amplitude of the online particle size detection value and the measurement signal-to-noise ratio, the confidence index of the feedback adjustment at the current moment is calculated; Based on the comprehensive confidence index of the media wear rate model and the material dynamic matching model, as well as the confidence index of the feedback adjustment amount, dynamic weights are calculated so that the higher the comprehensive confidence of the media wear rate model and the material dynamic matching model, the greater the weight of the forward adjustment amount; and the higher the confidence of the feedback adjustment amount, the greater the weight of the real-time feedback adjustment amount. The composite adjustment amount is obtained by weighting and summing the forward-looking pre-adjustment amount and the real-time feedback adjustment amount according to the dynamic weights. The composite adjustment amount is converted into a displacement drive command for the electromagnetic servo mechanism, which is then used as the composite adjustment command.

[0076] In this embodiment, the historical prediction error of the media wear rate model refers to comparing the predicted value of the cumulative wear rate by the media wear rate model with the actual measured value over a past period, and calculating the average absolute deviation or root mean square error between the two, which is used to evaluate the prediction accuracy of the model.

[0077] In this embodiment, the identification stability of the material dynamic matching model refers to the coefficient of variation or standard deviation of the identification results of the same batch of materials after multiple pulse excitations and identification of hardness coefficient, viscosity index and crushing energy consumption characteristics in a short period of time. The smaller the value, the more stable and reliable the model's identification results of material characteristics.

[0078] In this embodiment, the comprehensive confidence index of the media wear rate model and the material dynamic matching model at the current moment is calculated based on the historical prediction error of the media wear rate model and the identification stability of the material dynamic matching model. This is achieved by normalizing the historical prediction error and the identification stability separately, and then subtracting the weighted sum of the two from 1 to obtain a comprehensive confidence value ranging from 0 to 1. The larger the value, the more reliable the joint output of the two models. When calculating the comprehensive confidence, the weighting coefficient of the historical prediction error is 0.6, and the weighting coefficient of the identification stability is 0.4.

[0079] In this embodiment, the fluctuation amplitude and measurement signal-to-noise ratio of the online particle size detection value refer to the standard deviation of the continuously acquired online particle size detection value sequence as the fluctuation amplitude, and the ratio of the average value of the particle size detection value to the standard deviation of the background noise as the measurement signal-to-noise ratio. Both reflect the reliability of the current particle size detection.

[0080] In this embodiment, the confidence index of the feedback adjustment amount at the current moment, based on the fluctuation amplitude and measurement signal-to-noise ratio of the online granularity detection value, is calculated by normalizing the fluctuation amplitude and measurement signal-to-noise ratio respectively, taking the confidence index of the fluctuation amplitude contribution as an inversely proportional value, and the confidence index of the measurement signal-to-noise ratio contribution as a directly proportional value, and then weighted summing the two to obtain the confidence index of the feedback adjustment amount. The weighting coefficient for the confidence index of the fluctuation amplitude is 0.3, and the weighting coefficient for the confidence index of the measurement signal-to-noise ratio is 0.7.

[0081] In this embodiment, the calculation of dynamic weights based on the comprehensive confidence index of the medium wear rate model and the material dynamic matching model, as well as the confidence index of the feedback adjustment amount, refers to dividing the comprehensive confidence index by the sum of the comprehensive confidence index and the confidence index of the feedback adjustment amount to obtain the weight of the forward-looking pre-adjustment amount, and subtracting the weight of the forward-looking pre-adjustment amount from 1 to obtain the weight of the real-time feedback adjustment amount. The higher the comprehensive confidence, the greater the weight of the forward-looking pre-adjustment amount, and the higher the confidence of the feedback adjustment amount, the greater the weight of the real-time feedback adjustment amount.

[0082] In this embodiment, the composite adjustment amount is obtained by weighting and summing the forward-looking pre-adjustment amount and the real-time feedback adjustment amount according to the dynamic weight. This means multiplying the forward-looking pre-adjustment amount by the weight of the forward-looking pre-adjustment amount, and adding the real-time feedback adjustment amount by multiplying the weight of the real-time feedback adjustment amount to obtain the final fused adjustment amount value.

[0083] In this embodiment, converting the composite adjustment amount into a displacement drive command of the electromagnetic servo mechanism as a composite adjustment command means multiplying the composite adjustment amount by the displacement conversion coefficient to obtain the target displacement value of the separator, encapsulating the target displacement value into a position control command packet according to the communication protocol format of the servo driver, and sending it to the servo driver of the electromagnetic servo mechanism through the industrial bus.

[0084] Furthermore, the step of adjusting the separator gap to the target position by driving the electromagnetic servo mechanism connected to the separator according to the composite adjustment command includes: The composite adjustment command is converted into the target displacement value of the separator, and the target displacement value is sent to the servo driver of the electromagnetic servo mechanism; The servo driver generates a drive current, which drives the electromagnetic servo mechanism to move the separator. The feedback values ​​of the displacement sensors installed on the separator are collected in real time to form a closed-loop position control until the deviation between the actual displacement of the separator and the target displacement value is less than the preset positioning accuracy threshold.

[0085] In this embodiment, converting the composite adjustment command into the target displacement value of the separator means parsing the adjustment amount value from the received composite adjustment command data packet, adding the value to the current actual displacement value of the separator, and obtaining the target position value that the separator needs to achieve.

[0086] In this embodiment, sending the target displacement value to the servo driver of the electromagnetic servo mechanism means re-encapsulating the calculated target displacement value into a position command data packet according to the servo driver's communication protocol format, and sending it to the servo driver of the electromagnetic servo mechanism connected to the separator via industrial Ethernet or pulse sequence.

[0087] In this embodiment, the servo driver generating a drive current to drive the electromagnetic servo mechanism to move the separator means that after the servo driver receives the target displacement value, it compares it with the current position feedback value, calculates the required current command through the built-in proportional-integral-derivative control algorithm, and then converts the current command into a three-phase pulse width modulation signal to drive the power module. The output of the three-phase current corresponding to the command is sent to the linear motor of the electromagnetic servo mechanism, so that the motor generates electromagnetic force to drive the separator to move towards the target position.

[0088] In this embodiment, the real-time acquisition of feedback values ​​from displacement sensors installed on the separator constitutes a closed-loop position control until the deviation between the actual displacement of the separator and the target displacement value is less than the preset positioning accuracy threshold. This means that the servo driver reads the measurement values ​​of the optical or magnetic scale displacement sensors installed on the separator thousands of times per second, continuously compares the actual displacement value with the target displacement value, and adjusts the direction and magnitude of the drive current in real time according to the magnitude of the deviation, so that the separator gradually approaches the target position. This process is repeated until the absolute value of the difference between the actual displacement and the target displacement is less than the set positioning accuracy requirement.

[0089] In this embodiment, the preset positioning accuracy threshold refers to the maximum allowable deviation value preset according to the control accuracy requirements of the nano-sand mill for the separator gap. It is usually set between 0.005 mm and 0.01 mm. When the deviation between the actual displacement and the target displacement is less than this value, it is considered that the adjustment is in place.

[0090] Furthermore, it also includes collaborative protection steps for abnormal operating conditions: Real-time monitoring of the slurry temperature, grinding chamber pressure, and spindle motor current in the nano-sand mill. When the slurry temperature exceeds the preset temperature threshold, the grinding chamber pressure exceeds the preset pressure threshold, or the spindle motor current exceeds the preset current threshold, the abnormal protection mode is triggered. In abnormal protection mode, the weighted fusion process of forward pre-adjustment amount and real-time feedback adjustment amount is suspended, and the emergency command to increase the gap is forcibly executed to quickly increase the separator gap to the preset safe gap value. At the same time, a deceleration command is sent to the agitator drive motor to reduce the agitator linear speed to the preset safe linear speed. Once all monitoring parameters return to the normal range and remain within the preset time, the abnormal protection mode will exit, and the normal automatic gap adjustment process will resume.

[0091] In this embodiment, the preset temperature threshold refers to the highest allowable temperature value pre-set based on the temperature resistance limit of the grinding chamber material and seals of the nano-sand mill and the thermal stability requirements of the material. It is usually set to 60 to 80 degrees Celsius. When the slurry temperature exceeds this value, it may affect the safety of the equipment or the performance of the material.

[0092] In this embodiment, the preset pressure threshold refers to the maximum allowable pressure value preset according to the structure of the grinding chamber and the pressure bearing capacity of the seal. It is usually set to 0.3 to 0.5 MPa. When the pressure in the grinding chamber exceeds this value, there may be a risk of seal failure or equipment damage.

[0093] In this embodiment, the preset current threshold refers to the maximum allowable current value preset according to the rated current and thermal protection characteristics of the stirrer drive motor. It is usually set to 1.2 to 1.5 times the rated current of the motor. When the current of the spindle motor exceeds this value, it indicates that the motor may be overloaded.

[0094] In this embodiment, the gap increase emergency command refers to a special adjustment command automatically generated by the controller when the abnormal protection mode is triggered. This command forcibly pauses the weighted fusion process of the forward pre-adjustment amount and the real-time feedback adjustment amount, and directly outputs a control signal that makes the separator move rapidly in the direction of increasing the gap.

[0095] In this embodiment, rapidly increasing the separator gap to a preset safe gap value means executing an emergency command to increase the gap, controlling the electromagnetic servo mechanism to drive the separator to move at the maximum permissible speed until the separator gap reaches the preset safe gap value.

[0096] In this embodiment, the preset safety gap value refers to the gap value that is pre-set according to the maximum particle size of the grinding media and the structural characteristics of the equipment, which can ensure the smooth flow of the slurry and prevent the separator from getting stuck. It is usually set to 1.5 to 2 times the maximum gap during normal operation.

[0097] In this embodiment, the stirrer drive motor refers to the power equipment installed at the end of the main shaft of the nano-sand mill and used to drive the rotor of the stirrer to rotate. It is usually a variable frequency speed-regulating three-phase asynchronous motor or a permanent magnet synchronous motor.

[0098] In this embodiment, reducing the linear speed of the agitator to a preset safe linear speed means sending a deceleration command to the frequency converter of the agitator drive motor, so that the rotational speed of the agitator rotor decreases until the linear speed of the outer edge of the agitator reaches the preset safe linear speed value.

[0099] In this embodiment, the preset safe linear speed refers to the maximum permissible linear speed of the agitator that is pre-set according to the settling characteristics of the grinding media and the vibration requirements of the equipment, and can ensure the safety of the equipment under abnormal working conditions. It is usually set to 60% to 80% of the normal working linear speed.

[0100] In this embodiment, all monitoring parameters returning to the normal range and remaining within the preset time means that the slurry temperature value drops below the preset temperature threshold, the grinding chamber pressure value drops below the preset pressure threshold, and the spindle motor current value drops below the preset current threshold, and these three parameters remain within the normal range for more than 30 seconds.

[0101] In this embodiment, the normal automatic gap adjustment process refers to exiting the abnormal protection mode, restarting the weighted fusion calculation process of the forward pre-adjustment amount and the real-time feedback adjustment amount, and continuing to adjust the separator gap normally according to the composite adjustment command generated by the fusion.

[0102] The present invention also provides an embodiment of an automatic dynamic gap adjustment system for a nano-sand mill, comprising: The wear prediction module is used to collect the concentration sequence of media dust discharged after grinding by the nano-grinding mill in real time. By performing time series analysis on the media dust concentration sequence, a media wear rate model is constructed, and the equivalent particle size decay curve of the grinding media in the future preset time window is predicted based on the media wear rate model. The material identification module is used to apply a preset short-time pulse excitation to the material under test during the grinding start-up stage, collect the response data of the material under test under the preset short-time pulse excitation, and identify the hardness coefficient, viscosity index and crushing energy consumption characteristics of the material under test online based on the response data, and construct a dynamic matching model between the material and grinding parameters. The pre-adjustment calculation module is used to calculate the forward-looking pre-adjustment amount of the separator gap of the nano-sand mill at each moment within a future preset time window, based on the equivalent particle size decay curve and dynamic matching model. The feedback adjustment module is used to collect the online particle size detection value at the discharge port in real time, calculate the real-time deviation between the online particle size detection value and the target particle size value, and obtain the real-time feedback adjustment amount of the gap based on the real-time deviation. The instruction fusion module is used to weight and fuse the forward-looking pre-adjustment amount and the real-time feedback adjustment amount according to the preset dynamic weight to generate a composite adjustment instruction; The execution control module is used to drive the electromagnetic servo mechanism connected to the separator to adjust the separator gap to the target position according to the compound adjustment command.

[0103] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of this invention and its equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for automatically adjusting the dynamic gap of a nano-sand mill, characterized in that, include: The concentration sequence of media dust discharged after grinding by the nano-sand mill is collected in real time. By performing time series analysis on the media dust concentration sequence, a media wear rate model is constructed, and the equivalent particle size decay curve of the grinding media in the future preset time window is predicted based on the media wear rate model. During the grinding start-up phase, a preset short-time pulse excitation is applied to the test material, and the response data of the test material under the preset short-time pulse excitation is collected. Based on the response data, the hardness coefficient, viscosity index and crushing energy consumption characteristics of the test material are identified online, and a dynamic matching model between the material and grinding parameters is constructed. Based on the equivalent particle size decay curve and dynamic matching model, the forward-looking pre-adjustment amount of the separator gap of the nano-sand mill is calculated at each moment within a future preset time window. The online particle size detection value at the discharge port is collected in real time, the real-time deviation between the online particle size detection value and the target particle size value is calculated, and the real-time feedback adjustment amount of the gap is obtained based on the real-time deviation. The forward-looking pre-adjustment amount and the real-time feedback adjustment amount are weighted and fused according to the preset dynamic weights to generate a composite adjustment command; The electromagnetic servo mechanism connected to the separator is driven to move according to the compound adjustment command, and the separator gap is adjusted to the target position.

2. The method for automatic dynamic gap adjustment of a nano-sand mill according to claim 1, characterized in that, The steps for constructing a media wear rate model by real-time acquisition of the media dust concentration sequence discharged after grinding by a nano-grinding mill and time series analysis of the media dust concentration sequence include: A laser dust concentration sensor is installed on the discharge pipe downstream of the nano-sand mill discharge port to continuously collect the instantaneous value of the media dust concentration discharged with the slurry after grinding at a preset sampling frequency. The instantaneous values ​​of media dust concentration collected continuously are arranged in order of collection time to generate a media dust concentration time series; Outlier removal and smoothing filtering are performed on the media dust concentration time series to obtain the preprocessed media dust concentration series; The cumulative integral of the pretreated media dust concentration sequence is calculated to obtain the cumulative mass of media dust discharged per unit time. The cumulative wear rate of the media at the current moment is calculated based on the ratio of the cumulative discharged mass to the initial total mass of the grinding media. Using grinding operation time as the independent variable and cumulative wear rate of the media as the dependent variable, an exponential decay regression model was used for curve fitting to obtain the media wear rate model.

3. The method for automatic dynamic gap adjustment of a nano-sand mill according to claim 2, characterized in that, The steps for predicting the equivalent particle size decay curve of abrasive media within a future preset time window based on a media wear rate model include: The wear rate constant is extracted from the medium wear rate model, and the predicted cumulative wear rate is calculated at each moment within a future preset time window; Based on the predicted cumulative wear rate and the initial particle size of the medium, the equivalent remaining particle size at each moment within the future preset time window is calculated according to the mapping relationship between particle size and wear rate. The equivalent remaining particle size at each time point within a future preset time window is connected in chronological order to generate an equivalent particle size decay curve.

4. The method for automatic dynamic gap adjustment of a nano-sand mill according to claim 1, characterized in that, The steps of applying a preset short-duration pulse excitation to the test material during the grinding start-up phase and collecting the response data of the test material under the preset short-duration pulse excitation include: During the initial period after the grinding starts, the stirrer is controlled to apply an excitation signal according to a preset pulse waveform, which includes a square wave pulse sequence or a triangular wave sweep pulse. The instantaneous power response curve of the agitator drive motor, the instantaneous pressure fluctuation curve in the grinding chamber, and the instantaneous flow fluctuation curve at the discharge port are collected simultaneously. The instantaneous power response curve, instantaneous pressure fluctuation curve, and instantaneous flow fluctuation curve are used as the response data of the tested material under a preset short-time pulse excitation.

5. The method for automatic dynamic gap adjustment of a nano-sand mill according to claim 4, characterized in that, Based on the response data, the hardness coefficient, viscosity index, and crushing energy consumption characteristics of the tested material are identified online, and a dynamic matching model between the material and grinding parameters is constructed, including: Based on the pre-stored hardness-response mapping table, the peak power, power rise slope and power decay time constant extracted from the instantaneous power response curve during pulse excitation are mapped to the hardness coefficient of the material. The ratio of pressure pulsation amplitude to excitation frequency is extracted from the instantaneous pressure fluctuation curve and used as a characterization value of material viscosity. The characterization value of material viscosity is then normalized to obtain the viscosity index of the material. Extract the cumulative material throughput corresponding to the unit excitation energy input from the instantaneous flow fluctuation curve as the material crushing energy consumption efficiency value, and take the reciprocal of the crushing energy consumption efficiency value as the material crushing energy consumption characteristic. Based on hardness coefficient, viscosity index and crushing energy consumption characteristics, a dynamic matching model is constructed with material characteristics as input and recommended grinding process parameters as output.

6. The method for automatic dynamic gap adjustment of a nano-sand mill according to claim 1, characterized in that, Based on the equivalent particle size decay curve and dynamic matching model, the steps for calculating the forward-looking pre-adjustment amount of the separator gap at each time point within a future preset time window include: Extract the equivalent remaining particle size at each time point within a future preset time window from the equivalent particle size decay curve; Obtain the optimal grinding gap coefficient corresponding to the current material characteristics from the dynamic matching model; Based on the equivalent residual particle size and the optimal grinding gap coefficient, the basic pre-adjustment amount is calculated at each moment within the future preset time window; Based on the decay rate of the equivalent particle size decay curve, the basic pre-adjustment amount at each moment within the future preset time window is dynamically corrected to obtain the forward-looking pre-adjustment amount at each moment within the future preset time window.

7. The method for automatic dynamic gap adjustment of a nano-sand mill according to claim 1, characterized in that, The steps of weighting and fusing forward-looking pre-adjustment and real-time feedback adjustment according to preset dynamic weights to generate a composite adjustment command include: Based on the historical prediction error of the media wear rate model and the identification stability of the material dynamic matching model, the comprehensive confidence index of the media wear rate model and the material dynamic matching model at the current moment is calculated. Based on the fluctuation amplitude of the online particle size detection value and the measurement signal-to-noise ratio, the confidence index of the feedback adjustment at the current moment is calculated; Based on the comprehensive confidence index of the media wear rate model and the material dynamic matching model, as well as the confidence index of the feedback adjustment amount, dynamic weights are calculated so that the higher the comprehensive confidence of the media wear rate model and the material dynamic matching model, the greater the weight of the forward adjustment amount; and the higher the confidence of the feedback adjustment amount, the greater the weight of the real-time feedback adjustment amount. The composite adjustment amount is obtained by weighting and summing the forward-looking pre-adjustment amount and the real-time feedback adjustment amount according to the dynamic weights. The composite adjustment amount is converted into a displacement drive command for the electromagnetic servo mechanism, which is then used as the composite adjustment command.

8. The method for automatic dynamic gap adjustment of a nano-sand mill according to claim 1, characterized in that, The steps of adjusting the separator gap to the target position by driving the electromagnetic servo mechanism connected to the separator according to the composite adjustment command include: The composite adjustment command is converted into the target displacement value of the separator, and the target displacement value is sent to the servo driver of the electromagnetic servo mechanism; The servo driver generates a drive current, which drives the electromagnetic servo mechanism to move the separator. The feedback values ​​of the displacement sensors installed on the separator are collected in real time to form a closed-loop position control until the deviation between the actual displacement of the separator and the target displacement value is less than the preset positioning accuracy threshold.

9. The method for automatic dynamic gap adjustment of a nano-sand mill according to any one of claims 1 to 8, characterized in that, It also includes collaborative protection steps for abnormal operating conditions: Real-time monitoring of the slurry temperature, grinding chamber pressure, and spindle motor current in the nano-sand mill. When the slurry temperature exceeds the preset temperature threshold, the grinding chamber pressure exceeds the preset pressure threshold, or the spindle motor current exceeds the preset current threshold, the abnormal protection mode is triggered. In abnormal protection mode, the weighted fusion process of forward pre-adjustment amount and real-time feedback adjustment amount is suspended, and the emergency command to increase the gap is forcibly executed to quickly increase the separator gap to the preset safe gap value. At the same time, a deceleration command is sent to the agitator drive motor to reduce the agitator linear speed to the preset safe linear speed. Once all monitoring parameters return to the normal range and remain within the preset time, the abnormal protection mode will exit, and the normal automatic gap adjustment process will resume.

10. A dynamic gap automatic adjustment system for a nano-sand mill, characterized in that, include: The wear prediction module is used to collect the concentration sequence of media dust discharged after grinding by the nano-grinding mill in real time. By performing time series analysis on the media dust concentration sequence, a media wear rate model is constructed, and the equivalent particle size decay curve of the grinding media in the future preset time window is predicted based on the media wear rate model. The material identification module is used to apply a preset short-time pulse excitation to the material under test during the grinding start-up stage, collect the response data of the material under test under the preset short-time pulse excitation, and identify the hardness coefficient, viscosity index and crushing energy consumption characteristics of the material under test online based on the response data, and construct a dynamic matching model between the material and grinding parameters. The pre-adjustment calculation module is used to calculate the forward-looking pre-adjustment amount of the separator gap of the nano-sand mill at each moment within a future preset time window, based on the equivalent particle size decay curve and dynamic matching model. The feedback adjustment module is used to collect the online particle size detection value at the discharge port in real time, calculate the real-time deviation between the online particle size detection value and the target particle size value, and obtain the real-time feedback adjustment amount of the gap based on the real-time deviation. The instruction fusion module is used to weight and fuse the forward-looking pre-adjustment amount and the real-time feedback adjustment amount according to the preset dynamic weight to generate a composite adjustment instruction; The execution control module is used to drive the electromagnetic servo mechanism connected to the separator to adjust the separator gap to the target position according to the compound adjustment command.

Images