Intelligent regulation and control method applied to cement grinding grain composition

Abstract

The invention discloses an intelligent regulation and control method applied to cement grinding particle gradation, and relates to the technical field of cement grinding control, and the method comprises the following steps: S1, collecting grinding system operation parameters in real time, including mill current, powder concentrator rotating speed, feeding amount, grinding sound signals and online particle size detection data; s2, calculating core characteristic parameters based on the operation parameters, wherein the core characteristic parameters comprise an effective crushing power index, a dynamic fine powder retention rate and an in-mill material filling state index; s3, inputting the core characteristic parameters into a grading prediction model, and outputting predicted particle size distribution; s4, generating an adjustment instruction through a hierarchical control strategy based on the predicted particle size distribution; and S5, outputting the adjustment instruction to an execution mechanism. The particle size distribution is accurately predicted by collecting and analyzing the operation parameters of the grinding system in real time and combining core characteristic parameters, closed-loop intelligent regulation and control are achieved, the grinding efficiency, stability and product quality are improved, and meanwhile the control precision is optimized.

Description

Technical Field

[0001] The present invention relates to the technical field of cement grinding control, and specifically to an intelligent regulation method for particle size distribution applied to cement grinding. Background Art

[0002] The particle size distribution in the cement grinding process has a significant impact on the performance of the final product. Most existing grinding systems adopt fixed parameters or empirical control methods, lacking the ability to respond to dynamic operating conditions and being difficult to achieve intelligent regulation of particle size distribution. Especially when the properties of raw materials fluctuate or the equipment load changes, the adjustment lag is obvious, resulting in problems such as reduced fine powder recovery rate, increased energy consumption, and poor product stability.

[0003] At the same time, traditional control systems usually rely only on a single index, such as current or speed, and cannot comprehensively reflect the complex crushing and classification behaviors inside the mill. They lack a systematic modeling and feedback mechanism for crushing efficiency, fine powder retention, and material filling state, and are difficult to accurately approximate the target particle size distribution. Therefore, there is an urgent need for an intelligent regulation method for particle size distribution that integrates multi-source operating parameters, establishes a prediction model, and realizes closed-loop control to improve the regulation accuracy and operating efficiency of the cement grinding system. Summary of the Invention

[0004] Based on the above-mentioned disadvantages of the prior art, the purpose of the present invention is to provide an intelligent regulation method for particle size distribution applied to cement grinding to solve the above technical problems.

[0005] To achieve the above purpose, the present invention provides the following technical solution: An intelligent regulation method for particle size distribution applied to cement grinding, including: S1: Real-time collect the operating parameters of the grinding system, including mill current, separator speed, feeding amount, grinding sound signal, and on-line particle size detection data; S2: Calculate the core characteristic parameters based on the operating parameters. The core characteristic parameters include effective crushing work index, dynamic fine powder retention rate, and in-mill material filling state index; S3: Input the core characteristic parameters into the particle size distribution prediction model and output the predicted particle size distribution; S4: Generate adjustment instructions through a hierarchical control strategy based on the predicted particle size distribution; S5: Output the adjustment instructions to the actuator.

[0006] The present invention is further configured such that step S2 specifically includes: The effective crushing work index is calculated by the mechanical energy obtained by unit material within the integration period; The dynamic fine powder retention rate is determined by the ratio of the on-line particle size to the theoretical fine powder amount combined with the separator speed ratio; The in-mill material filling state index is characterized by the normalized difference between the grinding sound signal and the reference value.

[0007] The present invention is further configured such that the calculation logic of the effective breakage work index is as follows: , is the effective breakage work index, is the current time point, is the time window length, is the current value of the electric current, is the voltage value, is the mechanical efficiency of the motor, is the rotational speed of the main shaft of the mill, is the feeding rate, and the calculation logic of the mechanical efficiency of the motor is as follows: , is the maximum mechanical efficiency of the motor, is the optimal operating rotational speed of the mill, is the expansion factor; The calculation logic of the dynamic fine powder retention rate is as follows: , is the dynamic fine powder retention rate, is the proportion of particles smaller than measured by the on-line particle size analyzer, is the proportion of fine powder that should be in the theoretical equilibrium state smaller than , is the rotational speed of the classifier, is the reference critical rotational speed value of the classifier, is the preset threshold value; The calculation logic of the material filling state index in the mill is as follows: , is the material filling state index in the mill, is the modified linear unit function, is the current grinding sound power, is the reference grinding sound power, is the structural sound response coefficient of the mill, is the current material density.

[0008] The present invention is further configured such that step S3 specifically includes: Establish a first distribution function term reflecting the coarse particle distribution characteristics based on the effective breakage work index; Establish a second distribution function term reflecting the fine particle enrichment characteristics based on the dynamic fine powder retention rate; Establish a third distribution function term reflecting the coarse particle suppression characteristics based on the material filling state index in the mill; Perform weighted fusion on the first, second, and third distribution function terms, and generate a predicted particle size distribution through normalization processing; Optimize and train the parameters of the grading prediction model by using a loss function based on the distribution distance.

[0009] The present invention is further configured to predict the particle size distribution of the material particles by using a modified Rosin-Rammler distribution function based on the effective crushing work index. The calculation logic of the first distribution function term is as follows: , where is the cumulative distribution function of coarse particles, is the particle diameter, is the effective crushing work index, is the median particle size, is the distribution index, The calculation logic of is: is the scale factor, is the adjustment index, The calculation logic of is: is the reference distribution index, is the adjustment factor; The fine particle distribution is further corrected based on the dynamic fine powder retention rate. The calculation logic of the second distribution function term is as follows: , is the fine powder correction function, is the dynamic fine powder retention rate, is the characteristic particle size of the fine powder, is the distribution standard deviation; The secondary crushing probability of coarse particles is suppressed based on the material filling state in the mill. The calculation logic of the third distribution function term is as follows: , is the coarse particle suppression function, is the material filling state index in the mill, is the attenuation coefficient; The first, second, and third distribution function terms are weighted and fused, and the final predicted distribution is obtained by normalizing through the Softmax function. The calculation logic of the predicted distribution is as follows: , is the predicted particle size distribution, is the th weighting function, is the standard Softmax normalization function, , and are the weight coefficients; The calculation logic of the loss function is as follows: , is the loss function, is the predicted particle size distribution, is the actual particle size distribution, is the gradient coefficient, is the difference between the predicted particle size distribution and the actual particle size distribution, The calculation logic is as follows: , is the conveying amount allocated between and , and are the particle diameters of the predicted distribution and the actual distribution respectively; The parameter set of the grading prediction model includes the distribution function parameter and the weight coefficient , where , , the parameter set is expressed as: ; The update rule of the weight coefficient is as follows: , is the weight coefficient vector of the th iteration, is the weight learning rate, is the loss function gradient with respect to the weight coefficient, is the updated weight coefficient vector.

[0010] The present invention is further configured such that step S4 specifically includes: Obtain the current predicted particle distribution and the preset target particle distribution; Based on the difference between the predicted particle size distribution and the preset target particle size distribution, construct an adjustment function including multiple sub-control targets to obtain the adjustment amounts of the mill loading force, the classifier speed, and the feeding amount; Output an adjustment instruction vector, and the specific expression form of the adjustment instruction vector is: , is the adjustment instruction vector, is the adjustment amount of the mill loading force, is the adjustment amount of the classifier speed, is the adjustment amount of the feeding amount.

[0011] The present invention is further configured such that the calculation logic of the adjustment amount of the mill loading force is: , is the adjustment coefficient, is the effective crushing work index, is the predicted particle distribution, is the preset target particle size distribution; The calculation logic of the adjustment amount of the classifier speed is: , is the proportionality coefficient, is the structural similarity index, is the sign function, is the median particle size of the target particle distribution, is the median particle size of the current predicted particle size distribution; The feed rate adjustment amount has the following calculation logic: , is the sensitivity adjustment coefficient, is the in-mill material filling state index.

[0012] The present invention is further configured such that step S5 specifically includes: Converting the mill loading force adjustment instruction into a pressure control signal and outputting it to the mill pressure execution unit; Converting the classifier speed adjustment instruction into a speed control signal and outputting it to the classifier drive unit; Converting the feed rate adjustment instruction into a flow control signal and outputting it to the material conveying control unit.

[0013] The present invention is further configured such that the calculation logic of the mill loading force adjustment instruction is: , is the loading force set value, is the base loading force, is the mill loading force adjustment amount, is the mill structure coefficient; The calculation logic of the classifier speed adjustment instruction is: , is the time of the target set speed of the classifier, is the classifier speed adjustment amount; The calculation logic of the feed rate adjustment instruction is: , is the time of the feed rate, is the initial feed rate, is the response time constant, is the feed rate adjustment amount.

[0014] The present invention is further configured to further include dynamically adjusting the adjustment instruction based on the feedback information of the execution structure: Real-time monitoring of the change rate of the classifier current , when is greater than the preset threshold , triggering the loading force compensation, and the loading force compensation mechanism is: , is the adjusted mill loading force adjustment amount, is the adjustment step coefficient; Obtaining the actual particle size distribution through an on-line particle size analyzer , modify the parameters of the gradation prediction model every T minutes, and the logic for modifying the parameters of the gradation prediction model is as follows: , is the modified parameter of the gradation prediction model, is the difference between the predicted particle size distribution and the actual particle size distribution, is the gradient operator, is the learning rate; When the number of times of feed rate adjustment , trigger the adjustment of the reference grinding sound power , is the threshold of the preset adjustment times within a cycle, and the adjustment logic of the reference grinding sound power is as follows: , is the adjusted reference grinding sound power, is the adjustment amount of the grinding sound power, and the calculation logic of the adjustment amount of the grinding sound power is as follows: , is the average grinding sound power in the current cycle, is the smoothing coefficient; If , then reject this adjustment, adopt a fallback mechanism, and keep the original unchanged, where is the long-term average value of the grinding sound power, is the standard deviation of the grinding sound power, and are the periodic statistics based on the grinding sound signal.

[0015] The present invention provides an intelligent control method for the particle size gradation in cement grinding. Through S1: Real-time collection of the operating parameters of the grinding system, including mill current, classifier speed, feed rate, grinding sound signal, and online particle size detection data; S2: Calculating the core characteristic parameters based on the operating parameters, and the core characteristic parameters include effective crushing work index, dynamic fine powder retention rate, and grinding chamber material filling state index; S3: Inputting the core characteristic parameters into the gradation prediction model to output the predicted particle size distribution; S4: Generating adjustment instructions through a hierarchical control strategy based on the predicted particle size distribution; S5: Outputting the adjustment instructions to the actuator, and the beneficial effects generated include: 1. Improve the grinding efficiency and stability: By real-time collecting and analyzing the operating parameters of the grinding system, combined with multiple core characteristic parameters, such as effective crushing work index, dynamic fine powder retention rate, and grinding chamber material filling state index, it is possible to more accurately predict and adjust the particle size gradation, thereby significantly improving the efficiency and stability of the grinding process and avoiding the deficiency of over-relying on experience in the traditional method; 2. Optimize the prediction of particle size distribution: By establishing a multi-level distribution function term based on the effective work index of crushing, the dynamic retention rate of fine powder, and the material filling state index in the mill, and weighted fusion to generate the particle size distribution, it can accurately predict the distribution of each particle size component in the cement grinding process, and then provide a more accurate adjustment basis for the control system; 3. Achieve closed-loop intelligent regulation: Through the real-time feedback and dynamic update of the prediction model, a closed-loop regulation is formed under the feedback control of the actuator to effectively cope with various uncertain factors in the grinding process, such as changes in the mill loading force, the rotational speed of the separator, and the feeding amount, and optimize the training of the model through the loss function based on the particle size distribution difference, which can continuously improve the accuracy of the prediction model.

[0016] The above description is only an overview of the technical solution of this application. In order to be able to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features, and advantages of this application more obvious and understandable, the specific implementation manners of this application are specifically given below. Brief Description of the Drawings

[0017] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings. In the drawings: Figure 1 It is a flowchart of an intelligent control method for the particle size distribution of cement grinding shown in an exemplary embodiment of the present invention. Detailed Embodiments

[0018] The following will describe the embodiments of the present invention with reference to the drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for explaining the present invention, rather than for limiting the protection scope of the present invention.

[0019] It should be noted that the drawings provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. Therefore, only the components related to the present invention are shown in the drawings, rather than being drawn according to the number, shape, and size of the components in actual implementation. The type, number, and ratio of each component in its actual implementation can be an arbitrary change, and the component layout type may also be more complex.

[0020] In the following description, numerous specific details are set forth to provide a more thorough explanation of embodiments of the present invention. However, it will be apparent to those skilled in the art that embodiments of the present invention may be practiced without these specific details. In other instances, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the present invention.

[0021] Applied to the intelligent regulation method of particle size distribution in cement grinding, as Figure 1 shown, it includes: S1: Real-time collect the operating parameters of the grinding system, including mill current, classifier speed, feeding rate, grinding sound signal, and online particle size detection data; S2: Calculate the core characteristic parameters based on the operating parameters. The core characteristic parameters include effective crushing work index, dynamic fine powder retention rate, and grinding chamber material filling state index; S3: Input the core characteristic parameters into the particle size distribution prediction model and output the predicted particle size distribution; S4: Generate adjustment instructions based on the predicted particle size distribution through a hierarchical control strategy; S5: Output the adjustment instructions to the actuator.

[0022] The present invention is further configured such that step S2 specifically includes: The effective crushing work index is calculated by the mechanical energy obtained per unit of material within the integration period; the dynamic fine powder retention rate is determined by the ratio of the online particle size to the theoretical fine powder amount combined with the classifier speed ratio; the grinding chamber material filling state index is characterized by the normalized difference between the grinding sound signal and the reference value. The present invention is further configured such that the calculation logic of the effective crushing work index is: , is the effective crushing work index, is the current time point, is the time window length, is the current value, is the voltage value, is the motor mechanical efficiency, is the main shaft speed of the mill, is the feeding rate, and the motor mechanical efficiency has the calculation logic of: , is the maximum motor mechanical efficiency, is the optimal operating speed of the mill, is the expansion factor; the calculation logic of the dynamic fine powder retention rate is: , is the dynamic fine powder retention rate, is the proportion of particles less than measured by the online particle size analyzer, is the proportion of particles less than The proportion of fine powder should be is the rotational speed of the classifier, is the reference critical rotational speed value of the classifier, is the preset threshold; The calculation logic of the material filling state index inside the mill is: , is the material filling state index inside the mill, is the modified linear unit function, is the current grinding sound power, is the reference grinding sound power, is the structural sound response coefficient of the mill, is the current material density.

[0023] Specifically, the effective crushing work index is used to measure the mechanical energy obtained by unit material during the grinding process. The obtained mechanical energy represents the grinding efficiency of the mill. The higher the energy obtained by unit material at each moment, the higher the grinding efficiency of the mill; The motor mechanical efficiency is used to reflect the dependence of the motor mechanical efficiency on the rotational speed of the mill and reaches the highest efficiency near the optimal operating rotational speed; is used to control the sensitive width of the efficiency function to the rotational speed, with a value range of [50, 200] and the unit of rpm; The dynamic fine powder retention rate reflects the dynamic accumulation degree of fine particles in the grinding system inside the mill and is used to measure whether there is abnormal accumulation or a decrease in separation efficiency during the classification and grading of fine powder; is the preset threshold of particle size, used to characterize the boundary of fine powder, with a value range of [20, 30] and the unit of nanometer; The material filling state index inside the mill is used to dynamically evaluate the filling degree of materials inside the mill cylinder. By analyzing the change in the grinding sound power generated during the operation of the mill and combining the structural characteristics of the mill and the material density, it is inferred whether there is a non-optimal state such as too much or too little material at present; The grinding sound power is a physical index calculated after processing such as filtering and energy extraction of the grinding sound signal obtained by the sound signal acquisition device installed outside the mill. Its acquisition method belongs to the prior art and will not be elaborated here; By converting the key physical parameters of effective crushing work, dynamic fine powder retention rate, and material filling state inside the mill into mathematical expressions, it no longer depends on empirical judgment, making the understanding of the operating state of the grinding system change from qualitative to quantitative, providing a basis for intelligent prediction and control.

[0024] The present invention is further configured such that step S3 specifically includes: establishing a first distribution function term reflecting the coarse particle distribution characteristics based on the effective crushing work index; establishing a second distribution function term reflecting the fine particle enrichment characteristics based on the dynamic fine powder retention rate; establishing a third distribution function term reflecting the coarse particle inhibition characteristics based on the in-mill material filling state index; performing weighted fusion on the first, second, and third distribution function terms, and generating a predicted particle size distribution through normalization processing; and optimizing and training the parameters of the grading prediction model by using a loss function based on the distribution distance. The present invention is further configured to predict the particle size distribution of the material by using a modified Rosin-Rammler distribution function based on the effective crushing work index, and the calculation logic of the first distribution function term is: , where is the coarse particle cumulative distribution function,[[]]END]] is the particle diameter,[[]]END]] is the effective crushing work index,[[]]END]] is the median particle size,[[]]END]] is the distribution index,[[]]END]] The calculation logic of is: is the scale factor,[[]]END]] is the adjustment index,[[]]END]] The calculation logic of is: is the reference distribution index,[[]]END]] is the adjustment factor; further correct the fine particle distribution based on the dynamic fine powder retention rate, and the calculation logic of the second distribution function term is: , is the fine powder correction function,[[]]END]] is the dynamic fine powder retention rate,[[]]END]] is the fine powder characteristic particle size,[[]]END]] is the distribution standard deviation; inhibit the secondary crushing probability of the coarse particles based on the in-mill material filling state, and the calculation logic of the third distribution function term is: , is the coarse particle inhibition function,[[]]END]] is the in-mill material filling state index,[[]]END]] is the attenuation coefficient; Perform weighted fusion on the first, second, and third distribution function terms, and obtain the final predicted distribution through normalization by the Softmax function. The calculation logic of the predicted distribution is: , is the predicted particle size distribution,[[]]END]] is the th weighted function,[[]]END]] is the standard Softmax normalization function,[[]]END]] , and are the weight coefficients; the calculation logic of the loss function is: , is the loss function, is the predicted particle size distribution, is the actual particle size distribution, is the gradient coefficient, is the difference between the predicted particle size distribution and the actual particle size distribution, The calculation logic of is: , is the transfer amount allocated to and ; and are the particle diameters of the predicted distribution and the actual distribution respectively; the parameter set and of the grading prediction model includes the distribution function parameter and the weight coefficient , where , , The parameter set is expressed as: ; the update rule of the weight coefficient is: , is the weight coefficient vector at the th iteration, is the weight learning rate, is the loss function of the gradient of the weight coefficient, is the updated weight coefficient vector; Specifically, the first distribution function term is used to predict the cumulative distribution of the coarse particle part of the material particles; is used to control the position of the particle distribution; is used to control the steepness of the distribution; is used to map the effective crushing work index to the median particle size value with physical meaning through a power function, and the value range is [20, 1000]; is used to control the sensitivity of the median particle size to the change of crushing work, and the value range is [0.1, 0.6]; is used to control the baseline of the steepness of the particle distribution, and the value range is [0.5, 2.5]; is used to control the response degree of the distribution steepness to the energy consumption, and the value range is [0.1, 1]; , , and are fitting parameters. The initial fitting parameters in the first distribution function term of the grading prediction model can be obtained based on the prior art. Specifically, a certain number of particle size distribution samples of the grinding products are collected, and combined with the effective crushing work index , the nonlinear least square method is used to fit the model parameters, which is the prior art and will not be described here; the second distribution function term is used to correct the distribution of fine powder particle size segment in the cement grinding process to better reflect the retention of fine particles under actual operating conditions; Used to control the dispersion of fine powder distribution, it can be dynamically updated according to the current particle size distribution data obtained by the online particle size analyzer, for example, the particle size interval span is calculated and estimated based on the distribution morphology characteristics. , in nanometers; the third distribution function term reflects the dynamic regulation of the probability of secondary crushing of large particles based on the filling state of the material in the mill. During the grinding process, if the coarse particles are in a high filling state, they are easily protected by the cushion effect and are not easy to be effectively crushed. It is necessary to introduce a mechanism to inhibit the secondary crushing of coarse particles; It is used to control the sensitivity of the filling state to the suppression of coarse particles, and the value range is [0.001, 0.01]. The weighted distribution is normalized by the Softmax function to ensure that the final output is a legal probability distribution and the sum of all predicted values ​​is 1. , and It is used to adjust the contribution of each distribution function term to the predicted particle size distribution, and , and The sum is 1, , and The specific value is obtained by training the historical data. The weight training uses a neural network to construct a loss function based on the collected particle size distribution data and the corresponding mill operating data. By minimizing the difference between the predicted particle size distribution and the actual particle size distribution, the most suitable weight coefficient is optimized. This is the existing technology and will not be described in detail here. By integrating the three distribution function terms, the complex factors in multiple grinding processes can be comprehensively considered to improve the accuracy of particle size prediction. The loss function Used to measure the difference between the predicted and actual particle size distributions, as well as gradient constraints; Used to control the gradient constraint For the loss function The influence of, the value range is [0.0001,0.1]; weight coefficient The optimization is performed by gradient descent. The above weight coefficient update process calculates the loss function after each iteration. , and determine whether to terminate the iteration by any of the following convergence conditions: ,in is the preset convergence threshold; It is used to control the adjustment step size of the weight vector in each iteration, and the value range is [0.0001, 0.1]; accurate particle size distribution prediction can be used for feedback control to optimize the parameter settings in the grinding process and achieve more efficient grinding operations.

[0025] The present invention is further configured such that step S4 specifically includes: obtaining the current predicted particle distribution and the preset target particle distribution; based on the difference between the predicted particle size distribution and the preset target particle size distribution, constructing an adjustment function including multiple sub-control targets to obtain the adjustment amounts of the mill loading force, the classifier speed, and the feeding amount; outputting an adjustment instruction vector, and the specific manifestation form of the adjustment instruction vector is: , is the adjustment instruction vector, is the adjustment amount of the mill loading force, is the adjustment amount of the classifier speed, is the adjustment amount of the feeding amount; the present invention is further configured such that the calculation logic of the adjustment amount of the mill loading force is: , is the adjustment coefficient, is the effective crushing work index, is the predicted particle distribution, is the preset target particle size distribution; the calculation logic of the adjustment amount of the classifier speed is: , is the proportionality coefficient, is the structural similarity index, is the sign function, is the median particle size of the target particle distribution, is the median particle size of the current predicted particle size distribution; the calculation logic of the adjustment amount of the feeding amount is: , is the sensitivity adjustment coefficient, is the filling state index of the materials in the mill; specifically, the purpose of step S4 is to achieve intelligent control of the operating conditions of the mill. By comparing the difference between the current particle distribution and the target distribution, the adjustment instruction is dynamically calculated to accurately control the particle gradation in the mill; the current predicted particle distribution is predicted by the gradation prediction model, reflecting the particle size gradation under the current operating state of the mill; the preset target particle distribution is preset according to product requirements or empirical curves; obtaining the current predicted particle distribution and the preset target particle distribution is used to clarify the deviation degree between the current state and the target, providing a basis for regulation; the sub-control objectives include gradation shape optimization, overall fineness adjustment, and operation stability maintenance; the adjustment amount of the mill loading force determines the pressure in the mill and affects the crushing strength; the adjustment amount of the separator speed controls the fine powder separation efficiency and adjusts the particle retention time; the adjustment amount of the feeding rate affects the load of the mill and the grinding effect of the materials. Excessive feeding rate may cause the mill to be overloaded, while too low feeding rate will affect the grinding efficiency; is the divergence function, which is used to measure the information loss between the current predicted particle distribution and the preset target particle distribution. The larger the value, the more significant the difference; is used to control the adjustment amount of the mill loading force The adjustment range is [0.001, 0.1]; is used to determine the response speed of the separator speed to the particle size distribution difference, and the value range is [0.1, 2]; is used to control the adjustment sensitivity of the feeding rate, and the value range is [0.01, 0.05]; Through the above steps, the particle control accuracy is improved, and the energy consumption and over-grinding phenomenon during the working process are reduced.

[0026] The present invention is further configured that step S5 specifically includes: converting the mill loading force adjustment instruction into a pressure control signal and outputting it to the mill pressure execution unit; converting the separator speed adjustment instruction into a speed control signal and outputting it to the separator drive unit; converting the feeding rate adjustment instruction into a flow control signal and outputting it to the material conveying control unit; The present invention is further configured that the calculation logic of the mill loading force adjustment instruction is: , is the loading force set value, is the basic loading force, is the mill loading force adjustment amount, is the mill structure coefficient; the calculation logic of the separator speed adjustment instruction is: , is the time The target set speed of the separator at, is the separator speed adjustment amount; the calculation logic of the feeding rate adjustment instruction is: , is the time The feeding speed at, is the initial feeding speed, is the response time constant, is the adjustment amount of the feeding quantity; specifically, is the final target loading force obtained by the control system according to the adjustment amount; is the target rotational speed calculated according to the adjustment amount of the classifier rotational speed; is the target feeding speed obtained according to the adjustment amount of the feeding quantity; is used to adjust the response sensitivity of the mill, and its value range is [0.1, 2]; is used to control the response speed of the feeding quantity adjustment, and its value range is [0.1, 1]; By calculating the adjustment instructions for the mill loading force, classifier rotational speed and feeding quantity, the entire cement grinding process can be effectively controlled, and the grinding efficiency and cement particle size distribution can be optimized.

[0027] The present invention is further configured to further include dynamically adjusting the adjustment instruction based on the feedback information of the execution structure: The current change rate of the classifier current is monitored in real time , when is greater than the preset threshold , a loading force compensation is triggered, and the loading force compensation mechanism is: , is the adjusted mill loading force adjustment amount, is the adjustment step coefficient; The actual particle size distribution is obtained through an on-line particle size analyzer , and the parameters of the grading prediction model are corrected every T minutes. The modification logic of the grading prediction model parameters is: , is the modified grading prediction model parameter, is the difference between the predicted particle size distribution and the actual particle size distribution, is the gradient operator, is the learning rate; When the number of feeding quantity adjustments , the reference grinding sound power is adjusted, is the preset adjustment number threshold within the period, and the adjustment logic of the reference grinding sound power is: , is the adjusted reference grinding sound power, is the grinding sound power adjustment amount, and the calculation logic of the grinding sound power adjustment amount is: , is the average grinding sound power in the current period, is the smoothing coefficient; If , then this adjustment is rejected, and a fallback mechanism is adopted to keep the original unchanged, where is the long-term average value of the grinding sound power, is the standard deviation of the grinding sound power, and is the periodic statistic based on the grinding sound signal; specifically, the change rate of the current of the classifier is monitored in real time , The value reflects the dynamic change of the classifier current. When exceeds the preset threshold , it indicates that there is load imbalance or system abnormality in the classifier, and the loading force needs to be adjusted; is the sign function, which represents the direction of the current change rate. When the current change rate is a positive change rate, the sign function is 1, indicating an increase in the loading force. When the current change rate is a negative change rate, the sign function is -1, indicating a decrease in the loading force; is used to control the magnitude of the compensation amount, and its value range is [0.1, 1]; every T minutes, the actual particle size distribution is obtained through an on-line particle size analyzer , and based on the EMD error between it and the predicted particle size distribution , the parameters of the grading prediction model are corrected by using the gradient descent method according to the learning rate ; T is the preset time interval parameter, which is used to control the update frequency of the grading prediction model parameters, and its value range is [3, 30], and the unit is minutes; in the parameter modification logic of the grading prediction model is the gradient operator, which represents the partial derivative vector of the distribution function parameter in the parameter set , The specific manifestation form of is: , then The calculation logic of is: is used to control the step size of each correction, and its value range is [0.00001, 0.1]; whenever the adjustment times of the feeding amount within a unit cycle reach or exceed the preset threshold , it indicates that the material supply is adjusted frequently, and there is an offset in the material characteristics or operating state. At this time, the reference grinding sound power needs to be re-evaluated and adjusted appropriately; the adjustment logic of the reference grinding sound power is essentially an exponentially weighted average update to avoid drastic jumps; if the updated value deviates from the long-term mean by more than twice the standard deviation, that is, it satisfies: , then it is considered that the update is abnormal, and this update is rejected, and the original is still used; it is used to control the update amplitude of the reference grinding sound power, and its value range is [0.05, 0.3]; based on the dynamic adjustment mechanism of the feedback information, when an abnormality occurs, such as the change rate of the classifier current exceeding the threshold, the loading force compensation or the reference grinding sound power adjustment can be carried out in time to ensure that the grinding system can still maintain a good operating state under complex operating conditions and improve the robustness of the grinding system.​​

[0028] The above embodiments can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or a data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0029] It should be understood that the term "and / or" in this document is merely a description of the association relationship between associated objects, indicating that three relationships can exist. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone. Here, A and B can be singular or plural. In addition, the character " / " in this document generally represents an "or" relationship between the associated objects before and after, but it may also represent an "and / or" relationship, which can be specifically understood with reference to the context.

[0030] In this application, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or a similar expression refers to any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0031] It should be understood that in various embodiments of the present application, the sequence numbers of the above processes do not mean the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0032] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. Professional technicians can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of this application.

[0033] Those skilled in the art can clearly understand that for the convenience and conciseness of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments, and will not be elaborated herein.

[0034] In several embodiments provided in this application, it should be understood that the disclosed systems can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of devices or units can be electrical, mechanical, or other forms.

[0035] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they can be located in one place, or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0036] In addition, the functional units in each embodiment of this application can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit.

[0037] When the above-mentioned functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of this application. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical discs that can store program codes.

[0038] As described above, the above is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered by the protection scope of this application. Therefore, the protection scope of this application should be subject to the protection scope of the claims.

Claims

1. An intelligent regulation method for particle size distribution applied to cement grinding, characterized in that, Including: S1: Real-time collect the operating parameters of the grinding system, including mill current, classifier speed, feeding rate, grinding sound signal and on-line particle size detection data; S2: Calculate the core characteristic parameters based on the operating parameters. The core characteristic parameters include effective crushing work index, dynamic fine powder retention rate and in-mill material filling state index; S3: Input the core characteristic parameters into the grading prediction model and output the predicted particle size distribution; S4: Generate adjustment instructions through a hierarchical control strategy based on the predicted particle size distribution; S5: Output the adjustment instructions to the actuator.

2. The intelligent control method for particle size distribution applied to cement grinding according to claim 1, characterized in that Step S2 specifically includes: The effective crushing work index is calculated by the mechanical energy obtained per unit of material within the integration period; The dynamic fine powder retention rate is determined by the ratio of the on-line particle size to the theoretical fine powder amount combined with the classifier speed ratio; The in-mill material filling state index is characterized by the normalized difference between the grinding sound signal and the reference value.

3. The intelligent regulation method for particle size distribution applied to cement grinding according to claim 2, characterized in that, The calculation logic of the effective work index for crushing is as follows: , is the effective work index for crushing, is the current time point, is the time window length, is the current value of the electric current, is the current value of the voltage, is the mechanical efficiency of the motor, is the rotational speed of the main shaft of the mill, is the feeding rate, and the calculation logic of the mechanical efficiency of the motor is as follows: , is the maximum mechanical efficiency of the motor, is the optimal operating rotational speed of the mill, is the expansion factor; The calculation logic of the dynamic fine powder retention rate is as follows: , is the dynamic fine powder retention rate, is the proportion of particles smaller than measured by the on-line particle size analyzer, is the proportion of fine powder that should be smaller than under the theoretical equilibrium state, is the rotational speed of the classifier, is the reference critical rotational speed value of the classifier, is the preset threshold value; The calculation logic of the material filling state index in the mill is as follows: , is the material filling state index in the mill, is the modified linear unit function, is the current mill sound power, is the reference mill sound power, is the mill structure sound response coefficient, is the current material density.

4. The intelligent regulation method for particle size distribution applied to cement grinding according to claim 1, wherein Step S3 specifically includes: Establish a first distribution function term reflecting the coarse particle distribution characteristics based on the effective crushing work index; Establish a second distribution function term reflecting the fine particle enrichment characteristics based on the dynamic fine powder retention rate; Establish a third distribution function term reflecting the coarse particle suppression characteristics based on the in-mill material filling state index; Perform weighted fusion on the first, second and third distribution function terms, and generate the predicted particle size distribution through normalization processing; Optimize and train the parameters of the grading prediction model using a loss function based on distribution distance.

5. The intelligent control method for particle size distribution applied to cement grinding according to claim 4, characterized in that, Predict the particle size distribution of materials based on the effective work index of fragmentation using a modified Rosin-Rammler distribution function. The calculation logic of the first distribution function term is as follows: , where is the cumulative distribution function of coarse particles, is the particle diameter, is the effective work index of fragmentation, is the median particle size, is the distribution index, The calculation logic of is as follows: is the scale factor, is the adjustment index, The calculation logic of is as follows: is the reference distribution index, is the adjustment factor; Further correct the fine particle distribution based on the dynamic fine powder retention rate, and the calculation logic of the second distribution function term is as follows: , is the fine powder correction function, is the dynamic fine powder retention rate, is the characteristic particle size of the fine powder, is the distribution standard deviation; Inhibit the secondary crushing probability of coarse particles based on the material filling state in the mill. The calculation logic of the third distribution function term is as follows: , is the coarse particle inhibition function, is the material filling state index in the mill, is the attenuation coefficient; The first, second, and third distribution function terms will be weighted and fused, and the final prediction distribution will be obtained through normalization by the Softmax function. The calculation logic of the prediction distribution is as follows: , is the predicted particle size distribution, is the th term weighting function, is the standard Softmax normalization function, , and are the weight coefficients; The calculation logic of the loss function is as follows: , is the loss function, is the predicted particle size distribution, is the actual particle size distribution, is the gradient coefficient, is the difference between the predicted particle size distribution and the actual particle size distribution, The calculation logic of is as follows: is the transfer amount allocated between and , and are the particle diameters of the predicted distribution and the actual distribution, respectively; Gradation prediction model parameter set including distribution function parameters and weight coefficients , where , , the parameter set is in the form of: ; Weight coefficient The update rule is as follows: , is the weight coefficient vector for the th iteration, is the weight learning rate, is the loss function is the gradient of the loss function with respect to the weight coefficient, is the updated weight coefficient vector.

6. The intelligent control method for particle size distribution applied to cement grinding according to claim 1, wherein Step S4 specifically includes: Obtain the current predicted particle distribution and the preset target particle distribution; Based on the difference between the predicted particle size distribution and the preset target particle size distribution, construct an adjustment function including multiple sub-control targets to obtain the adjustment amounts of the mill loading force, classifier speed and feeding rate; Output adjustment instruction vector, and the specific manifestation form of the adjustment instruction vector is as follows: , is the adjustment instruction vector, is the adjustment amount of the mill loading force, is the adjustment amount of the classifier speed, is the adjustment amount of the feeding rate.

7. The intelligent control method for particle size distribution applied to cement grinding according to claim 6, characterized in that, The adjustment amount of the grinding mill loading force The calculation logic is as follows: , is the adjustment coefficient, is the effective crushing work index, is the predicted particle size distribution, is the preset target particle size distribution; The regulating amount of the classifier rotational speed has the following calculation logic: , is the proportionality coefficient, is the structural similarity index, is the sign function, is the median particle size of the target particle distribution, is the median particle size of the current predicted particle size distribution; The adjusted amount of the feeding quantity The calculation logic is as follows: , is the sensitivity adjustment coefficient, is the index of the material filling state in the mill.

8. The intelligent control method for particle size distribution applied to cement grinding according to claim 1, characterized in that Step S5 specifically includes: Convert the mill loading force adjustment instruction into a pressure control signal and output it to the mill pressure execution unit; Convert the classifier speed adjustment instruction into a speed control signal and output it to the classifier drive unit; Convert the feeding rate adjustment instruction into a flow control signal and output it to the material conveying control unit.

9. The intelligent control method for particle size distribution applied to cement grinding according to claim 8, wherein, The calculation logic of the mill loading force adjustment instruction is as follows: , is the set value of the loading force, is the basic loading force, is the adjustment amount of the mill loading force, is the mill structure coefficient; The calculation logic of the classifier speed adjustment command is as follows: , is the target set speed of the classifier at time , and is the classifier speed adjustment amount; The calculation logic of the feed rate adjustment command is as follows: , is the feed rate at time , is the initial feed rate, is the response time constant, is the feed rate adjustment amount.

10. The intelligent control method for particle size distribution applied to cement grinding according to claim 9, characterized in that, It also includes dynamically adjusting the adjustment instructions based on the feedback information of the execution structure: Real-time monitoring of the current change rate of the separator When is greater than the preset threshold , trigger the loading force compensation, and the loading force compensation mechanism is as follows: , is the adjusted amount of the mill loading force after compensation, is the adjustment step coefficient; Obtain the actual particle size distribution through an online particle size analyzer , correct the parameters of the gradation prediction model every T minutes, and the parameter modification logic of the gradation prediction model is as follows: , is the parameter of the modified gradation prediction model, is the difference between the predicted particle size distribution and the actual particle size distribution, is the gradient operator, is the learning rate; When the number of times of feed rate adjustment triggers the adjustment of the reference grinding sound power for the preset adjustment number threshold within a period, the adjustment logic of the reference grinding sound power is: , is the adjusted reference grinding sound power, is the grinding sound power adjustment amount, and the calculation logic of the grinding sound power adjustment amount is: , is the average grinding sound power in the current period, is the smoothing coefficient; If , reject this adjustment, adopt a rollback mechanism, and keep the original unchanged, where is the long-term average value of the grinding sound power, is the standard deviation of the grinding sound power, and is the periodic statistic based on the grinding sound signal.

Images