Intelligent adjusting and grinding system for grain composition in cement production

Abstract

The invention discloses a cement production grain composition intelligent adjustment grinding system, relates to an image optimization technology, and is used for solving the problem that grinding parameters cannot be dynamically adjusted due to the fact that the replacement risk cannot be warned in time in the raw material replacement process. A density detection channel is opened to detect the density of the raw material when the raw material is subjected to grinding detection; the anti-wear characteristic of the raw material is analyzed in combination with the temperature of the raw material; historical grinding parameters jointly influenced by the grinding power and pressure are called; the current grinding power and pressure are detected and corrected; and generating grinding parameter characteristics, comprehensively analyzing the rotating speed adjustment proportion with the wear-resistant characteristics of the raw materials, and optimizing the existing cement grinding rotating speed.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent control, and more specifically, to an intelligent grinding system for adjusting the particle size distribution of cement production. Background Art

[0002] As an essential basic material in construction engineering, the performance indicators of cement are affected by various factors, among which the particle size distribution has a significant impact on the strength, workability, and durability of cement. In the prior art, the grinding process of cement production mainly relies on traditional ball mills or roller presses to achieve the refinement of raw materials through mechanical means. However, due to the wide variety of raw materials and significant differences in physical properties (such as density, temperature, hardness, etc.), problems such as particle size distribution deviation, unstable grinding efficiency, and increased equipment wear are likely to occur during continuous production; The prior art has the following deficiencies: Currently, the existing cement grinding systems generally lack the ability to evaluate the real-time performance of raw materials. Especially during the process of raw material replacement, they cannot timely warn of the replacement risk, resulting in the inability to dynamically adjust the grinding parameters, affecting the consistency of product quality, lacking a data-driven intelligent adjustment mechanism. In the case of different abrasion resistances of raw materials, it is easy to cause an increase in grinding energy consumption or over-grinding, reducing the overall production efficiency. Therefore, an intelligent grinding system for adjusting the particle size distribution of cement production is proposed. Summary of the Invention

[0003] To overcome the above-mentioned defects of the prior art, the embodiments of the present invention provide an intelligent grinding system for adjusting the particle size distribution of cement production. By constructing an intelligent control architecture integrating raw material risk assessment, pre-grinding performance detection, and parameter adaptive adjustment, dynamically analyze its abrasion resistance characteristics based on physical parameters such as raw material density and temperature, and adaptively correct the grinding process parameters in combination with historical grinding data to solve the problems raised in the above background art.

[0004] To achieve the above purpose, the present invention provides the following technical solutions: An intelligent grinding system for adjusting the particle size distribution of cement production, comprising a data acquisition module, a raw material evaluation module, a grinding detection module, and a parameter adjustment module; The data acquisition module is used to retrieve the raw material storage information and stored historical data in the cement raw material warehouse, and transmit the raw material storage information and stored historical data to the raw material evaluation module; After receiving the raw material storage information and stored historical data, the raw material evaluation module analyzes the replacement risk of cement raw materials, sets different decision values according to the replacement risk of cement raw materials, and transmits the decision values to the grinding detection module; The grinding detection module determines whether to perform pre-grinding detection on the raw materials according to the decision value. When performing grinding detection on the raw materials, it activates the density detection channel, obtains the vibration frequency of the raw materials in the density detection channel through the connected vibration sensor, calculates the density of the raw materials, detects the temperature of the raw materials in the density detection channel, analyzes the abrasion resistance characteristics of the raw materials in combination with the raw material density, and transmits it to the parameter adjustment module; The parameter adjustment module calls the historical grinding parameters, detects the working power and grinding pressure of the grinding machine, corrects the historical grinding parameters, generates the grinding parameter characteristics, and adjusts the cement grinding speed in combination with the abrasion resistance characteristics of the raw materials.

[0005] In a preferred embodiment, the data acquisition module acquires the storage quantity and storage historical data of the raw materials. The storage historical data includes the raw material batch replacement time and the historical average replacement cycle interval in the past period of time; The ratio of the storage quantity of the raw materials to the preset safety inventory threshold is used as the inventory safety ratio. The time difference between the current time and the time of the last replacement of the raw materials is used as the cycle interval of this replacement. The ratio of the cycle interval of this replacement to the historical average replacement cycle interval is used as the replacement cycle ratio.

[0006] In a preferred embodiment, the raw material evaluation module comprehensively calculates the cement raw material replacement risk based on the inventory safety ratio and the replacement cycle ratio: The sum of the inventory safety ratio and the replacement cycle ratio is used as the logistic regression parameter of the raw materials, and a logistic regression model is constructed with the logistic regression parameter of the raw materials: , where z is the logistic regression parameter of the raw materials, e is the natural base, and L is the raw material replacement risk score.

[0007] In a preferred embodiment, the decision value is set for each raw material according to the raw material replacement risk score as follows: If the raw material replacement risk score is less than or equal to the preset risk level threshold, it is determined that the corresponding raw material is classified as a low risk level, and the decision value is set to 0; otherwise, the corresponding raw material is classified as a high risk level, and the decision value is set to 1; When the decision value of the raw material is 0, no pre-grinding detection is performed. When the decision value of the raw material is 1, pre-grinding detection is performed.

[0008] In a preferred embodiment, during the transportation of the raw materials, leakage occurs through the edge opening of the conveyor belt and enters the density detection channel. The vibration frequency of the raw materials is detected and the density of the raw materials is calculated. The calculation formula is: , where f is the vibration frequency of the raw materials, m is the mass of the raw materials, k is the elastic constant, and ρ is the density of the raw materials; The abrasion resistance characteristics of the raw materials are calculated by the K-means clustering analysis method using the raw material temperature and the raw material density. The specific steps are as follows: Step 1: Set the acquisition time. During the acquisition time, set multiple time points to collect the raw material temperature and raw material density, and perform normalization processing. Combine the normalized results into a feature vector, which is used as the input data; Step 2: Perform iterative calculation on the input data to generate the target discrimination cluster K; Step 3: Combine the input data in the target discrimination cluster K with a preset weighted threshold and perform weighted summation to obtain the raw material anti-wear feature score.

[0009] In a preferred embodiment, the specific steps for iterative calculation to generate the discrimination cluster are as follows: Mark the feature vector as , randomly set a clustering number K value, divide the feature vector into K discrimination clusters, randomly select K data points as the initial clustering centers, calculate the distances between the data points and each initial clustering center, and assign them to the nearest initial clustering center. Calculate their distances through the Euclidean distance, and the calculation formula is: , where and are the coordinates of the initial clustering center, is the i-th raw material temperature, is the i-th raw material density, is the i-th feature vector, is the K-th initial clustering center; after assigning the feature vector to the nearest discrimination cluster, recalculate the clustering center of each discrimination cluster, and the calculation formula is: , where is the number of feature vectors in discrimination cluster k, is the i-th raw material temperature in discrimination cluster k, is the i-th raw material density in discrimination cluster k, is the clustering center. Repeat the above operations until the clustering center reaches the preset maximum number of iterations and then stop to generate the target discrimination cluster K; The clustering center is the temperature mean and density mean of all raw material data sets within each discrimination cluster.

[0010] In a preferred embodiment, the historical grinding parameter in the parameter adjustment module is the historical grinding efficiency , and the historical grinding efficiency is obtained by taking the average value of the grinding efficiency of the grinding machine within the historical time; Collect the grinding power and grinding pressure within a past period of time and calculate their respective means as the historical grinding working power and historical grinding pressure, which are respectively marked as and ; Mark the difference between the real-time grinding working power and the historical grinding working power as the grinding power difference, which is marked as , the difference between the real-time grinding working pressure and the historical grinding pressure is used as the grinding pressure difference and marked as ; Generate the grinding parameter features by synthesizing the grinding power difference, the grinding pressure difference, and the historical grinding efficiency as follows: Obtain the grinding parameter features of the current state of the grinder through proportional adjustment. The calculation formula is: , where , are adjustment coefficients, is the grinding parameter feature.

[0011] In a preferred embodiment, generate the grinding parameter features by synthesizing the grinding power difference, the grinding pressure difference, and the historical grinding efficiency as follows: Calculate the corrected grinding efficiency by synthesizing the grinding parameter features and the wear resistance feature of the raw material: , where is a preset adjustment factor, is the corrected grinding efficiency; Calculate the rotational speed adjustment ratio according to the wear resistance feature score of the raw material: , where k is the rotational speed adjustment ratio.

[0012] In a preferred embodiment, adjust the current rotational speed of the pulverizer according to the rotational speed adjustment ratio. The calculation formula is: , where is the current rotational speed, is the adjusted rotational speed.

[0013] Technical effects and advantages of the intelligent adjustment pulverizing system for particle size distribution in cement production of the present invention: The present invention obtains the raw material storage information and the historical replacement batch data in the cement raw material warehouse, analyzes the raw material replacement risk to set the decision value, determines whether to enable the pre-grinding detection, opens the density detection channel to detect the raw material density when performing the grinding detection on the raw material, analyzes the wear resistance characteristics of the raw material in combination with the raw material temperature, retrieves the historical grinding parameters affected by both the grinding power and the pressure, detects the current grinding power and pressure and makes corrections, generates the grinding parameter features, comprehensively analyzes the rotational speed adjustment ratio with the wear resistance feature of the raw material, and optimizes the existing rotational speed of the cement grinding. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 is the structure diagram of the intelligent adjustment pulverizing system for particle size distribution in cement production of the present invention.

[0015] Figure 2 is the structure diagram of the density detection channel of the grinding detection module.

[0016] Figure 3 is the working principle diagram of the parameter adjustment module. Detailed implementation manners

[0017] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0018] The present invention obtains the raw material storage information and historical replacement batch data in the cement raw material warehouse, analyzes the raw material replacement risk to set a decision value, determines whether to enable pre-grinding detection, opens a density detection channel to detect the raw material density when performing grinding detection on the raw material, analyzes the abrasion resistance characteristics of the raw material in combination with the raw material temperature, retrieves the historical grinding parameters affected by both the grinding power and pressure, detects the current grinding power and pressure and corrects them, generates grinding parameter characteristics, comprehensively analyzes the rotation speed adjustment ratio with the raw material abrasion resistance characteristics, and optimizes the existing cement grinding rotation speed, realizing efficient and precise control, improving the intelligent level of raw material processing and the grinding efficiency.

[0019] Embodiment 1, an intelligent adjustment grinding system for cement production particle size distribution, as Figure 1 shown, includes a data acquisition module, a raw material evaluation module, a grinding detection module, and a parameter adjustment module, and the modules are electrically connected to each other; The functions of each module are as follows: The data acquisition module is used to retrieve the raw material storage information (raw material storage quantity) in the cement raw material warehouse and the stored historical data (raw material replacement batches in the past time), and transmit the raw material storage information and the stored historical data to the raw material evaluation module; After receiving the raw material storage information and the stored historical data, the raw material evaluation module analyzes the cement raw material replacement risk, sets different decision values according to the cement raw material replacement risk, and transmits the decision values to the grinding detection module; The grinding detection module determines whether to perform pre-grinding detection on the raw material according to the decision value. When performing grinding detection on the raw material, it opens a density detection channel (an opening at the edge of the conveyor belt to let each wave of raw materials leak in for detection), obtains the raw material vibration frequency in the density detection channel through the connected vibration sensor and calculates the raw material density (calculating the density through vibration), detects the raw material temperature in the density detection channel, analyzes the raw material abrasion resistance characteristics in combination with the raw material density, and transmits them to the parameter adjustment module; The parameter adjustment module calls historical grinding parameters (find a parameter affected by both grinding power and grinding pressure), detects the working power and grinding pressure of the grinder, corrects the historical grinding parameters, generates grinding parameter features, and adjusts the cement grinding speed in combination with the wear resistance features of the raw materials (calculate a speed adjustment ratio by combining the grinding parameter features and the wear resistance features of the raw materials, and then multiply the existing speed by the speed adjustment ratio to obtain the adjusted speed).

[0020] The specific implementation is as follows: The data acquisition module obtains the storage quantity and storage historical data of the raw materials in real time by connecting to the inventory management system of the cement raw material warehouse, including the raw material batch replacement time and the historical average replacement cycle interval in the past period. The data acquisition module transmits the collected storage quantity and storage historical data of the raw materials to the raw material evaluation module through the data interface.

[0021] The data acquisition module collects the storage quantity and storage historical data of the current raw materials in the cement raw material warehouse, provides basic data for the raw material evaluation module, is a prerequisite for raw material risk judgment, and affects the reliability of subsequent evaluation and decision-making.

[0022] It should be noted that the inventory management system of the cement raw material warehouse is a technology to ensure the timely, accurate supply of raw materials and effective inventory control during the cement production process. It not only comprehensively controls the incoming and outgoing, storage, and batch management of raw materials, but also can monitor information such as inventory levels and raw material quality in real time, providing support for production decision-making and material scheduling.

[0023] The raw material evaluation module receives the storage quantity and storage historical data of the raw materials from the data acquisition module, and comprehensively analyzes the risk of cement raw material replacement. The specific steps are as follows: Take the ratio of the storage quantity of the raw materials to the safety inventory threshold as the inventory safety ratio. If the inventory safety ratio is less than 1, it indicates a risk of raw material shortage. If the inventory safety ratio is greater than or equal to 1, it indicates sufficient inventory; take the difference between the current time and the time of the last replacement of the raw materials as the cycle interval of this replacement, and take the ratio of the cycle interval of this replacement to the historical average replacement cycle interval as the replacement cycle ratio. If the replacement cycle ratio is greater than 1, it indicates that the current raw materials have exceeded the normal replacement cycle. If the replacement cycle ratio is less than 1, it indicates that the raw material replacement is too frequent.

[0024] Sum the inventory safety ratio and the replacement cycle ratio as the logistic regression parameter of the raw materials, and construct a logistic regression model with the logistic regression parameter of the raw materials: , where z is the logistic regression parameter of the raw materials, e is the natural logarithm base, L is the calculation result of the logistic regression model corresponding to the raw materials, and take the calculation result of the logistic regression model of the raw materials as the raw material replacement risk score; The raw material replacement risk score is compared with the preset risk level threshold, and the raw materials are divided into three levels: low risk, medium risk, and high risk. For example, if the preset risk level threshold is 0.5 and the raw material replacement risk score is 0.3, then this raw material belongs to the low risk level.

[0025] If the raw material replacement risk score is less than or equal to the risk level threshold, then this raw material is classified as a low risk level; if the raw material replacement risk score is greater than the risk level threshold, then this raw material is classified as a high risk level.

[0026] Different decision values are set according to the risk level: if the raw material is classified as a low risk level, the decision value is set to 0; if the raw material is classified as a high risk level, the decision value is set to 1.

[0027] It should be noted that the risk level threshold is dynamically adjusted according to different production requirements and actual situations and is set by professionals, which will not be elaborated here; The raw material evaluation module comprehensively obtains the cement raw material replacement risk through the storage quantity and storage historical data of the raw materials, judges whether pre-grinding detection is needed, effectively reduces the system fluctuations and equipment impacts caused by raw material differences, improves the stability of the grinding process, ensures the consistency and controllability of the quality of cement products, sets different decision values through the raw material replacement risk, and transmits the decision value to the grinding detection module to judge whether to conduct pre-grinding detection on the raw materials.

[0028] The grinding detection module receives the decision value in the raw material evaluation module to decide whether pre-grinding detection is needed. When the decision value is 0, it is judged that pre-grinding detection is not carried out; when the decision value is 1, it is judged that pre-grinding detection is carried out.

[0029] During the transportation of the raw materials, slight leakage occurs through the edge opening of the conveyor belt and enters the density detection channel. The vibration sensor connected in the density detection channel detects the vibration frequency of the raw materials, and calculates the density of the raw materials through the vibration frequency of the raw materials. The calculation formula is: , where f is the vibration frequency of the raw materials, m is the mass of the raw materials, k is the elastic constant, and ρ is the density of the raw materials.

[0030] The change in the temperature of the raw materials affects the chemical reaction rate of the cement, and thus affects the grinding efficiency. In the density detection channel, the temperature of the raw materials is monitored and collected in real time through a temperature sensor, and the abrasion resistance characteristics of the raw materials, that is, the wear resistance of the raw materials during the grinding process, are comprehensively analyzed in combination with the density of the raw materials. The abrasion resistance characteristics of the raw materials are calculated through the K-means clustering analysis method. The specific steps are as follows: Step 1: Set the collection time. During the collection time, set multiple time points to collect the temperature and density of the raw materials and perform normalization processing. Combine the normalized results into a feature vector, which is used as the input data for the K-means clustering analysis of the abrasion resistance characteristics of the raw materials.

[0031] Step 2: Mark the feature vector as , randomly set a clustering number K value, divide the feature vectors into K distinct clusters, randomly select K data points as the initial cluster centers, calculate the distances between the data points and each initial cluster center, and assign them to the nearest initial cluster center. Calculate the distance through the Euclidean distance, and the calculation formula is: , where and are the coordinates of the initial cluster center, is the temperature of the i-th raw material, is the density of the i-th raw material, is the i-th feature vector, is the K-th initial cluster center; after the feature vectors are assigned to the nearest distinct clusters, recalculate the cluster centers of each distinct cluster, and the calculation formula is: , where is the number of feature vectors in distinct cluster k, is the temperature of the i-th raw material in distinct cluster k, is the density of the i-th raw material in distinct cluster k, is the cluster center. Repeat the above operations until the cluster center reaches the preset maximum number of iterations and then stop to generate the target distinct cluster K.

[0032] Step 3: Combine the input data in the target distinct cluster K with a preset weighted threshold and perform weighted summation to obtain the raw material anti-wear feature score.

[0033] If the raw material anti-wear feature of the cluster is relatively high, then the raw materials in this cluster require relatively high grinding pressure and rotational speed to be effectively pulverized. If the raw material anti-wear feature of the cluster is relatively low, then the raw materials in this cluster require relatively low grinding pressure and rotational speed to be effectively pulverized, avoiding excessive grinding and energy consumption waste.

[0034] It should be noted that a vibration sensor is a sensor used to measure the vibration characteristics of an object in a specific direction. It can sense the vibration or oscillation generated by the action of an external force and convert these physical changes into electrical signals for analysis; the k value in the raw material density calculation formula is calculated based on the physical model of the vibration sensor and calibrated in combination with actual data. For example, apply a known force F to the vibration sensor and measure the resulting displacement , then the k value is , which is set by professionals and will not be elaborated here; a temperature sensor is a device used to detect and measure temperature changes, converting temperature changes into electrical signals or other recognizable signals; the K-means clustering method is used to divide a dataset into several clusters, which can effectively divide raw materials into multiple clusters according to their density and temperature characteristics. Each cluster represents a group with similar anti-wear characteristics of raw materials. For example, cluster k1 contains raw materials with lower density and lower temperature, and cluster 1 contains raw materials with higher density and higher temperature.

[0035] K-means clustering analysis can group raw materials with similar anti-wear characteristics into the same cluster, analyze the anti-wear characteristics of raw materials in each cluster, improve the system response speed, enhance the regulation efficiency, provide a decision-making basis for the subsequent parameter adjustment module, optimize the working state of the grinder in real time according to the anti-wear characteristics of raw materials, reduce ineffective energy consumption, and ensure that the cement particle size distribution meets the standards.

[0036] The parameter adjustment module dynamically corrects by real-time detecting the working power and pressure of the grinder, combines the historical grinding efficiency, generates the characteristics of the current grinding parameters, calculates the rotation speed adjustment ratio based on the anti-wear characteristics score of raw materials, and dynamically optimizes the grinding rotation speed by integrating the grinding parameter characteristics and the anti-wear characteristics of raw materials.

[0037] The grinding efficiency, as a core parameter, reflects the overall efficiency of the grinding process. It is automatically obtained by the production management system according to the calculation formula of the output per unit energy consumption, and the corresponding average grinding efficiency is calculated within the historical time as the historical grinding working efficiency; the grinding working power and grinding pressure are real-time monitored through a power sensor and a pressure sensor, and the historical grinding working power and pressure are represented by the average value of the data collected by the sensor within a period of time; Calculate the difference between the real-time grinding working power and the historical grinding working power. The calculation formula is: , where, is the real-time grinding working power, is the historical grinding working power, is the grinding power difference; calculate the difference between the real-time grinding pressure and the historical grinding pressure. The calculation formula is: , where, is the real-time grinding pressure, is the historical grinding pressure, is the grinding pressure difference; Obtain the corrected efficiency of the current state through proportional adjustment. The calculation formula is: , where, , are adjustment coefficients, is the historical grinding efficiency, is the grinding parameter characteristic.

[0038] The grinding efficiency is adjusted by comprehensively analyzing the raw material anti-wear characteristics and the correction efficiency. The calculation formula is as follows: , where is a preset adjustment factor, is the corrected grinding efficiency; It should be noted that the value of the adjustment factor is set by professionals based on the clustering analysis results.

[0039] The rotation speed adjustment ratio is analyzed by comparing the corrected grinding efficiency with the historical grinding efficiency. The calculation formula is as follows: , where k is the rotation speed adjustment ratio. According to the rotation speed adjustment ratio, the rotation speed of the existing pulverizer is adjusted. The calculation formula is as follows: , where is the current rotation speed, is the adjusted rotation speed.

[0040] Through real-time detection and parameter correction, the system can quickly respond to changes in raw material characteristics and equipment status fluctuations, dynamically optimize the grinding parameters, and the dynamic adjustment mechanism directly acts on the refinement process of cement particles. For example, when the raw material anti-wear characteristics are high, the system automatically increases the rotation speed and pressure to ensure that the high-hardness raw materials are fully crushed and avoid excessive proportion of coarse particles caused by insufficient grinding. On the contrary, for raw materials with low anti-wear characteristics, the system appropriately reduces the rotation speed and pressure to reduce unnecessary energy consumption and equipment wear.

[0041] It should be noted that the production management system is responsible for uniformly managing the raw material warehousing information, recording the historical track of raw material replacement, and various operating parameters during the grinding process, such as grinding power, pressure, rotation speed, output, and energy consumption, etc.; the power sensor is an intelligent detection device used to detect the power output level of the equipment, communicates with the parameter adjustment module in real time, has a high-frequency sampling rate and high-precision output ability, can convert the instantaneous power into a standard signal for the parameter adjustment module to call; the pressure sensor is used to detect the vertical or radial pressure applied to the raw material in real time, can convert the pressure signal into a linear electrical signal and transmit it to the parameter adjustment module for generating the real-time grinding pressure.

[0042] 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.

[0043] Those of ordinary skill in the art can realize that the modules 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 of the technical solution and the invention constraints. 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.

[0044] In addition, in each embodiment of this application, the various functional modules can be integrated into one processing module, or each module can exist physically alone, or two or more modules can be integrated into one module.

[0045] As mentioned 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 in 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.

[0046] Finally: The above is only the preferred embodiment of the present invention and is not used to limit the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent grinding system for adjusting the particle size distribution in cement production, characterized in that, It includes a data acquisition module, a raw material evaluation module, a grinding detection module, and a parameter adjustment module; The data acquisition module is used to retrieve the raw material storage information and storage historical data in the cement raw material warehouse, and transmit the raw material storage information and storage historical data to the raw material evaluation module; After receiving the raw material storage information and storage historical data, the raw material evaluation module analyzes the cement raw material replacement risk, sets different decision values according to the cement raw material replacement risk, and transmits the decision values to the grinding detection module; The grinding detection module determines whether to perform pre-grinding detection on the raw material according to the decision value. When performing grinding detection on the raw material, it turns on the density detection channel, obtains the vibration frequency of the raw material in the density detection channel through the connected vibration sensor and calculates the raw material density, detects the temperature of the raw material in the density detection channel, analyzes the abrasion resistance characteristics of the raw material in combination with the raw material density, and transmits it to the parameter adjustment module; The parameter adjustment module calls the historical grinding parameters, detects the working power and grinding pressure of the grinding machine, corrects the historical grinding parameters and generates grinding parameter characteristics, and adjusts the cement grinding speed in combination with the abrasion resistance characteristics of the raw material.

2. The intelligent grinding system for adjusting the particle size distribution in cement production according to claim 1, characterized in that: The data acquisition module collects the storage quantity and storage historical data of the raw material. The storage historical data includes the raw material batch replacement time and the historical average replacement cycle interval in the past period of time; The ratio of the storage quantity of the raw material to the preset safety inventory threshold is used as the inventory safety ratio, the time difference between the current time and the last replacement time of the raw material is used as the replacement cycle interval for this time, and the ratio of the replacement cycle interval for this time to the historical average replacement cycle interval is used as the replacement cycle ratio.

3. The intelligent grinding system for adjusting the particle size distribution in cement production according to claim 2, characterized in that: The raw material evaluation module comprehensively calculates the cement raw material replacement risk based on the inventory safety ratio and the replacement cycle ratio: Sum the inventory safety ratio and the replacement cycle ratio as the logistic regression parameter of the raw material, and construct a logistic regression model with the logistic regression parameter of the raw material: , where z is the logistic regression parameter of the raw material, e is the natural base, and L is the raw material replacement risk score.

4. The intelligent grinding system for adjusting the particle size distribution in cement production according to claim 3, characterized in that: The decision values are set for each raw material according to the raw material replacement risk score as follows: If the raw material replacement risk score is less than or equal to the preset risk level threshold, it is determined that the corresponding raw material is classified as a low risk level, and the decision value is set to 0; otherwise, the corresponding raw material is classified as a high risk level, and the decision value is set to 1; When the decision value of the raw material is 0, pre-grinding detection is not performed. When the decision value of the raw material is 1, pre-grinding detection is performed.

5. The intelligent grinding system for adjusting the particle size distribution in cement production according to claim 4, characterized in that: During the transportation process, the raw materials leak through the edge opening of the conveyor belt and enter the density detection channel to detect the vibration frequency of the raw materials and calculate the raw material density. The calculation formula is as follows: , where f is the vibration frequency of the raw materials, m is the mass of the raw materials, k is the elastic constant, and ρ is the raw material density; The abrasion resistance characteristics of the raw material are calculated by the K-means clustering analysis method for the raw material temperature and raw material density. The specific steps are as follows: Step 1: Set the acquisition time. During the acquisition time, set multiple time points to collect the raw material temperature and raw material density and perform normalization processing. Combine the normalized results into a feature vector and use it as the input data; Step 2: Perform iterative calculation on the input data to generate the target discrimination cluster K; Step 3: Perform weighted summation on the input data in the target discrimination cluster K in combination with the preset weighted threshold to obtain the abrasion resistance characteristic score of the raw material.

6. The intelligent grinding system for adjusting the particle size distribution in cement production according to claim 1 is characterized in that: The specific steps for generating distinguishable clusters through iterative calculation are as follows: Label the feature vector as , randomly set a clustering number K value, divide the feature vectors into K distinct clusters, randomly select K data points as the initial clustering centers, calculate the distances between the data points and each initial clustering center, and assign them to the nearest initial clustering center. Calculate their distances through the Euclidean distance, and the calculation formula is: , where and are the coordinates of the initial clustering center, is the temperature of the i-th raw material, is the density of the i-th raw material, is the i-th feature vector, is the K-th initial clustering center; after allocating the feature vectors to the nearest distinct clusters, recalculate the clustering centers of each distinct cluster, and the calculation formula is: , where is the number of feature vectors in distinct cluster k, is the temperature of the i-th raw material in distinct cluster k, is the density of the i-th raw material in distinct cluster k, is the clustering center. Repeat the above operations until the clustering center reaches the preset maximum number of iterations and then stop to generate the target distinct cluster K; The cluster centers are the average temperature and average density of all raw material data sets within each distinguishable cluster.

7. The intelligent grinding system for adjusting the particle size distribution in cement production according to claim 1 is characterized in that: The historical grinding parameter in the parameter adjustment module is the historical grinding efficiency , and the historical grinding efficiency is obtained by taking the average value of the grinding efficiency of the grinding machine within the historical time; Collect the grinding power and grinding pressure over a period of time and calculate their respective means as the historical grinding working power and historical grinding pressure, respectively, marked as and ; The difference between the real-time grinding working power and the historical grinding working power is used as the grinding power difference and marked as , and the difference between the real-time grinding working pressure and the historical grinding pressure is used as the grinding pressure difference and marked as ; Generate grinding parameter features by integrating the difference in grinding power, the difference in grinding pressure, and the historical grinding efficiency as follows: The grinding parameter characteristics of the current state of the grinder are obtained through ratio adjustment, and the calculation formula is: , where , are adjustment coefficients, is the grinding parameter characteristic.

8. The intelligent grinding system for adjusting the particle size distribution in cement production according to claim 7 is characterized in that: Calculate the corrected grinding efficiency by integrating the grinding parameter characteristics and the wear resistance characteristics of the raw material: , where is a preset adjustment factor, is the corrected grinding efficiency; Calculate the rotational speed adjustment ratio based on the anti-wear characteristic score of the raw material: , where k is the rotational speed adjustment ratio.

9. The intelligent grinding system for adjusting the particle size distribution in cement production according to claim 8 is characterized in that: Adjust the current rotational speed of the pulverizer according to the adjustment ratio, and the calculation formula is: , where is the current rotational speed, is the adjusted rotational speed.

Images