A cathode carbon grinding equipment control method and system based on process optimization

Abstract

The present invention relates to the technical field of cathode carbon grinding equipment, and in particular to a cathode carbon grinding equipment control method and system based on process optimization, wherein the control method determines the state of cathode carbon powder according to the actual bulk density; if the state of cathode carbon powder is a coarse powder state, then the wear state image of the grinding medium in the cathode carbon grinding equipment is obtained, and the wear state of the grinding medium is judged according to the wear state image; if the state of cathode carbon powder is a fine powder state, then the real-time working parameters of a grading device in the cathode carbon grinding equipment are obtained, and the working state of the grading device is judged according to the real-time working parameters; the grinding process of cathode carbon is optimized according to the wear state of the grinding medium and the working state of the grading device. The present invention helps to improve the overall control level of the cathode carbon grinding process, thereby improving product quality and production efficiency.

Description

Technical Field

[0001] The invention relates to the technical field of cathode carbon grinding equipment, in particular to a cathode carbon grinding equipment control method and system based on process optimization. Background Art

[0002] As a key component of lithium-ion batteries, the physical properties of cathode carbon materials, such as powder particle size and bulk density, have a crucial impact on the performance of the battery. The control of traditional cathode carbon grinding equipment mainly relies on experience and manual operation, and has problems such as low control accuracy, low efficiency, and difficulty in real-time adaptation to raw material changes, which makes it difficult to stably control product quality and high production costs. Therefore, it is urgent to develop a method that can monitor and control the state of cathode carbon powder in real time and intelligently adjust the grinding process according to the powder state to improve product quality and production efficiency. In addition, in the prior art, some studies have attempted to use image processing technology to monitor the grinding process, but most of these methods focus on monitoring the equipment status and lack real-time feedback and control of product quality. Moreover, for key characteristics such as the bulk density of cathode carbon powder, most of the prior art uses offline measurement, which cannot meet the needs of real-time control. Summary of the invention

[0003] The present invention overcomes the shortcomings of the prior art and provides a cathode carbon grinding equipment control method and system based on process optimization.

[0004] To achieve the above-mentioned purpose, the technical solution adopted by the present invention is:

[0005] The first aspect of the present invention discloses a cathode carbon grinding equipment control method based on process optimization, comprising the following steps:

[0006] Acquire a real-time stacking state image of cathode carbon powder in a storage barrel of a cathode carbon grinding device at a preset time node, perform characteristic analysis on the real-time stacking state image, and acquire a stacking volume value of the cathode carbon powder in the storage barrel at the preset time node;

[0007] Obtaining the accumulated powder mass of the cathode carbon powder in the storage barrel at a preset time node, performing ratio processing on the accumulated powder mass and the accumulated volume value to obtain the actual loose density of the cathode carbon powder; determining the state of the cathode carbon powder according to the actual loose density;

[0008] If the cathode carbon powder is in a coarse powder state, then obtaining a wear state image of the grinding medium in the cathode carbon grinding device, and judging the wear state of the grinding medium according to the wear state image;

[0009] If the cathode carbon powder is in a fine powder state, then obtaining real-time working parameters of a grading device in the cathode carbon grinding equipment, and determining the working state of the grading device according to the real-time working parameters;

[0010] The cathode carbon grinding process is optimized according to the wear state of the grinding medium and the working state of the classifying device.

[0011] Preferably, the real-time accumulation state image is subjected to characteristic analysis to obtain the accumulation volume value of the cathode carbon powder in the storage barrel at a preset time node, specifically:

[0012] Acquire characteristic images of cathode carbon powder in various irregular stacking states through a big data network, generate characteristic model diagrams of cathode carbon powder in various irregular stacking states according to the corresponding characteristic images; and obtain the model volume value of each characteristic model diagram in advance based on a gridding method;

[0013] Constructing a graph network, and dividing a corresponding number of storage nodes in the graph network according to the number of models of the feature model graph, and storing each feature model graph on a corresponding storage node; and assigning a model volume value numerical label to the corresponding storage node according to the model volume value of each feature model graph;

[0014] Acquire a real-time stacking state image of cathode carbon powder in a storage barrel of a cathode carbon grinding device, and generate a real-time stacking state model diagram of the cathode carbon powder in the storage barrel according to the real-time stacking state image;

[0015] Importing the real-time accumulation state model graph into the graph network, calculating the overlap degree between the real-time accumulation state model graph and the feature model graphs on each storage node; comparing and analyzing the overlap degree between the real-time accumulation state model graph and the feature model graphs on each storage node, and comparing the feature model graph with the highest overlap degree with the real-time accumulation state model graph;

[0016] Mark the storage node where the characteristic model graph with the highest degree of overlap with the real-time stacking state model graph is located, obtain the model volume value numerical label of the marked storage node, and determine the volume value of the cathode carbon powder in the irregular stacking area in the storage barrel according to the obtained model volume value numerical label;

[0017] Obtaining a bottom surface position node of a real-time stacking state model diagram in the storage barrel, obtaining a volume scale reading of the bottom surface position node in the storage barrel, and obtaining a volume value of cathode carbon powder in a regular stacking area in the storage barrel according to the volume scale reading;

[0018] The volume value of the cathode carbon powder in the irregular stacking area and the volume value of the cathode carbon powder in the regular stacking area are summed to obtain the stacking volume value of the cathode carbon powder in the storage barrel at a preset time node.

[0019] Preferably, the state of the cathode carbon powder is determined according to the actual bulk density, specifically:

[0020] Obtaining the process accuracy requirements of the cathode carbon powder, and presetting the preset bulk density range of the cathode carbon powder according to the process accuracy requirements;

[0021] Obtaining an actual bulk density of cathode carbon powder, and determining whether the actual bulk density is within a preset bulk density range;

[0022] If the actual bulk density is within the preset bulk density range, the state of the cathode carbon powder at the preset time node is marked as a standard state;

[0023] If the actual bulk density is not within the preset bulk density range, determining whether the actual bulk density is lower than a minimum value within the preset bulk density range;

[0024] If the actual bulk density is lower than the minimum value of the preset bulk density range, the state of the cathode carbon powder at the preset time node will be marked as a coarse powder state; otherwise, the state of the cathode carbon powder at the preset time node will be marked as a fine powder state.

[0025] Preferably, the wear state of the grinding medium is judged according to the wear state image, specifically:

[0026] Obtaining a wear state image of the grinding medium in the cathode carbon grinding device, and constructing a wear state model diagram of the grinding medium according to the wear state image; and obtaining an unworn state model diagram of the grinding medium in the cathode carbon grinding device;

[0027] Constructing a voxel grid space, inputting the worn state model graph and the unworn state model graph into the voxel grid space, and making the centroid points of the worn state model graph and the unworn state model graph coincide with each other;

[0028] Obtaining a model volume value of a mutually overlapping portion between the wear state model diagram and the unworn state model diagram, and obtaining a model volume value of a non-overlapping portion between the wear state model diagram and the unworn state model diagram;

[0029] The wear coefficient of the grinding medium is obtained by performing ratio processing on the model volume value of the non-overlapping part between the wear state model diagram and the unworn state model diagram and the model volume value of the overlapping part between the wear state model diagram and the unworn state model diagram;

[0030] comparing the wear coefficient of the grinding medium with a preset coefficient threshold;

[0031] If the wear coefficient of the grinding medium is greater than a preset coefficient threshold, the wear state of the grinding medium in the cathode carbon grinding device is determined to be an over-wear state;

[0032] If the wear coefficient of the grinding medium is not greater than the preset coefficient threshold, the wear state of the grinding medium in the cathode carbon grinding device is determined to be a normal wear state.

[0033] Preferably, the working state of the grading device is determined according to the real-time working parameters, specifically:

[0034] Obtaining an electronic work log of the grading device, and obtaining historical work data corresponding to various fault states of the grading device during operation according to the electronic work log, which is defined as fault characteristic data;

[0035] Taking various fault states as fault state nodes and various fault characteristic data as variable nodes; making directed connections between each fault state node and each variable node to form a Bayesian network;

[0036] In the Bayesian network, the correlation frequency between each variable node and each fault state node is counted, and the conditional probability between each fault feature data and each fault state is calculated according to the correlation frequency between each variable node and each fault state node;

[0037] Construct a conditional probability table according to the conditional probability between each fault feature data and each fault state, construct a fault prediction model based on a deep learning network, import the conditional probability table into the fault prediction model for learning and training based on a back propagation function until the model parameters meet the preset requirements, thereby obtaining a trained fault prediction model;

[0038] Acquire real-time working parameters of a grading device in a cathode carbon grinding device, import the real-time working parameters into the trained fault prediction model for prediction, and acquire the fault probability of the grading device in the cathode carbon grinding device;

[0039] If the failure probability is greater than the preset failure probability value, the working state of the grading device is determined to be a failure state; if the failure probability is not greater than the preset failure probability value, the working state of the grading device is determined to be a normal state.

[0040] Preferably, the cathode carbon grinding process is optimized according to the wear state of the grinding medium and the working state of the classifying device, specifically:

[0041] If the wear state of the grinding medium is an over-wear state, it means that the grinding medium in the cathode carbon grinding equipment needs to be replaced, and a grinding medium over-wear warning message is generated;

[0042] If the working state of the grading device is a fault state, it means that the grading device needs to be repaired, and then a grading device fault warning information is generated.

[0043] Preferably, the grinding process of the cathode carbon is optimized according to the wear state of the grinding medium and the working state of the classifying device, and further comprises the following steps:

[0044] If the wear state of the grinding medium is a normal wear state and the working state of the classifying device is a normal state, then various actual grinding parameters of the cathode carbon grinding equipment are obtained;

[0045] and obtaining a preset grinding process scheme for the cathode carbon grinding equipment, and extracting various preset grinding parameters of the cathode carbon grinding equipment from the preset grinding process scheme;

[0046] Calculate the difference between each actual grinding parameter and the corresponding preset grinding parameter to obtain the deviation value of each actual grinding parameter; compare the deviation value of each actual grinding parameter with the preset deviation value threshold;

[0047] If there is a deviation value of one or more actual grinding parameters that is greater than the preset deviation value threshold, the actual grinding parameters whose deviation value is greater than the preset deviation value threshold are corrected based on the corresponding deviation value, and the cathode carbon grinding equipment is continued to perform grinding work after the correction is completed;

[0048] If the deviation values ​​of various actual grinding parameters are not greater than the preset deviation value threshold, the preset grinding process plan is calibrated as a process plan that needs to be adjusted and optimized, and the cathode carbon grinding equipment is controlled to stop grinding.

[0049] Among them, the variable nodes include temperature data, vibration frequency data, airflow intensity data, voltage data and current data; the fault status nodes include rotor breakage fault, blade deformation fault, motor overload or short circuit fault, speed sensor failure fault and abnormal airflow fault.

[0050] The actual grinding parameters include grinding time, grinding speed, grinding pressure and grinding temperature.

[0051] The second aspect of the present invention discloses a cathode carbon grinding equipment control system based on process optimization, the cathode carbon grinding equipment control system includes a memory and a processor, the memory stores a cathode carbon grinding equipment control method program, when the cathode carbon grinding equipment control method program is executed by the processor, all steps of the cathode carbon grinding equipment control method are implemented.

[0052] The present invention solves the technical defects existing in the background technology, and the present invention has the following beneficial effects: First, the loose density is calculated by analyzing the image of the cathode carbon powder accumulation state, and then the powder state is accurately judged, thereby achieving effective control of product quality. Secondly, the judgment of the wear of the grinding medium in the coarse powder state and the prediction of the working state of the grading device in the fine powder state can timely discover equipment problems. Furthermore, the process is optimized according to the wear state of the grinding medium and the working state of the grading device. In terms of ensuring the normal operation of the equipment, production delays caused by excessive wear of the grinding medium or failure of the grading device can be avoided. At the same time, the analysis and adjustment of the grinding parameters help to optimize the grinding process, improve product quality consistency, and promote continuous improvement of the process plan, enhance the controllability and stability of production, and improve production efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, drawings of other embodiments can be obtained based on these drawings without paying creative work.

[0054] Figure 1 This is the overall method flow chart of the cathode carbon grinding equipment control method;

[0055] Figure 2 This is a partial method flow chart of the cathode carbon grinding equipment control method;

[0056] Figure 3 This is the system block diagram of the cathode carbon grinding equipment control system. DETAILED DESCRIPTION

[0057] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0058] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0059] The first aspect of the present invention discloses a cathode carbon grinding equipment control method based on process optimization, such as Figure 1 As shown, the following steps are included:

[0060] S102: obtaining a real-time stacking state image of cathode carbon powder in a storage barrel of a cathode carbon grinding device at a preset time node, performing characteristic analysis on the real-time stacking state image, and obtaining a stacking volume value of the cathode carbon powder in the storage barrel at the preset time node;

[0061] S104: obtaining the accumulated powder mass of the cathode carbon powder in the storage barrel at a preset time node, performing ratio processing on the accumulated powder mass and the accumulated volume value to obtain the actual loose density of the cathode carbon powder; and determining the state of the cathode carbon powder according to the actual loose density;

[0062] S106: if the cathode carbon powder is in a coarse powder state, obtaining a wear state image of the grinding medium in the cathode carbon grinding device, and determining the wear state of the grinding medium according to the wear state image;

[0063] S108: If the cathode carbon powder is in a fine powder state, obtain real-time working parameters of a classifying device in the cathode carbon grinding equipment, and determine the working state of the classifying device according to the real-time working parameters;

[0064] S110: Optimizing the cathode carbon grinding process according to the wear state of the grinding medium and the working state of the classifying device.

[0065] It should be noted that the cathode carbon powder accumulation volume is calculated in real time through image technology. This method can obtain the important parameter of powder accumulation volume in a non-contact and fast manner, avoiding the interference that traditional measurement methods may cause to the powder. The bulk density is a key indicator reflecting the characteristics of the powder. The actual bulk density calculated in this way can accurately reflect the accumulation characteristics of the powder under the current production status. The state of the cathode carbon powder (coarse powder state or fine powder state, etc.) is determined based on the actual bulk density, which provides a basis for subsequent targeted process adjustments based on the powder state.

[0066] If the cathode carbon powder is in a coarse powder state, the wear state image of the grinding medium in the cathode carbon grinding equipment is obtained, and the wear state of the grinding medium is determined based on the wear state image. The wear of the grinding medium will affect the grinding effect. By analyzing the wear state of the grinding medium by image, the condition of the grinding medium can be intuitively understood. For example, if the grinding medium is severely worn, the grinding efficiency may be reduced and the powder particle size may be difficult to refine. This image-based wear state judgment can timely discover the problem of the grinding medium and provide an accurate reference for subsequent process tuning. If the cathode carbon powder is in a fine powder state, the real-time working parameters of the classifier in the cathode carbon grinding equipment are obtained, and the working state of the classifier is determined based on the real-time working parameters. The classifier plays a key role in the particle size screening of the powder during the powder production process. By obtaining real-time working parameters (such as air volume, rotation speed, etc.) to judge its working state, it can be timely found whether the classifier is operating normally. For example, if the air volume of the classifier is abnormal, it may lead to inaccurate screening of fine powder, thereby affecting product quality.

[0067] The cathode carbon grinding process is optimized according to the wear state of the grinding medium and the working state of the grading device. This is the core purpose of the entire method. By judging the wear state of the grinding medium and the working state of the grading device in the previous steps, the grinding process can be adjusted in a targeted manner. For example, if the grinding medium is severely worn, you can consider replacing the grinding medium or adjusting parameters such as the grinding time and mill speed; if the working state of the grading device is abnormal, you can adjust the air volume, speed and other parameters of the grading device, so as to achieve precise control of the particle size and loose density of the cathode carbon powder, and improve product quality and production efficiency.

[0068] The powder state is determined by real-time monitoring and calculation of the actual loose density of the cathode carbon powder, and then the wear state of the grinding medium and the working state of the grading device are judged and the corresponding process is adjusted according to the different powder states (coarse powder or fine powder). This helps to accurately control key characteristics such as the particle size and loose density of the powder, thereby improving the quality of cathode carbon products. Timely judgment of the wear state of the grinding medium and the working state of the grading device can avoid production delays caused by excessive wear of the grinding medium or failure of the grading device. For example, replacing or adjusting the grinding medium before the wear affects the grinding effect, and timely repairing the failure of the grading device can ensure the continuity of the production process and improve production efficiency.

[0069] The entire process is based on the acquisition and analysis of images and real-time working parameters, realizing intelligent control of cathode carbon grinding equipment. This control method does not rely on traditional experience and frequent manual intervention, and can automatically optimize the process according to the powder state and equipment state, reducing the impact of human factors on production and improving the stability and controllability of the production process.

[0070] Preferably, the real-time accumulation state image is subjected to characteristic analysis to obtain the accumulation volume value of the cathode carbon powder in the storage barrel at a preset time node, specifically:

[0071] Acquire characteristic images of cathode carbon powder in various irregular stacking states through a big data network, generate characteristic model diagrams of cathode carbon powder in various irregular stacking states according to the corresponding characteristic images; and obtain the model volume value of each characteristic model diagram in advance based on a gridding method;

[0072] Constructing a graph network, and dividing a corresponding number of storage nodes in the graph network according to the number of models of the feature model graph, and storing each feature model graph on a corresponding storage node; and assigning a model volume value numerical label to the corresponding storage node according to the model volume value of each feature model graph;

[0073] Acquire a real-time stacking state image of cathode carbon powder in a storage barrel of a cathode carbon grinding device, and generate a real-time stacking state model diagram of the cathode carbon powder in the storage barrel according to the real-time stacking state image;

[0074] Importing the real-time accumulation state model graph into the graph network, calculating the overlap degree between the real-time accumulation state model graph and the feature model graphs on each storage node; comparing and analyzing the overlap degree between the real-time accumulation state model graph and the feature model graphs on each storage node, and comparing the feature model graph with the highest overlap degree with the real-time accumulation state model graph;

[0075] Mark the storage node where the characteristic model graph with the highest degree of overlap with the real-time stacking state model graph is located, obtain the model volume value numerical label of the marked storage node, and determine the volume value of the cathode carbon powder in the irregular stacking area in the storage barrel according to the obtained model volume value numerical label;

[0076] Obtaining a bottom surface position node of a real-time stacking state model diagram in the storage barrel, obtaining a volume scale reading of the bottom surface position node in the storage barrel, and obtaining a volume value of cathode carbon powder in a regular stacking area in the storage barrel according to the volume scale reading;

[0077] The volume value of the cathode carbon powder in the irregular stacking area and the volume value of the cathode carbon powder in the regular stacking area are summed to obtain the stacking volume value of the cathode carbon powder in the storage barrel at a preset time node.

[0078] It should be noted that the rich data resources of the big data network are used to collect various possible powder accumulation state images, so as to construct a comprehensive feature model map. The model volume value of each feature model map is obtained in advance based on the gridding method. The gridding method is a method of dividing the model space into small grids to calculate the volume. This method can accurately obtain the volume of each feature model map and provide basic data for subsequent volume matching. The graph network is used as a data storage and management structure here, which is similar to the concept of graph database and can effectively organize and manage a large amount of feature model map data. At the same time, according to the model volume value of each feature model map, a model volume value numerical label is assigned to the corresponding storage node, so that each storage node has an information label related to the volume of the feature model map, which is convenient for subsequent query and comparison. The real-time accumulation state model map is imported into the graph network, and the degree of overlap between the real-time accumulation state model map and the feature model map on each storage node is calculated. The overlap degree calculation here involves the feature matching algorithm in image recognition technology, such as calculating the shape similarity and texture similarity between model maps. Then, the degree of overlap between the real-time stacking state model graph and the characteristic model graph on each storage node is compared and analyzed, and the characteristic model graph with the highest degree of overlap with the real-time stacking state model graph is compared. The purpose of this step is to find the model graph that is most similar to the real-time stacking state in the existing characteristic model graphs. The storage node where the characteristic model graph with the highest degree of overlap with the real-time stacking state model graph is located is recorded, and the model volume value numerical label of the marked storage node is obtained. The volume value of the cathode carbon powder in the irregular stacking area in the storage barrel is determined according to the obtained model volume value numerical label. Since the model volume value numerical label of the previous storage node is associated with the volume of the characteristic model graph, the volume of the powder in the irregular stacking area can be approximated by finding the storage node label corresponding to the most similar characteristic model graph. The bottom surface position node of the real-time stacking state model graph in the storage barrel is obtained, and the volume scale reading of the bottom surface position node in the storage barrel is obtained. The volume value of the cathode carbon powder in the regular stacking area in the storage barrel is obtained according to the volume scale reading. Here, the volume scale of the storage barrel itself is used to determine the volume of the regular accumulation part (e.g., the flat bottom accumulation part) according to the position of the powder on the bottom of the storage barrel. The volume value of the cathode carbon powder in the irregular accumulation area is summed with the volume value of the cathode carbon powder in the regular accumulation area to obtain the accumulation volume value of the cathode carbon powder in the storage barrel at the preset time node. By calculating the volumes of the irregular accumulation and regular accumulation areas separately and summing them, the complete powder accumulation volume is obtained.

[0079] In general, compared with the traditional method of calculating volume by simple geometric shapes, this method can better adapt to the actual situation of irregular powder accumulation. Since it is based on the matching of characteristic model graphs under various irregular accumulation states, this method can adapt to various complex accumulation states of cathode carbon powder in the storage barrel. Regardless of the different accumulation forms caused by factors such as feeding method, powder characteristics, or internal structure of the storage barrel, this method can be used to effectively calculate the volume, so that the results can be obtained in time when the powder accumulation volume is monitored in real time, meeting the real-time requirements of the production process.

[0080] In addition, the accurate powder stacking volume value is one of the key factors in calculating the bulk density. This precise volume measurement method can provide accurate data basis for subsequent calculation of bulk density, judgment of powder state and further process optimization, which helps to improve the overall control level of cathode carbon grinding process, thereby improving product quality and production efficiency.

[0081] Preferably, the state of the cathode carbon powder is determined according to the actual bulk density, such as Figure 2 As shown, specifically:

[0082] S202: Obtaining the process accuracy requirements of the cathode carbon powder, and presetting a preset bulk density range of the cathode carbon powder according to the process accuracy requirements;

[0083] S204: Obtaining the actual bulk density of the cathode carbon powder, and determining whether the actual bulk density is within a preset bulk density range;

[0084] S206: if the actual bulk density is within a preset bulk density range, marking the state of the cathode carbon powder at a preset time node as a standard state;

[0085] S208: If the actual bulk density is not within the preset bulk density range, determining whether the actual bulk density is lower than a minimum value of the preset bulk density range;

[0086] S210: If the actual bulk density is lower than the minimum value of the preset bulk density range, the state of the cathode carbon powder at the preset time node is marked as a coarse powder state; otherwise, the state of the cathode carbon powder at the preset time node is marked as a fine powder state.

[0087] It should be noted that different application scenarios (such as different types of battery manufacturing) have different precision requirements for the bulk density of cathode carbon powder. Relevant technicians set a reasonable range by obtaining process precision requirements. If the actual bulk density is within the preset bulk density range, the state of the cathode carbon powder at the preset time node will be marked as the standard state, which indicates that the bulk density of the cathode carbon powder currently produced meets the process requirements, the production process is in a normal state, and the particle size and other characteristics of the powder are within an acceptable range, which can meet the needs of subsequent production or application.

[0088] If the actual bulk density is not within the preset bulk density range, it is further determined whether the actual bulk density is lower than the minimum value of the preset bulk density range. This step is to distinguish whether the actual bulk density exceeds the range because it is too low (possibly related to the coarse powder state) or too high (possibly related to the fine powder state). Generally speaking, in the coarse powder state, due to the large particles, the gaps between the particles are relatively large, the mass of powder per unit volume is relatively small when naturally stacked, and the bulk density is low. In the fine powder state, fine particles can fill part of the gaps, and fine powder has a large specific surface area and strong interactions between particles, which makes it easy to form agglomerations, which will increase the mass of powder per unit volume to a certain extent, and the bulk density is relatively high.

[0089] After accurately judging whether the powder state is standard, coarse powder or fine powder, it can provide a clear basis for subsequent process adjustments. For example, if it is in a coarse powder state, the grinding equipment, grinding time, etc. can be adjusted according to the characteristics of the coarse powder; if it is in a fine powder state, the classification device can be optimized to improve product quality and production efficiency. In addition, accurate judgment of the powder state helps maintain the stability of the production process. By timely discovering whether the powder state deviates from the standard state and taking corresponding measures to adjust it, it can avoid producing a large number of products that do not meet the requirements, reduce fluctuations in the production process, and ensure the continuity and stability of production.

[0090] Preferably, the wear state of the grinding medium is judged according to the wear state image, specifically:

[0091] Obtaining a wear state image of the grinding medium in the cathode carbon grinding device, and constructing a wear state model diagram of the grinding medium according to the wear state image; and obtaining an unworn state model diagram of the grinding medium in the cathode carbon grinding device;

[0092] Constructing a voxel grid space, inputting the worn state model graph and the unworn state model graph into the voxel grid space, and making the centroid points of the worn state model graph and the unworn state model graph coincide with each other;

[0093] Obtaining a model volume value of a mutually overlapping portion between the wear state model diagram and the unworn state model diagram, and obtaining a model volume value of a non-overlapping portion between the wear state model diagram and the unworn state model diagram;

[0094] The wear coefficient of the grinding medium is obtained by performing ratio processing on the model volume value of the non-overlapping part between the wear state model diagram and the unworn state model diagram and the model volume value of the overlapping part between the wear state model diagram and the unworn state model diagram;

[0095] comparing the wear coefficient of the grinding medium with a preset coefficient threshold;

[0096] If the wear coefficient of the grinding medium is greater than a preset coefficient threshold, the wear state of the grinding medium in the cathode carbon grinding device is determined to be an over-wear state;

[0097] If the wear coefficient of the grinding medium is not greater than the preset coefficient threshold, the wear state of the grinding medium in the cathode carbon grinding device is determined to be a normal wear state.

[0098] It should be noted that the wear state image reflects the actual state of the current grinding media after use, while the unworn state model image serves as a benchmark. Both are important bases for wear analysis. The voxel grid space provides a unified spatial framework for the model image, which is convenient for volume calculation and comparison. By coinciding the center of mass points, it can be ensured that the two model images are analyzed in the same reference system, improving the accuracy of subsequent calculations. The wear coefficient is an indicator that can comprehensively reflect the degree of wear of the grinding media. It expresses the relative degree of wear by the volume ratio of the non-overlapping part (i.e. the worn or newly added part, mainly the worn part in the case of wear) to the overlapping part (i.e. the part that has not been worn). The preset coefficient threshold is a critical value determined based on experience or research on the wear limit of the grinding media, which is used to determine whether the wear state of the grinding media has reached a level that requires attention or measures.

[0099] If the wear coefficient of the grinding medium is greater than the preset coefficient threshold, the wear state of the grinding medium in the cathode carbon grinding equipment is determined to be an over-wear state, which means that the wear of the grinding medium is already serious and may have a significant impact on the grinding effect, and corresponding measures need to be taken, such as replacing the grinding medium. If the wear coefficient of the grinding medium is not greater than the preset coefficient threshold, the wear state of the grinding medium in the cathode carbon grinding equipment is determined to be a normal wear state, indicating that the current wear of the grinding medium is within an acceptable range and has little impact on the grinding process.

[0100] This method can timely detect whether the grinding medium is in an over-wear state, thereby providing timely warning for equipment maintenance. When it is determined to be in an over-wear state, timely measures can be taken, such as replacing the grinding medium, to avoid problems such as reduced grinding efficiency and unstable product quality caused by excessive wear of the grinding medium, which helps to ensure the stability of the production process and product quality. At the same time, the grinding process is adjusted according to the wear state of the grinding medium. For example, under normal wear conditions, grinding parameters such as grinding time and mill speed can be appropriately fine-tuned according to the degree of wear; under over-wear conditions, in addition to replacing the grinding medium, the entire grinding process can be re-evaluated to ensure the grinding effect and product quality and improve production efficiency.

[0101] Preferably, the working state of the grading device is determined according to the real-time working parameters, specifically:

[0102] Obtaining an electronic work log of the grading device, and obtaining historical work data corresponding to various fault states of the grading device during operation according to the electronic work log, which is defined as fault characteristic data;

[0103] Taking various fault states as fault state nodes and various fault characteristic data as variable nodes; making directed connections between each fault state node and each variable node to form a Bayesian network;

[0104] In the Bayesian network, the correlation frequency between each variable node and each fault state node is counted, and the conditional probability between each fault feature data and each fault state is calculated according to the correlation frequency between each variable node and each fault state node;

[0105] Construct a conditional probability table according to the conditional probability between each fault feature data and each fault state, construct a fault prediction model based on a deep learning network, import the conditional probability table into the fault prediction model for learning and training based on a back propagation function until the model parameters meet the preset requirements, thereby obtaining a trained fault prediction model;

[0106] Acquire real-time working parameters of a grading device in a cathode carbon grinding device, import the real-time working parameters into the trained fault prediction model for prediction, and acquire the fault probability of the grading device in the cathode carbon grinding device;

[0107] If the failure probability is greater than the preset failure probability value, the working state of the grading device is determined to be a failure state; if the failure probability is not greater than the preset failure probability value, the working state of the grading device is determined to be a normal state.

[0108] Among them, the variable nodes include temperature data, vibration frequency data, airflow intensity data, voltage data and current data; the fault status nodes include rotor breakage fault, blade deformation fault, motor overload or short circuit fault, speed sensor failure fault and abnormal airflow fault.

[0109] It should be noted that the electronic work record book contains various information about the past operation of the grading device. By screening out the working data under the fault state, these data can reflect the operating characteristics of the equipment when different faults occur, and provide basic data for subsequent fault analysis and prediction. Various fault states are used as fault state nodes, and various fault characteristic data (such as temperature data, vibration frequency data, airflow intensity data, voltage data, and current data) are used as variable nodes; each fault state node is connected to each variable node in a directed manner to form a Bayesian network. The Bayesian network is a probabilistic graph model. Through this directed connection, the causal relationship or dependency relationship between the variable node (fault characteristic data) and the fault state node can be expressed. For example, temperature data may have some correlation with motor overload or short circuit fault, and this correlation can be reflected through the directed edges in the network.

[0110] Conditional probability reflects the probability of a certain fault state occurring when certain variables (fault characteristic data) are known. For example, when the temperature rises, the probability of motor overload or short circuit failure is known. Calculating conditional probability by statistical correlation frequency is based on a large amount of historical data, so the probability obtained can better reflect the actual situation.

[0111] The real-time working parameters of the grading device in the cathode carbon grinding equipment (including temperature, vibration frequency, airflow intensity, voltage and current data) are obtained, and these real-time working parameters are imported into the trained fault prediction model for prediction to obtain the failure probability of the grading device in the cathode carbon grinding equipment. This step inputs the actual operating equipment parameters into the trained model, and the model will calculate the probability of failure of the current grading device based on the previously learned rules.

[0112] In summary, this method can predict the failure probability of the grading device based on the real-time working parameters, thereby realizing early warning of the failure of the grading device. When the failure probability is predicted to be high, maintenance measures such as inspection, repair or replacement of parts can be taken in time to avoid sudden equipment failures leading to production interruptions. At the same time, this prediction also helps to optimize the maintenance plan, reasonably arrange maintenance time and resources according to the actual operating status of the equipment, reduce maintenance costs, ensure the normal operation of the grading device, and thus improve the reliability of the entire cathode carbon grinding equipment and improve the consistency and quality of the product.

[0113] Preferably, the cathode carbon grinding process is optimized according to the wear state of the grinding medium and the working state of the classifying device, specifically:

[0114] If the wear state of the grinding medium is an over-wear state, it means that the grinding medium in the cathode carbon grinding equipment needs to be replaced, and a grinding medium over-wear warning message is generated;

[0115] If the working state of the grading device is a fault state, it means that the grading device needs to be repaired, and then a grading device fault warning information is generated.

[0116] Preferably, the grinding process of the cathode carbon is optimized according to the wear state of the grinding medium and the working state of the classifying device, and further comprises the following steps:

[0117] If the wear state of the grinding medium is a normal wear state and the working state of the classifying device is a normal state, then various actual grinding parameters of the cathode carbon grinding device are obtained;

[0118] and obtaining a preset grinding process scheme for the cathode carbon grinding equipment, and extracting various preset grinding parameters of the cathode carbon grinding equipment from the preset grinding process scheme;

[0119] Calculate the difference between each actual grinding parameter and the corresponding preset grinding parameter to obtain the deviation value of each actual grinding parameter; compare the deviation value of each actual grinding parameter with the preset deviation value threshold;

[0120] If there is a deviation value of one or more actual grinding parameters that is greater than the preset deviation value threshold, the actual grinding parameters whose deviation value is greater than the preset deviation value threshold are corrected based on the corresponding deviation value, and the cathode carbon grinding equipment is continued to perform grinding work after the correction is completed;

[0121] If the deviation values ​​of various actual grinding parameters are not greater than the preset deviation value threshold, the preset grinding process plan is calibrated as a process plan that needs to be adjusted and optimized, and the cathode carbon grinding equipment is controlled to stop grinding.

[0122] The actual grinding parameters include grinding time, grinding speed, grinding pressure and grinding temperature.

[0123] It should be noted that if the wear state of the grinding medium is excessive wear, it means that the grinding medium has been severely worn and cannot effectively perform grinding work. In this case, the grinding medium in the cathode carbon grinding equipment needs to be replaced, so the grinding medium excessive wear warning information is generated. This warning information can remind the operator to replace the grinding medium in time to ensure the grinding effect and product quality.

[0124] If the working state of the grading device is a fault state, it indicates that there is a problem with the grading device and it needs to be repaired. At this time, a fault warning message of the grading device is generated. This helps the operator to quickly locate the problem and perform repairs in time to avoid the fault from further affecting the production process.

[0125] If the wear state of the grinding medium is normal and the working state of the grading device is normal, the main components of the equipment are in normal working state, and further analysis is needed to determine whether the grinding process is optimal. The preset grinding parameters are standard parameters set according to ideal production requirements. The deviation values ​​of each actual grinding parameter are compared with the preset deviation value threshold. The preset deviation value threshold is a pre-set limit used to determine whether the deviation is within an acceptable range. If the deviation value of one or more actual grinding parameters is greater than the preset deviation value threshold, this indicates that some grinding parameters deviate too much from the ideal state, which may affect product quality or production efficiency. Based on the corresponding deviation value, the actual grinding parameters with deviation values ​​greater than the preset deviation value threshold are corrected. For example, if the grinding time deviation is too large, adjust the grinding time; if the grinding speed deviation is too large, adjust the grinding speed, etc. After the correction is completed, continue to make the cathode carbon grinding equipment to perform grinding work to ensure that the grinding process is carried out in the optimal state.

[0126] If the deviation values ​​of the actual grinding parameters are not greater than the preset deviation value threshold, this means that although the equipment is operating normally, the current preset grinding process may have room for optimization. The preset grinding process is marked as a process that needs to be adjusted and optimized, and the cathode carbon grinding equipment is controlled to stop grinding, so as to adjust and optimize the process to improve production efficiency or product quality.

[0127] In summary, by monitoring and early warning the wear state of the grinding medium and the working state of the grading device, problems in the equipment can be discovered in time and corresponding measures can be taken. Timely replacement of excessively worn grinding media and timely repair of grading device failures can avoid production interruptions or product quality degradation caused by equipment component problems and ensure the normal operation of the equipment. When the main components of the equipment are working normally, the actual grinding parameters are compared and analyzed with the preset grinding parameters to timely discover slight deviations in the grinding process. By correcting the actual grinding parameters whose deviation values ​​are greater than the preset deviation value threshold, the grinding process can be made closer to the ideal state, thereby improving the consistency of product quality and ensuring that the product meets the requirements. When the deviation values ​​of various actual grinding parameters are within an acceptable range but there is still room for optimization, the preset grinding process plan is calibrated as the process plan that needs to be adjusted and optimized, and the grinding work is stopped for adjustment. This practice helps to continuously optimize the grinding process, improve production efficiency, reduce production costs, and enable the entire cathode carbon grinding process to adapt to different production needs or continuously improving product quality requirements. The entire tuning process is based on the precise monitoring and analysis of equipment status and grinding parameters, making the production process more controllable. Whether it is the maintenance of equipment components or the adjustment of the grinding process, it helps to reduce fluctuations in the production process and improve production stability, thereby improving the company's production efficiency.

[0128] The cathode carbon grinding equipment control method may further include the following steps:

[0129] Purity characteristic images of cathode carbon powder under various preset purities are obtained through a big data network, and feature extraction processing is performed on each purity characteristic image based on the principal component analysis method to obtain the covariance matrix of each purity characteristic image, and obtain the covariance matrix corresponding to various preset purities;

[0130] Constructing a database, importing the covariance matrices corresponding to various preset purities into the database to obtain a purity feature database; and regularly updating the purity feature database;

[0131] Acquire a surface feature image of cathode carbon powder in a storage barrel of a cathode carbon grinding device at a preset time node, perform feature extraction processing on the surface feature image based on a principal component analysis method, and obtain a covariance matrix corresponding to the surface feature image;

[0132] Extracting covariance matrices corresponding to various preset purities from the purity feature database, and calculating the F norm between the covariance matrix corresponding to the surface feature image and the covariance matrix corresponding to the various preset purities;

[0133] Extracting the maximum F norm from the calculated F norms, obtaining a preset purity corresponding to the maximum F norm, and using the preset purity corresponding to the maximum F norm as the actual purity of the cathode carbon powder in the storage barrel of the cathode carbon grinding equipment;

[0134] Determine whether the actual purity of the cathode carbon powder is within the preset range; if not, it means that the cathode carbon grinding equipment needs to be deeply cleaned, and generate cathode carbon grinding equipment cleaning warning information.

[0135] It should be noted that the covariance matrix corresponding to various preset purities is imported into the database to construct a purity feature database. This database is the basis for subsequent judgment of the actual purity of the cathode carbon powder. Since new data or more accurate data may appear during the production process, it is necessary to update this database regularly to ensure the accuracy and effectiveness of the database. The F norm (Frobenius norm) is an indicator for measuring the difference between matrices. By calculating this norm, the degree of difference between the surface feature image and the images of different preset purities can be obtained. Determine whether the actual purity of the cathode carbon powder is within the preset range. If it is not within the preset range, this may mean that there are some problems in the production process, such as equipment contamination, indicating that the cathode carbon grinding equipment needs to be deeply cleaned and the cathode carbon grinding equipment cleaning warning information is generated. This warning information can remind the operator to maintain the equipment in time to ensure that the purity of the product meets the requirements and avoid unqualified product purity due to equipment contamination and other problems, thereby improving the stability of product quality.

[0136] The cathode carbon grinding equipment control method may further include the following steps:

[0137] At several time points, the actual particle size information of the cathode carbon powder produced by the cathode carbon grinding equipment is continuously collected by a laser particle size analyzer; and the collection time point corresponding to each actual particle size information is recorded;

[0138] The Hermite curve method is introduced, and the actual granularity information collected and its corresponding time nodes are determined as data points. The granularity values ​​corresponding to the first and last time nodes are used as the starting and end point function values, and the first-order derivative values ​​of the starting and end points are determined by difference.

[0139] The Hermite curve equation for each time interval is constructed according to the first-order derivative values ​​of the determined starting point and end point, the t value is selected in each interval and substituted into the equation to calculate the particle size value, and finally the actual particle size Hermite curve is drawn using the obtained coordinate points;

[0140] Obtaining a preset process accuracy requirement for cathode carbon powder, and determining a preset particle size range threshold of the cathode carbon powder according to the preset process accuracy requirement;

[0141] Importing the preset particle size range threshold of the cathode carbon powder into the actual particle size Hermite curve, determining a definition domain according to the preset particle size range threshold, and judging whether the boundary of the actual particle size Hermite curve is completely within the definition domain;

[0142] If the boundary of the actual particle size Hermite curve is completely within the definition domain, it means that the particle size of the cathode carbon powder meets the standard; otherwise, it means that the particle size of the cathode carbon powder does not meet the standard, and a particle size non-standard warning message is generated.

[0143] It should be noted that the Hermite curve equation can construct a smooth curve based on the given starting point, end point, and derivative information at the starting point and end point, and accurately describe the relationship between the change of particle size over time in each time interval. The curve drawn in this way can intuitively show the dynamic change trend of the particle size of the cathode carbon powder during the entire acquisition time period. The preset particle size range threshold is imported into the actual particle size Hermite curve, and the definition domain is determined according to the threshold to determine whether the boundary of the actual particle size Hermite curve is completely within the definition domain. Here, the particle size range required by the process is converted into the definition domain of the curve, and the relationship between the curve boundary and the definition domain is judged to evaluate whether the product particle size meets the requirements. If the boundary of the actual particle size Hermite curve is completely within the definition domain, it means that the particle size of the cathode carbon powder meets the standard; otherwise, it means that the particle size of the cathode carbon powder does not meet the standard, and a warning message of particle size non-compliance is generated. This judgment method is based on the relationship between the overall curve and the preset range, and can comprehensively consider the changes in particle size during the entire acquisition time period, rather than just the data at a certain point in time, so as to more accurately evaluate product quality. The particle size range threshold determined based on the preset process accuracy requirements is compared with the actual particle size Hermite curve, and whether the particle size meets the standard is determined by whether the curve boundary is within the definition domain. This evaluation method comprehensively considers the particle size changes during the entire acquisition period. Compared with only looking at the particle size data at a single time point, it can more accurately determine whether the product meets the quality requirements. When it is determined that the particle size does not meet the standard, an early warning message is generated to quickly take measures, such as adjusting the parameters of the grinding equipment (grinding time, grinding speed, etc.), so as to avoid the production of a large number of unqualified products, reduce production losses, and improve production efficiency.

[0144] The second aspect of the present invention discloses a cathode carbon grinding equipment control system based on process optimization, such as Figure 3 As shown, the cathode carbon grinding equipment control system includes a memory 50 and a processor 101. The memory 50 stores a cathode carbon grinding equipment control method program. When the cathode carbon grinding equipment control method program is executed by the processor 101, all steps of the cathode carbon grinding equipment control method are implemented.

[0145] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A cathode carbon grinding equipment control method based on process optimization, characterized in that: The following steps are involved: Acquire a real-time stacking state image of cathode carbon powder in a storage barrel of a cathode carbon grinding device at a preset time node, perform characteristic analysis on the real-time stacking state image, and acquire a stacking volume value of the cathode carbon powder in the storage barrel at the preset time node; Obtaining the accumulated powder mass of the cathode carbon powder in the storage barrel at a preset time node, performing ratio processing on the accumulated powder mass and the accumulated volume value to obtain the actual loose density of the cathode carbon powder; determining the state of the cathode carbon powder according to the actual loose density; If the cathode carbon powder is in a coarse powder state, then obtaining a wear state image of the grinding medium in the cathode carbon grinding device, and judging the wear state of the grinding medium according to the wear state image; If the cathode carbon powder is in a fine powder state, then obtaining real-time working parameters of a grading device in the cathode carbon grinding equipment, and determining the working state of the grading device according to the real-time working parameters; The cathode carbon grinding process is optimized according to the wear state of the grinding medium and the working state of the classifying device.

2. The cathode carbon grinding equipment control method based on process optimization according to claim 1 is characterized in that: The real-time accumulation state image is analyzed for characteristics to obtain the accumulation volume value of the cathode carbon powder in the storage barrel at a preset time node, specifically: Acquire characteristic images of cathode carbon powder in various irregular stacking states through a big data network, generate characteristic model diagrams of cathode carbon powder in various irregular stacking states according to the corresponding characteristic images; and obtain the model volume value of each characteristic model diagram in advance based on a gridding method; Constructing a graph network, and dividing a corresponding number of storage nodes in the graph network according to the number of models of the feature model graph, and storing each feature model graph on a corresponding storage node; and assigning a model volume value numerical label to the corresponding storage node according to the model volume value of each feature model graph; Acquire a real-time stacking state image of cathode carbon powder in a storage barrel of a cathode carbon grinding device, and generate a real-time stacking state model diagram of the cathode carbon powder in the storage barrel according to the real-time stacking state image; Importing the real-time accumulation state model graph into the graph network, calculating the overlap degree between the real-time accumulation state model graph and the feature model graphs on each storage node; comparing and analyzing the overlap degree between the real-time accumulation state model graph and the feature model graphs on each storage node, and comparing the feature model graph with the highest overlap degree with the real-time accumulation state model graph; Mark the storage node where the characteristic model graph with the highest degree of overlap with the real-time stacking state model graph is located, obtain the model volume value numerical label of the marked storage node, and determine the volume value of the cathode carbon powder in the irregular stacking area in the storage barrel according to the obtained model volume value numerical label; Obtaining a bottom surface position node of a real-time stacking state model diagram in the storage barrel, obtaining a volume scale reading of the bottom surface position node in the storage barrel, and obtaining a volume value of cathode carbon powder in a regular stacking area in the storage barrel according to the volume scale reading; The volume value of the cathode carbon powder in the irregular stacking area and the volume value of the cathode carbon powder in the regular stacking area are summed to obtain the stacking volume value of the cathode carbon powder in the storage barrel at a preset time node.

3. The cathode carbon grinding equipment control method based on process optimization according to claim 1 is characterized in that: The state of the cathode carbon powder is determined according to the actual bulk density, specifically: Obtaining the process accuracy requirements of the cathode carbon powder, and presetting the preset bulk density range of the cathode carbon powder according to the process accuracy requirements; Obtaining an actual bulk density of cathode carbon powder, and determining whether the actual bulk density is within a preset bulk density range; If the actual bulk density is within the preset bulk density range, the state of the cathode carbon powder at the preset time node is marked as a standard state; If the actual bulk density is not within the preset bulk density range, determining whether the actual bulk density is lower than a minimum value within the preset bulk density range; If the actual bulk density is lower than the minimum value of the preset bulk density range, the state of the cathode carbon powder at the preset time node will be marked as a coarse powder state; otherwise, the state of the cathode carbon powder at the preset time node will be marked as a fine powder state.

4. The cathode carbon grinding equipment control method based on process optimization according to claim 1 is characterized in that: The wear state of the grinding medium is determined according to the wear state image, specifically: Obtaining a wear state image of the grinding medium in the cathode carbon grinding device, and constructing a wear state model diagram of the grinding medium according to the wear state image; and obtaining an unworn state model diagram of the grinding medium in the cathode carbon grinding device; Constructing a voxel grid space, inputting the worn state model graph and the unworn state model graph into the voxel grid space, and making the centroid points of the worn state model graph and the unworn state model graph coincide with each other; Obtaining a model volume value of a mutually overlapping portion between the wear state model diagram and the unworn state model diagram, and obtaining a model volume value of a non-overlapping portion between the wear state model diagram and the unworn state model diagram; The wear coefficient of the grinding medium is obtained by performing ratio processing on the model volume value of the non-overlapping part between the wear state model diagram and the unworn state model diagram and the model volume value of the overlapping part between the wear state model diagram and the unworn state model diagram; comparing the wear coefficient of the grinding medium with a preset coefficient threshold; If the wear coefficient of the grinding medium is greater than a preset coefficient threshold, the wear state of the grinding medium in the cathode carbon grinding device is determined to be an over-wear state; If the wear coefficient of the grinding medium is not greater than the preset coefficient threshold, the wear state of the grinding medium in the cathode carbon grinding device is determined to be a normal wear state.

5. The cathode carbon grinding equipment control method based on process optimization according to claim 1 is characterized in that: The working state of the grading device is determined according to the real-time working parameters, specifically: Obtaining an electronic work log of the grading device, and obtaining historical work data corresponding to various fault states of the grading device during operation according to the electronic work log, which is defined as fault characteristic data; Taking various fault states as fault state nodes and various fault characteristic data as variable nodes; making directed connections between each fault state node and each variable node to form a Bayesian network; In the Bayesian network, the correlation frequency between each variable node and each fault state node is counted, and the conditional probability between each fault feature data and each fault state is calculated according to the correlation frequency between each variable node and each fault state node; Construct a conditional probability table according to the conditional probability between each fault feature data and each fault state, construct a fault prediction model based on a deep learning network, import the conditional probability table into the fault prediction model for learning and training based on a back propagation function until the model parameters meet the preset requirements, thereby obtaining a trained fault prediction model; Acquire real-time working parameters of a grading device in a cathode carbon grinding device, import the real-time working parameters into the trained fault prediction model for prediction, and acquire the fault probability of the grading device in the cathode carbon grinding device; If the failure probability is greater than the preset failure probability value, the working state of the grading device is determined to be a failure state; if the failure probability is not greater than the preset failure probability value, the working state of the grading device is determined to be a normal state.

6. The cathode carbon grinding equipment control method based on process optimization according to claim 1 is characterized in that: The cathode carbon grinding process is optimized according to the wear state of the grinding medium and the working state of the classifier, specifically: If the wear state of the grinding medium is an over-wear state, it means that the grinding medium in the cathode carbon grinding equipment needs to be replaced, and a grinding medium over-wear warning message is generated; If the working state of the grading device is a fault state, it means that the grading device needs to be repaired, and then a grading device fault warning information is generated.

7. A cathode carbon grinding equipment control method based on process optimization according to claim 6, characterized in that: The cathode carbon grinding process is optimized according to the wear state of the grinding medium and the working state of the classifying device, and the following steps are also included: If the wear state of the grinding medium is a normal wear state and the working state of the classifying device is a normal state, then various actual grinding parameters of the cathode carbon grinding equipment are obtained; and obtaining a preset grinding process scheme for the cathode carbon grinding equipment, and extracting various preset grinding parameters of the cathode carbon grinding equipment from the preset grinding process scheme; Calculate the difference between each actual grinding parameter and the corresponding preset grinding parameter to obtain the deviation value of each actual grinding parameter; compare the deviation value of each actual grinding parameter with the preset deviation value threshold; If there is a deviation value of one or more actual grinding parameters that is greater than the preset deviation value threshold, the actual grinding parameters whose deviation value is greater than the preset deviation value threshold are corrected based on the corresponding deviation value, and the cathode carbon grinding equipment is continued to perform grinding work after the correction is completed; If the deviation values ​​of various actual grinding parameters are not greater than the preset deviation value threshold, the preset grinding process plan is calibrated as a process plan that needs to be adjusted and optimized, and the cathode carbon grinding equipment is controlled to stop grinding.

8. The cathode carbon grinding equipment control method based on process optimization according to claim 5 is characterized in that: The variable nodes include temperature data, vibration frequency data, airflow intensity data, voltage data and current data; the fault status nodes include rotor breakage fault, blade deformation fault, motor overload or short circuit fault, speed sensor failure fault and abnormal airflow fault.

9. The cathode carbon grinding equipment control method based on process optimization according to claim 7 is characterized in that: The actual grinding parameters include grinding time, grinding speed, grinding pressure and grinding temperature.

10. A cathode carbon grinding equipment control system based on process optimization, characterized in that: The cathode carbon grinding equipment control system includes a memory and a processor, wherein the memory stores a cathode carbon grinding equipment control method program. When the cathode carbon grinding equipment control method program is executed by the processor, the cathode carbon grinding equipment control method steps as described in any one of claims 1 to 9 are implemented.

Images