Optimal control method for rubber intelligent manufacturing based on multi-source monitoring data

Abstract

The present application relates to the field of rubber manufacturing technology, and provides a rubber intelligent manufacturing optimization control method based on multi-source monitoring data. The method includes: an interactive feeding and crushing mechanism; identifying incoming materials and obtaining feeding attribute information; obtaining multi-source monitoring data and generating a simulation model; calling the simulation model to train the monitoring model, using the hidden danger monitoring model to identify the feeding attribute information and obtain hidden danger indicators; inputting the hidden danger indicators into the controller for control optimization and outputting control parameters; the controller controls the feeding and crushing mechanism according to the control parameters. The present application solves the technical problem of the splashing and collision of crushed particles caused by the uniform and unchanging processing force when the feeding and crushing mechanism in the rubber manufacturing process processes materials with higher hardness, which in turn causes component jamming and crushing safety hazards. It achieves the effect of improving the working performance and safety of the feeding and crushing mechanism by adjusting the control parameters flexibly and adaptively in combination with the type of feeding.

Description

Technical Field

[0001] The present application relates to the field of chemical material manufacturing technology, specifically to the field of rubber manufacturing technology, and in particular to a rubber intelligent manufacturing optimization control method based on multi-source monitoring data. Background Art

[0002] With the rapid development of the rubber manufacturing industry, demands for rubber product quality and production efficiency are constantly increasing. This is particularly true during the feeding and crushing phase of rubber manufacturing. Due to the diverse hardness of materials and the complexity of the processing, traditional feeding and crushing control methods often fail to meet the efficiency, precision, and safety requirements of modern rubber manufacturing. When the feed material is hard, the uniform processing force of traditional feeding and crushing mechanisms can lead to splashing and collision of crushed particles, which not only affects the stability of the production process but can also cause equipment components to jam, posing a safety hazard. Summary of the Invention

[0003] This application provides a rubber intelligent manufacturing optimization control method based on multi-source monitoring data, aiming to solve the technical problems of component jamming and crushing safety hazards caused by the splashing and collision of crushed particles due to the uniform and unchanging processing force when the feeding and crushing mechanism in the rubber manufacturing process processes materials with higher hardness.

[0004] In view of the above problems, the present application provides a rubber intelligent manufacturing optimization control method based on multi-source monitoring data.

[0005] The present application provides a rubber intelligent manufacturing optimization control method based on multi-source monitoring data, the method comprising: a feeding and crushing mechanism of interactive rubber processing equipment; identifying materials entering the feeding and crushing mechanism, and obtaining feeding attribute information, wherein the feeding attribute information includes feeding physical properties, feeding chemical properties and feeding crushing properties; obtaining multi-source monitoring data of the feeding and crushing mechanism, and generating a feeding and crushing simulation model of the feeding and crushing mechanism using the multi-source monitoring data; calling the feeding and crushing simulation model to train a crushing conflict hidden danger monitoring model, and using the crushing conflict hidden danger monitoring model to identify the feeding attribute information and obtain a first conflict hidden danger indicator; inputting the first conflict hidden danger indicator into a crushing optimization controller for control optimization, and outputting a crushing optimization control parameter, wherein the crushing optimization controller includes a preset conflict hidden danger indicator; the crushing optimization controller controls the feeding and crushing mechanism according to the crushing optimization control parameter.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0007] The aforementioned intelligent rubber manufacturing optimization control method based on multi-source monitoring data identifies materials entering the feeder-crushing mechanism and obtains information on their properties, including their physical and chemical properties, as well as their crushing characteristics. Subsequently, multi-source monitoring data, such as temperature, pressure, and motor speed, is collected during the feeder-crushing mechanism's operation. This data is used to construct a feeder-crushing simulation model that simulates the actual processing process. This simulation model is then used to train a crushing conflict risk monitoring model. This model identifies which material properties or processing conditions may lead to crushing conflicts. Once potential risks are identified, a conflict risk indicator is generated. This conflict risk indicator is then input into the crushing optimization controller. This controller, with a preset acceptable risk level, performs optimization calculations based on the input indicators and outputs optimal crushing control parameters. Finally, the crushing optimization controller adjusts the feeder-crushing mechanism's operation based on these optimized parameters, ensuring efficient production while minimizing the risk of crushing conflicts and improving the safety and stability of the entire process.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0010] Figure 1 1 is a flow chart of a rubber intelligent manufacturing optimization control method based on multi-source monitoring data in one embodiment;

[0011] Figure 2 Schematic diagram of the process of training a crushing conflict hidden danger monitoring model for a rubber intelligent manufacturing optimization control method based on multi-source monitoring data in one embodiment. DETAILED DESCRIPTION

[0012] The embodiment of the present application provides a rubber intelligent manufacturing optimization control method based on multi-source monitoring data to solve the technical problem of component jamming and crushing safety hazards caused by splashing and collision of crushed particles due to uniform and unchanging processing force when the feeding and crushing mechanism in the rubber manufacturing process processes materials with higher hardness.

[0013] The following will be combined with the accompanying drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only some of the embodiments of this application, not all of them. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0014] It should be noted that the terms "including" and "having" are intended to cover non-exclusive inclusions. For example, a process, method, system, product or server that includes a series of steps or units is not necessarily limited to those steps or units clearly listed, but may include other steps or modules that are not clearly listed or are inherent to these processes, methods, products or devices. Example 1

[0015] like Figure 1 As shown, the present application provides a rubber intelligent manufacturing optimization control method based on multi-source monitoring data, the method comprising:

[0016] Feeding and crushing mechanism of interactive rubber processing equipment.

[0017] In the rubber manufacturing industry, the feeder crushing mechanism plays a crucial role. It's an essential step in the rubber product production process, responsible for finely crushing and mixing various raw materials to provide a uniform supply for subsequent processes. However, with increasing production demands, the diversity of raw materials is becoming increasingly prominent. However, the use of a uniform processing force to handle this diverse range of materials can lead to insufficient processing force, resulting in particle splashing and collisions during the crushing process. This wastes raw materials and can also cause equipment components to jam and damage, seriously impacting production continuity and safety.

[0018] In the embodiments of the present application, the feeding and pulverizing mechanism is a key component of the rubber processing equipment. It is responsible for receiving and processing rubber raw materials or other materials, and mechanically pulverizing them into the desired particle size. The system terminal interacts with the feeding and pulverizing mechanism via wireless communication. During this interaction, the feeding and pulverizing mechanism receives signals from the system terminal and pulverizes the raw materials.

[0019] Identify the material entering the feeding and crushing mechanism, and obtain the feeding property information, wherein the feeding property information includes the feeding physical properties, the feeding chemical properties and the feeding crushing properties.

[0020] In one embodiment, during the rubber processing process, the system terminal identifies the material entering the feed and crushing mechanism. This identification process not only confirms the material type but also requires obtaining detailed information about the material's properties. This information includes the material's physical properties, chemical properties, and crushing characteristics. Physical properties describe the material's basic physical properties, such as hardness, density, and shape. Understanding these physical properties helps the system terminal determine the force, speed, and method of crushing used by the feed and crushing mechanism to ensure effective and efficient crushing. Chemical properties describe the material's chemical composition, stability, and potential chemical reactions with other substances. This information is crucial for preventing undesirable chemical changes or the generation of harmful substances during the crushing process. Crushing characteristics describe potential characteristics of the material during the crushing process, such as its ease of crushing, dust formation, and equipment clogging. Understanding these characteristics helps the system terminal optimize operating parameters to minimize clogging, improve crushing efficiency, and reduce dust generation. Identifying these characteristics enables more precise control of the crushing process, ensuring safety and efficiency in rubber processing.

[0021] Multi-source monitoring data of the material feeding and crushing mechanism is acquired, and a material feeding and crushing simulation model of the material feeding and crushing mechanism is generated using the multi-source monitoring data.

[0022] In one embodiment, during the rubber processing process, to optimize the efficiency and safety of the feed and pulverization mechanism, the system terminal acquires multi-source monitoring data from the mechanism. This data, derived from various sensors and monitoring devices, includes mechanical motion parameters, temperature, pressure, vibration, and more. This data provides the system terminal with comprehensive information on the real-time operating status of the feed and pulverization mechanism. Subsequently, key features relevant to the feed and pulverization performance are extracted from the multi-source monitoring data. The physical framework of the simulation model is then designed based on the physical structure and operating principle of the feed and pulverization mechanism. For example, if the feed and pulverization mechanism uses rotating blades for pulverization, the simulation model needs to simulate the blade's rotational motion and the interaction between the material and the blade. Once the physical framework is designed, the system terminal establishes a corresponding mathematical framework based on the physical framework. This includes mathematical equations and formulas describing processes such as material flow, blade cutting, and energy transfer. The system terminal then sets simulation model parameters based on the extracted key features. These parameters include blade rotation speed, feed rate, and material hardness. The simulation model's boundary conditions are then determined, such as the location, shape, and size of the material inlet and outlet, as well as the simulation time. An appropriate algorithm is then selected for simulation calculations based on the complexity of the simulation model and the computing resources available. For example, the discrete element method (DEM) is used to simulate the movement and interaction of material particles, or computational fluid dynamics (CFD) is used to simulate material flow. After the parameters and algorithms are set, the system terminal verifies and calibrates the simulation model using historical operating data from the feed and pulverizer. This involves comparing the simulation results with actual production data to verify the model's predictive accuracy and reliability. If errors or deficiencies are found in the model, the model is adjusted and optimized to improve its predictive performance. This adjustment process includes modifying the model structure and verifying the appropriateness of the model parameter settings. Otherwise, the system terminal outputs the current simulation model to generate a feed and pulverizer simulation model. This model simulates the actual feed and pulverizer operation process and can reflect the impact of different operating parameters and material properties on the pulverization effect. By adjusting the model parameters, the pulverization effect under different conditions can be predicted, providing guidance for actual production.

[0023] Furthermore, the present application provides that after obtaining the feeding and crushing simulation model, the method further includes:

[0024] According to the feeding and crushing simulation model, the internal space of the crushing chamber is identified; the dangerous area of ​​the internal space of the crushing chamber is marked, and the marked area is output; the feeding and crushing simulation model is called to simulate with the feeding attribute sample data set, the crushing damage data of the marked area is monitored, and the crushing monitoring sample data set is generated.

[0025] Optionally, during the feed milling process, to ensure production safety and efficiency, the system terminal first identifies critical areas within the feed milling equipment, namely the internal space within the milling chamber, within the feed milling simulation model. This space is the primary location for material crushing and is also a potential hazardous area. The system terminal then identifies hazardous areas within the milling chamber. This process, based on the equipment's physical structure, operating principles, and past experience, identifies areas prone to damage or safety incidents due to material impact, blade wear, or equipment failure. These areas are specifically marked, forming designated areas. Once designated areas are identified, the system terminal performs a simulation using the established feed milling simulation model. This simulation uses a sample data set of feed properties as input. This data set contains various possible feed conditions, such as material type, particle size, humidity, temperature, and feed rate. These properties are key factors influencing the milling process. Through simulation, the system terminal monitors damage data in these designated areas during the milling process, such as wear, temperature changes, and pressure distribution. The system terminal then compiles the monitored data into a milling monitoring sample data set. This dataset not only records various data from the simulation process but also reflects the actual operating conditions and potential risks in the identified areas. By analyzing this dataset, we can further understand the operating performance of the internal space within the crushing chamber, achieve comprehensive monitoring and optimization of the feed crushing process, and improve production safety and efficiency.

[0026] The feeding and crushing simulation model is called to train a crushing conflict hidden danger monitoring model, and the feeding attribute information is identified using the crushing conflict hidden danger monitoring model to obtain a first conflict hidden danger indicator.

[0027] In one embodiment, during the material-adding and pulverizing production process, to prevent and mitigate potential conflict risks, the system terminal utilizes an established material-adding and pulverizing simulation model to train a pulverizing conflict risk monitoring model. Specifically, the system terminal uses the material-adding and pulverizing simulation model to simulate various material-adding scenarios and collects a sample data set of pulverizing monitoring samples generated in these scenarios. This pulverizing monitoring sample data set is then assigned a conflict risk level based on expert advice and historical experience. The system terminal then uses this data to train the pulverizing conflict risk monitoring model. After training, the system terminal develops a pulverizing conflict risk monitoring model capable of identifying potential conflict risks within material-adding attribute information. The system terminal then inputs the material-adding attribute information into this model, which analyzes and determines this information based on its learned knowledge and outputs a first conflict risk indicator. These indicators intuitively reflect the potential conflict risk level under current material-adding conditions, enabling the system terminal to take timely measures to avoid or mitigate conflict risks, thereby improving safety and stability during the production process.

[0028] Further, if Figure 2 As shown, the present application provides a method for calling the feeding and crushing simulation model to train a crushing conflict hidden danger monitoring model, and the method further includes:

[0029] A feeding attribute sample data set is set, and the feeding and crushing simulation model is called to simulate with the feeding attribute sample data set to obtain a crushing monitoring sample data set corresponding to the feeding attribute sample data set; the feeding attribute sample data set and the crushing monitoring sample data set as well as identification information for identifying the degree of conflict hidden danger are used for training, and a crushing conflict hidden danger monitoring model is output; and a first conflict hidden danger indicator is output according to the crushing conflict hidden danger monitoring model.

[0030] Preferably, to build a model capable of monitoring potential conflicts during the charging and comminution process, the system terminal first obtains a sample data set of charging attributes and, using the same method described above, runs a simulation run using the charging and comminution simulation model to generate a comminution monitoring sample data set. The system terminal then combines the charging attribute sample data set, the comminution monitoring sample data set, and the identification information indicating the degree of potential conflicts to form a complete training data set. This training data set is then divided into a training set, a validation set, and a test set. The number of nodes in the input, hidden, and output layers of a multilayer perceptron (MLP) is then determined based on the complexity of the problem. The number of input layer nodes should match the number of features in the charging attributes and comminution monitoring data, and the number of output layer nodes should match the classification of potential conflicts. The system terminal then randomly initializes the weights and biases of the neural network and iteratively trains using the training set. During training, the system terminal randomly extracts a batch of sample data from the training set each time and inputs this data into the neural network, calculating the output through forward propagation. A loss function is then used to calculate the difference between the output and the identification information indicating the degree of potential conflicts. Furthermore, a backpropagation algorithm is used to calculate the gradient, and an optimizer is used to update the network parameters. Repeat the above steps until the preset number of training rounds is reached. After each training iteration, the system terminal uses the validation set to evaluate the performance of the model, which is done by calculating the accuracy on the validation set. If the accuracy on the validation set begins to decrease, it means that overfitting has occurred, and training should be stopped to prevent overfitting. After all training is completed, the system terminal uses the test set to evaluate the final performance of the model. If the performance meets the requirements, the current model is output to generate a crushing conflict hidden danger monitoring model. Otherwise, the weights and biases are reinitialized, or the structure of the multi-layer perceptron is adjusted, and retraining is performed until the performance meets the requirements. When there is feed attribute data input, the crushing conflict hidden danger monitoring model can quickly analyze this data and output a first conflict hidden danger indicator to avoid or reduce the occurrence of conflict hidden dangers.

[0031] Furthermore, the present application provides an output crushing conflict hidden danger monitoring model, and the method further includes:

[0032] A plurality of crushing granularity samples are set; a first feedback network layer is generated according to the plurality of crushing granularity samples; feedback training is performed on the crushing conflict hidden danger monitoring model according to the first feedback network layer to obtain an optimized crushing conflict hidden danger monitoring model.

[0033] Optionally, when building a pulverization conflict potential monitoring model, the system terminal first sets up multiple samples of different pulverization particle sizes. These samples represent different pulverization results, namely, the different particle sizes achieved during the pulverization process. Particle size directly affects the pulverization efficiency and equipment performance, and is therefore a key factor in assessing pulverization conflict potentials. The system terminal then designs a feedback network layer using an Elman neural network, as it can process time-dependent data and capture the impact of particle size changes on equipment status. This feedback network layer generates a corresponding feedback signal based on the input pulverization particle size samples. The parameters of this feedback network layer are then initialized. These parameters, including weights and biases, are adjusted during the training process to optimize network performance. The system terminal then trains the feedback network layer using the multiple pulverization particle size samples as input data. The output is calculated through forward propagation, and a loss function is used to calculate the difference between the predicted results and the actual labels. An optimizer is then used to update the network parameters to reduce the prediction error. The training process is repeated until the preset number of training rounds is reached. The system terminal then connects the trained feedback network layer to the pulverization conflict potential monitoring model. Ensure that the shattering conflict potential monitoring model can receive input from the feedback network layer and adjust its output accordingly. Furthermore, the system terminal determines how the feedback network layer provides feedback to the shattering conflict potential monitoring model, using the feedback network layer's output as one of the monitoring model's inputs. For each shattering granularity sample, the system terminal inputs it into the feedback network layer. Based on the input granularity sample and its internal parameters, the feedback network layer calculates and outputs a feedback signal. This signal reflects the potential conflict potential or changes in device status at the current granularity. The feedback signal generated by the feedback network layer is then used to correct and optimize the monitoring model's output. By adjusting the monitoring model's parameters and structure, the model can better utilize the feedback signal to predict and identify conflict potentials. The feedback training process is repeated to gradually optimize the monitoring model's performance. In each iteration, the system terminal evaluates the model's performance based on the feedback signal and the monitoring model's predictions, and adjusts the model parameters to improve prediction results. The optimized model is then validated using an independent validation dataset to evaluate its performance in real-world applications. If the model performance meets the requirements, the optimized shattering conflict potential monitoring model is output. This model not only considers the impact of feed properties on the pulverization process but also incorporates the effect of particle size variations on potential collisions, enabling a more comprehensive and accurate assessment of potential collisions during the pulverization process. The optimized model can more accurately predict and identify potential collisions at different particle sizes, improving the safety and stability of the pulverization process.

[0034] Obtain the required crushing granularity, and the optimized crushing conflict hidden danger monitoring model performs feedback optimization on the first conflict hidden danger indicator according to the required crushing granularity, and outputs a second conflict hidden danger indicator; and inputs the second conflict hidden danger indicator into the crushing optimization controller for control optimization.

[0035] Optionally, during the pulverization production process, the system terminal determines the required product pulverization particle size, a key parameter in the production process. To ensure the safety and efficiency of the production process, the system terminal utilizes an optimized pulverization conflict risk monitoring model to monitor potential conflict risks. Specifically, the system terminal inputs the required pulverization particle size into the first feedback network layer to generate a feedback signal. This feedback signal is then combined with the first conflict risk indicator and input into the pulverization conflict risk monitoring model. Based on the knowledge learned during the feedback training process, the pulverization conflict risk monitoring model uses the input feedback signal to perform feedback optimization on the first conflict risk indicator, thereby obtaining a second conflict risk indicator that is more accurate and meets current production requirements. The system terminal then inputs this second conflict risk indicator into the pulverization optimization controller. Based on this indicator, combined with current production conditions and equipment status, the controller performs control optimization, adjusting and optimizing production parameters such as feeding rate and pulverizer speed to minimize conflict risks and improve production safety and efficiency.

[0036] Furthermore, the present application provides that after training the crushing conflict hidden danger monitoring model, the method further includes:

[0037] A temperature monitoring module is set up, and temperature monitoring data is obtained according to the temperature monitoring module; a temperature impact analysis is performed based on the temperature monitoring data and the added material attribute information to obtain a temperature impact index; a second feedback network layer is established based on the temperature impact index, and feedback optimization is performed on the crushing conflict hidden danger monitoring model based on the second feedback network layer.

[0038] Preferably, after the training of the crushing conflict hidden danger monitoring model is completed, the system terminal sets up the temperature monitoring module. This step is an important step to ensure the stability and safety of the crushing process. Through the temperature monitoring module, the system terminal obtains the temperature data of the crushing process in real time. These data are crucial for analyzing the impact of temperature on the crushing process. After obtaining the temperature monitoring data, the system terminal organizes the temperature monitoring data and the feeding attribute information to ensure that their timestamps match. Subsequently, a preliminary analysis is performed on the temperature data and the feeding attribute data to understand their changing trends and ranges. Among them, the changes in temperature under different feeding attributes and whether the temperature fluctuates significantly when certain attributes change are analyzed in particular. Afterwards, based on the preliminary analysis results, the system terminal calculates the average value and fluctuation rate of the temperature at different feeding attribute levels as the temperature impact index. Then, using the same method as the aforementioned construction of the first feedback network layer, a second feedback network layer is constructed based on the temperature impact index. The role of this network layer is to provide feedback and optimization to the crushing conflict hidden danger monitoring model based on the temperature impact index. Finally, the system terminal performs the same feedback optimization operation again. Through continuous learning and adjustment, it further optimizes the crushing conflict hidden danger monitoring model to ensure that the production process is carried out within a controllable temperature range and reduce the conflict hidden dangers caused by temperature changes.

[0039] The first conflict hidden danger index is input into a crushing optimization controller for control optimization, and crushing optimization control parameters are output, wherein the crushing optimization controller includes a preset conflict hidden danger index.

[0040] In one embodiment, to ensure production safety and efficiency, after detecting potential conflict risks and calculating a first conflict risk indicator, the system terminal inputs this first conflict risk indicator into a pulverization optimization controller for control optimization. The pulverization optimization controller includes a simulation component that contains simulation models of key equipment, such as the crushing roller assembly. This simulation component can simulate various conditions in the actual production process, including key indicators such as crushing performance, energy consumption, and temperature under different production parameters. When the first conflict risk indicator is input into the pulverization optimization controller, it adjusts production parameters such as the crushing roller speed and feed rate based on this indicator to generate pulverization control parameters. The pulverization optimization controller then performs multiple simulation runs using the internal simulation component. These simulation runs evaluate the degree of conflict risk under different parameter combinations, as well as other key indicators such as production efficiency and energy consumption. Through these simulation evaluations, the pulverization optimization controller finds a set of control parameters that meets the preset conflict risk indicators. This set of parameters can optimize production efficiency while ensuring production safety. Once these control parameters are found, the pulverization optimization controller outputs them to the system terminal to generate the optimized pulverization control parameters.

[0041] Furthermore, the present application provides inputting the first conflict potential indicator into a crushing optimization controller to perform control optimization and output crushing optimization control parameters, and the method further includes:

[0042] The pulverizing optimization controller includes a crushing roller group, and the crushing roller group includes a first crushing roller and a second crushing roller; the first conflict hidden danger index is input into the pulverizing optimization controller, and the first crushing roller and the second crushing roller are controlled and optimized with the preset conflict hidden danger index as the adaptation target, and the pulverizing optimization control parameters are output; wherein the parameter indicators of the pulverizing optimization control parameters include feed speed, feed amount and roller group rotation speed.

[0043] Optionally, to ensure operational safety and efficiency, the system terminal uses a crushing optimization controller to manage and optimize the operation of the crushing roller assembly. The crushing roller assembly within this crushing optimization controller also includes a first crushing roller and a second crushing roller. When the system terminal inputs the first conflict risk indicator into the crushing optimization controller, it compares the input first conflict risk indicator with the preset conflict risk indicator within the controller to determine whether the current conflict risk is within an acceptable range. If the first conflict risk indicator exceeds the preset range, the controller initiates a control optimization process. First, the controller adjusts parameters related to the crushing roller assembly operation, such as feed rate, feed volume, and the rotational speeds of the first and second crushing rollers. After each parameter adjustment, the controller uses an internal simulation component to simulate and evaluate the operation of the first and second crushing rollers. This simulation process considers various factors, such as material properties and mechanical efficiency, to predict the impact of the adjusted parameter combination on the conflict risk indicator. The controller then calculates the new conflict risk indicator using the crushing conflict risk monitoring model. Based on the simulation evaluation results, the controller determines whether the current parameter combination meets the preset conflict risk indicator requirements. If the conditions are not met, the controller adjusts the parameters again and repeats the simulation and evaluation process until a satisfactory parameter combination is found. Once the controller finds a parameter combination that meets the pre-set conflict risk indicators, the system terminal outputs this set of parameters as the crushing optimization control parameters. This set of parameters will directly guide the operation of the first and second crushing rollers in actual production.

[0044] The pulverization optimization controller controls the feeding and pulverizing mechanism according to the pulverization optimization control parameters.

[0045] In one embodiment, after generating optimized pulverization control parameters, the pulverization optimization controller directly applies these parameters to control the operation of the feed pulverization mechanism. Specifically, the pulverization optimization controller directly adjusts the operation of the feed pulverization mechanism based on the output optimized pulverization control parameters, such as feed speed, feed amount, and the rotational speed of the crushing roller group. For example, if the control parameters indicate a need to reduce the feed speed or feed amount, the controller will send corresponding instructions to the feed pulverization mechanism to ensure that the actual feed amount matches the preset value. Similarly, if the control parameters require adjustment of the rotational speed of the crushing roller group, the controller will also adjust the output of the feed pulverization mechanism accordingly to achieve precise control of the rotational speed.

[0046] Furthermore, the present application provides that the feeding and crushing mechanism also includes a material drying device, and the method further includes:

[0047] Identify the material viscosity index based on the feed attribute information; if the material viscosity index is greater than a preset viscosity index, train a crushing blockage hazard monitoring model based on the feed crushing simulation model;

[0048] Preferably, the system terminal identifies the viscosity index of the material based on the feeding attribute information. This viscosity index describes the degree of viscosity of the material under specific conditions, and is of great significance for predicting whether the material is prone to clogging during the crushing process. If the identified material viscosity index is greater than the preset viscosity index, this means that the material has a higher risk of clogging during the crushing process. In order to discover and prevent this clogging hazard in advance, the system terminal uses the feeding and crushing simulation model to train a crushing clogging hazard monitoring model. This training process is similar to the aforementioned training of the crushing conflict hazard monitoring model. A crushing clogging hazard monitoring model is trained using the feeding attribute sample data set and the crushing monitoring sample data set as well as identification information that identifies the degree of clogging hazard. This crushing clogging hazard monitoring model can learn the mapping relationship between the material viscosity index and the clogging hazard. Once the training is completed, the crushing clogging hazard monitoring model can monitor the viscosity index of the material in real time and predict whether there is a clogging hazard.

[0049] The feed attribute information is identified using the crushing blockage hazard monitoring model to obtain a first blockage hazard index; if the first blockage hazard index is greater than a preset blockage hazard index, the material drying equipment is activated to dry the material.

[0050] Preferably, the system terminal inputs the material's feeding attribute information into the crushing blockage hazard monitoring model, and the crushing blockage hazard monitoring model identifies and analyzes this information based on the learned knowledge, and outputs a first blockage hazard index. This indicator directly reflects the possibility of the material being blocked during the crushing process. Subsequently, this first blockage hazard index is compared with the preset blockage hazard index. If the first blockage hazard index is greater than the preset blockage hazard index, it means that the material has a high risk of blockage, and immediate measures need to be taken to reduce this risk. At this point, the system terminal activates the material drying equipment to dry the material. The drying process can reduce the viscosity of the material, thereby reducing the possibility of blockage during the crushing process.

[0051] Furthermore, the present application provides activating the material drying device to dry the material, the method further comprising:

[0052] A safety viscosity index is generated with the difference between the first clogging hazard index and the preset clogging hazard index as a target; the safety viscosity index is converted into a moisture content, and a preset moisture content is output. The preset moisture content is used as the drying target of the material drying equipment for drying, and after drying, the material falls into the crushing chamber of the feeding and crushing mechanism.

[0053] Optionally, to prevent blockage risks during the crushing process, the system terminal first calculates the difference between the first blockage risk indicator and a preset blockage risk indicator. This difference is used to generate a safe viscosity index. This safe viscosity index is an intermediate value that represents the viscosity at which the material can be safely crushed without blockage. The system then converts this safe viscosity index to moisture content. This is because the viscosity of a material is often closely related to its moisture content. Through moisture content conversion, the system terminal can determine a specific preset moisture content that can be used to guide drying operations. This conversion is based on a viscosity index-moisture content correlation database, established based on experimental and historical data, which records viscosity indexes and corresponding moisture contents. The system terminal then uses this preset moisture content as the drying target for the material drying equipment and performs drying. When the material reaches the preset moisture content, the drying process is complete and the material is fed into the crushing chamber of the feed crushing mechanism for crushing.

[0054] In summary, the embodiments of the present application have at least the following technical effects:

[0055] The present embodiment collects and analyzes feed attribute information, combines it with multi-source monitoring data to establish a feed pulverization simulation model, and then trains a pulverization conflict risk monitoring model. The pulverization conflict risk monitoring model identifies material conflict risk indicators. Based on these conflict risk indicators, a pulverization optimization controller performs control optimization and outputs optimized pulverization control parameters to achieve precise control of the feed pulverization mechanism. Furthermore, when the material viscosity exceeds a preset indicator, the material drying device is activated. By calculating a safe viscosity indicator and converting it to a preset moisture content, the material is dried to reduce the risk of clogging. When optimizing the monitoring model, not only is a feed attribute sample dataset used, but also different pulverization particle size samples are set up, and a feedback network layer is constructed to train and optimize the model for feedback, thereby improving monitoring accuracy. These technical effects collectively address the technical issues of component jamming and pulverization safety hazards caused by uniform processing force when the feed pulverization mechanism processes hard materials in rubber manufacturing. This improves the performance and safety of the feed pulverization mechanism by enabling flexible and adaptive control parameter adjustment based on the feed type.

[0056] It should be noted that the order in which the embodiments of the present application are presented is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. Furthermore, the foregoing descriptions of specific embodiments of this specification are provided. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential sequence shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0057] The above description is only a preferred embodiment of the present application and is not intended to limit the present application. Any modifications, equivalent replacements, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

[0058] This specification and drawings are merely illustrative of the present application and are intended to cover any and all modifications, variations, combinations, or equivalents within the scope of this application. Obviously, those skilled in the art may make various modifications and variations to this application without departing from the scope of this application. Thus, this application is intended to include such modifications and variations as fall within the scope of this application and its equivalents.

Claims

1. A rubber intelligent manufacturing optimization control method based on multi-source monitoring data, characterized in that: The method comprises: Feeding and crushing mechanism of interactive rubber processing equipment; Identifying the material entering the feeding and crushing mechanism and obtaining the material attribute information, wherein the material attribute information includes the material physical properties, the material chemical properties and the material crushing properties; Acquiring multi-source monitoring data of the material feeding and crushing mechanism, and generating a material feeding and crushing simulation model of the material feeding and crushing mechanism using the multi-source monitoring data; calling the charging and crushing simulation model to train a crushing conflict hidden danger monitoring model, and using the crushing conflict hidden danger monitoring model to identify the charging attribute information to obtain a first conflict hidden danger indicator; Inputting the first conflict hidden danger indicator into a crushing optimization controller to perform control optimization and output crushing optimization control parameters, wherein the crushing optimization controller includes a preset conflict hidden danger indicator; The pulverization optimization controller controls the feeding and pulverizing mechanism according to the pulverization optimization control parameters; The method of calling the feeding and crushing simulation model to train the crushing conflict hidden danger monitoring model includes: Setting a material feeding attribute sample data set, calling the material feeding and crushing simulation model to simulate the material feeding attribute sample data set, and obtaining a crushing monitoring sample data set corresponding to the material feeding attribute sample data set; Using the feed attribute sample data set and the crushing monitoring sample data set as well as identification information identifying the degree of conflict hidden dangers for training, outputting a crushing conflict hidden danger monitoring model; Outputting a first conflict potential risk indicator according to the crushing conflict potential risk monitoring model; Outputting a crushing conflict hidden danger monitoring model, the method further includes: Set up multiple crushing particle size samples; generating a first feedback network layer according to the plurality of crushing granularity samples, and performing feedback training on the crushing conflict hidden danger monitoring model according to the first feedback network layer to obtain an optimized crushing conflict hidden danger monitoring model; Obtaining a demand crushing granularity, and using the optimized crushing conflict hidden danger monitoring model to perform feedback optimization on the first conflict hidden danger indicator according to the demand crushing granularity, and outputting a second conflict hidden danger indicator; The second conflict potential risk index is input into the crushing optimization controller to perform control optimization.

2. The method according to claim 1, wherein After obtaining the feeding and crushing simulation model, the method further includes: identifying the internal space of the pulverizing chamber according to the feeding and pulverizing simulation model; Marking the dangerous area of ​​the internal space of the crushing chamber and outputting the marked area; The feed crushing simulation model is called to perform simulation using the feed attribute sample data set, and crushing damage data of the marked area is monitored to generate a crushing monitoring sample data set.

3. The method according to claim 1, wherein The feeding and crushing mechanism also includes a material drying device, including: Identifying the material viscosity index based on the material attribute information; If the material viscosity index is greater than a preset viscosity index, a crushing blockage hidden danger monitoring model is trained according to the material adding and crushing simulation model; Using the crushing blockage hazard monitoring model to identify the feed attribute information, and obtain a first blockage hazard indicator; If the first clogging potential risk index is greater than a preset clogging potential risk index, the material drying equipment is activated to dry the material.

4. The method according to claim 3, wherein Activating the material drying equipment to dry the material, the method comprising: generating a safety viscosity index based on the difference between the first blocking hazard index and the preset blocking hazard index; The safety viscosity index is converted into a moisture content, and a preset moisture content is output. The preset moisture content is used as the drying target of the material drying equipment for drying, and after drying, the material falls into the grinding chamber of the feeding and grinding mechanism.

5. The method according to claim 1, wherein The method includes: inputting the first conflict hidden danger indicator into a crushing optimization controller to perform control optimization and outputting crushing optimization control parameters. The pulverization optimization controller includes a crushing roller group, and the crushing roller group includes a first crushing roller and a second crushing roller; inputting the first conflict hidden danger index into the crushing optimization controller, optimizing the control of the first crushing roller and the second crushing roller with the preset conflict hidden danger index as the adaptation target, and outputting crushing optimization control parameters; The parameter indicators of the crushing optimization control parameters include feed speed, feed amount and roller group rotation speed.

6. The method according to claim 1, wherein After training the crushing conflict hidden danger monitoring model, the method further includes: Setting a temperature monitoring module and obtaining temperature monitoring data according to the temperature monitoring module; Performing temperature impact analysis based on the temperature monitoring data and the added material attribute information to obtain a temperature impact index; A second feedback network layer is established according to the temperature impact index, and feedback optimization is performed on the crushing conflict hidden danger monitoring model according to the second feedback network layer.

Images