Automatic mixing control method and system for rubber and plastic products

Abstract

The application relates to the technical field of production control, and discloses an automatic mixing regulation method and system for rubber and plastic products, which comprises the following steps: collecting mixing environment data, mixing formula data and corresponding rubber and plastic product performance, inputting the data into an artificial intelligence model for learning, and outputting an initial optimal mixing scheme; before the initial optimal mixing scheme is executed, the actual quantity of each material used for producing a target rubber and plastic product and the total moisture content in the material are measured, measurement data is obtained, the performance of the target rubber and plastic product is predicted, and an adjustment instruction for the initial optimal mixing scheme is generated; based on the adjustment instruction, a secondary optimal mixing scheme and a regulation signal for controlling a mixing device are generated; and the mixing device automatically controls the mixing device to execute a mixing operation according to the secondary optimal mixing scheme based on the regulation signal; and the application can improve the quality of rubber and plastic products.

Description

Technical Field

[0001] This application relates to the field of production control technology, and in particular to an automatic mixing and control method and system for rubber and plastic products. Background Technology

[0002] In the production of rubber and plastic products, the mixing process is one of the key steps determining product performance and quality. Traditional rubber and plastic mixing processes mainly rely on manual experience and mechanical parameter-based feeding. Given the frequent changes in raw material types, operating conditions, and equipment status, this process suffers from technical pain points such as poor control precision, large quality fluctuations, high energy consumption, and significant material waste. Especially when facing the demands of manufacturing multi-category, small-batch, and rapidly changeover rubber and plastic products, existing control methods are insufficient to meet the technological upgrade requirements of flexible production and intelligent manufacturing.

[0003] A similar prior art is Chinese patent application CN118884812A, which provides a control system and method for automatic mixing of materials used in premix production. This system integrates modules for proportion prediction, feeding metering and control, mixing control, and quality detection to achieve precise prediction and adjustment of raw material supply, optimized control of the mixing process, and real-time and offline detection of finished product quality. The proportion prediction unit predicts the raw material proportions and feeding order based on historical data; the feeding metering unit and feeding control unit accurately calculate and control the raw material feeding amount and rate in real time; the mixing control module uses genetic algorithms and fuzzy PID control to optimize mixing parameters; the quality detection module ensures product quality and provides feedback for optimized proportions through online and offline detection; and an integrated interactive module manages the operation of each module and displays the results. However, this application does not consider the real-time changes in the moisture content of the materials during mixing, and therefore cannot accurately control the actual amount of materials fed.

[0004] Similar prior art includes Chinese patent application CN119828474A, which discloses a method, apparatus, equipment, and storage medium for controlling powder mixing and processing. The method includes: real-time acquisition of target data during the powder mixing and processing process; inputting the target data into a prediction model to obtain probability distribution predictions of the state of the mixing production line at several future time steps, wherein the prediction model is pre-constructed using a dynamic Bayesian network based on the mixing production line; optimizing the real-time control strategy using a model predictive control algorithm based on the probability distribution predictions and the target data to obtain control commands; and controlling the mixing production line according to the control commands. However, this application lacks dynamic prediction and adaptive control, and cannot accurately control the actual amount of material added when the moisture content changes in real time.

[0005] Therefore, this application provides an automatic mixing and control method and system for rubber and plastic products. Summary of the Invention

[0006] To address the aforementioned technical issues, this application provides an automatic mixing and control method and system for rubber and plastic products. By accurately measuring and feeding forward compensation for key disturbances such as material moisture content and quantity, it achieves intelligent and precise control over the quality of rubber and plastic products.

[0007] In a first aspect, this application provides an automatic mixing and control method for rubber and plastic products, the method comprising: Collect mixing environment data, mixing formula data and corresponding rubber and plastic product performance, input them into an artificial intelligence model for learning, and the artificial intelligence model outputs the optimal ratio data of rubber and plastic products and the operating parameters of the mixing equipment based on the learning results to form an initial optimal mixing scheme. Before executing the initial optimal mixing scheme, the actual quantity of each material used in the production of the target rubber and plastic product and the total moisture content in the material are measured and the measurement data are obtained. The artificial intelligence model predicts the performance of the target rubber and plastic product based on the measurement data, and generates adjustment instructions for the initial optimal mixing scheme based on the predicted performance of the target rubber and plastic product. Based on the adjustment command, a secondary optimal mixing scheme and a control signal for controlling the mixing device are generated. Based on the control signal, the mixing device automatically controls itself to perform mixing operations according to the secondary optimal mixing scheme.

[0008] In conjunction with the first aspect, in the first implementation of the first aspect of this application, during the execution of the mixing operation, multiple sensors deployed inside and outside the mixing device collect mixing status data in real time and feed it back to the artificial intelligence model, outputting a new mixing scheme and generating a new control signal to adjust the current mixing operation process.

[0009] In conjunction with the first aspect, in a second implementation of the first aspect of this application, measuring the total moisture content in the material includes: The moisture content of the first material is obtained by measuring the surface moisture of the first material using a first method. The first material is photographed to obtain an image of the first material, and the image of the first material is analyzed to obtain the second moisture content of the first material; The temperature change of the first material before mixing is measured by a second method, and the third moisture content of the first material is estimated based on the temperature change. The total moisture content of the first material is estimated based on the first moisture content, the second moisture content, and the third moisture content of the first material.

[0010] In conjunction with the first aspect, in the third implementation of the first aspect of this application, estimating the third moisture content of the first material includes: At one end of the conveyor belt, a first temperature of the first material is measured by a first temperature sensor, and a heating device installed on the conveyor belt is activated to start heating the first material. At the other end of the conveyor belt, a second temperature of the first material is measured by a second temperature sensor. Calculate the temperature difference between the second temperature and the first temperature, and estimate the third moisture content of the first material based on a preset relationship between the material temperature difference and moisture content.

[0011] In conjunction with the first aspect, in the fourth implementation of the first aspect of this application, measuring the actual quantity of each material dispensed includes: An image capturing device is installed on the mixing device. The image capturing device captures a second material image of the material during the mixing process in real time, and performs preprocessing and normalization processing. The first feature and the second feature are extracted from the second material image based on a convolutional neural network. The first feature includes edges and texture, and the second feature includes particle shape, size and number. Based on the first feature and the second feature, the material type and the quantity of material to be delivered are calculated to obtain each material type and its corresponding actual quantity to be delivered. Each material type and its corresponding actual quantity are compared with the target optimal ratio data, and the comparison result is obtained. If the comparison result does not match the target optimal ratio data, the quantity of each material is recalculated and transmitted to the mixing equipment through a control signal to automatically adjust the quantity of subsequent materials.

[0012] In conjunction with the first aspect, in the fifth implementation of the first aspect of this application, the mixing operation further includes: Obtain the current value of the motor in the mixing equipment, where the current value represents the load status of the mixing equipment; The machine learning model predicts the performance of the mixed rubber and plastic products based on the relationship between the historical current values ​​of the mixing equipment and the optimal ratio data. It also outputs the predicted performance of the target rubber and plastic products by learning the relationship between the current values ​​and the performance of the rubber and plastic products. If the predicted performance of the target rubber and plastic product does not meet the design requirements, the operating parameters of the mixing equipment are controlled, and the amount of material added is adjusted. The operating parameters include increasing or decreasing moisture and increasing or decreasing additives.

[0013] In conjunction with the first aspect, in the sixth implementation of the first aspect of this application, image analysis is performed on the first material image to obtain the second moisture content of the first material, including: The first material image is processed using a convolutional neural network, and material features in the first material image are automatically extracted. The material features include the particle size, shape, color, and texture of the material. A deep learning model is trained based on historical material image data and corresponding moisture content data. The material features are input into the pre-trained deep learning model, and the deep learning model calculates the moisture content of the material features as the second moisture content of the first material. The moisture content is the proportion of moisture in the first material.

[0014] In conjunction with the first aspect, in the seventh implementation of the first aspect of this application, a quadratic optimal mixing scheme is generated, including: Using any one of the liquid materials in the secondary optimal mixing scheme as the third material, the actual quantity of the third material is calculated based on the relationship between the preset operating parameters of the mixing equipment and the quantity of material remaining on the mixing equipment.

[0015] In conjunction with the first aspect, in the eighth implementation of the first aspect of this application, the relationship between the operating parameters of the mixing equipment and the amount of material remaining on the mixing equipment changes over time.

[0016] Secondly, this application provides an automatic mixing and control system for rubber and plastic products, the system comprising: The training unit is used to collect mixing environment data, mixing formula data and corresponding rubber and plastic product performance, and input them into the artificial intelligence model for learning. The artificial intelligence model outputs the optimal ratio data of rubber and plastic products under the current mixing environment and the operating parameters of the mixing equipment based on the learning results, thus forming the initial optimal mixing scheme. An adjustment unit is used to measure the actual quantity of each material used in the production of the target rubber and plastic product and the total moisture content of the material before executing the initial optimal mixing scheme, and to obtain measurement data. The artificial intelligence model predicts the performance of the target rubber and plastic product based on the measurement data, and generates adjustment instructions for the initial optimal mixing scheme based on the predicted performance of the target rubber and plastic product. The mixing unit is used to generate a secondary optimal mixing scheme and a control signal for controlling the mixing device based on the adjustment command. The mixing device automatically controls the mixing device to perform mixing operations according to the secondary optimal mixing scheme based on the control signal.

[0017] Compared with the prior art, the beneficial effects of the present invention are at least as follows: The technical solution provided in this application achieves precise control of the rubber and plastic mixing process through a multi-level dynamic optimization mechanism. Based on an artificial intelligence model, it learns the relationship between the mixing environment, formulation, and product performance to generate an initial optimal mixing scheme. Before execution, multi-dimensional real-time detection, including the actual amount of material added and the calculation of total moisture content using surface moisture sensing, image analysis, and temperature change methods, predicts product performance and dynamically generates adjustment instructions to form a secondary optimal mixing scheme. During the mixing process, a sensor network provides real-time feedback of status data, and the correlation between motor current value monitoring load status and product performance is continuously optimized. This technology improves adaptability to fluctuations in material characteristics, especially real-time changes in moisture content, solving the quality fluctuation problem caused by material addition errors and moisture effects in traditional methods, thus improving the quality of rubber and plastic products. Simultaneously, closed-loop intelligent control reduces manual intervention, lowers material waste and energy consumption, meets the flexible production needs of multiple categories and small batches, and provides effective technical support for the intelligent manufacturing upgrade of the rubber and plastics industry. Attached Figure Description

[0018] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1 This is a schematic diagram of one embodiment of an automatic mixing and control method for rubber and plastic products in this application. Figure 2 This is a schematic diagram of an embodiment of measuring the total moisture content in materials according to this application. Figure 3 This is a schematic diagram of an embodiment of measuring the actual quantity of each material in this application. Figure 4 This is a schematic diagram of one embodiment of an automatic mixing and control system for rubber and plastic products in this application. Detailed Implementation

[0020] This application provides an automatic mixing and control method and system for rubber and plastic products. The terms "first," "second," "third," "fourth," etc. (if present) in the specification, claims, and accompanying drawings of this application are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments described herein can be implemented in a sequence other than that illustrated or described herein. Furthermore, the terms "comprising" or "having" and any variations thereof are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or device that includes a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or devices.

[0021] For ease of understanding, the specific process of the embodiments of this application is described below. Please refer to [link / reference]. Figure 1 One embodiment of the automatic mixing and control method for rubber and plastic products in this application includes: Step S1: Collect mixing environment data, mixing formula data and corresponding rubber and plastic product performance, input them into the artificial intelligence model for learning, and the artificial intelligence model outputs the optimal ratio data of rubber and plastic products under the current mixing environment and the operating parameters of the mixing equipment based on the learning results, thus forming the initial optimal mixing scheme.

[0022] Specifically, environmental data during the mixing process is monitored in real time using sensors such as temperature, humidity, and air pressure sensors. This environmental data includes temperature, humidity, air pressure, and vibration at the mixing site. Environmental data affects material properties and mixing effects, making real-time monitoring of environmental changes essential. Simultaneously, based on the production requirements of the target rubber and plastic products, information such as the required material types, the proportions of each material, and batch numbers is collected and recorded. Furthermore, testing equipment and quality inspection systems are used to acquire mixing environmental data, mixing formula data, and corresponding performance data of the rubber and plastic products. Performance data includes mechanical properties and rheological properties. Mechanical properties include tensile strength and hardness, while rheological properties include viscosity and flowability. The collected data is then input into an artificial intelligence model, which can be a deep neural network, regression analysis model, etc. This model learns from historical data to identify the relationship between environmental factors, material proportions, and product performance. The goal of the artificial intelligence model is to output the optimal mixing proportions and mixing equipment operating parameters to predict the optimal mixing scheme under different environmental conditions, thereby improving the quality of rubber and plastic products.

[0023] Step S2: Before executing the initial optimal mixing scheme, measure the actual quantity of each material used in the production of the target rubber and plastic product and the total moisture content in the material, and obtain the measurement data. The artificial intelligence model predicts the performance of the target rubber and plastic product based on the measurement data, and generates adjustment instructions for the initial optimal mixing scheme based on the predicted performance of the target rubber and plastic product.

[0024] Specifically, the actual quantity of each material is monitored in real time using weighing sensors, flow meters, or image recognition devices installed on the mixing device. Image recognition technology, such as convolutional neural networks, can also be used to obtain information such as the number, size, and shape of material particles from real-time images and compare them with a preset formula to determine material error. Multiple moisture measurement methods are employed, such as surface moisture sensors, temperature change methods, or image analysis, to measure the moisture content of each material in real time. For example, surface moisture sensors can measure the moisture on the surface of rubber, plastics, or other materials, while temperature change methods can estimate the internal moisture content of the material. Accurate measurement of the material's quantity... Controlling the quantity and moisture content of materials ensures precise control of the mixing process, preventing quality fluctuations caused by material errors. All collected data on material quantity and moisture content are recorded and transmitted to an artificial intelligence (AI) model. Based on the measurement data, the AI ​​model analyzes the actual material quantity and moisture content, combining it with formula data to predict the mechanical and rheological properties of the resulting rubber and plastic product. Based on the prediction results, the AI ​​model generates adjustment instructions. If the predicted target performance fails to meet expectations, the model automatically adjusts the material quantity or mixing parameters, such as stirring time and speed. By dynamically predicting the performance of rubber and plastic products and adjusting the mixing ratios, the flexibility of the production process is improved, ensuring that each batch meets design requirements.

[0025] Step S3: Based on the adjustment command, generate a secondary optimal mixing scheme and a control signal for controlling the mixing device. Based on the control signal, the mixing device automatically controls the mixing device to perform mixing operations according to the secondary optimal mixing scheme.

[0026] Specifically, based on the adjustment instructions generated by the artificial intelligence model in step S2 above, the mixing ratio data is recalculated and optimized. For example, if the artificial intelligence model predicts poor performance of the rubber and plastic product, it will require increasing the proportion of a certain material or adjusting the order of material addition, generating a new mixing scheme, namely the secondary optimal mixing scheme. According to the secondary optimal mixing scheme, the artificial intelligence model will generate specific control signals and transmit them to the control system of the mixing device. The control signals typically include, but are not limited to, parameters such as the amount of material added, stirring speed, and mixing time. After receiving the control signals, the mixing device automatically adjusts its operating parameters, such as controlling the material addition rate and adjusting the stirring time. The mixing equipment will execute the mixing operation according to the secondary optimal mixing scheme to ensure that the final mixture meets the quality requirements of the rubber and plastic product. By dynamically adjusting the new optimal mixing scheme, the mixing process can be optimized at any time based on real-time data, reducing manual intervention and achieving precise control of the mixing process, thereby improving the quality of the rubber and plastic product.

[0027] Through the coordination of the above steps, this application achieves intelligent and precise control over the quality of rubber and plastic products.

[0028] It is understood that the executing entity of this application can be an automatic mixing and control system for rubber and plastic products, or it can be a terminal or a server; the specific implementation is not limited here. This application's embodiments use a server as an example for illustration.

[0029] Furthermore, during the mixing operation, multiple sensors deployed inside and outside the mixing device collect mixing status data in real time and feed it back to the artificial intelligence model to output new mixing schemes and generate new control signals to adjust the current mixing operation process.

[0030] Specifically, during the mixing operation, multiple sensors deployed inside and outside the mixing device collect mixing status data in real time. These sensors include, but are not limited to, temperature sensors, humidity sensors, pressure sensors, flow sensors, and acceleration sensors. Various state parameters during the mixing process, such as mixing temperature, humidity, material flowability, and mixing uniformity, are monitored in real time. The data collected by the sensors is transmitted to an artificial intelligence model via a data transmission channel. The AI ​​model, based on the real-time mixed status data, combined with historical data and current mixing environment data, recalculates and predicts the optimal adjustment parameters for the mixing scheme. During the training of the machine learning model used to predict the optimal adjustment parameters, the training data specifically includes the mixing status data collected in real time by the sensors, such as process parameters like mixing temperature, humidity, material flowability, and mixing uniformity; current mixing environment data, such as on-site temperature, humidity, and air pressure; and corresponding historical mixing scheme data. Historical proportions and operating parameters, as output labels, represent the historically optimal adjustment parameters corresponding to these input variables. Specifically, these are operational commands such as material quantity adjustments, stirring speed corrections, and mixing time adjustments that have been validated under specific states and environmental conditions. By using real-time status data, environmental data, and historical scheme data as input features, and the corresponding optimal adjustment parameters as prediction targets, the AI ​​model learns and establishes a mapping relationship between the current mixing state and the optimal control strategy. Through learning feedback information during the mixing process, the AI ​​model can dynamically optimize and adjust parameters such as proportions, feeding sequence, and stirring rate for the current mixing operation. Based on the new mixing scheme output by the model, new control signals are generated and transmitted to the control system of the mixing device. Upon receiving the new control signals, the mixing device automatically adjusts the mixing operation process, such as adjusting control parameters like material feeding rate, stirring time, and mixing temperature, to ensure that the mixing process conforms to the real-time predicted optimal scheme. By combining multi-sensor feedback and AI model optimization, the mixing operation can be monitored and adjusted in real time, ensuring the quality of each batch of mixed products, reducing human intervention, and improving production precision.

[0031] Further, see Figure 2As shown, measuring the total moisture content in a material includes: Step S21: Measuring the surface moisture of a first material using a first method and obtaining the first moisture content of the first material; Step S22: Taking an image of the first material and performing image analysis on the first material image to obtain the second moisture content of the first material; Step S23: Measuring the temperature change of the first material before mixing using a second method and estimating the third moisture content of the first material based on the temperature change; Step S24: Estimating the total moisture content of the first material based on the first moisture content, the second moisture content, and the third moisture content of the first material; wherein, the first material represents any solid material used in the production of the target rubber and plastic product.

[0032] Specifically, the surface moisture of a first material is first measured using a first method. The first material is any solid material used in the production of the target rubber and plastic product, such as rubber, plastic, powder, or other solid materials. The surface moisture is measured using a surface moisture sensor, such as a microwave sensor or a capacitance sensor. These sensors can accurately measure the water content on the surface of material particles and express it as a percentage of surface moisture. Specifically, to correlate the sensor signal with the moisture content, a calibration experiment is conducted using materials with known moisture content to establish a mathematical relationship between the signal and the moisture content. The sensor signal, such as the intensity of reflected microwave signals or capacitance value, is processed and analyzed to convert it into a moisture value, which represents the mass or proportion of moisture. The first material is then photographed to acquire its image, and image processing techniques and deep learning algorithms are used to analyze the image data. Analysis reveals that the morphology, size, and color of material particles are extracted from images. These features are then used to deduce the moisture content of the material, i.e., the moisture content of the material is calculated through image analysis. For example, in highly absorbent plastic materials, plastic particles with higher moisture content absorb water, resulting in a smoother surface, reduced friction between particles, and a tighter bond between them. Particles with higher moisture content also tend to have more regular shapes. Especially when the moisture acts as a binder between particles, materials with higher moisture content, such as wet plastics, often have a darker surface color due to their higher gloss. However, when the moisture content is low, the surface of plastic particles usually appears rougher, and irregular shapes are more common, especially in dry particles. The surface is also looser, with potentially larger spaces between particles, and the color may be lighter or uneven.

[0033] The second moisture content is predicted using an artificial intelligence model based on the relationship between the material's image features and historical moisture data. The temperature change of the first material before mixing is measured using a second method. This can be achieved by a temperature sensor installed on the material conveyor belt, measuring the temperature change before and after heating. The specific process will be described in detail below. Based on the temperature difference and a pre-defined material-moisture relationship model, the third moisture content of the material can be estimated. Specifically, the temperature difference of the material is correlated with its moisture content; materials with larger temperature differences generally have lower moisture content, while materials with smaller temperature differences indicate higher moisture content. Finally, based on the first, second, and third moisture contents, the total moisture content of the first material is estimated. The calculation of the total moisture content provides a more comprehensive moisture measurement result, laying the foundation for subsequent adjustments and optimizations to the mixing ratio data, ensuring precise control of the material's moisture content during the mixing process.

[0034] Further, estimating the third moisture content of the first material includes: measuring the first temperature of the first material at one end of the conveyor belt using a first temperature sensor, activating a heating device installed on the conveyor belt to start heating the first material, measuring the second temperature of the first material at the other end of the conveyor belt using a second temperature sensor; calculating the temperature difference between the second temperature and the first temperature, and estimating the third moisture content of the first material based on a preset relationship between the material temperature difference and the moisture content.

[0035] Specifically, the first temperature of the first material is first measured by a first temperature sensor installed at one end of the material conveyor belt. The first temperature sensor monitors the initial temperature of the material on the conveyor belt in real time. Then, the heating device installed on the conveyor belt is activated to heat the first material. The heating device can gradually increase the material temperature through heating radiation, hot air, or other heating methods, thereby promoting the evaporation of moisture in the material. Next, at the other end of the conveyor belt, a second temperature sensor measures the second temperature of the first material. The second temperature sensor is used to measure the temperature of the material after the heating process. The temperature difference between the second temperature and the first temperature reflects the degree of moisture evaporation during the heating process. Materials with more moisture need to absorb more heat for evaporation, so their temperature change is smaller, while materials with less moisture will have a larger temperature change during the heating process. Therefore, based on a preset model of the relationship between material temperature difference and moisture content, the third moisture content of the first material can be calculated based on the measured temperature difference value. This model establishes a correlation between temperature difference and moisture content, which is usually obtained through practical application in projects and continues to be used in subsequent production processes. Finally, considering the relationship between temperature difference and moisture content, the third moisture content of the first material is estimated, thereby providing a basis for subsequent adjustment of mixing ratio data and avoiding instability in the ratio or fluctuation in product quality caused by moisture errors.

[0036] Further, see Figure 3 As shown, measuring the actual quantity of each material includes: Step S31: Installing an image capturing device on the mixing device, the image capturing device captures a second material image of the material during the mixing process in real time, and performs preprocessing and normalization processing; Step S32: Extracting a first feature and a second feature from the second material image based on a convolutional neural network, the first feature including edges and texture, and the second feature including particle shape, size, and quantity; Step S33: Calculating the material type and material quantity based on the first and second features to obtain each material type and its corresponding actual quantity; Step S34: Comparing each material type and its corresponding actual quantity with the target optimal ratio data and obtaining the comparison result. If the comparison result does not conform to the target optimal ratio data, recalculating the quantity of each material and transmitting it to the mixing device through a control signal to automatically adjust the quantity of subsequent materials; The second material image includes the shape, size, color, quantity, and particle distribution of the material, and the target optimal ratio data refers to the optimal ratio data used for the production of the target rubber and plastic product.

[0037] Specifically, an image capturing device installed on the mixing device can capture a second material image of each material in real time during the mixing process. The first material image is captured before the material is added to the mixing device, and the second material image is captured inside the mixing device during the mixing process. The image capturing device includes a camera or image acquisition device, which can continuously acquire the state of the material during the mixing process. The captured second material images are first preprocessed and normalized to eliminate uneven lighting, image noise, or other environmental interference factors to ensure that the image data quality is suitable for subsequent analysis. The preprocessing process includes image denoising, brightness adjustment, contrast enhancement, and size normalization to improve the standardization and quality of the second material images. Then, based on a convolutional neural network, a first feature and a second feature are extracted from the second material images. The first feature includes low-level features such as the material's edges and textures, which belong to local information in the image. The second feature includes high-level features such as particle shape, particle size, particle quantity, and particle distribution. The first and second features are morphological features of the material extracted by a deep learning model. Based on the extracted first and second features, the type of material and the actual quantity added can be deduced. Specifically, the trained artificial intelligence model can infer the quantity of each material based on the relationship between the material characteristics and the type and quantity of the material in the image. Then, the calculated quantity of each material type and its corresponding quantity is compared with the pre-set target optimal ratio data, which refers to the optimal material ratio for the production of the target rubber and plastic product. If the comparison result shows that the actual quantity of the material does not match the target ratio, the quantity of each material to be added is recalculated, and a corresponding control signal is generated and transmitted to the mixing equipment to automatically adjust the quantity of subsequent materials to ensure that each material is added according to the optimal ratio. Image recognition can accurately identify and calculate the quantity of materials added, ensuring the accuracy of the material ratio and avoiding uneven mixing or fluctuations in the quality of rubber and plastic products due to feeding errors.

[0038] Furthermore, during the mixing operation, the process also includes: acquiring the current value of the motor in the mixing equipment, which represents the load status of the mixing equipment; the machine learning model predicts the performance of the mixed rubber and plastic product based on the relationship between the historical current value of the mixing equipment and the optimal ratio data, and also outputs the predicted performance of the target rubber and plastic product by learning the relationship between the current value and the performance of the rubber and plastic product; if the predicted performance of the target rubber and plastic product does not meet the design requirements, the operating parameters of the mixing equipment are controlled to adjust the amount of material added, including increasing or decreasing moisture and increasing or decreasing additives.

[0039] Specifically, during the mixing operation, the load status of the equipment is monitored in real time by monitoring the motor current value in the mixing equipment. The current value is a key parameter during the operation of the mixing equipment, reflecting its load level. A high current value indicates a heavy load on the mixing equipment, suggesting that the material is viscous or too much material has been added. Conversely, a low current value indicates a light load and good material flowability. Real-time monitoring of the motor current value effectively provides insight into the operating status of the mixing process. The machine learning model performs predictive analysis based on the relationship between the historical current values ​​of the mixing equipment and the optimal proportion data. The machine learning model learns the relationship between historical current values ​​and the performance of rubber and plastic products, and, combined with the mixing formula data, predicts the performance of the mixed rubber and plastic products, such as strength, toughness, and... Indicators such as flowability are specifically defined as follows: In scenarios where machine learning models predict the performance of rubber and plastic products, the training data consists of input variables including historical current values ​​of the motor in the mixing equipment, optimal proportion data in the initial optimal mixing scheme, and specific mixing formula data. These variables collectively characterize the load state and material composition of the mixing process. The output data are historical rubber and plastic product performance indicators corresponding to these input variables, specifically including mechanical properties such as tensile strength and hardness, and rheological indicators such as viscosity and flowability. By using historical current values, optimal proportion data, and formula data as input features, and the corresponding rubber and plastic product performance as the prediction target, the machine learning model can learn and establish a mapping relationship between the load state reflected by the current value and the final product quality. Through learning from historical data, the machine learning model can infer the correlation between current values ​​and product performance, thereby predicting whether the performance of the rubber and plastic product meets the design requirements under the current mixing operation. If the predicted target rubber and plastic product performance fails to meet the design requirements, the operating parameters of the mixing equipment are automatically adjusted based on the prediction results, such as adjusting the amount of material added, increasing or decreasing moisture, and adjusting the amount of additives, to optimize the mixing results. Specific operations include increasing or decreasing the moisture content during the mixing process, or adjusting the dosage of certain additives according to the requirements of rubber and plastic products, to ensure that the final performance of the rubber and plastic products meets the standards. By combining current value monitoring, machine learning prediction, and operating parameter adjustment, this method can precisely control every operation in the mixing process and adjust the ratio and dosage in a timely manner based on the prediction results, thereby ensuring the quality of the final rubber and plastic products.

[0040] Further, image analysis is performed on the first material image to obtain the second moisture content of the first material, including: processing the first material image using a convolutional neural network and automatically extracting material features from the first material image, the material features including particle size, shape, color and texture of the material; training a deep learning model based on historical material image data and corresponding moisture content data, inputting the material features into the pre-trained deep learning model, and the deep learning model calculating the moisture content of the material features as the second moisture content of the first material, the moisture content being the moisture ratio in the first material.

[0041] Specifically, the first step is to perform image analysis on the first material image. This is done by processing the image using a convolutional neural network (CNN) to extract material features. Specifically, the CNN can extract low-level features such as edges, textures, and shapes through multiple convolutional operations, and then progressively extract higher-level features such as particle size, shape, color, and texture. Material features are crucial information closely related to moisture content, as moisture content affects the material's appearance, such as particle gloss, color intensity, and particle aggregation. Through multi-layer feature extraction by the CNN, these features can be accurately captured, serving as the basis for subsequent moisture estimation. Furthermore, based on historical material image data and corresponding moisture content data, a model is trained using deep learning methods to learn the relationship between material features and moisture content. Specifically, using historical image data and the moisture content labels of these images, a deep learning model is trained through supervised learning to identify the mapping relationship between different material characteristics and moisture content. The trained deep learning model can automatically calculate the moisture content of materials based on new material image data, that is, predict the moisture ratio of materials through extracted material features. This method can predict moisture content using image data without direct contact with the material, avoiding the limitations of traditional moisture measurement methods. Especially when dealing with particulate materials and materials with complex shapes, this method can provide more flexible and accurate results, thereby achieving precise moisture control in the mixing process.

[0042] Furthermore, generating a secondary optimal mixing scheme includes: taking any liquid material in the secondary optimal mixing scheme as the third material, and calculating the actual quantity of the third material based on the relationship between the preset operating parameters of the mixing equipment and the quantity of material remaining on the mixing equipment, wherein the relationship between the operating parameters of the mixing equipment and the quantity of material remaining on the mixing equipment changes over time.

[0043] Specifically, based on the previously generated optimal mixing scheme, any one of the liquid materials in the scheme is used as the third material for further calculation and adjustment. Liquid materials usually have a significant impact on the performance of rubber and plastic products during the mixing process, so it is necessary to accurately calculate their dosage. Based on the preset operating parameters of the mixing equipment, combined with the actual state of the material in the mixing equipment, especially the amount of material left on the mixing equipment, the actual dosage of the third material is calculated. Specifically, the operating parameters of the mixing equipment, such as the feeding rate, stirring speed, and heating temperature, may affect the fluidity and residual amount of the material inside the equipment. Therefore, it is necessary to estimate the amount of material left on the equipment based on the historical operating status of the mixing equipment and the physical characteristics of the material. Residual material refers to the amount of material that was not completely mixed or removed in the previous batch mixing operation. This amount of material accumulates over time and affects the amount of material added in the current batch mixing operation. Therefore, it is necessary to calculate the amount of residual material on the equipment to accurately control the amount of each material added during the mixing process and avoid over- or under-addition of materials. Simultaneously, to ensure the accuracy of the mixing ratio, the relationship between the operating parameters of the mixing equipment and the amount of residual material on the equipment is dynamically adjusted over time. For example, as the number of mixing cycles increases, the residual material will change, and the mixing equipment may need to be cleaned or the feeding rate adjusted. Therefore, this relationship needs to be fed back to the mixing system in real time to ensure that the feeding rate can be optimized and adjusted according to changes in the residual material each time, thus ensuring the quality of the final product.

[0044] The above describes an automatic mixing and control method for rubber and plastic products in the embodiments of this application. The following describes an automatic mixing and control system for rubber and plastic products in the embodiments of this application. Please refer to [link / reference]. Figure 4 One embodiment of the automatic mixing and control system for rubber and plastic products in this application includes: The training unit is used to collect mixing environment data, mixing formula data and corresponding rubber and plastic product performance, and input them into the artificial intelligence model for learning. Based on the learning results, the artificial intelligence model outputs the optimal ratio data of rubber and plastic products under the current mixing environment and the operating parameters of the mixing equipment to form the initial optimal mixing scheme. The adjustment unit is used to measure the actual quantity of each material used in the production of the target rubber and plastic product and the total moisture content in the material before executing the initial optimal mixing scheme. The artificial intelligence model predicts the performance of the target rubber and plastic product based on the measurement data and generates adjustment instructions for the initial optimal mixing scheme based on the predicted performance of the target rubber and plastic product. The mixing unit is used to generate a secondary optimal mixing scheme and control signals for controlling the mixing device based on adjustment instructions. The mixing device automatically controls the mixing device to perform mixing operations according to the secondary optimal mixing scheme based on the control signals.

[0045] Through the synergistic cooperation of the above-mentioned components, this application further realizes intelligent and precise control over the quality of rubber and plastic products.

[0046] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0047] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0048] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. An automatic mixing and control method for rubber and plastic products, characterized in that, The method includes: Collect mixing environment data, mixing formula data and corresponding rubber and plastic product performance, input them into an artificial intelligence model for learning, and the artificial intelligence model outputs the optimal ratio data of rubber and plastic products and the operating parameters of the mixing equipment based on the learning results to form an initial optimal mixing scheme. Before executing the initial optimal mixing scheme, the actual quantity of each material used in the production of the target rubber and plastic product and the total moisture content in the material are measured and the measurement data are obtained. The artificial intelligence model predicts the performance of the target rubber and plastic product based on the measurement data, and generates adjustment instructions for the initial optimal mixing scheme based on the predicted performance of the target rubber and plastic product. Based on the adjustment command, a secondary optimal mixing scheme and a control signal for controlling the mixing device are generated. Based on the control signal, the mixing device automatically controls itself to perform mixing operations according to the secondary optimal mixing scheme.

2. The automatic mixing and control method for rubber and plastic products according to claim 1, characterized in that, During the mixing operation, multiple sensors deployed inside and outside the mixing device collect mixing status data in real time and feed it back to the artificial intelligence model, outputting new mixing schemes and generating new control signals to adjust the current mixing operation process.

3. The automatic mixing and control method for rubber and plastic products according to claim 1, characterized in that, Measuring the total moisture content in the material includes: The moisture content of the first material is obtained by measuring the surface moisture of the first material using a first method. The first material is photographed to obtain an image of the first material, and the image of the first material is analyzed to obtain the second moisture content of the first material; The temperature change of the first material before mixing is measured by a second method, and the third moisture content of the first material is estimated based on the temperature change. The total moisture content of the first material is estimated based on the first moisture content, the second moisture content, and the third moisture content of the first material.

4. The automatic mixing and control method for rubber and plastic products according to claim 3, characterized in that, Estimating the third moisture content of the first material includes: At one end of the conveyor belt, a first temperature of the first material is measured by a first temperature sensor, and a heating device installed on the conveyor belt is activated to start heating the first material. At the other end of the conveyor belt, a second temperature of the first material is measured by a second temperature sensor. Calculate the temperature difference between the second temperature and the first temperature, and estimate the third moisture content of the first material based on a preset relationship between the material temperature difference and moisture content.

5. The automatic mixing and control method for rubber and plastic products according to claim 1, characterized in that, Measure the actual quantity of each material dispensed, including: An image capturing device is installed on the mixing device. The image capturing device captures a second material image of the material during the mixing process in real time, and performs preprocessing and normalization processing. The first feature and the second feature are extracted from the second material image based on a convolutional neural network. The first feature includes edges and texture, and the second feature includes particle shape, size and number. Based on the first feature and the second feature, the material type and the quantity of material to be delivered are calculated to obtain each material type and its corresponding actual quantity to be delivered. Each material type and its corresponding actual quantity are compared with the target optimal ratio data, and the comparison result is obtained. If the comparison result does not match the target optimal ratio data, the quantity of each material is recalculated and transmitted to the mixing equipment through a control signal to automatically adjust the quantity of subsequent materials.

6. The automatic mixing and control method for rubber and plastic products according to claim 1, characterized in that, The mixing operation also includes: During the mixing operation, the current value of the motor in the mixing equipment is acquired, and the current value represents the load status of the mixing equipment. The machine learning model predicts the performance of the mixed rubber and plastic products based on the relationship between the historical current values ​​of the mixing equipment and the optimal ratio data. It also outputs the predicted performance of the target rubber and plastic products by learning the relationship between the current values ​​and the performance of the rubber and plastic products. If the predicted performance of the target rubber and plastic product does not meet the design requirements, the operating parameters of the mixing equipment are controlled, and the amount of material added is adjusted.

7. The automatic mixing and control method for rubber and plastic products according to claim 3, characterized in that, Image analysis is performed on the image of the first material to obtain the second moisture content of the first material, including: The first material image is processed using a convolutional neural network, and material features in the first material image are automatically extracted. The material features include the particle size, shape, color, and texture of the material. A deep learning model is trained based on historical material image data and corresponding moisture content data. The material features are input into the pre-trained deep learning model, and the deep learning model calculates the moisture content of the material features as the second moisture content of the first material. The moisture content is the proportion of moisture in the first material.

8. The automatic mixing and control method for rubber and plastic products according to claim 1, characterized in that, Generate the second-optimal mixing scheme, including: Using any one of the liquid materials in the secondary optimal mixing scheme as the third material, the actual quantity of the third material is calculated based on the relationship between the preset operating parameters of the mixing equipment and the quantity remaining on the mixing equipment.

9. The automatic mixing and control method for rubber and plastic products according to claim 8, characterized in that, The relationship between the operating parameters of the mixing equipment and the amount of material remaining on the mixing equipment changes over time.

10. An automatic mixing and control system for rubber and plastic products, used to implement the automatic mixing and control method for rubber and plastic products as described in any one of claims 1-9, characterized in that, The system includes: The training unit is used to collect mixing environment data, mixing formula data and corresponding rubber and plastic product performance, and input them into the artificial intelligence model for learning. The artificial intelligence model outputs the optimal ratio data of rubber and plastic products under the current mixing environment and the operating parameters of the mixing equipment based on the learning results, thus forming the initial optimal mixing scheme. An adjustment unit is used to measure the actual quantity of each material used in the production of the target rubber and plastic product and the total moisture content of the material before executing the initial optimal mixing scheme, and to obtain measurement data. The artificial intelligence model predicts the performance of the target rubber and plastic product based on the measurement data, and generates adjustment instructions for the initial optimal mixing scheme based on the predicted performance of the target rubber and plastic product. The mixing unit is used to generate a secondary optimal mixing scheme and a control signal for controlling the mixing device based on the adjustment command. The mixing device automatically controls the mixing device to perform mixing operations according to the secondary optimal mixing scheme based on the control signal.

Images