Automatic control system and method for natural rubber processing production line

Through automatic control system and deep learning technology, the rubber block dehydration process is analyzed, and the mixing leaf speed is optimized, which solves the problem of low dehydration efficiency in the existing technology, and improves production efficiency and product quality.

CN118941873AInactive Publication Date: 2024-11-12NINGBO LAILI MECHANICAL & ELECTRICAL IND CO LTD

Patent Information

Application Number
CN202411121608.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-08-15
Publication Date
2024-11-12
Estimated Expiration
Not applicable · inactive patent

AI Technical Summary

Technical Problem

The dehydration process in existing natural rubber processing is inefficient and lacks the ability to dynamically adjust according to actual working conditions, resulting in limited production efficiency.

Method used

The automatic control system is adopted to obtain the rubber block dehydration working image through the camera, the sensor collects the dehydration environment data and the stirring leaf speed value, and use deep learning technology to extract feature vectors and correlation analysis to determine whether the stirring leaf speed needs to be adjusted to optimize the dehydration process.

Benefits of technology

It improves the dehydration efficiency, avoids excessive or insufficient stirring, enhances production efficiency and product quality stability, and reduces resource waste.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118941873A_ABST
    Figure CN118941873A_ABST
Patent Text Reader

Abstract

The invention relates to the field of natural rubber processing, and particularly discloses an automatic control system and method for a natural rubber processing production line. The method comprises the following steps: firstly, acquiring a rubber block dehydration working image acquired by a camera, rubber block dehydration working environment condition data acquired by a sensor and stirring speed values of stirring blades at a plurality of preset time points acquired by the sensor, and then performing feature extraction and correlation analysis on the three parts by using a deep learning technology to obtain a rubber block dehydration result; finally, a classification result is obtained through a classifier to judge whether the speed of the stirring blades needs to be adjusted or not, so that the speed of the stirring blades is adjusted in time, excessive or insufficient stirring is avoided, and the dehydration efficiency is improved to the maximum extent.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present application relates to the field of natural rubber processing, and more specifically, to an automatic control system and method for a natural rubber processing production line. Background Art

[0002] Natural rubber is a polyisoprene cis polymer extracted from the sap of the rubber tree. This sap is an emulsion containing organic components suspended in water. During the production process, dehydration is usually required to remove water from the rubber mass.

[0003] Current methods mainly accelerate the evaporation of water by stirring and using hot air blowers. However, existing dehydration methods usually rely on preset hot air blower settings and lack the ability to dynamically adjust according to actual working conditions. This deficiency greatly reduces the efficiency of the dehydration process, thereby limiting the room for improvement in production efficiency.

[0004] Therefore, an automatic control system and method for a natural rubber processing production line is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The embodiment of the present application provides an automatic control system and method for a natural rubber processing production line, which first obtains the rubber block dehydration working image collected by a camera, the rubber block dehydration working environment condition data collected by a sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor, and then uses deep learning technology to perform feature extraction and association analysis on the three, and finally obtains the classification result through a classifier to determine whether the stirring blade speed needs to be adjusted, so as to adjust the stirring blade speed in time to avoid excessive or insufficient stirring, thereby maximizing the dehydration efficiency.

[0006] According to one aspect of the present application, there is provided an automatic control system for a natural rubber processing production line, comprising: A natural rubber processing data acquisition module, used to acquire the rubber block dehydration working image acquired by the camera, the rubber block dehydration working environment condition data acquired by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points acquired by the sensor; A natural rubber processing data extraction module, used for extracting a multi-modal correlation feature vector of a rubber block dehydration state and an optimized correlation feature vector of a stirring blade stirring speed from the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor; The stirring blade speed adjustment judgment module is used to judge whether the stirring blade speed needs to be adjusted based on the multi-modal correlation feature vector of the rubber block dehydration state and the optimized stirring blade stirring speed correlation feature vector.

[0007] According to another aspect of the present application, there is provided an automatic control method for a natural rubber processing production line, comprising: Acquire the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor; Extracting a multimodal correlation feature vector of a rubber block dehydration state and an optimized correlation feature vector of a stirring blade stirring speed from the rubber block dehydration working image acquired by the camera, the rubber block dehydration working environment condition data acquired by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points acquired by the sensor; Based on the multi-modal correlation feature vector of the dehydration state of the rubber block and the optimized correlation feature vector of the stirring blade stirring speed, it is determined whether the stirring blade speed needs to be adjusted.

[0008] Compared with the prior art, the present application provides an automatic control system and method for a natural rubber processing production line, which first obtains the rubber block dehydration working image collected by a camera, the rubber block dehydration working environment condition data collected by a sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the three, and finally obtains the classification result through a classifier to determine whether the stirring blade speed needs to be adjusted, so as to adjust the stirring blade speed in time to avoid excessive or insufficient stirring, thereby maximizing the dehydration efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] By describing the embodiments of the present application in more detail in conjunction with the accompanying drawings, the above and other purposes, features and advantages of the present application will become more apparent. The accompanying drawings are used to provide a further understanding of the embodiments of the present application and constitute a part of the specification. Together with the embodiments of the present application, they are used to explain the present application and do not constitute a limitation of the present application. In the accompanying drawings, the same reference numerals generally represent the same components or steps.

[0010] Figure 1 It is a block diagram of an automatic control system of a natural rubber processing production line according to an embodiment of the present application.

[0011] Figure 2 It is a block diagram of a natural rubber processing data extraction module in an automatic control system of a natural rubber processing production line according to an embodiment of the present application.

[0012] Figure 3It is a block diagram of a stirring speed feature extraction unit in an automatic control system of a natural rubber processing production line according to an embodiment of the present application.

[0013] Figure 4 It is a block diagram of a stirring blade speed adjustment judgment module in an automatic control system of a natural rubber processing production line according to an embodiment of the present application.

[0014] Figure 5 The figure is a flow chart of an automatic control method for a natural rubber processing production line according to an embodiment of the present application.

[0015] Figure 6 is a block diagram of an electronic device according to an embodiment of the present application. DETAILED DESCRIPTION

[0016] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0017] Figure 1 FIG. 1 is a block diagram of an automatic control system for a natural rubber processing production line according to an embodiment of the present application. Figure 1 As shown, the automatic control system 100 of the natural rubber processing production line according to the embodiment of the present application includes: a natural rubber processing data acquisition module 110, which is used to acquire the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor; a natural rubber processing data extraction module 120, which is used to extract the rubber block dehydration state multimodal correlation feature vector and the optimized stirring blade stirring speed correlation feature vector from the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor; a stirring blade speed adjustment judgment module 130, which is used to judge whether the stirring blade speed needs to be adjusted based on the rubber block dehydration state multimodal correlation feature vector and the optimized stirring blade stirring speed correlation feature vector.

[0018] In the automatic control system 100 of the above-mentioned natural rubber processing production line, the natural rubber processing data acquisition module 110 is used to obtain the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor and the stirring speed value of the stirring blade at multiple predetermined time points collected by the sensor. It should be understood that natural rubber is a polyisoprene cis polymer extracted from the sap of rubber trees. This sap is an emulsion containing a variety of organic components and suspended in water. In the production process of rubber, one of the key steps is to remove the moisture in the rubber block, which is usually achieved by dehydration. The current dehydration method mainly relies on the combined action of stirring and hot air blower to accelerate the evaporation process of water. However, the existing technology often relies only on the preset hot air blower setting, lacking the ability to dynamically adjust according to the actual working conditions. This limitation significantly reduces the efficiency of the dehydration process and limits the potential for improving efficiency in the production process. Therefore, in the technical solution of the present application, by acquiring the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor, and combining with deep learning technology, it is determined whether the stirring blade speed needs to be adjusted, thereby ensuring the high efficiency of the dehydration process in rubber production, avoiding excessive or insufficient stirring, and maximizing the dehydration efficiency.

[0019] Specifically, the camera can accurately reflect the state and changes of the rubber block at each stage, such as the degree of dehydration and shape changes, by capturing images of the rubber block in real time during the dehydration process. The data on the working environment conditions of the rubber block dehydration collected by the sensor include multiple parameters such as temperature, humidity, and air pressure. These data can provide an in-depth understanding of the environmental background of the dehydration work and help the system analyze various factors that affect the dehydration quality of the rubber block. After feature extraction and semantic association processing, the environmental data forms a detailed environmental feature vector, which provides semantic information under environmental conditions for the control system, and helps to adjust and optimize the environmental control strategy in the production process. In addition, the sensor also regularly records the stirring speed values ​​of the stirring blade at multiple time points. These data directly reflect the operating state and speed changes of the blade during the stirring process. Through the feature extraction and encoding processing of the stirring speed, the system can obtain the key dynamic characteristics of the stirring process and provide real-time operating data and feedback for the dehydration and processing of the rubber block. The purpose of acquiring camera images, environmental condition data, and stirring speed values ​​is to build an intelligent natural rubber processing production line control system. Through real-time data collection, feature extraction, and comprehensive analysis, the production process can be finely monitored and controlled, thereby improving production efficiency, ensuring product quality, and reducing adverse changes and waste in the production process.

[0020] In the automatic control system 100 of the natural rubber processing production line, the natural rubber processing data extraction module 120 is used to extract the multimodal correlation feature vector of the dehydration state of the rubber block and the optimized correlation feature vector of the stirring blade stirring speed from the dehydration working image of the rubber block collected by the camera, the dehydration working environment condition data of the rubber block collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor. It should be understood that extracting the correlation feature vector of the dehydration state of the rubber block and the stirring blade speed means fusing and analyzing the information of the above different data sources to form a comprehensive feature description, thereby realizing real-time prediction, abnormality detection and automatic control of the production process, thereby improving production efficiency, reducing costs, and ensuring the stability and consistency of product quality.

[0021] Figure 2 FIG. 1 is a block diagram of a natural rubber processing data extraction module in an automatic control system of a natural rubber processing production line according to an embodiment of the present application. Figure 2 As shown, in a specific embodiment of the present application, the natural rubber processing data extraction module 120 includes: a dehydration work image feature extraction unit 121, which is used to perform feature extraction on the rubber block dehydration work image collected by the camera to obtain a rubber block dehydration state feature vector; an environmental condition data feature extraction unit 122, which is used to perform feature extraction on the rubber block dehydration work environmental condition data collected by the sensor to obtain a rubber block dehydration work environment semantic association feature vector; a rubber block dehydration state feature association unit 123, which is used to associate the rubber block dehydration state feature vector with the rubber block dehydration work environment semantic association feature vector to obtain the rubber block dehydration state multimodal association feature vector; a stirring speed feature extraction unit 124, which is used to perform feature extraction on the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor to obtain the optimized stirring blade stirring speed association feature vector.

[0022] It should be understood that in the rubber industry, the dehydration process of rubber blocks directly affects the quality and performance of the final product, so it must be carefully analyzed and monitored. However, the images collected by the camera can provide rich information about the surface state of the rubber block. This information includes features such as the color, texture, shape and surface moisture of the rubber block. Through image processing technology, quantitative descriptions of these features can be extracted from the image, such as the distribution of color and the surface texture of the rubber block. These features can reflect the degree and uniformity of dehydration of the rubber block, and have a crucial impact on product quality. Through feature extraction, complex image data can be converted into a higher level of abstract feature representation, which is easier to analyze and compare.

[0023] Furthermore, feature extraction of the working environment condition data of rubber block dehydration collected by sensors is to deeply understand and effectively control the environmental variables in the industrial production process, thereby optimizing product quality and production efficiency. The data collected by the sensor covers multidimensional information of various environmental conditions in the process of rubber block dehydration. Through feature extraction, representative and semantically related feature vectors can be extracted from these raw data, such as the average temperature change trend or humidity fluctuation range in a certain period of time. These feature vectors can more intuitively reflect the degree of correlation between the change of environmental conditions and the dehydration effect of rubber blocks. It should be understood that in industrial production environments, sensors usually generate a large amount of data at a high frequency, and the direct application of raw data to real-time control and decision-making may be inefficient. Through feature extraction, a large amount of data can be compressed into fewer but more information-rich feature vectors, which can be used to establish prediction models or monitoring systems to help engineers and operators better understand the changes in environmental conditions in the production process and adjust production parameters in time to ensure the stability of product quality and production efficiency.

[0024] Furthermore, the dehydration state feature vector of the rubber block and the semantic association feature vector of the dehydration working environment of the rubber block are associated in order to fully understand and accurately describe the comprehensive influence of various factors in the dehydration process of the rubber block, so as to optimize production control and improve product quality. Among them, the dehydration state feature vector of the rubber block contains visual features extracted from the camera image, such as color distribution, surface texture and shape. These feature vectors reflect the actual physical state and changes of the rubber block during the dehydration process. By associating with the feature vectors in the dehydration working environment condition data, the direct influence of environmental factors such as temperature, humidity, and air pressure on the surface state of the rubber block can be explored, and the relationship between these influencing factors and the dehydration effect can be further identified and quantified. The semantic association feature vector of the dehydration working environment provides semantic information about the production environment conditions, such as the stability of the environment, the trend of change, and the possible impact on the dehydration process. These feature vectors can be extracted from the data collected by the sensor, and the potential causal relationship and trend law can be revealed by analyzing the interaction between the changes in environmental variables and the dehydration state characteristics of the rubber block. By associating these two types of feature vectors, multimodal description and comprehensive analysis of the dehydration process can be achieved.

[0025] In particular, feature extraction of the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor is to deeply understand the dynamic changes in the stirring process and optimize the stirring operation and the quality of the final product. The stirring speed of the stirring blade is a key operating parameter that affects the mixing, drying and forming of rubber blocks. The stirring speed data at multiple predetermined time points collected by the sensor reflects the motion state and speed change of the stirring blade in different time periods. Through feature extraction, key features of various stirring processes, such as average speed, maximum speed, speed change rate, etc., can be extracted from these raw speed data. These features can reveal the dynamic behavior and stirring effect of the stirring blade during the entire process. The extraction of the optimized stirring blade stirring speed associated feature vector can help analyze important parameters such as mixing uniformity, reaction rate and energy transfer efficiency during the stirring process. Different stirring speeds have a significant effect on the particle distribution, component mixing uniformity and reaction rate of the rubber block.

[0026] In a specific embodiment of the present application, the dehydration work image feature extraction unit 121 includes: passing the rubber block dehydration work image collected by the camera through a rubber block dehydration work target detection network to obtain a rubber block dehydration state interest region; passing the rubber block dehydration state interest region through a rubber block dehydration state feature encoder based on a salient target detector to obtain the rubber block dehydration state feature vector.

[0027] It should be understood that during the dehydration process of the rubber block, the image captured by the camera may contain a large amount of information, such as the overall appearance of the rubber block, surface texture and color changes. However, for actual industrial applications, usually only the changes in specific areas of the rubber block surface are of concern, such as possible cracks, bubbles or uneven colors. Through the target detection network, these areas of interest, namely the areas of interest of the dehydration state of the rubber block, can be accurately located and marked in the image. The target detection network can identify key features of the surface of the rubber block, such as areas of uneven dehydration or problem areas that require extra attention. Specifically, the rubber block dehydration working target detection network is used to process the rubber block dehydration working image captured by the camera using the following target detection formula to obtain the area of ​​interest of the dehydration state of the rubber block; Among them, the target detection formula is: ;in, This is the image of the rubber block dehydration process captured by the camera. is the anchor frame, The region of interest for the dehydrated state of the rubber block, Indicates classification, Indicates regression.

[0028] Furthermore, the region of interest of the dehydration state of the rubber block is passed through a feature encoder of the dehydration state of the rubber block based on a salient object detector, aiming to extract key dehydration state features from the visual image and convert them into feature vectors with semantic information. It should be understood that during the dehydration process of the rubber block, the salient object detector can identify the most informative and influential regions in the image. These regions usually include important features on the surface of the rubber block, such as cracks, color changes, lumps or other unevenness, which directly affect the final quality and performance of the product. Through salient object detection, these key regions of interest can be accurately located and extracted, rather than all parts of the entire image, thereby effectively reducing the complexity of data processing and analysis. The use of a feature encoder based on a salient object detector can improve the real-time monitoring capability of the dehydration state of the rubber block during the production process. By continuously identifying and updating the feature vectors of the region of interest, timely response and adjustment to the changes in the surface quality of the rubber block during the dehydration process can be achieved. This real-time monitoring not only helps to improve the consistency and quality stability of the product, but also reduces the defective rate and resource waste in production, thereby reducing costs and enhancing the competitiveness of enterprises. Specifically, each layer of the rubber block dehydration state feature encoder based on the salient object detector performs the following on the input data in the forward pass of the layer: convolution processing is performed on the input data using a first convolution kernel to obtain a first convolution feature map; convolution processing is performed on the first convolution feature map using a second convolution kernel to obtain a second convolution feature map, wherein the size of the first convolution kernel is larger than the size of the second convolution kernel; pooling processing is performed on the second convolution feature map to obtain a pooling feature map; and activation processing is performed on the pooling feature map to obtain an activation feature map; wherein the output of the last layer of the rubber block dehydration state feature encoder based on the salient object detector is the rubber block dehydration state feature vector, and the input of the first layer of the rubber block dehydration state feature encoder based on the salient object detector is the rubber block dehydration state region of interest.

[0029] In a specific embodiment of the present application, the environmental condition data feature extraction unit 122 includes: passing the rubber block dehydration working environment condition data collected by the sensor through a rubber block dehydration working environment context encoder including an embedded layer to obtain a plurality of rubber block dehydration working environment feature vectors; and cascading the plurality of rubber block dehydration working environment feature vectors to obtain the rubber block dehydration working environment semantic association feature vector.

[0030] It should be understood that during the dehydration process of rubber blocks, environmental conditions such as temperature, humidity, air pressure and other factors have an important influence on the dehydration speed, moisture content and final quality of the rubber blocks. The environmental data collected by the sensor provides real-time numerical information, but a single data point is difficult to reveal the overall trend and impact of environmental changes. Through the rubber block dehydration working environment context encoder, these environmental condition data can be integrated and encoded and converted into multiple feature vectors, each of which captures the key features and change laws of environmental variables. The embedding layer plays a key role in the encoding process. It can map high-dimensional environmental data into a low-dimensional semantic space, thereby retaining important information of the data while reducing redundancy and noise. These embedded feature vectors can not only provide the correlation and impact between environmental variables, but also reflect the comprehensive impact of environmental conditions on the overall effect of the rubber block dehydration process. Specifically, the rubber block dehydration working environment condition data collected by the sensor is segmented to obtain a word sequence; the embedding layer of the rubber block dehydration working environment context encoder including the embedding layer is used to map each word in the word sequence into a word embedding vector to obtain a sequence of word embedding vectors; the converter-based Bert model of the rubber block dehydration working environment context encoder including the embedding layer is used to perform global context semantic encoding on the sequence of word embedding vectors to obtain multiple rubber block dehydration working environment feature vectors.

[0031] Furthermore, multiple feature vectors of the working environment of rubber block dehydration are cascaded to comprehensively consider and express the comprehensive impact and semantic association of various environmental factors on the overall production process. The key to this method is to integrate environmental features of different levels and types in order to more comprehensively understand and manage the environmental conditions in the process of rubber block dehydration. Among them, the feature vector of the working environment of rubber block dehydration includes environmental data such as temperature, humidity, air pressure, wind speed, etc. obtained from various sensors and monitoring devices. Each feature vector represents the environmental state in a specific time period and reflects the numerical information and trend of environmental variables. Cascading these feature vectors means connecting them in time sequence or event sequence to form a longer sequence to capture the dynamic process and long-term impact of environmental changes. The cascading operation is not just a simple connection of feature vectors, but a semantic association of environmental data is established in the feature space. This means that the connection between feature vectors not only considers their time series relationship, but also their functions and influences in the entire production process. For example, how the environmental state of the previous time period affects the production efficiency or product quality of the subsequent time period can be reflected and analyzed by cascading feature vectors.

[0032] Figure 3 FIG. 1 is a block diagram of a stirring speed feature extraction unit in an automatic control system of a natural rubber processing production line according to an embodiment of the present application. Figure 3 As shown, in a specific embodiment of the present application, the stirring speed feature extraction unit 124 includes: a stirring speed data preprocessing subunit 1241, which is used to perform data preprocessing on the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor to obtain a stirring blade stirring speed association matrix; a stirring blade stirring speed convolution encoding subunit 1242, which is used to pass the stirring blade stirring speed association matrix through a stirring blade stirring speed feature encoder based on a convolutional neural network to obtain a stirring blade stirring speed association feature vector; a stirring blade stirring speed feature optimization subunit 1243, which is used to perform feature distribution deviation quantization compensation on the stirring blade stirring speed association feature vector to obtain an optimized stirring blade stirring speed association feature vector.

[0033] It should be understood that the stirring blade speed data collected by the sensor may have multiple sampling time points and frequencies, which directly reflects the operating status and dynamic characteristics of the stirring blade during the stirring process. However, the original speed data may be affected by noise, fluctuations or sampling deviations, which is not conducive to direct analysis and understanding. Therefore, the first task of data preprocessing is to clean and denoise the data to ensure that the extracted speed value accurately reflects the changing trend in the actual stirring operation.

[0034] Furthermore, the stirring blade stirring speed association matrix is ​​passed through a stirring blade stirring speed feature encoder based on a convolutional neural network, aiming to transform complex time series data into high-level, semantically rich feature representations. The core of this method is to use the powerful feature extraction capability of convolutional neural networks to extract key patterns and change trends from the dynamic stirring speed association matrix to support more accurate and efficient process control and quality management. Among them, the stirring blade stirring speed association matrix contains speed correlation information between multiple time points, usually represented in the form of a two-dimensional array. These data reflect the complex patterns of speed changes during the stirring process, such as synchronization, volatility or hysteresis, which have an important impact on the stirring effect and the quality of the final product. Through convolutional neural networks, the spatial and temporal local patterns in these time series data can be effectively captured and learned, thereby extracting abstract, high-level feature representations. The stirring blade stirring speed feature encoder based on convolutional neural networks can realize automatic learning and feature extraction of complex data, avoiding the deficiency of manually designing feature extractors in traditional methods. Through the end-to-end training process, the network can learn the optimal feature representation method based on the input stirring blade stirring speed association matrix, so that the output stirring blade stirring speed association feature vector is more expressive and discriminative. Specifically, each layer of the stirring blade stirring speed feature encoder based on the convolutional neural network is used to perform convolution processing, mean pooling processing based on the local feature matrix and nonlinear activation processing on the input data in the forward transfer of the layer to output the stirring blade stirring speed association feature vector by the last layer of the stirring blade stirring speed feature encoder based on the convolutional neural network, wherein the input of the stirring blade stirring speed feature encoder based on the convolutional neural network is the stirring blade stirring speed association matrix.

[0035] In particular, in the technical solution of the present application, it is considered that the dehydration state feature vector of the rubber block is obtained based on the image data through the target detection network and the significant target detector, while the stirring blade stirring speed associated feature vector is obtained based on the speed value collected by the sensor through the association matrix and the convolutional neural network. These two data sources have different data characteristics and acquisition methods. The dehydration state feature vector of the rubber block is encoded by the significant target detector, while the stirring blade stirring speed associated feature vector is encoded by the convolutional neural network. Different feature extraction methods may lead to inconsistent distribution of feature vectors in feature space. The dehydration state feature of the rubber block may not directly involve time series data, while the stirring blade stirring speed value is arranged according to the time dimension. If there is a difference in the processing of the time series between the two, it may lead to inconsistency of the feature vector in the time dimension. This leads to a feature distribution shift deviation of the stirring blade stirring speed associated feature vector relative to the multimodal associated feature vector of the dehydration state of the rubber block, which makes the fused stirring blade speed adjustment judgment feature vector have extreme category deviation when the classification judgment is performed by the classifier, resulting in a decrease in the accuracy of the classification result. Based on this, in the technical solution of the present application, based on the shift information between the multimodal associated feature vector of the rubber block dehydration state and the associated feature vector of the stirring blade stirring speed, the characteristic distribution deviation of the associated feature vector of the stirring blade stirring speed is quantitatively compensated to obtain an optimized associated feature vector of the stirring blade stirring speed.

[0036] The method comprises: performing quantitative compensation for the characteristic distribution deviation of the stirring blade stirring speed associated feature vector to obtain an optimized stirring blade stirring speed associated feature vector, including: calculating a shift difference representation vector between the multimodal associated feature vector of the dehydration state of the rubber block and the stirring blade stirring speed associated feature vector; performing probabilistic activation function-based probabilistic operation on each eigenvalue of the shift difference representation vector to obtain a probabilistic shift difference representation vector; inputting the stirring blade stirring speed associated feature vector into a classifier to obtain an initial classification probability value; determining a back-propagation compensation representation vector based on a comparison between the eigenvalues ​​of each position of the probabilistic shift difference representation vector and the initial classification probability value and a weight matrix of the classifier; calculating the natural exponential function value of each eigenvalue of the back-propagation compensation representation vector to obtain a back-propagation compensation class representation vector; calculating a position-based point multiplication between the back-propagation compensation class representation vector and the stirring blade stirring speed associated feature vector to obtain a point multiplication result vector, and calculating a weighted sum between the point multiplication result vector and the stirring blade stirring speed associated feature vector to obtain the optimized stirring blade stirring speed associated feature vector.

[0037] The characteristic distribution deviation of the stirring blade stirring speed associated characteristic vector is quantitatively compensated by the following optimization formula to obtain an optimized stirring blade stirring speed associated characteristic vector; Wherein, the optimization formula is: ;in, The first multi-modal correlation feature vector representing the dehydration state of the rubber block The eigenvalues ​​at the positions, represents the stirring blade stirring speed associated feature vector, The first represents the stirring blade stirring speed associated feature vector. The eigenvalues ​​at the positions, The first vector representing the probabilistic shift difference representation vector The eigenvalues ​​at the positions, represents the length of the stirring blade stirring speed associated feature vector, and denote the first weighted hyperparameter and the second weighted hyperparameter respectively, represents the weight matrix of the classifier, represents the initial classification probability value, It means point multiplication by position. represents matrix multiplication, represents the symbolic function, Represents the optimized stirring blade stirring speed associated feature vector.

[0038] That is, for the problem of distribution shift deviation between feature vectors, the feature distribution deviation of the stirring blade stirring speed-related feature vector is quantitatively compensated based on the shift information between the multimodal associated feature vector of the dehydration state of the rubber block and the stirring blade stirring speed-related feature vector, which is obtained by calculating the difference between the multimodal associated feature vector of the dehydration state of the rubber block and the stirring blade stirring speed-related feature vector to obtain the shift difference representation vector, so as to quantify the deviation of the feature distribution. The shift difference representation vector reveals the alignment of the two feature vectors in the feature space, which provides a basis for subsequent compensation. Then, the probability activation function is applied to each eigenvalue of the shift difference representation vector for probabilistic processing, so that the differential value is converted into a probability distribution, so that the model can understand the deviation between the features in a probabilistic form. Then, the original stirring blade stirring speed-related feature vector is input into the classifier to obtain the initial classification probability value, wherein the initial classification probability value provides the model with the basic classification performance index before feature compensation. Then, the back propagation compensation representation vector is determined by combining the probabilistic shift difference representation vector and the initial classification probability value, as well as the weight matrix of the classifier. That is, the model's own weight and bias information are used to compensate the feature vector to reduce the classification error. Then, the natural exponential function value is calculated for each eigenvalue of the back-propagation compensation representation vector, thereby enhancing the expression range of the eigenvalue and providing a mechanism for enlarging or reducing the feature compensation. Finally, the optimized feature vector associated with the stirring blade stirring speed is obtained by calculating the dot product result between the back-propagation compensation class representation vector and the feature vector associated with the stirring blade stirring speed, and then calculating the weighted sum with the original feature vector associated with the stirring blade stirring speed. It should be understood that the dot product and weighted sum operations combine the compensation information and the original feature information to generate an adjusted feature vector, aiming to reduce the extreme deviation of the category and improve the accuracy of classification. The entire data processing process achieves the goal of optimizing the feature representation from identifying the feature distribution shift deviation to compensating for these deviations. This process not only improves the classifier's adaptability to changes in feature distribution, but also enhances the model's discrimination between different categories. Through the feature distribution deviation quantization compensation mechanism, the model can more accurately capture the subtle differences between feature vectors, thereby achieving higher accuracy and better generalization performance in classification tasks.

[0039] In a specific embodiment of the present application, the stirring speed data preprocessing subunit 1241 includes: arranging the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor according to the time dimension to obtain a stirring blade stirring speed input vector; multiplying the stirring blade stirring speed input vector by its transpose to obtain the stirring blade stirring speed association matrix.

[0040] It should be understood that the stirring blade plays a key role in industrial production, and its speed has a direct impact on the uniformity of the mixture and the quality of the final product. The speed data collected by the sensor at regular intervals usually exists in the form of a time series, and each time point records the operating status of the stirring blade at a specific moment. Arranging the speed values ​​of these time points according to the time dimension means organizing them in chronological order to form a speed sequence arranged along the time axis. The formation of the stirring blade stirring speed input vector can be achieved through a simple one-dimensional array or a time series. This vectorized representation makes the speed value at each time point an element of the vector, forming a continuous sequence.

[0041] Furthermore, the stirring blade stirring speed input vector is multiplied by its transpose to obtain the stirring blade stirring speed association matrix in order to analyze and quantify the mutual relationship and correlation between the stirring speeds at different time points. This operation is based on the matrix multiplication principle in linear algebra, which can help reveal the structure and pattern of speed changes during the stirring process and provide an important data basis for process optimization and quality control. Among them, the stirring blade stirring speed input vector is a one-dimensional array, in which each element represents the stirring speed value at a specific time point. By multiplying this vector by its transpose, a square matrix, namely the stirring blade stirring speed association matrix, can be obtained. Each element of this matrix represents the product of the speed values ​​at two different time points, reflecting the intrinsic correlation or similarity between them.

[0042] In the automatic control system 100 of the natural rubber processing production line, the stirring blade speed adjustment judgment module 130 is used to judge whether the stirring blade speed needs to be adjusted based on the multimodal correlation feature vector of the dehydration state of the rubber block and the optimized stirring blade stirring speed correlation feature vector. It should be understood that based on the multimodal correlation feature vector of the dehydration state of the rubber block and the optimized stirring blade stirring speed correlation feature vector, the rubber block needs to go through a dehydration stage during the production process, and the moisture content in this process has an important impact on the quality and performance of the final product. The multimodal sensor can simultaneously obtain various physical property data of the rubber block, such as temperature, humidity, density, etc., which constitute the multimodal correlation feature vector of the dehydration state of the rubber block. These feature vectors can reflect the state changes of the rubber block at different dehydration stages and provide information about the quality and maturity of the rubber. The optimized stirring blade stirring speed correlation feature vector reflects the mode and characteristics of the speed change during the stirring process. By encoding and analyzing the stirring blade speed data based on a convolutional neural network or other deep learning models, the correlation and change trend between the speeds can be extracted. By comprehensively analyzing the multimodal correlation feature vectors of the dehydration state of the rubber block and the optimized feature vectors of the stirring blade stirring speed, a prediction model or decision support system can be established. These systems can monitor the dehydration state of the rubber block and the speed dynamics during the stirring process in real time, and automatically determine whether the current stirring blade speed is suitable for the current production requirements through data analysis and model calculation. If the feature vector shows that there is a correlation change or deviation between the stirring blade speed and the dehydration state of the rubber block, the system can issue an alarm or recommend speed adjustment to ensure the stability of the rubber block dehydration process and the consistency of quality. The comprehensive analysis based on the feature vector not only improves the intelligent management level of the production process, but also reduces human errors and improves production efficiency. This data-driven approach is not limited to real-time monitoring and adjustment, but also provides in-depth insights and foundations for long-term process optimization, enabling rubber production to maintain a competitive advantage in a highly competitive market environment.

[0043] Figure 4 FIG. 1 is a block diagram of a stirring blade speed adjustment judgment module in an automatic control system of a natural rubber processing production line according to an embodiment of the present application. Figure 4 As shown, in a specific embodiment of the present application, the stirring blade speed adjustment judgment module 130 includes: a rubber block dehydration data feature fusion unit 131, which is used to fuse the multimodal association feature vector of the rubber block dehydration state and the optimized stirring blade stirring speed association feature vector to obtain a stirring blade speed adjustment judgment feature vector; a stirring blade speed classification judgment unit 132, which is used to pass the stirring blade speed adjustment judgment feature vector through a classifier to obtain a classification result, and the classification result is used to judge whether the stirring blade speed needs to be adjusted.

[0044] It should be understood that the fusion of the multimodal correlation feature vector of the dehydration state of the rubber block and the optimized feature vector associated with the stirring blade stirring speed is intended to comprehensively consider multiple key factors in the stirring process, so as to accurately evaluate whether the current stirring blade speed needs to be adjusted. This method of generating a comprehensive feature vector takes advantage of the richness of multimodal data and the advantages of deep learning models, providing intelligent decision support for stirring operations in industrial production.

[0045] Furthermore, the feature vector for determining the speed adjustment of the stirring blade integrates important information about the dehydration state of the rubber block and the change in stirring speed. The feature vector for determining the speed adjustment of the stirring blade reflects the dynamic state and potential problems of the current stirring process by integrating key features extracted from multiple sensor data and deep learning models. For example, the feature vector may include the moisture content of the rubber block, the temperature change trend, and the fluctuation pattern of the stirring blade speed, which are key factors affecting production quality and efficiency. By inputting the feature vector into the classifier for processing, a classification result can be generated based on a pre-trained model or rule system. Using the classification result to determine whether the stirring blade speed needs to be adjusted is essentially an application of a decision support system. If the classification result shows that the current stirring blade speed deviates from the optimal state or indicates that quality problems may occur, the system can automatically issue an alarm or suggest that the operator adjust the speed. This automated feedback mechanism not only reduces human subjective judgment and errors, but also enables real-time production process control and optimization. By establishing a decision support system based on feature vectors and classifiers, enterprises can achieve intelligent management of the production process. This system can not only improve production efficiency and product quality, but also reduce resource waste and costs, and enhance enterprise competitiveness.

[0046] In summary, the embodiment of the present application first obtains the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor, and then uses deep learning technology to perform feature extraction and correlation analysis on the three. Finally, the classification result is obtained through the classifier to determine whether the stirring blade speed needs to be adjusted, so as to adjust the stirring blade speed in time to avoid excessive or insufficient stirring, thereby maximizing the dehydration efficiency.

[0047] As described above, the automatic control system 100 of the natural rubber processing production line according to the embodiment of the present application can be implemented in various terminal devices. In one example, the automatic control system 100 of the natural rubber processing production line can be integrated into the terminal device as a software module and / or a hardware module. For example, the automatic control system 100 of the natural rubber processing production line can be a software module in the operating system of the terminal device, or can be an application developed for the terminal device; of course, the automatic control system 100 of the natural rubber processing production line can also be one of the many hardware modules of the terminal device.

[0048] Alternatively, in another example, the automatic control system 100 of the natural rubber processing production line and the terminal device may also be separate devices, and the automatic control system 100 of the natural rubber processing production line may be connected to the terminal device via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0049] Figure 5 FIG. 1 is a flow chart of an automatic control method for a natural rubber processing production line according to an embodiment of the present application. Figure 5 As shown, according to the automatic control method of the natural rubber processing production line of the embodiment of the present application, it includes: S110, acquiring the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor; S120, extracting the rubber block dehydration state multimodal correlation feature vector and the optimized stirring blade stirring speed correlation feature vector from the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor; S130, judging whether the stirring blade speed needs to be adjusted based on the rubber block dehydration state multimodal correlation feature vector and the optimized stirring blade stirring speed correlation feature vector.

[0050] Here, those skilled in the art will appreciate that the specific operations of each step in the automatic control method for the natural rubber processing production line have been described in the above reference. Figures 1 to 4 The automatic control system of the natural rubber processing production line has been described in detail, and therefore, its repeated description will be omitted.

[0051] Below, reference Figure 6 To describe an electronic device according to an embodiment of the present application.

[0052] like Figure 6As shown, the electronic device 10 includes an input device 11, an input interface 12, a central processing unit 13, a memory 14, an output interface 15, an output device 16, and a bus 17. The input interface 12, the central processing unit 13, the memory 14, and the output interface 15 are connected to each other through the bus 17, and the input device 11 and the output device 16 are connected to the bus 17 through the input interface 12 and the output interface 15, respectively, and then connected to other components of the electronic device 10.

[0053] Specifically, the input device 11 receives input information from the outside and transmits the input information to the central processing unit 13 through the input interface 12; the central processing unit 13 processes the input information based on the computer executable instructions stored in the memory 14 to generate output information, temporarily or permanently stores the output information in the memory 14, and then transmits the output information to the output device 16 through the output interface 15; the output device 16 outputs the output information to the outside of the electronic device 10 for user use.

[0054] In one embodiment, Figure 6 The electronic device 10 shown can be implemented as a network device, which may include: a memory configured to store programs; a processor configured to run the programs stored in the memory to execute any one of the automatic control methods for natural rubber processing production lines described in the above embodiments.

[0055] According to an embodiment of the present application, the process described above with reference to the flowchart can be implemented as a computer software program. For example, an embodiment of the present application includes a computer program product, which includes a computer program tangibly contained on a machine-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from a network, and / or installed from a removable storage medium.

[0056] It will be appreciated by those skilled in the art that all or some of the steps, systems, and functional modules / units in the methods disclosed above may be implemented as software, firmware, hardware, and appropriate combinations thereof. In hardware implementations, the division between the functional modules / units mentioned in the above description does not necessarily correspond to the division of physical components; for example, a physical component may have multiple functions, or a function or step may be performed by several physical components in cooperation. Some or all physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or implemented as hardware, or implemented as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on a computer-readable medium, which may include a computer storage medium (or non-transitory medium) and a communication medium (or temporary medium). As known to those skilled in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information (such as computer-readable instructions, data structures, program modules, or other data). Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those skilled in the art that communication media typically contain computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0057] It is to be understood that the above embodiments are merely exemplary embodiments used to illustrate the principles of the present application, but the present application is not limited thereto. For those skilled in the art, various modifications and improvements can be made without departing from the spirit and substance of the present application, and these modifications and improvements are also considered to be within the scope of protection of the present application.

Claims

1. An automatic control system for a natural rubber processing production line. It is characterized in that include: A natural rubber processing data acquisition module, used to acquire the rubber block dehydration working image acquired by the camera, the rubber block dehydration working environment condition data acquired by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points acquired by the sensor; A natural rubber processing data extraction module, used for extracting a multi-modal correlation feature vector of a rubber block dehydration state and an optimized correlation feature vector of a stirring blade stirring speed from the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor; The stirring blade speed adjustment judgment module is used to judge whether the stirring blade speed needs to be adjusted based on the multi-modal correlation feature vector of the rubber block dehydration state and the optimized stirring blade stirring speed correlation feature vector.

2. The automatic control system of the natural rubber processing production line according to claim 1, It is characterized in that The natural rubber processing data extraction module comprises: A dehydration working image feature extraction unit, used for performing feature extraction on the rubber block dehydration working image collected by the camera to obtain a dehydration state feature vector of the rubber block; An environmental condition data feature extraction unit, used for extracting features from the rubber block dehydration working environmental condition data collected by the sensor to obtain a semantically associated feature vector of the rubber block dehydration working environment; A rubber block dehydration state feature association unit, used for associating the rubber block dehydration state feature vector with the rubber block dehydration working environment semantic association feature vector to obtain the rubber block dehydration state multimodal association feature vector; The stirring speed feature extraction unit is used to extract features from the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor to obtain the optimized stirring blade stirring speed associated feature vector.

3. The automatic control system of the natural rubber processing production line according to claim 2, It is characterized in that The dehydration working image feature extraction unit comprises: The rubber block dehydration working image collected by the camera is passed through a rubber block dehydration working target detection network to obtain a region of interest of the rubber block dehydration state; The rubber block dehydration state region of interest is passed through a rubber block dehydration state feature encoder based on a salient target detector to obtain the rubber block dehydration state feature vector.

4. The automatic control system of the natural rubber processing production line according to claim 3, It is characterized in that The environmental condition data feature extraction unit comprises: Passing the rubber block dehydration working environment condition data collected by the sensor through a rubber block dehydration working environment context encoder including an embedding layer to obtain a plurality of rubber block dehydration working environment feature vectors; The plurality of rubber block dehydration working environment feature vectors are cascaded to obtain the rubber block dehydration working environment semantic association feature vector.

5. The automatic control system of the natural rubber processing production line according to claim 4, It is characterized in that The stirring speed feature extraction unit comprises: A stirring speed data preprocessing subunit, used for performing data preprocessing on the stirring speed values ​​of the stirring blade at a plurality of predetermined time points collected by the sensor to obtain a stirring blade stirring speed correlation matrix; A stirring blade stirring speed convolutional encoding subunit, used for passing the stirring blade stirring speed association matrix through a stirring blade stirring speed feature encoder based on a convolutional neural network to obtain a stirring blade stirring speed association feature vector; The stirring blade stirring speed characteristic optimization subunit is used to perform characteristic distribution deviation quantitative compensation on the stirring blade stirring speed associated characteristic vector to obtain an optimized stirring blade stirring speed associated characteristic vector.

6. The automatic control system of the natural rubber processing production line according to claim 5, It is characterized in that The stirring speed data preprocessing subunit comprises: Arrange the stirring speed values ​​of the stirring blade at a plurality of predetermined time points collected by the sensor according to the time dimension to obtain a stirring blade stirring speed input vector; The stirring blade stirring speed input vector is multiplied by its transpose to obtain the stirring blade stirring speed association matrix.

7. The automatic control system of the natural rubber processing production line according to claim 6, It is characterized in that The stirring blade stirring speed characteristic optimization subunit comprises: Calculating a shift difference representation vector between the multi-modal associated feature vector of the dehydration state of the rubber block and the associated feature vector of the stirring speed of the stirring blade; Probabilize each eigenvalue of the shifted differential representation vector based on a probability activation function to obtain a probabilized shifted differential representation vector; Inputting the stirring blade stirring speed associated feature vector into a classifier to obtain an initial classification probability value; Determine a back-propagation compensation representation vector based on a comparison between the feature values ​​of each position of the probabilistic shifted differential representation vector and the initial classification probability value and a weight matrix of the classifier; Calculating the natural exponential function value of each eigenvalue of the back-propagation compensation representation vector to obtain a back-propagation compensation class representation vector; The position-based point multiplication between the back-propagation compensation class representation vector and the stirring blade stirring speed associated feature vector is calculated to obtain a point product result vector, and the weighted sum between the point product result vector and the stirring blade stirring speed associated feature vector is calculated to obtain the optimized stirring blade stirring speed associated feature vector.

8. The automatic control system of the natural rubber processing production line according to claim 7, It is characterized in that The stirring blade speed adjustment judgment module includes: A rubber block dehydration data feature fusion unit, used for fusing the multi-modal correlation feature vector of the rubber block dehydration state and the optimized stirring blade stirring speed correlation feature vector to obtain a stirring blade speed adjustment judgment feature vector; The stirring blade speed classification and judgment unit is used to pass the stirring blade speed adjustment judgment feature vector through a classifier to obtain a classification result, and the classification result is used to judge whether the stirring blade speed needs to be adjusted.

9. An automatic control method for a natural rubber processing production line, It is characterized in that include: Acquire the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor; Extracting a multimodal correlation feature vector of a rubber block dehydration state and an optimized correlation feature vector of a stirring blade stirring speed from the rubber block dehydration working image acquired by the camera, the rubber block dehydration working environment condition data acquired by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points acquired by the sensor; Based on the multi-modal correlation feature vector of the dehydration state of the rubber block and the optimized correlation feature vector of the stirring blade stirring speed, it is determined whether the stirring blade speed needs to be adjusted.

10. The automatic control method for a natural rubber processing production line according to claim 9, It is characterized in that Extracting a multimodal correlation feature vector of the dehydration state of the rubber block and an optimized correlation feature vector of the stirring blade stirring speed from the rubber block dehydration working image collected by the camera, the rubber block dehydration working environment condition data collected by the sensor, and the stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor, including: Performing feature extraction on the rubber block dehydration working image collected by the camera to obtain a rubber block dehydration state feature vector; Performing feature extraction on the rubber block dehydration working environment condition data collected by the sensor to obtain a semantically associated feature vector of the rubber block dehydration working environment; Associating the dehydration state feature vector of the rubber block with the semantic association feature vector of the dehydration working environment of the rubber block to obtain a multimodal association feature vector of the dehydration state of the rubber block; The stirring speed values ​​of the stirring blade at multiple predetermined time points collected by the sensor are subjected to feature extraction to obtain the optimized stirring blade stirring speed associated feature vector.

Citation Information

Patent Citations

  • Intelligent preparation method and system of modified PBT material

    CN116690823A

  • High-performance mould-proof alcohol type beautifying edge-closing glue and preparation method thereof

    CN116716079A

  • Intelligent toxic and harmful gas alarm system for nitric acid preparation

    CN116935586A

  • Production method of additive for formaldehyde-removing antibacterial fabric

    CN117758517A

  • Energy management control system for preparation of lithium hexafluorophosphate and control method therefor

    WO2024000800A1

Cited By

  • Automatic control method and system for drying furnace

    CN120686755A

  • A method and system for automatic control of a drying furnace

    CN120686755B