Automatic control system and method for permanent magnet material smelting
Through deep learning technology, the temperature, pressure and magnetic induction strength during the smelting process are analyzed, and the automated temperature control of permanent magnet material smelting is achieved, which solves the problems of inefficiency and unstable quality in the traditional smelting process, and improves the smelting efficiency and product performance.
Patent Information
- Application Number
- CN202411048087.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-01
- Publication Date
- 2024-11-19
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
The smelting process of traditional permanent magnet materials relies on manual operation and experience, is inefficient and susceptible to human factors and smelting conditions, resulting in unstable smelting quality and fluctuations in product performance.
Using artificial intelligence technology based on deep learning to analyze and process the temperature, pressure and magnetic induction intensity of permanent magnet materials of the melting furnace, dig out the dynamic changes of timing characteristics, and intelligently use the timing response information of the pressure and magnetic induction intensity of the melting furnace to the temperature. Temperature adjustment to achieve automated control.
It reduces the difficulty and error of manual operation, improves smelting efficiency, and ensures the stability of smelting quality and consistency of product performance.
Smart Images

Figure CN118980254A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of intelligent control technology, and more specifically, to an automatic control system and method for smelting permanent magnetic materials. Background Art
[0002] Permanent magnetic materials, also known as hard magnetic materials, refer to magnetic materials that can retain their strong magnetism for a very long period of time after being magnetized by some external magnetic field, and have a high coercive force. Specifically, when subjected to external interference from magnetic and non-magnetic fields, permanent magnetic materials can maintain their permanent magnetic properties unchanged and have a high coercive force.
[0003] As an important functional material, permanent magnetic materials are widely used in electric vehicles, wind power generation, electronic products and other fields. In the production process of permanent magnetic materials, smelting is a key process link, and its quality directly affects the performance and quality of the final product. However, the traditional permanent magnetic material smelting process mainly relies on manual operation and experience judgment, which is not only inefficient, but also easily affected by human factors and smelting conditions, resulting in unstable smelting quality and fluctuations in product performance. Therefore, an automatic control system and method for permanent magnetic material smelting is expected. Summary of the invention
[0004] 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 permanent magnetic material smelting, which uses artificial intelligence technology based on deep learning to analyze and process the temperature and pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material during the smelting process of the permanent magnetic material, respectively dig out the time-series dynamic change characteristics of the temperature, pressure and magnetic induction intensity, and intelligently adjust the smelting furnace temperature based on the time-series response information of the smelting furnace pressure and the magnetic induction intensity of the permanent magnetic material relative to the smelting furnace temperature. In this way, the automatic temperature control of the permanent magnetic material smelting process is realized, the difficulty and error of manual operation are reduced, and the smelting efficiency is improved.
[0005] Accordingly, according to one aspect of the present application, an automatic control system for permanent magnetic material smelting is provided, comprising:
[0006] A smelting parameter monitoring module, used to obtain the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at multiple predetermined time points within a predetermined time period collected by a temperature sensor, a pressure sensor and a Hall effect sensor;
[0007] A smelting parameter time series encoding module, used for respectively extracting time series features of the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the plurality of predetermined time points to obtain a temperature time series feature vector, a pressure time series feature vector and a magnetic induction intensity time series feature vector;
[0008] A timing responsiveness analysis module, used for performing timing responsiveness analysis on the temperature timing characteristic vector, the pressure timing characteristic vector and the magnetic induction intensity timing characteristic vector to obtain a temperature response control characteristic vector;
[0009] A control instruction generation module is used to determine the control instruction of the internal temperature of the smelting furnace based on the temperature response control characteristic vector.
[0010] In the above-mentioned automatic control system for permanent magnetic material smelting, the smelting parameter timing encoding module includes: a timing arrangement unit, which is used to arrange the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the multiple predetermined time points in chronological order to obtain a temperature timing input vector, a pressure timing input vector and a magnetic induction intensity timing input vector; a timing pattern feature extraction unit, which is used to extract the timing pattern features of the temperature timing input vector, the pressure timing input vector and the magnetic induction intensity timing input vector to obtain the temperature timing feature vector, the pressure timing feature vector and the magnetic induction intensity timing feature vector.
[0011] In the above-mentioned automatic control system for permanent magnetic material smelting, the timing pattern feature extraction unit is used to: respectively pass the temperature timing input vector, the pressure timing input vector and the magnetic induction intensity timing input vector through a timing pattern feature extractor based on a one-dimensional convolution layer to obtain the temperature timing feature vector, the pressure timing feature vector and the magnetic induction intensity timing feature vector.
[0012] In the above-mentioned automatic control system for permanent magnetic material smelting, the timing responsiveness analysis module includes: a responsiveness calculation unit, which is used to respectively calculate the responsiveness estimates of the pressure timing characteristic vector and the magnetic induction intensity timing characteristic vector relative to the temperature timing characteristic vector to obtain a pressure-temperature responsiveness characteristic matrix and a magnetic induction intensity temperature response characteristic matrix; a responsiveness correlation interaction analysis unit, which is used to perform correlation interaction analysis on the pressure-temperature responsiveness characteristic matrix and the magnetic induction intensity temperature response characteristic matrix to obtain the temperature response control characteristic vector.
[0013] In the above-mentioned automatic control system for permanent magnetic material smelting, the responsiveness correlation interaction analysis unit is used to: arrange the pressure-temperature responsiveness characteristic matrix and the magnetic induction intensity temperature response characteristic matrix into a temperature global responsiveness characteristic diagram and then obtain the temperature response control characteristic vector through a responsiveness correlation feature extractor based on a convolutional data network model.
[0014] In the above-mentioned automatic control system for permanent magnet material smelting, the control instruction generation module includes: a feature optimization unit, which is used to measure the local distribution difference of the temperature response control feature vector based on the weight learning parameter to obtain an optimized temperature response control feature vector; a classification decision unit, which is used to pass the optimized temperature response control feature vector through a classifier-based temperature controller to obtain the control instruction, and the control instruction is used to indicate whether the internal temperature of the smelting furnace at the current time point should be increased, decreased or maintained.
[0015] In the above-mentioned automatic control system for permanent magnet material smelting, the feature optimization unit includes: a classification weight parameter determination subunit, which is used to determine the weight learning parameter matrix of the classifier; a vectorization subunit, which is used to extract each column vector in the weight learning parameter matrix to obtain a set of weight learning parameter column vectors; a metric correlation prediction factor calculation subunit, which is used to calculate the metric correlation prediction factor between each weight learning parameter column vector in the set of weight learning parameter column vectors and the temperature response control feature vector to obtain a metric correlation prediction factor modulation vector; a manifold modulation subunit, which is used to use the metric correlation prediction factor modulation vector as a weight vector, and calculate the position point multiplication between the metric correlation prediction factor modulation vector and the temperature response control feature vector to obtain the optimized temperature response control feature vector.
[0016] In the above-mentioned automatic control system for permanent magnet material smelting, the metric correlation prediction factor calculation subunit is used to: calculate the metric correlation prediction factor between each weight learning parameter column vector in the set of weight learning parameter column vectors and the temperature response control feature vector respectively using the following prediction factor calculation formula to obtain the metric correlation prediction factor modulation vector, wherein the prediction factor calculation formula is:
[0017]
[0018] Among them, V i represents each weight learning parameter column vector in the set of weight learning parameter column vectors, V c represents the temperature response control eigenvector, It means subtracting by position. is the superposition norm of the two norms of the eigenvectors, ||·|| F is the Frobenius norm of the eigenvector, M p represents the weight learning parameter matrix, exp(·) represents the exponential operation with the natural constant e as the base, V p represents the metric correlation predictor modulation vector.
[0019] In the above-mentioned automatic control system for permanent magnetic material smelting, the classification decision unit is used to: use the fully connected layer of the classifier-based temperature controller to fully connect the optimized temperature response control feature vector to obtain a fully connected encoded feature vector; input the fully connected encoded feature vector into the Softmax classification function of the classifier-based temperature controller to obtain the probability value of the optimized temperature response control feature vector belonging to each classification label, wherein the classification labels include the internal temperature of the smelting furnace at the current time point should increase, the internal temperature of the smelting furnace at the current time point should decrease, and the internal temperature of the smelting furnace at the current time point should be maintained; and determine the classification label corresponding to the largest of the probability values as the control instruction.
[0020] According to another aspect of the present application, there is provided an automatic control method for permanent magnetic material smelting, which comprises:
[0021] Acquiring the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at a plurality of predetermined time points within a predetermined time period collected by the temperature sensor, the pressure sensor and the Hall effect sensor;
[0022] Extracting time series features of the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the plurality of predetermined time points to obtain a temperature time series feature vector, a pressure time series feature vector and a magnetic induction intensity time series feature vector;
[0023] Performing time series responsiveness analysis on the temperature time series characteristic vector, the pressure time series characteristic vector and the magnetic induction intensity time series characteristic vector to obtain a temperature response control characteristic vector;
[0024] Based on the temperature response control characteristic vector, a control instruction for the internal temperature of the smelting furnace is determined.
[0025] Compared with the prior art, the automatic control system and method for permanent magnetic material smelting provided by the present application adopts artificial intelligence technology based on deep learning to analyze and process the temperature and pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material during the smelting process of the permanent magnetic material, respectively dig out the time-series dynamic change characteristics of the temperature, pressure and magnetic induction intensity, and intelligently adjust the smelting furnace temperature based on the time-series response information of the smelting furnace pressure and the magnetic induction intensity of the permanent magnetic material relative to the smelting furnace temperature. In this way, the automatic temperature control of the permanent magnetic material smelting process is realized, the difficulty and error of manual operation are reduced, and the smelting efficiency is improved. BRIEF DESCRIPTION OF THE DRAWINGS
[0026] 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.
[0027] Figure 1 It is a block diagram of an automatic control system for permanent magnetic material smelting according to an embodiment of the present application.
[0028] Figure 2 Schematic diagram of the architecture of an automatic control system for permanent magnet material smelting according to an embodiment of the present application.
[0029] Figure 3 It is a block diagram of a smelting parameter timing encoding module in an automatic control system for permanent magnet material smelting according to an embodiment of the present application.
[0030] Figure 4 It is a block diagram of a timing responsiveness analysis module in an automatic control system for permanent magnetic material smelting according to an embodiment of the present application.
[0031] Figure 5 The figure is a flow chart of an automatic control method for melting permanent magnet materials according to an embodiment of the present application. DETAILED DESCRIPTION
[0032] 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.
[0033] Figure 1 It is a block diagram of an automatic control system for permanent magnetic material smelting according to an embodiment of the present application. Figure 2 FIG. 1 is a schematic diagram of the architecture of an automatic control system for permanent magnetic material smelting according to an embodiment of the present application. Figure 1 and Figure 2As shown, according to the embodiment of the present application, the automatic control system 100 for smelting permanent magnetic materials includes: a smelting parameter monitoring module 110, which is used to obtain the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at multiple predetermined time points within a predetermined time period collected by a temperature sensor, a pressure sensor and a Hall effect sensor; a smelting parameter timing encoding module 120, which is used to extract the timing characteristics of the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the multiple predetermined time points to obtain a temperature timing characteristic vector, a pressure timing characteristic vector and a magnetic induction intensity timing characteristic vector; a timing responsiveness analysis module 130, which is used to perform timing responsiveness analysis on the temperature timing characteristic vector, the pressure timing characteristic vector and the magnetic induction intensity timing characteristic vector to obtain a temperature response control characteristic vector; and a control instruction generation module 140, which is used to determine the control instruction of the internal temperature of the smelting furnace based on the temperature response control characteristic vector.
[0034] As mentioned in the above background technology, permanent magnetic materials are key functional materials in the fields of electric vehicles, wind power generation, electronic products, etc. Melting is the core process in the production process of permanent magnetic materials, which is crucial to the performance and quality of the final product. However, the traditional smelting method relies on manual labor and experience, is inefficient and easily affected by human and smelting conditions, which may lead to unstable quality and product performance fluctuations. In view of the above technical problems, the technical concept of the present application is to use artificial intelligence technology based on deep learning to analyze and process the temperature, pressure and magnetic induction intensity of the smelting furnace during the smelting process of permanent magnetic materials, respectively dig out the time-series dynamic change characteristics of temperature, pressure and magnetic induction intensity, and intelligently adjust the smelting furnace temperature based on the time-series response information of the smelting furnace pressure and the magnetic induction intensity of the permanent magnetic material relative to the smelting furnace temperature. In this way, the automatic temperature control of the permanent magnetic material smelting process is realized, the difficulty and error of manual operation are reduced, and the smelting efficiency is improved.
[0035] In the above-mentioned automatic control system 100 for smelting permanent magnetic materials, the smelting parameter monitoring module 110 is used to obtain the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at multiple predetermined time points within a predetermined time period collected by the temperature sensor, the pressure sensor and the Hall effect sensor. It should be understood that the temperature, pressure and magnetic induction intensity of the permanent magnetic material inside the smelting furnace are important parameters in the smelting process. Specifically, the temperature inside the smelting furnace determines the melting, flow and reaction rate of the material during the smelting process, and has a vital influence on the composition, structure and performance of the final product. The pressure inside the smelting furnace affects the gas flow, atmosphere control and oxidation prevention during the smelting process, and plays a key role in the purity and quality stability of the product. The magnetic induction intensity of the permanent magnetic material reflects the magnetic properties and structural characteristics inside the material, and is an important indicator for evaluating the performance of the material. In other words, the temperature and pressure inside the smelting furnace have an important influence on the magnetic induction intensity of the permanent magnetic material. Based on this, in the technical solution of the present application, sensors are used to collect the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at multiple predetermined time points within a predetermined time period, and then by analyzing the influence of the temperature and pressure inside the smelting furnace on the magnetic induction intensity of the permanent magnetic material, feedback control of the internal temperature of the smelting furnace can be achieved, thereby realizing real-time monitoring and intelligent adjustment of the smelting process.
[0036] In the above-mentioned automatic control system 100 for permanent magnetic material smelting, the smelting parameter time series encoding module 120 is used to extract time series features of the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the plurality of predetermined time points to obtain a temperature time series feature vector, a pressure time series feature vector and a magnetic induction intensity time series feature vector. Specifically, Figure 3 FIG. 1 is a block diagram of a melting parameter timing coding module in an automatic control system for permanent magnet material melting according to an embodiment of the present application. Figure 3 As shown, the smelting parameter timing encoding module 120 includes: a timing arrangement unit 121, which is used to arrange the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the multiple predetermined time points in chronological order to obtain a temperature timing input vector, a pressure timing input vector and a magnetic induction intensity timing input vector; a timing pattern feature extraction unit 122, which is used to extract the timing pattern features of the temperature timing input vector, the pressure timing input vector and the magnetic induction intensity timing input vector to obtain the temperature timing feature vector, the pressure timing feature vector and the magnetic induction intensity timing feature vector.
[0037] Specifically, the time series arrangement unit 121 is used to arrange the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the plurality of predetermined time points in time order to obtain a temperature time series input vector, a pressure time series input vector and a magnetic induction intensity time series input vector. It should be understood that time is an important factor in the smelting process. Considering that the temperature, pressure and magnetic induction intensity in the smelting process are parameters that change with time, there is a certain time series relationship in the time dimension. Therefore, in order to better mine the time series dynamic change characteristics of these parameters, the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the plurality of predetermined time points are arranged in time order to retain the time series relationship of the data, form an input vector with time series characteristics, and more accurately reflect the dynamic change characteristics of the parameters in the smelting process, thereby providing a reliable data basis for subsequent time series feature extraction.
[0038] Specifically, the time series pattern feature extraction unit 122 is used to extract the time series pattern features of the temperature time series input vector, the pressure time series input vector and the magnetic induction intensity time series input vector to obtain the temperature time series feature vector, the pressure time series feature vector and the magnetic induction intensity time series feature vector. In a specific example of the present application, the encoding method for extracting the time series pattern features of the temperature time series input vector, the pressure time series input vector and the magnetic induction intensity time series input vector is to pass the temperature time series input vector, the pressure time series input vector and the magnetic induction intensity time series input vector through a time series pattern feature extractor based on a one-dimensional convolution layer to obtain the temperature time series feature vector, the pressure time series feature vector and the magnetic induction intensity time series feature vector. It should be understood that the one-dimensional convolution layer has a good feature extraction capability in time series data processing, and can effectively extract the time series feature representation of the data from the time series data, so as to better understand the time series change pattern and law in the data. In the technical solution of the present application, the time series pattern feature extractor based on the one-dimensional convolution layer performs sliding convolution operations on the temperature time series input vector, the pressure time series input vector and the magnetic induction intensity time series input vector by using a one-dimensional convolution kernel to extract the local time series correlation patterns and trends of each parameter data, effectively capturing information such as the change trend and periodic change of each parameter data in the time dimension, thereby more accurately reflecting the time series dynamic change characteristics of temperature, pressure and magnetic induction intensity during the smelting process, and improving the model's ability to analyze and control key parameters in the smelting process.
[0039] In the above-mentioned automatic control system 100 for permanent magnetic material smelting, the timing responsiveness analysis module 130 is used to perform timing responsiveness analysis on the temperature timing characteristic vector, the pressure timing characteristic vector and the magnetic induction intensity timing characteristic vector to obtain the temperature response control characteristic vector. Specifically, Figure 4 FIG. 1 is a block diagram of a timing responsiveness analysis module in an automatic control system for permanent magnetic material smelting according to an embodiment of the present application. Figure 4 As shown, the timing responsiveness analysis module 130 includes: a responsiveness calculation unit 131, used to respectively calculate the responsiveness estimates of the pressure timing characteristic vector and the magnetic induction intensity timing characteristic vector relative to the temperature timing characteristic vector to obtain a pressure-temperature responsiveness characteristic matrix and a magnetic induction intensity temperature response characteristic matrix; a responsiveness correlation interaction analysis unit 132, used to perform correlation interaction analysis on the pressure-temperature responsiveness characteristic matrix and the magnetic induction intensity temperature response characteristic matrix to obtain the temperature response control characteristic vector.
[0040] Specifically, the responsiveness calculation unit 131 is used to respectively calculate the responsiveness estimates of the pressure time series characteristic vector and the magnetic induction intensity time series characteristic vector relative to the temperature time series characteristic vector to obtain the pressure temperature responsiveness characteristic matrix and the magnetic induction intensity temperature response characteristic matrix. It should be understood that in the smelting process, there is a complex interaction relationship between temperature, pressure and magnetic induction intensity. For example, the change in temperature directly affects the magnetic induction intensity of the permanent magnetic material; at the same time, there is also a mutual influence relationship between temperature and pressure, and there is a certain correlation. Therefore, in order to better understand the interaction relationship between temperature and pressure and magnetic induction intensity, in the technical solution of the present application, the responsiveness estimates of the pressure time series characteristic vector and the magnetic induction intensity time series characteristic vector relative to the temperature time series characteristic vector are calculated respectively to reveal the degree of correlation between the parameters, effectively characterize the sensitivity of pressure and magnetic induction intensity to temperature changes, deeply understand the mutual influence relationship between different parameters, and then optimize the temperature control strategy based on this mutual influence relationship, and improve the model's understanding and control capabilities of the smelting process.
[0041] Specifically, the responsiveness correlation interaction analysis unit 132 is used to perform correlation interaction analysis on the pressure-temperature responsiveness feature matrix and the magnetic induction intensity temperature response feature matrix to obtain the temperature response control feature vector. In a specific example of the present application, the processing method for performing correlation interaction analysis on the pressure-temperature responsiveness feature matrix and the magnetic induction intensity temperature response feature matrix is to arrange the pressure-temperature responsiveness feature matrix and the magnetic induction intensity temperature response feature matrix into a temperature global responsiveness feature map and then pass through a responsiveness correlation feature extractor based on a convolutional data network model to obtain the temperature response control feature vector. It should be understood that in order to comprehensively consider the sensitivity of pressure and magnetic induction intensity to temperature changes, in the technical solution of the present application, the pressure-temperature responsiveness feature matrix and the magnetic induction intensity temperature responsiveness feature matrix are arranged along the channel dimension to spatially integrate the responsiveness relationship between different parameters, providing a comprehensive data basis for subsequent correlation interaction analysis. Subsequently, the temperature global responsiveness feature map is processed by a responsiveness correlation feature extractor based on a convolutional data network model to further explore the deep-level correlation features between parameters. Among them, the convolutional data network model has excellent performance in processing image data and can effectively extract feature information from the image. In the technical solution of the present application, the convolutional data network model is used to perform a convolution operation on the temperature global responsiveness characteristic map, which can further explore the deep correlation and interactivity between temperature, pressure and magnetic induction intensity, and capture the complex interaction pattern and correlation between pressure and magnetic induction intensity relative to temperature, so as to more accurately understand and control the temperature changes in the smelting process and improve the stability and efficiency of the smelting process.
[0042] In the above-mentioned automatic control system 100 for permanent magnetic material smelting, the control instruction generation module 140 is used to determine the control instruction of the internal temperature of the smelting furnace based on the temperature response control feature vector. In a specific example of the present application, the control instruction generation module 140 includes: a feature optimization unit, which is used to measure the local distribution difference of the temperature response control feature vector based on the weight learning parameter to obtain an optimized temperature response control feature vector; a classification decision unit, which is used to pass the optimized temperature response control feature vector through a temperature controller based on a classifier to obtain the control instruction, and the control instruction is used to indicate whether the internal temperature of the smelting furnace at the current time point should be increased, decreased or maintained.
[0043] Specifically, the feature optimization unit is used to measure the local distribution difference of the temperature response control feature vector based on the weight learning parameter to obtain the optimized temperature response control feature vector. In particular, in the technical solution of the present application, it is considered that the temperature, pressure and magnetic induction intensity inside the smelting furnace may be different due to factors such as the thermodynamic properties of the material in the furnace, the combustion efficiency, the external environment, etc., resulting in inconsistent local distribution of the time series input vector at different time points. At the same time, the time series pattern feature extractor based on the one-dimensional convolution layer may not be able to fully capture all subtle changes related to the internal state of the smelting furnace, especially when the changes are subtle or there is noise. And when calculating the responsiveness estimation of pressure and magnetic induction intensity relative to temperature, there may be differences in the coupling relationship and response speed between different feature vectors due to the complexity of the physical process. In addition, when processing the global responsiveness feature map, the response feature detector based on the convolutional data network model may be more sensitive to certain features and ignore other equally important features, resulting in local distribution differences in the temperature response control feature vector. However, the temperature response control feature vector has local distribution differences, which will cause the class harmony deviation relative to the predetermined class label of the classifier when it is classified by the classifier, affecting the classification iteration effect. In order to solve this problem, in the technical solution of the present application, the local distribution difference measurement of the temperature response control feature vector is further performed based on the weight learning parameter.
[0044] Specifically, the feature optimization unit includes: a classification weight parameter determination subunit, used to determine the weight learning parameter matrix of the classifier; a vectorization subunit, used to extract each column vector in the weight learning parameter matrix to obtain a set of weight learning parameter column vectors; a metric correlation prediction factor calculation subunit, used to calculate the metric correlation prediction factor between each weight learning parameter column vector in the set of weight learning parameter column vectors and the temperature response control feature vector to obtain a metric correlation prediction factor modulation vector; a manifold modulation subunit, used to use the metric correlation prediction factor modulation vector as a weight vector, calculate the position point multiplication between the metric correlation prediction factor modulation vector and the temperature response control feature vector to obtain the optimized temperature response control feature vector.
[0045] More specifically, the metric correlation prediction factor calculation subunit is used to calculate the metric correlation prediction factor between each weight learning parameter column vector in the set of weight learning parameter column vectors and the temperature response control feature vector using the following prediction factor calculation formula to obtain the metric correlation prediction factor modulation vector, wherein the prediction factor calculation formula is:
[0046]
[0047] Among them, Vi represents each weight learning parameter column vector in the set of weight learning parameter column vectors, V c represents the temperature response control eigenvector, It means subtracting by position. is the superposition norm of the two norms of the eigenvectors, ||·|| F is the Frobenius norm of the eigenvector, M p represents the weight learning parameter matrix, exp(·) represents the exponential operation with the natural constant e as the base, V p represents the metric correlation predictor modulation vector.
[0048] Here, considering that the temperature response control feature vector has local distribution differences, it will cause a class harmony offset relative to the predetermined class label of the classifier when it is classified by the classifier, affecting the classification iteration effect. Based on this, in the technical solution of the present application, the temperature response control feature vector is measured for local distribution differences based on weight learning parameters, and the superposition paradigm of the two norms between each column vector in the weight learning parameter matrix of the classifier and the temperature response control feature vector is used as the subclass label related measurement integral factor and the matrix norm of the weight learning parameter of the classifier is used as the differential factor, and then the temperature response control feature vector is bidirectionally linearly modulated using the association paradigm between the subclass label related measurement integral factor and the differential factor, so as to improve the local feature distribution consistency of the temperature response control feature vector by adaptively stretching and compressing the temperature response control feature vector, thereby reducing the harmony offset of the temperature response control feature vector relative to the predetermined class label of the classifier, thereby improving the accuracy of the classification result.
[0049] Specifically, the classification decision unit is used to pass the optimized temperature response control feature vector through a temperature controller based on a classifier to obtain the control instruction, and the control instruction is used to indicate whether the temperature inside the smelting furnace at the current time point should be increased, decreased, or maintained. It should be understood that the temperature controller can judge how the temperature inside the smelting furnace should change based on the feature information in the optimized temperature response control feature vector based on the classification algorithm, and generate corresponding control instructions based on the judgment result, so as to perform precise temperature control based on the control instructions. Through this control method, the unstable product quality caused by temperature fluctuations during the smelting process can be effectively reduced, and the production efficiency and quality of the product can be improved. At the same time, since this control method is based on real-time temperature response data, the smelting process can be adjusted and optimized in real time, further improving the flexibility and controllability of the smelting process.
[0050] Specifically, the classification decision unit is used to: use the fully connected layer of the classifier-based temperature controller to fully connect encode the optimized temperature response control feature vector to obtain a fully connected encoded feature vector; input the fully connected encoded feature vector into the Softmax classification function of the classifier-based temperature controller to obtain the probability value of the optimized temperature response control feature vector belonging to each classification label, wherein the classification labels include the internal temperature of the smelting furnace at the current time point should increase, the internal temperature of the smelting furnace at the current time point should decrease, and the internal temperature of the smelting furnace at the current time point should be maintained; and determine the classification label corresponding to the largest of the probability values as the control instruction.
[0051] In summary, the automatic control system for permanent magnetic material smelting according to the embodiment of the present application is explained, which uses artificial intelligence technology based on deep learning to analyze and process the temperature and pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material during the smelting process of the permanent magnetic material, respectively dig out the time-series dynamic change characteristics of the temperature, pressure and magnetic induction intensity, and intelligently adjust the smelting furnace temperature based on the time-series response information of the smelting furnace pressure and the magnetic induction intensity of the permanent magnetic material relative to the smelting furnace temperature. In this way, the automatic temperature control of the permanent magnetic material smelting process is realized, the difficulty and error of manual operation are reduced, and the smelting efficiency is improved.
[0052] Figure 5 FIG. 1 is a flow chart of an automatic control method for melting permanent magnetic materials according to an embodiment of the present application. Figure 5 As shown, the automatic control method for smelting permanent magnetic materials according to the embodiment of the present application includes the steps of: S110, acquiring the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at multiple predetermined time points within a predetermined time period collected by the temperature sensor, the pressure sensor and the Hall effect sensor; S120, performing time series feature extraction on the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the multiple predetermined time points to obtain a temperature time series feature vector, a pressure time series feature vector and a magnetic induction intensity time series feature vector; S130, performing time series responsiveness analysis on the temperature time series feature vector, the pressure time series feature vector and the magnetic induction intensity time series feature vector to obtain a temperature response control feature vector; S140, determining the control instruction of the internal temperature of the smelting furnace based on the temperature response control feature vector.
[0053] Here, those skilled in the art will appreciate that the specific operations of each step in the above-mentioned automatic control method for permanent magnetic material smelting have been described in the above reference. Figures 1 to 4 The description of the automatic control system for permanent magnetic material melting has been described in detail, and therefore, its repeated description will be omitted.
[0054] In the above embodiments, the description of each embodiment has its own emphasis. For parts that are not described or recorded in detail in a certain embodiment, please refer to the relevant description of other embodiments. In the several embodiments provided by the present invention, it should be understood that the disclosed devices, systems and methods can be implemented in other ways. For example, the system embodiments described above are only schematic. For example, the module division is only a logical function division, and there may be other division methods in actual implementation. The modules described as separate components may or may not be physically separated, and the components displayed as modules may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment.
[0055] In addition, each functional module in each embodiment of the present invention may be integrated into one processing unit, each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of hardware plus software functional modules.
[0056] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference to a figure in a claim should not be considered as limiting the claim to which it relates.
[0057] In addition, it is obvious that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. The multiple units stated in the system claims can also be implemented by one unit through software or hardware.
[0058] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.
Claims
1. An automatic control system for permanent magnetic material smelting, It is characterized in that include: A smelting parameter monitoring module, used to obtain the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at multiple predetermined time points within a predetermined time period collected by a temperature sensor, a pressure sensor and a Hall effect sensor; A smelting parameter time series encoding module, used for respectively extracting time series features of the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the plurality of predetermined time points to obtain a temperature time series feature vector, a pressure time series feature vector and a magnetic induction intensity time series feature vector; A timing responsiveness analysis module, used for performing timing responsiveness analysis on the temperature timing characteristic vector, the pressure timing characteristic vector and the magnetic induction intensity timing characteristic vector to obtain a temperature response control characteristic vector; A control instruction generation module is used to determine the control instruction of the internal temperature of the smelting furnace based on the temperature response control characteristic vector.
2. The automatic control system for permanent magnetic material smelting according to claim 1, It is characterized in that The smelting parameter timing coding module includes: A time sequence arrangement unit, used for arranging the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the plurality of predetermined time points in time sequence to obtain a temperature time sequence input vector, a pressure time sequence input vector and a magnetic induction intensity time sequence input vector; A timing pattern feature extraction unit is used to extract the timing pattern features of the temperature timing input vector, the pressure timing input vector and the magnetic induction intensity timing input vector to obtain the temperature timing feature vector, the pressure timing feature vector and the magnetic induction intensity timing feature vector.
3. The automatic control system for permanent magnetic material smelting according to claim 2, It is characterized in that The temporal pattern feature extraction unit is used to: The temperature time series input vector, the pressure time series input vector and the magnetic induction intensity time series input vector are respectively passed through a time series pattern feature extractor based on a one-dimensional convolutional layer to obtain the temperature time series feature vector, the pressure time series feature vector and the magnetic induction intensity time series feature vector.
4. The automatic control system for permanent magnetic material smelting according to claim 3, It is characterized in that The timing responsiveness analysis module comprises: A responsiveness calculation unit, used to respectively calculate the responsiveness estimates of the pressure time series characteristic vector and the magnetic induction intensity time series characteristic vector relative to the temperature time series characteristic vector to obtain a pressure-temperature responsiveness characteristic matrix and a magnetic induction intensity-temperature responsiveness characteristic matrix; A responsiveness correlation interaction analysis unit is used to perform correlation interaction analysis on the pressure-temperature responsiveness characteristic matrix and the magnetic induction intensity temperature response characteristic matrix to obtain the temperature response control characteristic vector.
5. The automatic control system for permanent magnetic material smelting according to claim 4, It is characterized in that The responsiveness association interaction analysis unit is used to: The pressure-temperature responsiveness characteristic matrix and the magnetic induction intensity temperature responsiveness characteristic matrix are arranged into a temperature global responsiveness characteristic diagram and then passed through a responsiveness association feature extractor based on a convolutional data network model to obtain the temperature response control characteristic vector.
6. The automatic control system for permanent magnetic material smelting according to claim 5, It is characterized in that The control instruction generating module comprises: A feature optimization unit, used for performing a local distribution difference measurement on the temperature response control feature vector based on a weight learning parameter to obtain an optimized temperature response control feature vector; The classification decision unit is used to pass the optimized temperature response control feature vector through a temperature controller based on a classifier to obtain the control instruction, wherein the control instruction is used to indicate whether the internal temperature of the smelting furnace at the current time point should be increased, decreased or maintained.
7. The automatic control system for permanent magnetic material smelting according to claim 6, It is characterized in that The feature optimization unit comprises: A classification weight parameter determination subunit, used to determine a weight learning parameter matrix of the classifier; A vectorization subunit, used for extracting each column vector in the weight learning parameter matrix to obtain a set of weight learning parameter column vectors; A metric correlation prediction factor calculation subunit, used to calculate a metric correlation prediction factor between each weight learning parameter column vector in the set of weight learning parameter column vectors and the temperature response control feature vector to obtain a metric correlation prediction factor modulation vector; The manifold modulation subunit is used to use the metric correlation prediction factor modulation vector as a weight vector, calculate the position point multiplication between the metric correlation prediction factor modulation vector and the temperature response control feature vector to obtain the optimized temperature response control feature vector.
8. The automatic control system for permanent magnetic material smelting according to claim 7, It is characterized in that The metric correlation prediction factor calculation subunit is used to: The metric correlation prediction factor between each weight learning parameter column vector in the set of the weight learning parameter column vector and the temperature response control feature vector is calculated respectively using the following prediction factor calculation formula to obtain the metric correlation prediction factor modulation vector, wherein the prediction factor calculation formula is: Among them, V i represents each weight learning parameter column vector in the set of weight learning parameter column vectors, V c represents the temperature response control eigenvector, It means subtracting by position. is the superposition norm of the two norms of the eigenvectors, ||·|| F is the Frobenius norm of the eigenvector, M p represents the weight learning parameter matrix, exp(·) represents the exponential operation with the natural constant e as the base, V p represents the metric correlation predictor modulation vector.
9. The automatic control system for permanent magnetic material smelting according to claim 8, It is characterized in that The classification decision unit is used to: Performing full connection encoding on the optimized temperature response control feature vector using a fully connected layer of the classifier-based temperature controller to obtain a fully connected encoded feature vector; Inputting the fully connected encoded feature vector into the Softmax classification function of the classifier-based temperature controller to obtain probability values of the optimized temperature response control feature vector belonging to each classification label, wherein the classification labels include the internal temperature of the smelting furnace at the current time point should be increased, the internal temperature of the smelting furnace at the current time point should be decreased, and the internal temperature of the smelting furnace at the current time point should be maintained; The classification label corresponding to the largest probability value among the probability values is determined as the control instruction.
10. An automatic control method for melting permanent magnetic materials, It is characterized in that include: Acquiring the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at a plurality of predetermined time points within a predetermined time period collected by the temperature sensor, the pressure sensor and the Hall effect sensor; Extracting time series features of the internal temperature of the smelting furnace, the internal pressure of the smelting furnace and the magnetic induction intensity of the permanent magnetic material at the plurality of predetermined time points to obtain a temperature time series feature vector, a pressure time series feature vector and a magnetic induction intensity time series feature vector; Performing time series responsiveness analysis on the temperature time series characteristic vector, the pressure time series characteristic vector and the magnetic induction intensity time series characteristic vector to obtain a temperature response control characteristic vector; Based on the temperature response control characteristic vector, a control instruction for the internal temperature of the smelting furnace is determined.
Citation Information
Cited By
Data acquisition and analysis system and method based on web crawler technology
CN119558291A
