Intelligent temperature control method and system for high-silica glass fiber production line

Abstract

The invention discloses an intelligent temperature control method and system for a high-silica glass fiber production line, and relates to the technical field of high-silica glass fiber production.The intelligent temperature control method comprises the following steps that firstly, temperature sensors arranged at all nodes in a furnace hearth of a smelting furnace collect temperature data in real time, and on the basis of a finite element analysis algorithm, the temperature data are obtained; a regional temperature field model is constructed in combination with temperature data, content data of trace impurities in the raw materials are acquired in real time by adopting online detection equipment, the viscosity change trend of the melt under the current raw materials can be analyzed, the current temperature threshold value is dynamically adjusted according to the viscosity change, and the limitation of a traditional fixed temperature threshold value is broken through; the temperature control can better adapt to real-time fluctuation of trace impurities in raw materials, and proper temperature control can be realized under different raw material conditions, so that the qualified rate of products is effectively improved, the loss rate of the raw materials is reduced, manual intervention is reduced, the timeliness and accuracy of temperature control are improved, and the production cost is reduced. And the production efficiency and the product quality are further improved.

Description

Technical Field

[0001] This invention relates to the field of high-silica glass fiber production technology, specifically to a method and system for intelligent temperature control in a high-silica glass fiber production line. Background Technology

[0002] High-silica glass fiber is a special type of glass fiber with a SiO2 content exceeding 96%. It possesses high temperature resistance (long-term operating temperature can reach over 1000℃), excellent chemical stability, and electrical insulation properties, and is widely used in high-end fields such as aerospace, high-temperature filtration, and fireproofing and heat insulation. A high-silica glass fiber production line is a continuous production system that realizes the entire process from glass raw material melting, fiber drawing, forming to finished product processing. Its core processes include raw material melting (melting raw materials such as quartz sand into a homogeneous melt in a high-temperature furnace), fiber drawing (drawing molten glass into fibers through a spinneret), and high-temperature sintering (promoting the rearrangement and crystallization of SiO2 components in amorphous glass through gradient heating).

[0003] In the production process of high-silica glass fiber, the furnace is a key piece of equipment. Temperature fluctuations in the furnace can trigger a chain reaction, directly affecting product quality and production efficiency. Existing temperature control methods use single-point temperature measurement to characterize the regional temperature field, ignoring the temperature gradient formed by thermal convection within the furnace. This can lead to localized overheating, causing bubbles in the glass melt, or insufficient temperature, resulting in fiber breakage during drawing. Furthermore, the temperature threshold is a fixed value, failing to consider the real-time fluctuations of trace impurities in the raw materials. When the impurity content increases, the melt viscosity changes significantly at the same temperature. Relying solely on fixed temperature adjustment can lead to large deviations in the drawing diameter, resulting in more defective products and increased raw material loss.

[0004] Therefore, in view of this, the present invention proposes a method and system for intelligent temperature control of a high-silica glass fiber production line to make up for and improve the deficiencies of the prior art. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention provides an intelligent temperature control method and system for a high-silica glass fiber production line, thereby resolving the corresponding technical issues raised in the background section.

[0006] To achieve the above objectives, the technical solution adopted by this invention is: a method for intelligent temperature control in a high-silica glass fiber production line, comprising the following steps: Step 1: Temperature data is collected in real time by temperature sensors installed at various nodes in the furnace chamber. Based on the finite element analysis algorithm, a regional temperature field model is constructed by combining the temperature data. At the same time, online detection equipment is used to obtain the content data of trace impurities in the raw materials in real time. Step 2: Obtain historical content data of trace impurities, historical viscosity data of melt, and corresponding historical temperature thresholds from the historical database. Calculate the correlation coefficient between historical content data and historical viscosity data using correlation analysis. Based on a deep neural network, construct a correlation model between content and viscosity. Combine this with real-time acquired content data to analyze the viscosity change trend of melt under the current trace impurity content in the raw material. Step 3: Based on the viscosity change trend of the melt and the corresponding historical temperature threshold, dynamically adjust the current temperature threshold, compare the temperature data of each region in the constructed regional temperature field model with the dynamically adjusted current temperature threshold, and generate corresponding operation instructions based on the comparison results to regulate the temperature of the furnace.

[0007] Preferably, in step two, the correlation analysis method is the Pearson correlation coefficient method.

[0008] As a preferred approach, the specific process of constructing a regional temperature field model based on the finite element analysis algorithm and temperature data is as follows: S101. The internal space of the furnace chamber is divided using regular hexahedral elements. Let the length, width, and height of the furnace chamber in three-dimensional space be L, W, and H, respectively. Divide it into... Each small hexahedral unit has a length, width, and height of [missing information]. , , ; S102. Set nodes at the vertices of each small hexahedral unit region, and collect temperature data of each node in real time using temperature sensors. ,in, N represents the total number of nodes. Using the finite element analysis algorithm, the temperature distribution within each small hexahedral element is approximated by a linear interpolation function based on the temperature data of each node. The calculation results are then integrated to construct a regional temperature field model.

[0009] As a preferred method, the specific process for calculating the correlation coefficient between historical content data and historical viscosity data using correlation analysis is as follows: S201. Obtain historical content data of trace impurities from the historical database. Historical viscosity data of melt and the corresponding historical temperature thresholds, among which, n represents the number of data samples, and historical content data and historical viscosity data are organized into corresponding data pairs. ; Calculate the mean of historical content data respectively and the mean of historical viscosity data Its formula is: ; ; Calculate the covariance of historical content data and historical viscosity data. Its formula is: ; Calculate the standard deviation of historical content data separately. Standard deviation of historical viscosity data Its formula is: ; ; S202. Calculate the correlation coefficient r between historical content data and historical viscosity data according to the Pearson correlation coefficient formula. The formula is as follows: The correlation coefficient r ranges from 1 to 10. The closer the absolute value of r is to 1, the stronger the linear correlation between the two variables; the closer the absolute value of r is to 0, the weaker the linear correlation between the two variables.

[0010] As a preferred approach, the specific process for analyzing the viscosity change trend of the melt under the current trace impurity content in the raw material is as follows: S301. Obtain historical data on trace impurities, historical viscosity data of melt, and corresponding historical temperature thresholds from the historical database, and integrate them into a historical dataset. Divide the historical dataset into a training set and a validation set in a 7:3 ratio. Construct a correlation model between content and viscosity based on a deep neural network. Use the training set to train the correlation model. Continuously adjust the weights and bias parameters of the correlation model through the backpropagation algorithm so that the predicted value of the model gradually approaches the true value. Use the validation set to validate the trained correlation model. S302, The content data acquired in real time The data is input into the correlation model, which then outputs the predicted viscosity data. Its formula is: ,in, It is the weight matrix of each layer of the deep neural network. is the bias vector of each layer, and f is the activation function; S303. Compare the predicted viscosity data with the viscosity data of the melt under the same trace impurity content in the historical database, and analyze the viscosity change trend of the melt under the current trace impurity content in the raw material.

[0011] As a preferred approach, the specific process of comparing the temperature data of each region in the constructed regional temperature field model with the dynamically adjusted current temperature threshold is as follows: Based on the viscosity variation trend of the melt and the corresponding historical temperature threshold, the current temperature threshold is dynamically adjusted. Based on a preset temperature deviation range, the temperature data of each region in the constructed regional temperature field model are then used. With the dynamically adjusted current temperature threshold and preset temperature deviation range A comparison is performed, and based on the comparison results, corresponding operation instructions are generated to adjust the furnace temperature. The comparison is as follows: like If the temperature in the area is too high, an operation command to reduce the temperature in the area will be generated accordingly. like If the temperature in the area is too low, an operation command to raise the temperature in the area will be generated accordingly. like If the temperature in the area is normal, no temperature adjustment is required.

[0012] A temperature intelligent control system for a high-silica glass fiber production line includes a data acquisition unit, a change analysis unit, and an instruction generation unit. The data acquisition unit is used to collect temperature data in real time through temperature sensors installed at various nodes in the furnace chamber. Based on the finite element analysis algorithm, it constructs a regional temperature field model by combining the temperature data. At the same time, it uses online detection equipment to obtain the content data of trace impurities in the raw materials in real time and sends it to the change analysis unit. The change analysis unit is used to acquire temperature data and the content data of trace impurities in the raw materials, and to acquire historical content data of trace impurities, historical viscosity data of melt, and corresponding historical temperature thresholds from the historical database. It calculates the correlation coefficient between historical content data and historical viscosity data through correlation analysis, and constructs a correlation model between content and viscosity based on a deep neural network. Combined with the real-time acquired content data, it analyzes the viscosity change trend of melt under the current content of trace impurities in the raw materials, and sends it to the instruction generation unit. The instruction generation unit is used to obtain the viscosity change trend of the melt, and dynamically adjust the current temperature threshold according to the viscosity change trend of the melt and the corresponding historical temperature threshold. It also compares the temperature data of each region in the constructed regional temperature field model with the dynamically adjusted current temperature threshold, and generates corresponding operation instructions based on the comparison results to regulate the temperature of the furnace.

[0013] Compared with existing technologies, the beneficial effects of this invention are as follows: By collecting temperature data in real time, a regional temperature field model is constructed, and the content data of trace impurities in the raw materials is obtained. Historical content data of trace impurities, historical viscosity data of the melt, and corresponding historical temperature thresholds are obtained from historical databases. The correlation coefficient between historical content data and historical viscosity data is calculated, and a correlation model between content and viscosity is constructed. The viscosity change trend of the melt under the current content of trace impurities in the raw materials is analyzed. Combined with the corresponding historical temperature thresholds, the current temperature threshold is dynamically adjusted. The temperature data of each region in the constructed regional temperature field model is compared with the dynamically adjusted current temperature threshold. Based on the comparison results, corresponding operation instructions are generated to regulate the temperature of the furnace. This can overcome the limitations of traditional fixed temperature thresholds, enabling temperature control to better adapt to the real-time fluctuations of trace impurities in the raw materials. It ensures that appropriate temperature control can be achieved under different raw material conditions, thereby effectively improving the product qualification rate and reducing the raw material loss rate. At the same time, the corresponding operation instructions are automatically generated based on the comparison results to precisely regulate the temperature of the furnace. This intelligent operation method reduces manual intervention, improves the timeliness and accuracy of temperature control, and further enhances production efficiency and product quality. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the overall structure of a preferred embodiment of the present invention; Figure 2 This is a schematic diagram of the intelligent control system shown in this invention. Detailed Implementation

[0015] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0016] Embodiments of the present invention: Please refer to Figures 1 to 2 As shown, a method for intelligent temperature control in a high-silica glass fiber production line includes the following steps: Step 1: Temperature data is collected in real time by temperature sensors installed at various nodes in the furnace chamber. Based on the finite element analysis algorithm, a regional temperature field model is constructed by combining the temperature data. At the same time, online detection equipment is used to obtain the content data of trace impurities in the raw materials in real time. Step 2: Obtain historical content data of trace impurities, historical viscosity data of melt, and corresponding historical temperature thresholds from the historical database. Calculate the correlation coefficient between historical content data and historical viscosity data using correlation analysis. Based on a deep neural network, construct a correlation model between content and viscosity. Combine this with real-time acquired content data to analyze the viscosity change trend of melt under the current trace impurity content in the raw material. Step 3: Based on the viscosity change trend of the melt and the corresponding historical temperature threshold, dynamically adjust the current temperature threshold, compare the temperature data of each region in the constructed regional temperature field model with the dynamically adjusted current temperature threshold, and generate corresponding operation instructions based on the comparison results to regulate the temperature of the furnace.

[0017] In step two, the correlation analysis method is the Pearson correlation coefficient method.

[0018] The specific process of constructing a regional temperature field model based on the finite element analysis algorithm and temperature data is as follows: S101. The internal space of the furnace chamber is divided using regular hexahedral elements. Let the length, width, and height of the furnace chamber in three-dimensional space be L, W, and H, respectively. Divide it into... Each small hexahedral unit has a length, width, and height of [missing information]. , , ; S102. Set nodes at the vertices of each small hexahedral unit region, and collect temperature data of each node in real time using temperature sensors. ,in, N represents the total number of nodes. Using the finite element analysis algorithm, the temperature distribution within each small hexahedral element is approximated by a linear interpolation function based on the temperature data of each node. The calculation results are then integrated to construct a regional temperature field model.

[0019] The specific process for calculating the correlation coefficient between historical content data and historical viscosity data using correlation analysis is as follows: S201. Obtain historical content data of trace impurities from the historical database. Historical viscosity data of melt and the corresponding historical temperature thresholds, among which, n represents the number of data samples, and historical content data and historical viscosity data are organized into corresponding data pairs. ; Calculate the mean of historical content data respectively and the mean of historical viscosity data Its formula is: ; ; Calculate the covariance of historical content data and historical viscosity data. Its formula is: ; Calculate the standard deviation of historical content data separately. Standard deviation of historical viscosity data Its formula is: ; ; S202. Calculate the correlation coefficient r between historical content data and historical viscosity data according to the Pearson correlation coefficient formula. The formula is as follows: The correlation coefficient r ranges from 1 to 10. The closer the absolute value of r is to 1, the stronger the linear correlation between the two variables; the closer the absolute value of r is to 0, the weaker the linear correlation between the two variables.

[0020] The specific process for analyzing the viscosity change trend of the melt under the current trace impurity content in the raw material is as follows: S301. Obtain historical data on trace impurities, historical viscosity data of melt, and corresponding historical temperature thresholds from the historical database, and integrate them into a historical dataset. Divide the historical dataset into a training set and a validation set in a 7:3 ratio. Construct a correlation model between content and viscosity based on a deep neural network. Use the training set to train the correlation model. Continuously adjust the weights and bias parameters of the correlation model through the backpropagation algorithm so that the predicted value of the model gradually approaches the true value. Use the validation set to validate the trained correlation model. S302, The content data acquired in real time The data is input into the correlation model, which then outputs the predicted viscosity data. Its formula is: ,in, It is the weight matrix of each layer of the deep neural network. is the bias vector of each layer, and f is the activation function; S303. Compare the predicted viscosity data with the viscosity data of the melt under the same trace impurity content in the historical database, and analyze the viscosity change trend of the melt under the current trace impurity content in the raw material.

[0021] The specific process of comparing the temperature data of each region in the constructed regional temperature field model with the dynamically adjusted current temperature threshold is as follows: Based on the viscosity variation trend of the melt and the corresponding historical temperature threshold, the current temperature threshold is dynamically adjusted. Based on a preset temperature deviation range, the temperature data of each region in the constructed regional temperature field model are then used. With the dynamically adjusted current temperature threshold and preset temperature deviation range A comparison is performed, and based on the comparison results, corresponding operation instructions are generated to adjust the furnace temperature. The comparison is as follows: like If the temperature in the area is too high, an operation command to reduce the temperature in the area will be generated accordingly. like If the temperature in the area is too low, an operation command to raise the temperature in the area will be generated accordingly. like If the temperature in the area is normal, no temperature adjustment is required.

[0022] A temperature intelligent control system for a high-silica glass fiber production line includes a data acquisition unit, a change analysis unit, and an instruction generation unit. The data acquisition unit is used to collect temperature data in real time through temperature sensors installed at various nodes in the furnace chamber. Based on the finite element analysis algorithm, it constructs a regional temperature field model by combining the temperature data. At the same time, it uses online detection equipment to obtain the content data of trace impurities in the raw materials in real time and sends it to the change analysis unit. The change analysis unit is used to acquire temperature data and the content data of trace impurities in the raw materials. It also acquires historical content data of trace impurities, historical viscosity data of the melt, and corresponding historical temperature thresholds from the historical database. It calculates the correlation coefficient between historical content data and historical viscosity data through correlation analysis and constructs a correlation model between content and viscosity based on a deep neural network. Combined with the real-time acquired content data, it analyzes the viscosity change trend of the melt under the current content of trace impurities in the raw materials and sends it to the instruction generation unit. The instruction generation unit is used to obtain the viscosity change trend of the melt, and dynamically adjust the current temperature threshold based on the viscosity change trend of the melt and the corresponding historical temperature threshold. It also compares the temperature data of each region in the constructed regional temperature field model with the dynamically adjusted current temperature threshold, and generates corresponding operation instructions based on the comparison results to regulate the temperature of the furnace.

[0023] By collecting temperature data in real time, a regional temperature field model is constructed. Simultaneously, data on the content of trace impurities in the raw materials is acquired. Historical data on the content of trace impurities, the historical viscosity of the melt, and the corresponding historical temperature thresholds are obtained from a historical database. The correlation coefficient between historical content and viscosity data is calculated, and a correlation model between content and viscosity is constructed. The viscosity change trend of the melt under the current trace impurity content in the raw materials is analyzed. Combined with the corresponding historical temperature thresholds, the current temperature threshold is dynamically adjusted. The temperature data of each region in the constructed regional temperature field model is compared with the dynamically adjusted current temperature threshold. Based on the comparison results, corresponding operation commands are generated to regulate the furnace temperature. This overcomes the limitations of traditional fixed temperature thresholds, allowing temperature control to better adapt to real-time fluctuations in trace impurities in the raw materials. It ensures suitable temperature control under different raw material conditions, effectively improving product qualification rate and reducing raw material loss rate. Furthermore, the automatic generation of corresponding operation commands based on the comparison results enables precise temperature control of the furnace. This intelligent operation method reduces manual intervention, improves the timeliness and accuracy of temperature control, and further enhances production efficiency and product quality.

[0024] The size of the interval and threshold is set to facilitate comparison. The size of the threshold depends on the amount of sample data and the number of bases set by those skilled in the art for each set of sample data; as long as it does not affect the ratio between the parameter and the quantized value.

[0025] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation. In the two embodiments provided in this application, it should be understood that the disclosed apparatus and system can be implemented in other ways; for example, the apparatus embodiments described above are merely illustrative, and the division of modules is merely a logical functional division. In actual implementation, there may be other division methods, such as multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed; furthermore, the coupling or direct coupling or communication connection between the shown or discussed mutuals can be through some interfaces, and the indirect coupling or communication connection between the apparatus or modules can be electrical, mechanical or other forms. The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for intelligent temperature control in a high-silica glass fiber production line, characterized in that, Includes the following steps: Step 1: Temperature data is collected in real time by temperature sensors installed at various nodes in the furnace chamber. Based on the finite element analysis algorithm, a regional temperature field model is constructed by combining the temperature data. At the same time, online detection equipment is used to obtain the content data of trace impurities in the raw materials in real time. Step 2: Obtain historical content data of trace impurities, historical viscosity data of melt, and corresponding historical temperature thresholds from the historical database. Calculate the correlation coefficient between historical content data and historical viscosity data using correlation analysis. Based on a deep neural network, construct a correlation model between content and viscosity. Combine this with real-time acquired content data to analyze the viscosity change trend of melt under the current trace impurity content in the raw material. Step 3: Based on the viscosity change trend of the melt and the corresponding historical temperature threshold, dynamically adjust the current temperature threshold, compare the temperature data of each region in the constructed regional temperature field model with the dynamically adjusted current temperature threshold, and generate corresponding operation instructions based on the comparison results to regulate the temperature of the furnace.

2. The intelligent temperature control method for a high-silica glass fiber production line according to claim 1, characterized in that, In step two, the correlation analysis method is the Pearson correlation coefficient method.

3. The intelligent temperature control method for a high-silica glass fiber production line according to claim 2, characterized in that, The specific process of constructing a regional temperature field model based on the finite element analysis algorithm and temperature data is as follows: S101. The internal space of the furnace chamber is divided using regular hexahedral elements. Let the length, width, and height of the furnace chamber in three-dimensional space be L, W, and H, respectively. Divide it into... Each small hexahedral unit has a length, width, and height of [missing information]. , , ; S102. Set nodes at the vertices of each small hexahedral unit region, and collect temperature data of each node in real time using temperature sensors. ,in, N represents the total number of nodes. Using the finite element analysis algorithm, the temperature distribution within each small hexahedral element is approximated by a linear interpolation function based on the temperature data of each node. The calculation results are then integrated to construct a regional temperature field model.

4. The intelligent temperature control method for a high-silica glass fiber production line according to claim 3, characterized in that, The specific process for calculating the correlation coefficient between historical content data and historical viscosity data using correlation analysis is as follows: S201. Obtain historical content data of trace impurities from the historical database. Historical viscosity data of melt and the corresponding historical temperature thresholds, among which, n represents the number of data samples, and historical content data and historical viscosity data are organized into corresponding data pairs. ; Calculate the mean of historical content data respectively and the mean of historical viscosity data Its formula is: ； ； Calculate the covariance of historical content data and historical viscosity data. Its formula is: ； Calculate the standard deviation of historical content data separately. Standard deviation of historical viscosity data Its formula is: ； ； S202. Calculate the correlation coefficient r between historical content data and historical viscosity data according to the Pearson correlation coefficient formula. The formula is as follows: The correlation coefficient r ranges from 1 to 10. The closer the absolute value of r is to 1, the stronger the linear correlation between the two variables; the closer the absolute value of r is to 0, the weaker the linear correlation between the two variables.

5. The intelligent temperature control method for a high-silica glass fiber production line according to claim 4, characterized in that, The specific process for analyzing the viscosity change trend of the melt under the current trace impurity content in the raw material is as follows: S301. Obtain historical data on trace impurities, historical viscosity data of melt, and corresponding historical temperature thresholds from the historical database, and integrate them into a historical dataset. Divide the historical dataset into a training set and a validation set in a 7:3 ratio. Construct a correlation model between content and viscosity based on a deep neural network. Use the training set to train the correlation model. Continuously adjust the weights and bias parameters of the correlation model through the backpropagation algorithm so that the predicted value of the model gradually approaches the true value. Use the validation set to validate the trained correlation model. S302, The content data acquired in real time The data is input into the correlation model, which then outputs the predicted viscosity data. Its formula is: ,in, It is the weight matrix of each layer of the deep neural network. is the bias vector of each layer, and f is the activation function; S303. Compare the predicted viscosity data with the viscosity data of the melt under the same trace impurity content in the historical database, and analyze the viscosity change trend of the melt under the current trace impurity content in the raw material.

6. The intelligent temperature control method for a high-silica glass fiber production line according to claim 5, characterized in that, The specific process of comparing the temperature data of each region in the constructed regional temperature field model with the dynamically adjusted current temperature threshold is as follows: Based on the viscosity variation trend of the melt and the corresponding historical temperature threshold, the current temperature threshold is dynamically adjusted. Based on a preset temperature deviation range, the temperature data of each region in the constructed regional temperature field model are then used. With the dynamically adjusted current temperature threshold and preset temperature deviation range A comparison is performed, and based on the comparison results, corresponding operation instructions are generated to adjust the furnace temperature. The comparison is as follows: like If the temperature in the area is too high, an operation command to reduce the temperature in the area will be generated accordingly. like If the temperature in the area is too low, an operation command to raise the temperature in the area will be generated accordingly. like If the temperature in the area is normal, no temperature adjustment is required.

7. A temperature intelligent control system for a high-silica glass fiber production line, applied to the temperature intelligent control method for a high-silica glass fiber production line as described in any one of claims 1-6, characterized in that, It includes a data acquisition unit, a change analysis unit, and an instruction generation unit; The data acquisition unit is used to collect temperature data in real time through temperature sensors installed at various nodes in the furnace chamber. Based on the finite element analysis algorithm, it constructs a regional temperature field model by combining the temperature data. At the same time, it uses online detection equipment to obtain the content data of trace impurities in the raw materials in real time and sends it to the change analysis unit. The change analysis unit is used to acquire temperature data and the content data of trace impurities in the raw materials, and to acquire historical content data of trace impurities, historical viscosity data of melt, and corresponding historical temperature thresholds from the historical database. It calculates the correlation coefficient between historical content data and historical viscosity data through correlation analysis, and constructs a correlation model between content and viscosity based on a deep neural network. Combined with the real-time acquired content data, it analyzes the viscosity change trend of melt under the current content of trace impurities in the raw materials, and sends it to the instruction generation unit. The instruction generation unit is used to obtain the viscosity change trend of the melt, and dynamically adjust the current temperature threshold according to the viscosity change trend of the melt and the corresponding historical temperature threshold. It also compares the temperature data of each region in the constructed regional temperature field model with the dynamically adjusted current temperature threshold, and generates corresponding operation instructions based on the comparison results to regulate the temperature of the furnace.

Images