Glass liquid temperature control device and control method

Abstract

The application provides a glass liquid temperature control device and control method, relates to the technical field of glass manufacturing, and comprises a glass liquid flow guide groove, a heating assembly, a temperature detection assembly, a process parameter detection assembly and a control part. The glass liquid flow guide groove is sequentially divided into a feeding section, a conveying section and a melting section along the length direction, and the three sections are all provided with the heating assembly, the temperature detection assembly and the process parameter detection assembly. The heating assembly comprises a main heating part and an auxiliary heating part and is used for heating the glass liquid; the temperature detection assembly is used for detecting the temperature of the glass liquid; and the process parameter detection assembly is used for detecting the flow rate and viscosity of the glass liquid. The control part is electrically connected with each assembly respectively, is used for obtaining the glass liquid discharge amount of the melting furnace, and calculates the total heating power based on the corresponding PID control model according to the glass liquid temperature, flow rate and viscosity of the target groove section, and then determines and executes the corresponding heating strategy. The crystallization of the glass liquid in the glass liquid flow guide groove is avoided, and the quality of the glass product is improved.

Description

Technical Field

[0001] This invention relates to the field of glass manufacturing technology, and in particular to a glass melt temperature control device and control method. Background Technology

[0002] As a core material in fields such as construction, electronics, and automobiles, float glass faces ever-increasing demands for more refined, efficient, and stable production processes. However, in current float glass production, after the high-temperature molten glass is discharged from the furnace, it must be transported and diverted through flow channels before entering the tin bath for float forming. During this process, the high-temperature molten glass easily produces wollastonite crystals at the edges of the flow channels. These crystals cannot enter the tin bath with the molten glass; instead, they cause the molten glass to cool rapidly, further promoting crystal formation and agglomeration. In extreme cases, this can even cause the glass strip in the spreading area of ​​the tin bath to detach from the diaphragm, resulting in severe plate swaying, directly endangering production safety and reducing product qualification rates.

[0003] Therefore, there is an urgent need for a glass melt temperature control device and method to prevent glass melt from crystallizing in the flow channel and improve the quality of glass products. Summary of the Invention

[0004] This invention provides a glass melt temperature control device and method to prevent glass melt from crystallizing in the flow channel.

[0005] In a first aspect, the present invention provides a glass melt temperature control device, comprising: A molten glass guide channel is used to transport molten glass. The molten glass guide channel includes a feeding section, a conveying section, and a melting section that are connected sequentially along its length. The feeding section, conveying section, and melting section are all equipped with heating components, temperature detection components, and process parameter detection components. The heating assembly includes a main heating element and an auxiliary heating element, used to heat the molten glass; The temperature detection component is used to detect the temperature of the molten glass; the temperature of the molten glass includes the edge temperature, the bottom temperature, and the middle temperature. The process parameter detection component is used to detect the flow rate and viscosity of the molten glass; The control component is electrically connected to the heating assembly, temperature detection assembly, and process parameter detection assembly, respectively. It is used to obtain the glass melt output of the melting furnace, and calculate the total heating power of the target tank section based on the glass melt output and the glass melt temperature, flow rate, and viscosity of the target tank section according to the glass melt temperature PID control model corresponding to the target tank section. The heating strategy of the heating assembly of the target tank section is determined according to the total heating power. The target tank section is a feeding section, a conveying section, or a crystal melting section.

[0006] According to certain embodiments of the present invention, the temperature detection component includes an edge temperature detection element, a bottom temperature detection element, and a middle temperature detection element; the edge temperature detection element is embedded in the side wall of the molten glass guide channel and is used to detect the edge temperature of the molten glass; the bottom temperature detection element is embedded in the bottom surface of the molten glass guide channel and is used to detect the bottom temperature of the molten glass; the middle temperature detection element is disposed at the top of the molten glass guide channel with its detection end facing the middle of the molten glass and is used to detect the middle temperature of the molten glass.

[0007] According to certain embodiments of the present invention, the feeding section, conveying section and melting section each include a strong heating zone at the edge and a weak heating zone in the middle along their radial cross sections. The main heating element is disposed in the strong heating zone at the edge, and the auxiliary heating element is disposed in the weak heating zone in the middle.

[0008] In a second aspect, the present invention provides a method for controlling the temperature of molten glass, comprising: Obtain the crystallization-affecting parameters of the target tank section; the crystallization-affecting parameters of the glass melt include the glass melt output rate, glass melt temperature, flow rate, and viscosity. Based on the crystallization influence parameters, the total heating power of the target tank section is calculated using the glass melt temperature PID control model corresponding to the target tank section. The heating strategy of the heating components in the target tank segment is determined based on the total heating power, and the heating components in the target tank segment are controlled to execute the heating strategy.

[0009] According to the technical solutions provided in certain embodiments of the present invention, the total heating power of the target tank segment is calculated based on the crystallization influence parameters and a PID control model for the glass melt temperature corresponding to the target tank segment, including: Determine the corresponding PID control model for the molten glass temperature based on the target tank segment; The crystallization influence parameters are input into the glass melt temperature PID control model corresponding to the target tank section. The preset target values ​​of the glass melt edge temperature, bottom temperature, and middle temperature of the target tank section are used as the control benchmark. The deviation is calculated based on the glass melt output, glass melt flow rate, and viscosity of the melting furnace. The rate of change of glass melt temperature is used as a constraint. The deviation is adjusted proportionally, integrally, and derivatively through the PID algorithm to calculate the total heating power of the target tank section.

[0010] According to the technical solutions provided in certain embodiments of the present invention, determining the heating strategy of the heating assembly of the target tank segment based on the total heating power includes: If the total heating power is less than or equal to the first preset power threshold, then the auxiliary heating element of the target tank section is controlled to operate independently according to the total heating power; If the total heating power is greater than a first preset power threshold and less than a second preset power threshold, then the main heating element of the target tank segment is controlled to operate independently according to the total heating power; the second preset power threshold is greater than the first preset power threshold. If the total heating power is greater than or equal to the second preset power threshold, the total heating power is allocated to the main heating element and auxiliary heating element of the target tank segment according to the preset power weight, and the main heating element and auxiliary heating element are controlled to operate synchronously according to the allocated heating power.

[0011] According to the technical solutions provided in some embodiments of the present invention, the method further includes: The target crystallization risk prediction model is retrieved from the model database based on the glass type and production conditions in the tank section to be predicted. The model database pre-stores crystallization risk prediction models corresponding to different glass types and production conditions. The crystallization risk prediction model is obtained by training the model with the sample crystallization influence parameters of the glass melt under different glass types and production conditions as the model input and the sample crystallization rate corresponding to the sample crystallization influence parameters as the model output. Obtain the crystallization influence parameters corresponding to the cell segment to be predicted, input the crystallization influence parameters corresponding to the cell segment to be predicted into the target crystallization risk prediction model, and obtain the crystallization rate of the cell segment to be predicted; Based on the crystallization rate, a crystallization control strategy is determined, and crystallization control actions are performed on the cell segment to be predicted based on the crystallization control strategy.

[0012] According to the technical solutions provided in certain embodiments of the present invention, a crystallization control strategy is determined based on the crystallization rate, including: If the crystallization rate is greater than the first preset crystallization rate threshold and less than the second preset crystallization rate threshold, then the preset target value of the glass melt edge temperature of the predicted tank section is increased according to a preset gradient, and the glass melt flow rate is increased according to a first preset ratio; the second preset crystallization rate threshold is greater than the first preset crystallization rate threshold. If the crystallization rate is greater than or equal to the second preset crystallization rate threshold, the main heating element of the melting section is controlled to operate at full power, and the auxiliary heating elements of the feeding section, conveying section and melting section are controlled to start synchronously, and the flow rate of the glass melt is increased according to the second preset ratio; the second preset ratio is greater than the first preset ratio.

[0013] According to the technical solutions provided in some embodiments of the present invention, the method further includes: Obtain a glass melt crystallization sample dataset, which includes multiple sets of one-to-one corresponding glass melt sample crystallization influence parameters and sample crystallization rates; Using the sample crystallization influence parameters as model input and the sample crystallization rate as model output, a network model is trained to obtain the crystallization risk prediction model.

[0014] According to certain embodiments of the present invention, a network model is trained using the sample crystallization influence parameter as model input and the sample crystallization rate as model output to obtain the crystallization risk prediction model, including: The dataset of glass melt crystallization samples was divided into a training set and a test set. Using the crystallization influence parameters of the training set as model input and the crystallization rate of the training set as model output, the network model is trained to obtain the intermediate risk prediction model. The sample crystallization influence parameters of the test set are input into the intermediate risk prediction model to obtain the predicted crystallization rate; The crystallization error is calculated based on the sample crystallization rate of the test set and the predicted crystallization rate; The network weight parameters of the intermediate risk prediction model are adjusted based on the crystallization error until the crystallization error is less than or equal to a preset crystallization error threshold, thereby obtaining the crystallization risk prediction model.

[0015] In summary, this invention provides a glass melt temperature control device, including a glass melt guide channel for conveying glass melt. The glass melt guide channel includes a feeding section, a conveying section, and a melting section connected sequentially along its length. Each of the feeding section, conveying section, and melting section is equipped with a heating component, a temperature detection component, and a process parameter detection component. The heating component includes a main heating element and an auxiliary heating element for heating the glass melt. The temperature detection component detects the glass melt temperature. The glass melt temperature includes edge temperature, bottom temperature, and center temperature. The process parameter detection component is used to detect the flow rate and viscosity of molten glass; the control component, which is electrically connected to the heating component, temperature detection component and process parameter detection component respectively, is used to obtain the molten glass output from the melting furnace, and calculate the total heating power of the target tank section based on the molten glass output and the temperature, flow rate and viscosity of the molten glass in the target tank section, based on the PID control model of the molten glass temperature corresponding to the target tank section, and determine the heating strategy of the heating component of the target tank section based on the total heating power; the target tank section is a feeding section, a conveying section or a melting section. This invention equips the feeding section, conveying section, and melting section of the molten glass channel with heating components, temperature detection components, and process parameter detection components. The control unit is electrically connected to each of these components to form a closed-loop control system. By acquiring the edge, bottom, and center temperatures, flow rates, viscosity, and furnace discharge rates of the molten glass in each section, and combining this with the corresponding PID control model for each section, the total heating power is calculated. Based on the required total heating power, the corresponding heating strategy is executed, achieving independent temperature control of the feeding, conveying, and melting sections. This ensures that the molten glass temperature in these sections prevents wollastonite crystallization and melts existing crystals, thereby improving the quality of the glass products.

[0016] It should be understood that the descriptions of technical features, technical solutions, beneficial effects, or similar language in this invention do not imply that all features and advantages can be achieved in any single embodiment. Rather, it is understood that the description of a feature or beneficial effect means that a specific technical feature, technical solution, or beneficial effect is included in at least one embodiment. Therefore, the descriptions of technical features, technical solutions, or beneficial effects in this specification do not necessarily refer to the same embodiment. Furthermore, the technical features, technical solutions, and beneficial effects described in this embodiment can be combined in any suitable manner. Those skilled in the art will understand that embodiments can be implemented without one or more specific technical features, technical solutions, or beneficial effects of a particular embodiment. In other embodiments, additional technical features and beneficial effects may be identified in specific embodiments that do not embody all embodiments. Attached Figure Description

[0017] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a schematic diagram of a glass melt temperature control device provided in an embodiment of the present invention; Figure 2 A schematic flowchart of a glass melt temperature control method provided in an embodiment of the present invention; Figure 3 This is a flowchart illustrating step S2 provided in an embodiment of the present invention.

[0019] The text labels in the image represent: 1. Glass melt guide channel; 11. Feeding section; 12. Conveying section; 13. Crystal melting section; 2. Heating components; 21. Main heating element; 22. Auxiliary heating element; 31. Edge temperature detection element; 32. Middle temperature detection element; 4. Process parameter detection components. Detailed Implementation

[0020] To enable those skilled in the art to better understand the technical solutions of the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. This description is merely illustrative and explanatory, and should not be construed as limiting the scope of protection of the present invention in any way. Specifically, the described embodiments are only some embodiments of the present invention, not all embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort should fall within the scope of protection of the present invention.

[0021] It should be noted that similar reference numerals and letters in the following figures denote similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such process, method, product, or apparatus.

[0022] As mentioned in the background section, in view of the problems in the prior art, this embodiment provides a glass melt temperature control device, including: The glass liquid guide channel 1 is used to transport glass liquid. The glass liquid guide channel 1 includes a feeding section 11, a conveying section 12 and a melting section 13 connected sequentially along its length. The feeding section 11, the conveying section 12 and the melting section 13 are all equipped with a heating component 2, a temperature detection component and a process parameter detection component 4. Heating assembly 2 includes a main heating element 21 and an auxiliary heating element 22, used to heat the molten glass; The temperature sensing component is used to detect the temperature of the molten glass; the molten glass temperature includes the edge temperature, bottom temperature, and center temperature. Process parameter detection component 4 is used to detect the flow rate and viscosity of molten glass; The control unit is electrically connected to the heating component 2, the temperature detection component, and the process parameter detection component 4, respectively. It is used to obtain the glass melt output of the melting furnace, and calculate the total heating power of the target tank based on the glass melt output and the glass melt temperature, flow rate, and viscosity of the target tank section according to the glass melt temperature PID control model corresponding to the target tank section. The heating strategy of the heating component 2 of the target tank section is determined according to the total heating power. The target tank section is the feeding section 11, the conveying section 12, or the crystal melting section 13.

[0023] For details, please refer to Figure 1The glass melt temperature control device includes a glass melt guide channel 1, a heating component 2, a temperature detection component, a process parameter detection component 4, and control components. The glass melt guide channel 1 is made of cast refractory material (such as high-alumina corundum bricks, fused zirconia corundum bricks, etc.), and its function is to transport the glass melt from the melting furnace to the tin bath. The glass melt guide channel 1 includes a feeding section 11, a conveying section 12, and a crystal melting section 13 connected sequentially along its length. The feeding section 11 is connected to the outlet of the melting furnace and is used to receive the glass melt flowing out of the melting furnace and achieve stable flow. The conveying section 12 is the main conveying channel for the glass melt, used for continuous transport of the glass melt. The crystal melting section 13 is located near the inlet of the tin bath. Because the crystal melting section 13 is farthest from the melting furnace and close to the relatively low-temperature area of ​​the tin bath, heat loss is more severe, and the risk of crystallization is higher than that of the feeding section 11 and the conveying section 12. To reduce the risk of crystallization, this embodiment includes heating components 2, temperature detection components, and process parameter detection components 4 in the feeding section 11, conveying section 12, and melting section 13. The heating component 2 includes a main heating element 21 and an auxiliary heating element 22 for heating the molten glass. In this embodiment, the main heating element 21 is a silicon molybdenum rod, and the auxiliary heating element 22 is a far-infrared radiation electric heating tape. The temperature detection component detects the temperature of the molten glass, including edge temperature, bottom temperature, and center temperature. The temperature detection component can detect the glass temperature using platinum-rhodium thermocouples and high-temperature infrared thermometers at measuring points in the molten glass flow channel 1, or it can use a distributed fiber optic temperature measurement system to continuously and uninterruptedly detect the temperature of the molten glass along the entire length of the flow channel 1. The process parameter detection component 4 detects the flow rate and viscosity of the molten glass. In this embodiment, the arrangement of the process parameter detection component 4 balances detection accuracy with non-interference with the flow of the molten glass, avoiding dead zones or affecting the homogenization of the molten glass. Electromagnetic flowmeters are used for flow rate detection, with one unit each in the feeding section 11, conveying section 12, and melting section 13, installed on the sidewall of the glass melt guide trough 1, with the detection probe aimed at the central region of the glass melt. Online rotational viscometers are used for viscosity detection, also installed on the sidewalls of the feeding section 11, conveying section 12, and melting section 13. The control unit is electrically connected to the heating assembly 2, temperature detection assembly, and process parameter detection assembly 4, respectively, to obtain the glass melt output from the melting furnace. Based on the glass melt output and the glass melt temperature of the target section obtained through the temperature detection assembly, and using the flow rate and viscosity obtained through the process parameter detection assembly 4, the total heating power of the target section is calculated based on a PID control model corresponding to the glass melt temperature of the target section. The heating strategy of the heating assembly 2 for the target section is then determined based on the total heating power. The target section is either the feeding section 11, conveying section 12, or melting section 13. In this embodiment, the control unit is a DCS system (Distributed Control System).

[0024] This invention provides heating components 2, temperature detection components, and process parameter detection components 4 in the feeding section 11, conveying section 12, and melting section 13 of the molten glass flow channel 1. A control unit is electrically connected to these components. The control unit acquires the edge, bottom, and center temperatures, flow rates, viscosity, and furnace discharge rates of the molten glass in each section. It then calculates the total heating power using the corresponding PID control model for each section and executes the appropriate heating strategy based on the required total heating power. This allows for independent detection and control of the molten glass parameters in the feeding section 11, conveying section 12, and melting section 13. This maintains the molten glass temperature in these sections within a suitable range, effectively suppressing wollastonite crystallization and melting the precipitated crystals, thus preventing quality defects caused by crystallization and improving the quality of the finished glass product.

[0025] In a preferred embodiment, the temperature detection assembly includes an edge temperature detection element 31, a bottom temperature detection element, and a middle temperature detection element 32; the edge temperature detection element 31 is embedded in the side wall of the glass melt guide channel 1 and is used to detect the edge temperature of the glass melt; the bottom temperature detection element is embedded in the bottom surface of the glass melt guide channel 1 and is used to detect the bottom temperature of the glass melt; the middle temperature detection element 32 is disposed at the top of the glass melt guide channel 1 with its detection end facing the middle of the glass melt and is used to detect the middle temperature of the glass melt.

[0026] Specifically, such as Figure 1 As shown, the temperature detection assembly includes an edge temperature detector 31, a bottom temperature detector, and a middle temperature detector 32. The edge temperature detector 31 uses a platinum-rhodium thermocouple, embedded 100-120mm into the side wall of the glass melt guide channel 1, and the measured temperature is taken as the edge temperature of the glass melt. The bottom temperature detector also uses a platinum-rhodium thermocouple, embedded 80-100mm into the bottom surface of the glass melt guide channel 1 (not shown in the figure), and the measured temperature is taken as the bottom temperature of the glass melt. The middle temperature detector 32 uses a high-temperature resistant non-contact infrared thermometer, mounted on a high-temperature resistant bracket above the glass melt guide channel 1, with the lens facing the middle region of the glass melt. The non-contact detection of the high-temperature resistant non-contact infrared thermometer avoids interference with the flow of the glass melt, and the measured temperature is taken as the middle temperature of the glass melt. In the feeding section 11 and the conveying section 12, one set of edge temperature detectors 31 is arranged every 0.5m, one set of bottom temperature detectors is arranged every 1m, and one middle temperature detector 32 is arranged every 2m. In the melting crystal section 13, one set of temperature detection devices 31 is arranged every 0.3m at the edge, one set of temperature detection devices is arranged every 0.3m at the bottom, and one set of temperature detection devices 32 is arranged every 2m in the middle. However, the arrangement can be adjusted according to requirements, and no specific limit is made here.

[0027] In a preferred embodiment, the feeding section 11, the conveying section 12 and the melting section 13 each include a strong heating zone at the edge and a weak heating zone in the middle along their radial cross sections. The main heating element 21 is disposed in the strong heating zone at the edge and the auxiliary heating element 22 is disposed in the weak heating zone in the middle.

[0028] Specifically, such as Figure 1 As shown, the radial cross-sections of the feeding section 11, conveying section 12, and melting section 13 all include a strong heating zone at the edges and a weak heating zone in the middle. During the flow of molten glass, the edge region, due to its direct contact with the wall of the molten glass guide channel 1, has a large heat dissipation area and rapid heat loss, easily forming local low-temperature areas and a higher risk of crystallization. The molten glass in the middle region is relatively concentrated, with a large heat capacity and a relatively uniform and stable temperature. The heating power and thermal radiation intensity of the main heating element 21 (silicon molybdenum rod) are greater than those of the auxiliary heating element 22 (far-infrared radiation type electric heating tape). Therefore, arranging the main heating element 21 along the outer wall of the strong heating zone at the edges of the molten glass guide channel 1 can specifically compensate for the heat loss at the edges, quickly increase the temperature of the edge region, and suppress crystallization at the edges. Placing the auxiliary heating element 22 on the outer wall of the weak heating zone in the middle can keep the molten glass in the middle warm or heat it slightly.

[0029] This embodiment also provides a method for controlling the temperature of molten glass, including: S1. Obtain the crystallization-affecting parameters of the target tank section; the crystallization-affecting parameters of the glass melt include the glass melt output rate, glass melt temperature, flow rate and viscosity. S2. Based on the crystallization influence parameters, calculate the total heating power of the target tank section using the glass melt temperature PID control model corresponding to the target tank section; S3. Determine the heating strategy of the heating component 2 in the target tank section based on the total heating power, and control the heating component 2 in the target tank section to execute the heating strategy.

[0030] Specifically, such as Figure 2As shown, the amount of molten glass discharged from the furnace changes the residence time of the molten glass in the molten glass guide channel 1, and the change in residence time affects the stability of the molten glass temperature. The flow rate of the molten glass directly affects the heat transfer efficiency and composition uniformity within the molten glass guide channel 1. If the flow rate is too slow, uneven heat accumulation in the molten glass will occur, forming localized supercooled areas that induce crystallization. If the flow rate is too fast, it will affect the sufficient heat transfer, making it difficult to maintain a stable temperature of the molten glass and indirectly increasing the risk of crystallization. The viscosity of the molten glass directly affects the crystallization rate. Therefore, the amount of molten glass discharged from the furnace, the molten glass temperature, the flow rate, and the viscosity are used as parameters affecting crystallization. The controller calculates the total heating power of the target channel (i.e., the heating power required to avoid crystallization) based on the above crystallization parameters and the PID control model of the molten glass temperature corresponding to the target channel. Then, based on the total heating power, the heating strategy of the heating component 2 in the target channel (a heating strategy adapted to the target channel to avoid crystallization in the target channel) can be determined, and the heating component 2 in the target channel is controlled to execute the heating strategy.

[0031] In a preferred embodiment, S2, based on the crystallization influence parameters, the total heating power of the target tank section is calculated using a PID control model for the glass melt temperature corresponding to the target tank section, including: S201. Determine the corresponding PID control model for the glass melt temperature based on the target tank section; Specifically, such as Figure 3 As shown, the feeding section 11 receives the molten glass flowing from the melting furnace, resulting in significant temperature fluctuations but relatively convenient heat replenishment and a low risk of crystallization. The conveying section 12 is the main channel for conveying molten glass, requiring stable temperature and uniform heat dissipation. The crystallization section 13 is furthest from the melting furnace and closest to the low-temperature tin bath, resulting in the fastest heat dissipation and the highest risk of crystallization. Furthermore, the preset target values ​​for the edge, bottom, and middle temperatures of the molten glass in each section are different, thus requiring different PID control models for the molten glass temperature. Therefore, before inputting crystallization-affecting parameters into the PID control model for molten glass temperature and calculating the total heating power, it is necessary to first determine the corresponding PID control model for molten glass temperature to ensure that it is compatible with the actual operating conditions of the target section. This avoids deviations in the calculation of total heating power due to model misuse, which could affect the control effect of molten glass temperature.

[0032] S202. Input the crystallization influence parameters into the PID control model of the glass melt temperature corresponding to the target tank section. Use the preset target values ​​of the glass melt edge temperature, bottom temperature and middle temperature of the target tank section as the control benchmark. Calculate the deviation based on the glass melt output, glass melt flow rate and viscosity of the melting furnace. Use the glass melt temperature change rate as a constraint. Adjust the deviation proportionally, integrally and derivatively through the PID algorithm to calculate the total heating power of the target tank section.

[0033] Specifically, such as Figure 3As shown, after determining the glass melt temperature PID control model corresponding to the target tank section, the crystallization influence parameters of the target tank section can be input into the glass melt temperature PID control model. The glass melt temperature PID control model converts the crystallization influence parameters into executable heating power commands. Specifically, after the crystallization influence parameters are input into the glass melt temperature PID control model corresponding to the target tank section, the preset target values ​​of the glass melt edge temperature, bottom temperature, and middle temperature of the target tank section are used as the control benchmark. Based on the glass melt output, glass melt flow rate, and viscosity of the melting furnace, the deviation is calculated according to the following formulas (1) and (2): Formula (1) in, For the temperature deviation of the molten glass, , , These are the weighting coefficients for edge temperature, bottom temperature, and middle temperature, respectively. , , These are the preset target values ​​for the edge temperature, bottom temperature, and middle temperature, respectively. , , These are the measured values ​​for the edge temperature, bottom temperature, and middle temperature, respectively.

[0034] Formula (2) in, This is the temperature deviation correction factor. , , These are correction factors for the glass melt output rate, flow rate, and viscosity of the melting furnace, respectively. The rated output of molten glass from the melting furnace. This refers to the measured value of the glass molten material output from the melting furnace. The rated flow rate of the molten glass. This is the measured value of the glass melt flow rate. The rated viscosity of the molten glass; This is the measured value of the glass melt viscosity.

[0035] Then, based on the above glass melt temperature deviation and temperature deviation correction coefficient, and constrained by the glass melt temperature change rate, the deviation is adjusted proportionally, integrally, and derivatively using a PID algorithm, and the total heating power of the target tank section is calculated according to the following formula (3): Formula (3) in, The total heating power of the target tank section. This is the proportionality coefficient. The integral coefficient is... The differential coefficients are... To control time, For the temperature deviation of the molten glass, This is the temperature deviation correction factor. This represents the rate of change in the temperature of the molten glass.

[0036] In a preferred embodiment, S3, determining the heating strategy of the heating component 2 for the target tank segment based on the total heating power includes: If the total heating power is less than or equal to the first preset power threshold, the auxiliary heating element 22 of the target tank section is controlled to operate independently according to the total heating power. If the total heating power is greater than the first preset power threshold and less than the second preset power threshold, then the main heating element 21 of the target tank section is controlled to operate independently according to the total heating power; the second preset power threshold is greater than the first preset power threshold. If the total heating power is greater than or equal to the second preset power threshold, the total heating power is allocated to the main heating element 21 and auxiliary heating element 22 of the target tank section according to the preset power weight, and the main heating element 21 and auxiliary heating element 22 are controlled to operate synchronously according to the allocated heating power.

[0037] Specifically, once the total heating power is determined, the heating strategy for the heating component 2 of the target tank segment can be determined. If the total heating power is less than or equal to a first preset power threshold, it indicates that the current glass melt temperature deviation is small, requiring only minor supplementary heating to meet process requirements. In this case, the auxiliary heating element 22 of the target tank segment is controlled to operate independently according to the total heating power, while the main heating element 21 remains in standby mode. If the total heating power is greater than the first preset power threshold but less than a second preset power threshold, it indicates that the current glass melt temperature deviation is large, requiring greater power supplementary heating to quickly approach the preset target temperature value. In this case, the main heating element 21 of the target tank segment is controlled to operate independently according to the total heating power, while the auxiliary heating element 22 remains in standby mode, utilizing the high power output characteristics of the main heating element 21 to improve temperature response speed. If the total heating power is greater than or equal to the second preset power threshold, it indicates that the current glass melt temperature deviation is very large, and a single heating element cannot meet the current requirements. In this case, the total heating power is allocated to the main heating element 21 and the auxiliary heating element 22 of the target tank segment according to preset power weights, and the main heating element 21 and the auxiliary heating element 22 are controlled to operate synchronously according to the allocated heating power. The second preset power threshold is greater than the first preset power threshold.

[0038] In a preferred embodiment, the method further includes: The target crystallization risk prediction model is retrieved from the model database based on the glass type and production conditions in the tank section to be predicted. The model database pre-stores crystallization risk prediction models corresponding to different glass types and production conditions. The crystallization risk prediction model is obtained by training the model with the sample crystallization influence parameters of the molten glass under different glass types and production conditions as the model input and the sample crystallization rate corresponding to the sample crystallization influence parameters as the model output. Obtain the crystallization influence parameters corresponding to the cell segment to be predicted, input the crystallization influence parameters corresponding to the cell segment to be predicted into the target crystallization risk prediction model, and obtain the crystallization rate of the cell segment to be predicted; Based on the crystallization rate, a crystallization control strategy is determined, and crystallization control actions are performed on the tank segment to be predicted based on the crystallization control strategy.

[0039] Specifically, reversing the crystallization of molten glass by raising the temperature after crystallization not only significantly increases energy consumption and prolongs the production cycle, but also may lead to defects such as stones and streaks inside the glass due to the difficulty in completely eliminating crystallization products, seriously affecting the quality of finished glass and production stability. Furthermore, equipment nodules caused by crystallization can lead to production interruptions and increased maintenance costs. Therefore, predicting the crystallization risk of molten glass in advance and implementing prediction and prevention before crystallization occurs is of great significance for ensuring continuous and stable glass production and improving product qualification rates. Specifically, due to significant differences in the chemical composition, viscosity characteristics, upper limit temperature of crystallization, and crystallization tendency of different glass types, as well as significant differences in temperature field distribution, flow state, residence time, and cooling rate under different production conditions, the crystallization risk of molten glass varies. Therefore, it is necessary to call the target crystallization risk prediction model from the model database based on the glass type and production conditions in the tank to be predicted. The model database pre-stores crystallization risk prediction models corresponding to different glass types and production conditions. The crystallization risk prediction model is obtained by training a model using sample crystallization influence parameters of molten glass under different glass types and production conditions as input and the sample crystallization rate corresponding to the sample crystallization influence parameters as output. After determination, it is necessary to obtain the crystallization influence parameters corresponding to the section of the tank to be predicted. These parameters are then input into the target crystallization risk prediction model to obtain the crystallization rate of the section. The crystallization rate refers to the proportion of the mass (or volume) of crystals in the section to be predicted to the total mass (or volume), reflecting the degree of transformation from an amorphous state to a crystalline state. Different crystallization rates indicate different crystallization risks; therefore, appropriate crystallization control strategies can be developed for different crystallization rates, and crystallization control actions can be implemented for the section to be predicted based on these strategies to prevent crystallization before it occurs.

[0040] In a preferred embodiment, a crystallization control strategy is determined based on the crystallization rate, including: If the crystallization rate is greater than the first preset crystallization rate threshold and less than the second preset crystallization rate threshold, then the preset target value of the glass melt edge temperature of the predicted tank section is increased according to the preset gradient, and the glass melt flow rate is increased according to the first preset ratio; the second preset crystallization rate threshold is greater than the first preset crystallization rate threshold. If the crystallization rate is greater than or equal to the second preset crystallization rate threshold, the main heating element 21 of the melting crystal section 13 is controlled to operate at full power, and the auxiliary heating elements 22 of the feeding section 11, the conveying section 12 and the melting crystal section 13 are controlled to start synchronously, and the flow rate of the glass melt is increased according to the second preset ratio; the second preset ratio is greater than the first preset ratio.

[0041] Specifically, if the crystallization rate is less than the first preset crystallization rate threshold, it indicates that the risk of glass melt crystallization in the current prediction tank section is very low, the production conditions are stable and controllable, and no operation is required. If the crystallization rate is greater than the first preset crystallization rate threshold but less than the second preset crystallization rate threshold, it indicates that there is a certain risk of glass melt crystallization in the prediction tank section. If not controlled in time, crystallization is likely to occur. In this case, the preset target value of the glass melt edge temperature in the prediction tank section is increased according to the preset gradient, and the glass melt flow rate is increased according to the first preset ratio. If the crystallization rate is greater than or equal to the second preset crystallization rate threshold, it indicates that the risk of glass melt crystallization in the prediction tank section is extremely high, approaching or reaching the critical crystallization condition, and large-scale crystallization is very likely to occur. In this case, the main heating element 21 of the melting section 13 is controlled to operate at full power, and the auxiliary heating elements 22 of the feeding section 11, conveying section 12, and melting section 13 are controlled to start synchronously, and the glass melt flow rate is increased according to the second preset ratio. The second preset ratio is greater than the first preset ratio. The second preset crystallization rate threshold is greater than the first preset crystallization rate threshold.

[0042] In a preferred embodiment, the method further includes: Obtain a glass melt crystallization sample dataset, which includes multiple sets of one-to-one corresponding glass melt sample crystallization influencing parameters and sample crystallization rates; Using the sample crystallization influence parameters as model input and the sample crystallization rate as model output, a network model is trained to obtain a crystallization risk prediction model.

[0043] Specifically, to predict the crystallization risk of molten glass, a crystallization risk prediction model needs to be established beforehand. This model is trained using historical data of crystallization influence parameters (samples of molten glass under different glass types and production conditions) as input and historical data of crystallization rates (samples of crystallization rates) corresponding to these parameters as output. Therefore, the first step is to obtain the data needed for model training: a molten glass crystallization sample dataset. This dataset includes multiple sets of one-to-one corresponding molten glass sample crystallization influence parameters and sample crystallization rates. Then, by using the sample crystallization influence parameters as input and the sample crystallization rates as output, the network model is trained to obtain the crystallization risk prediction model.

[0044] Furthermore, using the sample crystallization influence parameters as model input and the sample crystallization rate as model output, a network model is trained to obtain a crystallization risk prediction model, including: The glass melt crystallization sample dataset was divided into a training set and a test set; Using the crystallization influence parameters of the training set samples as model input and the crystallization rate of the training set samples as model output, the network model is trained to obtain the intermediate risk prediction model. Input the crystallization impact parameters of the test set into the intermediate risk prediction model to obtain the predicted crystallization rate; The crystallization error is calculated based on the sample crystallization rate and the predicted crystallization rate of the test set; The network weight parameters of the intermediate risk prediction model are adjusted based on the crystallization error until the crystallization error is less than or equal to the preset crystallization error threshold, thus obtaining the crystallization risk prediction model.

[0045] Specifically, after obtaining the glass melt crystallization sample dataset, a model can be trained to obtain a crystallization risk prediction model. The specific steps are as follows: First, the glass melt crystallization sample dataset is divided into a training set and a test set, which serve the functions of model training and model performance verification, respectively. Second, the crystallization influence parameters of the training set samples are used as model input, and the crystallization rate of the training set samples is used as model output to train the network model, obtaining an intermediate risk prediction model. Then, the crystallization influence parameters of the test set samples are input into the intermediate risk prediction model to obtain the predicted crystallization rate. The crystallization error is then calculated based on the crystallization rate of the test set samples and the predicted crystallization rate. If the crystallization error is greater than a preset crystallization error threshold, it indicates that the performance of the current intermediate risk prediction model is substandard. The network weight parameters of the intermediate risk prediction model should be adjusted based on the crystallization error, and a new predicted crystallization rate should be output based on the adjusted intermediate risk prediction model, and a new crystallization error should be calculated until the crystallization error is less than or equal to the preset crystallization error threshold, thus obtaining the crystallization risk prediction model.

[0046] The glass melt temperature control device provided by this invention equips the feeding section 11, conveying section 12, and melting section 13 of the glass melt guide tank 1 with heating components 2, temperature detection components, and process parameter detection components 4, respectively, and is uniformly electrically connected and controlled by a controller. The controller collects the glass melt temperature, flow rate, viscosity, and furnace discharge rate of each tank section, calculates the total heating power based on the corresponding PID control model, and executes the corresponding heating strategy. This achieves independent detection of glass melt parameters and independent temperature control of the glass melt in the feeding section 11, conveying section 12, and melting section 13, keeping the temperature of each tank section within a suitable range. This suppresses wollastonite crystallization and melts existing crystals, reducing quality defects caused by crystallization and improving the quality of the finished glass product. Simultaneously, a crystallization risk prediction model is used to predict and control crystallization risks in advance, ensuring continuous and stable production and improving the glass product qualification rate.

[0047] To facilitate understanding by those skilled in the art, the working process of the glass melt temperature control device provided by the present invention is further as follows: Molten glass is conveyed in the molten glass guide channel 1, sequentially passing through the feeding section 11, the conveying section 12, and the melting section 13. Temperature sensors at the edge, bottom, and center of the molten glass are collected in each section. The flow rate and viscosity of the molten glass are detected by the process parameter detection component 4. The control unit acquires the molten glass discharge rate from the melting furnace, the edge, bottom, and center temperatures collected by the temperature sensors, and the flow rate and viscosity detected by the process parameter detection component 4. Based on the current target section (feeding section 11, conveying section 12, or melting section 13), the control unit calls the corresponding molten glass temperature PID control model. The molten glass discharge rate, temperature, flow rate, and viscosity are input as crystallization influence parameters into the model. Using preset target values ​​for each temperature as the control benchmark, and combined with temperature change rate constraints, the PID algorithm adjusts the deviation to calculate the total heating power required for the target section. A graded heating strategy is then executed based on the total heating power. Simultaneously, the control unit matches and calls the corresponding target crystallization risk prediction model from the model database based on the glass type and production conditions of the cell segment to be predicted (feeding section 11, conveying section 12, or melting section 13); inputs the crystallization influence parameters of the cell segment to be predicted into the target crystallization risk prediction model to obtain the corresponding crystallization rate; determines the crystallization prevention and control strategy based on the crystallization rate; and performs crystallization prevention and control actions on the cell segment to be predicted based on the crystallization prevention and control strategy.

[0048] This article uses specific examples to illustrate the principles and implementation methods of the present invention. The descriptions of the above embodiments are only for the purpose of helping to understand the method and core ideas of the present invention. The above descriptions are only preferred embodiments of the present invention. It should be noted that due to the limitations of textual expression, and the objective existence of infinite specific structures, those skilled in the art can make several improvements, modifications, or changes without departing from the principles of the present invention, and can also combine the above technical features in an appropriate manner; these improvements, modifications, changes, or combinations, or the direct application of the inventive concept and technical solution to other situations without modification, should all be considered within the scope of protection of the present invention.

Claims

1. A glass melt temperature control device, characterized in that, include: A glass melt guide channel (1) is used to transport glass melt. The glass melt guide channel (1) includes a feeding section (11), a conveying section (12) and a melting section (13) connected sequentially along its length. The feeding section (11), the conveying section (12) and the melting section (13) are all equipped with a heating component (2), a temperature detection component and a process parameter detection component (4). The heating assembly (2) includes a main heating element (21) and an auxiliary heating element (22) for heating the molten glass; The temperature detection component is used to detect the temperature of the molten glass; the temperature of the molten glass includes the edge temperature, the bottom temperature, and the middle temperature. The process parameter detection component (4) is used to detect the flow rate and viscosity of the molten glass; The control component is electrically connected to the heating component (2), the temperature detection component and the process parameter detection component (4) respectively. It is used to obtain the output of glass melt from the melting furnace, and calculate the total heating power of the target tank based on the glass melt temperature PID control model corresponding to the target tank according to the output of glass melt from the melting furnace and the glass melt temperature, flow rate and viscosity of the target tank section. The heating strategy of the heating component (2) of the target tank section is determined according to the total heating power. The target tank section is the feeding section (11), the conveying section (12) or the crystal melting section (13).

2. The glass melt temperature control device according to claim 1, characterized in that, The temperature detection assembly includes an edge temperature detection element (31), a bottom temperature detection element, and a middle temperature detection element (32). The edge temperature detection element (31) is embedded in the side wall of the glass liquid guide channel (1) and is used to detect the edge temperature of the glass liquid. The bottom temperature detection element is embedded in the bottom surface of the glass liquid guide channel (1) and is used to detect the bottom temperature of the glass liquid. The middle temperature detection element (32) is located at the top of the glass liquid guide channel (1) with its detection end facing the middle of the glass liquid and is used to detect the middle temperature of the glass liquid.

3. The glass melt temperature control device according to claim 1, characterized in that, The feeding section (11), conveying section (12) and melting section (13) each have a radial cross-section including a strong heating zone at the edge and a weak heating zone in the middle. The main heating element (21) is located in the strong heating zone at the edge, and the auxiliary heating element (22) is located in the weak heating zone in the middle.

4. A method for controlling the temperature of molten glass, characterized in that, The method is implemented using the glass melt temperature control device as described in any one of claims 1-3, and includes: Obtain the crystallization-affecting parameters of the target tank section; the crystallization-affecting parameters of the glass melt include the glass melt output rate, glass melt temperature, flow rate, and viscosity. Based on the crystallization influence parameters, the total heating power of the target tank section is calculated using the glass melt temperature PID control model corresponding to the target tank section. The heating strategy of the heating component (2) in the target tank segment is determined based on the total heating power, and the heating component (2) in the target tank segment is controlled to execute the heating strategy.

5. The glass melt temperature control method according to claim 4, characterized in that, Based on the crystallization influence parameters, the total heating power of the target tank section is calculated using a PID control model for the glass melt temperature corresponding to the target tank section, including: Determine the corresponding PID control model for the molten glass temperature based on the target tank segment; The crystallization influence parameters are input into the glass melt temperature PID control model corresponding to the target tank section. The preset target values ​​of the glass melt edge temperature, bottom temperature, and middle temperature of the target tank section are used as the control benchmark. The deviation is calculated based on the glass melt output, glass melt flow rate, and viscosity of the melting furnace. The glass melt temperature change rate is used as a constraint. The deviation is adjusted proportionally, integrally, and derivatively through the PID algorithm to calculate the total heating power of the target tank section.

6. The glass melt temperature control method according to claim 4, characterized in that, The heating strategy for the heating assembly (2) of the target tank segment is determined based on the total heating power, including: If the total heating power is less than or equal to the first preset power threshold, then the auxiliary heating element (22) of the target tank section is controlled to operate independently according to the total heating power; If the total heating power is greater than the first preset power threshold and less than the second preset power threshold, then the main heating element (21) of the target tank section is controlled to operate independently according to the total heating power; the second preset power threshold is greater than the first preset power threshold; If the total heating power is greater than or equal to the second preset power threshold, the total heating power is allocated to the main heating element (21) and auxiliary heating element (22) of the target tank segment according to the preset power weight, and the main heating element (21) and auxiliary heating element (22) are controlled to operate synchronously according to the allocated heating power.

7. The glass melt temperature control method according to claim 4, characterized in that, The method further includes: The target crystallization risk prediction model is retrieved from the model database based on the glass type and production conditions in the tank section to be predicted. The model database pre-stores crystallization risk prediction models corresponding to different glass types and production conditions. The crystallization risk prediction model is obtained by training the model with the sample crystallization influence parameters of the glass melt under different glass types and production conditions as the model input and the sample crystallization rate corresponding to the sample crystallization influence parameters as the model output. Obtain the crystallization influence parameters corresponding to the cell segment to be predicted, input the crystallization influence parameters corresponding to the cell segment to be predicted into the target crystallization risk prediction model, and obtain the crystallization rate of the cell segment to be predicted; Based on the crystallization rate, a crystallization control strategy is determined, and crystallization control actions are performed on the cell segment to be predicted based on the crystallization control strategy.

8. The method for controlling the temperature of molten glass according to claim 7, characterized in that, Based on the crystallization rate, a crystallization control strategy is determined, including: If the crystallization rate is greater than the first preset crystallization rate threshold and less than the second preset crystallization rate threshold, then the preset target value of the glass melt edge temperature of the predicted tank section is increased according to a preset gradient, and the glass melt flow rate is increased according to a first preset ratio; the second preset crystallization rate threshold is greater than the first preset crystallization rate threshold. If the crystallization rate is greater than or equal to the second preset crystallization rate threshold, the main heating element (21) of the melting crystal section (13) is controlled to operate at full power, and the auxiliary heating elements (22) of the feeding section (11), the conveying section (12) and the melting crystal section (13) are controlled to start synchronously, and the flow rate of the glass melt is increased according to the second preset ratio; the second preset ratio is greater than the first preset ratio.

9. The glass melt temperature control method according to claim 7, characterized in that, The method further includes: Obtain a glass melt crystallization sample dataset, which includes multiple sets of one-to-one corresponding glass melt sample crystallization influence parameters and sample crystallization rates; Using the sample crystallization influence parameters as model input and the sample crystallization rate as model output, a network model is trained to obtain the crystallization risk prediction model.

10. The glass melt temperature control method according to claim 9, characterized in that, Using the sample crystallization influence parameters as model input and the sample crystallization rate as model output, a network model is trained to obtain the crystallization risk prediction model, including: The dataset of glass melt crystallization samples was divided into a training set and a test set. Using the crystallization influence parameters of the training set as model input and the crystallization rate of the training set as model output, the network model is trained to obtain the intermediate risk prediction model. The sample crystallization influence parameters of the test set are input into the intermediate risk prediction model to obtain the predicted crystallization rate; The crystallization error is calculated based on the sample crystallization rate of the test set and the predicted crystallization rate; The network weight parameters of the intermediate risk prediction model are adjusted based on the crystallization error until the crystallization error is less than or equal to a preset crystallization error threshold, thereby obtaining the crystallization risk prediction model.

Images