Method, device and system for controlling deposition morphology of quartz glass

Abstract

The invention discloses a quartz glass deposition morphology control method, device and system, and relates to the technical field of glass synthesis, the method comprises the following steps: obtaining morphology images and process parameters of a blank in a quartz glass deposition furnace at a plurality of time points; selecting a deposition surface area and a side surface area of the blank from the morphology image; the deposition surface area is subjected to fitting processing to form a deposition surface fitting curve, a first corresponding relation is established, the side face area is subjected to fitting processing to form a side face fitting curve, and a second corresponding relation is established; according to the obtained camera parameters, the actual profile curve, the longitudinal growth speed and the curvature of the blank are calculated; performing full-view reconstruction according to the deposition surface fitting curve and the side surface fitting curve to obtain a full-view contour line of the blank; and based on the growth speed, the curvature, the first corresponding relation, the second corresponding relation and the full view contour line, controlling and adjusting process parameters so as to reach a consistency evaluation standard. According to the scheme, the control precision of the shape of the blank and the preparation efficiency of the quartz glass are improved.

Description

Technical Field

[0001] This application relates to the field of glass synthesis technology, and in particular to a method, apparatus and system for controlling the deposition morphology of quartz glass. Background Technology

[0002] In the rapid development of materials technology, synthetic quartz glass, as a preferred material for high-precision applications, has been widely used in various fields, such as aerospace, semiconductors, fiber optic communication, photovoltaics, metallurgy, chemicals, electric light sources, communications, and light industry, due to its characteristics such as high temperature resistance, corrosion resistance, high strength, high light transmittance, low coefficient of thermal expansion, and high optical uniformity. Synthetic quartz glass is produced by using an oxyhydrogen flame or plasma as a heat source, where silicon-containing precursor materials undergo a chemical reaction in a high-temperature melting furnace to generate SiO2 microparticles. These microparticles then undergo deposition, melting, and vitrification to ultimately form quartz glass. With the continuous upgrading of downstream products towards higher integration, higher precision, and longer lifespan, stringent requirements are placed on the morphological indicators of quartz glass, such as surface flatness, thickness uniformity, and microstructure density. Therefore, research on the control of quartz glass deposition morphology is particularly important in order to improve product precision, performance stability, and service life.

[0003] Currently, in the process of quartz glass deposition, the morphology of the quartz glass preform is generally controlled by on-site observation and monitoring video. Since the preparation of quartz glass preform is a continuous and uninterrupted process, it takes a long time. It is difficult for the naked eye to observe the subtle changes in the preform in a short period of time and to control the deposition surface by manually marking it. This results in different preform morphologies, which affect the optical and mechanical properties of quartz glass. Consequently, the morphology control accuracy of the preform is poor, leading to low quartz glass preparation efficiency. Summary of the Invention

[0004] The purpose of this application is to provide a method, apparatus and system for controlling the deposition morphology of quartz glass, so as to solve the technical problem that poor control accuracy of the morphology of the blank leads to low efficiency in the preparation of quartz glass.

[0005] To achieve the above objectives, this application provides the following solution: In a first aspect, this application provides a method for controlling the deposition morphology of quartz glass, the method comprising: The morphological images and process parameters of the billet in the quartz glass deposition furnace at multiple time points were acquired; the morphological images were acquired by an industrial camera after the quartz glass deposition process was started. Select the deposition surface region and side region of the billet from the topographic image; The deposition surface region is fitted into a deposition surface fitting curve, and a first correspondence between the fitting coefficient of the deposition surface fitting curve and the process parameters is established. The side region is fitted into a side fitting curve, and a second correspondence between the fitting coefficient of the side fitting curve and the process parameters is established. Obtain camera parameters, and calculate the actual contour curve, growth rate, and curvature of the billet based on the camera parameters; The overall outline of the billet is obtained by reconstructing the overall shape based on the fitted curve of the deposition surface and the fitted curve of the side surface. Based on the growth rate, curvature, first correspondence, second correspondence, and overall outline, the process parameters are controlled and adjusted to achieve the consistency evaluation standard.

[0006] Secondly, embodiments of this application provide a device for controlling the deposition morphology of quartz glass, the device comprising: A quartz glass deposition furnace is provided, wherein a deposition chamber is provided inside the quartz glass deposition furnace, the deposition chamber is provided with at least one observation window, a deposition heat source interface, a silicon-containing precursor interface and an exhaust vent, the deposition heat source interface and the silicon-containing precursor interface are located at the top of the deposition chamber; the exhaust vent is located at the bottom of the quartz glass deposition furnace; A deposition heat source, wherein the deposition heat source is located at the deposition heat source interface; An industrial camera, wherein the industrial camera is positioned outside the observation window; An active mechanism is located on the central axis of the quartz glass deposition furnace. The active mechanism includes a lifting frame and a rotating mechanism, and the billet is placed on the active mechanism. The process parameter acquisition device includes a differential pressure gauge and a thermometer. The differential pressure gauge is located on the side of the exhaust port near the top of the quartz glass deposition furnace and is used to collect the pressure difference inside the furnace. The thermometer is located on the top of the quartz glass deposition furnace and is used to collect the deposition temperature.

[0007] Thirdly, embodiments of this application provide a control system for the deposition morphology of quartz glass, the system comprising: computer equipment and a control device for the deposition morphology of quartz glass; The quartz glass deposition morphology control device is used to: after the quartz glass deposition furnace is started, provide a high-temperature environment through the deposition heat source, and allow the silicon-containing precursor to enter the furnace through the silicon-containing precursor interface. After the quartz glass deposition process begins, collect morphology images and process parameters of the billet in the quartz glass deposition furnace at multiple time points through an industrial camera and transmit them to the computer device. The computer equipment is used to: acquire morphological images and process parameters of a billet at multiple time points within a quartz glass deposition furnace; select the deposition surface region or side region of the billet from the morphological images; fit the deposition surface region into a deposition surface fitting curve and establish a first correspondence between the fitting coefficient of the deposition surface fitting curve and the process parameters; fit the side region into a side fitting curve and establish a second correspondence between the fitting coefficient of the side fitting curve and the process parameters; acquire camera parameters and calculate the actual contour curve, longitudinal growth rate, and curvature of the billet based on the camera parameters; reconstruct the overall appearance based on the deposition surface fitting curve and the side fitting curve to obtain the overall outline of the billet; and control and adjust the process parameters based on the growth rate, curvature, first correspondence, second correspondence, and overall outline to achieve a consistency evaluation standard.

[0008] According to the specific embodiments provided in this application, the following technical effects are disclosed: Compared with existing technologies, this application acquires billet morphology images at multiple time points using an industrial camera and simultaneously obtains process parameters. Compared with manual observation, it can accurately capture subtle changes in a short period of time, solving the problem that subtle changes are difficult to detect with the naked eye. The application selects the deposition surface region and side region from the morphology images and performs fitting processing to obtain deposition surface fitting curves and side fitting curves, transforming the practical problem into a mathematical model. This facilitates the subsequent accurate establishment of the correspondence between fitting coefficients and process parameters, replacing manual labeling control, avoiding errors from manual labeling, and improving the control accuracy of billet morphology. Furthermore, the application combines camera parameters to calculate the actual contour curve, growth rate, and curvature, achieving accurate conversion from image to actual size, providing a quantitative basis for subsequent billet consistency evaluation. Moreover, by reconstructing the entire morphology, a complete overall contour line is obtained, allowing for intuitive judgment of the overall shape of the billet, solving the problem of morphological inconsistencies caused by manual observation. Finally, based on multi-dimensional data such as growth rate, curvature, correspondence, and overall contour line, the application adjusts process parameters to ensure that the billet meets consistency standards, thereby improving the optical and mechanical properties of quartz glass, increasing the control accuracy of billet morphology, and further improving the preparation efficiency of quartz glass. Attached Figure Description

[0009] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0010] Figure 1 This is a schematic diagram of the preparation process of quartz glass according to an embodiment of this application; Figure 2 A schematic diagram of a device for controlling the deposition morphology of quartz glass provided in an embodiment of this application; Figure 3 A schematic flowchart illustrating a method for controlling the deposition morphology of quartz glass according to an embodiment of this application; Figure 4 A flowchart illustrating a method for establishing a first correspondence between the fitting coefficients of a deposition surface fitting curve and process parameters, provided in an embodiment of this application. Figure 5 A schematic diagram of a first deposition surface image provided for an embodiment of this application; Figure 6 A schematic diagram of the sedimentary surface outline corresponding to the first sedimentary surface image provided in an embodiment of this application; Figure 7 A schematic diagram of the sedimentary surface fitting curve corresponding to the first sedimentary surface image provided in an embodiment of this application; Figure 8 A schematic diagram of a second deposition surface image provided for an embodiment of this application; Figure 9 A schematic diagram of the sedimentary surface outline corresponding to the second sedimentary surface image provided in an embodiment of this application; Figure 10 A schematic diagram of the sedimentation surface fitting curve corresponding to the second sedimentation surface image provided in an embodiment of this application; Figure 11 This is a structural schematic diagram of the first side view provided in an embodiment of this application; Figure 12 A schematic diagram of the side outline of a first side image provided in an embodiment of this application; Figure 13 A schematic diagram of the side fitting curve of the first side image provided in an embodiment of this application; Figure 14 A schematic diagram of the side outline of a second side image provided in an embodiment of this application; Figure 15 A schematic diagram of the side fitting curve of a second side image provided in an embodiment of this application; Figure 16 A schematic diagram showing the overall outline of a blank provided in an embodiment of this application; Figure 17 A schematic diagram of a method for controlling the deposition morphology of quartz glass according to an embodiment of this application; Figure 18 This is a schematic diagram of the structure of a computer device provided in an embodiment of this application.

[0011] Explanation of reference numerals in the attached figures: Quartz glass deposition furnace-10, deposition chamber-11, observation window-12, deposition heat source interface-13, silicon-containing precursor interface-14, exhaust vent-15, industrial camera-20; moving mechanism-30, differential pressure gauge-40; thermometer-50; billet-60; deposition surface-61; side surface-62. Detailed Implementation

[0012] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0013] To make the above-mentioned objectives, features and advantages of this application more apparent and understandable, the application will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0014] As mentioned in the background section, please refer to Figure 1 As shown, in the preparation process of quartz glass, a hydrogen-oxygen flame or plasma is used as a heat source. The silicon-containing precursor material undergoes a chemical reaction in a high-temperature melting furnace to generate SiO2 particles. These particles are then subjected to deposition, melting, and vitrification treatments to finally form quartz glass.

[0015] In the process of preparing quartz glass, the morphology of the quartz glass preform is generally controlled by visual observation or video monitoring, supplemented by manual marking. Due to the unique nature of the quartz glass preform preparation process—a continuous and uninterrupted process—the production cycle typically takes 3 to 60 days, or even longer, depending on the preparation method and product size requirements. The preform growth rate also varies, ranging from only 1mm to 20mm per hour. It is difficult to accurately observe subtle changes in shape with the naked eye within a short period. Manual marking for deposition surface control further complicates this process, resulting in inconsistent preform morphologies. This leads to poor longitudinal and radial uniformity, reducing the light transmittance and strength of the quartz glass, and severely impacting its optical and mechanical properties. Furthermore, adjustments to process parameters directly affect the morphology. Adjustments to parameters such as temperature and pressure do not immediately show results in the preform's shape; the effects take time to become apparent. This requires workers to anticipate shape changes and precisely adjust parameters, demanding a high level of experience from frontline process personnel, resulting in low production efficiency and a low number of qualified products.

[0016] To address the aforementioned shortcomings, this application provides a method for controlling the deposition morphology of quartz glass. Compared to existing technologies, this method acquires multiple time-point images of the billet morphology using an industrial camera and simultaneously obtains process parameters. This allows for precise capture of subtle changes over a short period, overcoming the difficulty of visually detecting minute variations. Furthermore, by selecting the deposition surface and side regions from the morphology images and performing fitting processes on them respectively, a deposition surface fitting curve and a side region fitting curve are obtained. This transforms the practical problem into a mathematical model, facilitating the accurate establishment of the correspondence between the fitting coefficients and process parameters. This method replaces manual labeling control and avoids errors inherent in manual labeling. This improved the control precision of the billet morphology; by combining camera parameters to calculate the actual contour curve, growth rate, and curvature, it achieved a precise conversion from image to actual size, providing a quantitative basis for subsequent consistency evaluation of the billet; and by obtaining a complete overall contour line through overall reconstruction, it is possible to intuitively judge the overall shape of the billet, solving the problem of morphological inconsistencies caused by manual observation; finally, based on multi-dimensional data such as growth rate, curvature, correspondence, and overall contour line, the process parameters were adjusted to ensure that the billet meets the consistency standard, thereby improving the optical and mechanical properties of quartz glass, increasing the control precision of the billet morphology, and further improving the preparation efficiency of quartz glass.

[0017] Please see Figure 2 As shown, Figure 2 This is a schematic diagram of the structure of the control device for the deposition morphology of quartz glass provided in the embodiments of this application. The control device for the deposition morphology of quartz glass includes: a quartz glass deposition furnace 10, a deposition heat source, an industrial camera 20, a moving mechanism 30, and an industrial parameter acquisition device. The quartz glass deposition furnace 10 is provided with a deposition chamber 11, which is provided with at least one observation window 12, a deposition heat source interface 13, a silicon-containing precursor interface 14, and an exhaust vent 15.

[0018] The deposition heat source interface 13 and the silicon precursor interface 14 are located at the top of the deposition chamber 11; the exhaust vent 15 is located at the bottom of the quartz glass deposition furnace 10; the deposition heat source is located at the deposition heat source interface 13; the industrial camera 20 is located outside the observation window 12. The movable mechanism 30 is located on the central axis of the quartz glass deposition furnace 10. The movable mechanism 30 includes a lifting frame and a rotating mechanism. The billet 60 is placed on the movable mechanism 30. The billet 60 includes a deposition surface 61 and a side surface 62. The process parameter acquisition equipment includes: a differential pressure gauge 40 and a thermometer 50; the differential pressure gauge 40 is located on the side of the exhaust vent near the top of the quartz glass deposition furnace 10 and is used to collect the pressure difference inside the furnace; the thermometer 50 is located at the top of the quartz glass deposition furnace 10 and is used to collect the deposition temperature.

[0019] It should be noted that the size of the quartz glass deposition furnace is no less than 500mm×500mm×500mm; the deposition heat source can be set at any position on the top of the quartz glass deposition furnace, as long as it can provide a stable and reliable heat source for the quartz glass preparation process; the silicon-containing precursor interface is used to place the silicon-containing precursor material, and the distance between the silicon-containing precursor interface and the deposition heat source interface can be customized according to actual needs, for example, it can be no less than 30mm.

[0020] The distance between the deposition surface of the preform and the furnace top in the quartz glass deposition furnace is no less than 50 mm, and the distance between the side of the preform and the inner wall of the furnace is no less than 30 mm. An exhaust vent is located at the bottom of the quartz glass deposition furnace, with a distance of no less than 10 mm between the vent and the bottom of the furnace. The distance between the differential pressure gauge interface and the top of the exhaust vent is no less than 20 mm. A temperature measurement interface is located at the top of the quartz glass deposition furnace to house a temperature measuring instrument, such as an infrared thermometer, used to measure the deposition temperature inside the furnace. The infrared temperature measuring interface can be located anywhere on the furnace top. A differential pressure gauge interface is used to house a differential pressure gauge for measuring the pressure difference inside the furnace.

[0021] The aforementioned observation window is located at the observation window interface of the quartz glass deposition furnace. The observation window interface is not lower than the deposition substrate. The size of the observation window is not less than Φ50mm or 50mm×50mm, and not greater than 1 / 4 of the size of the quartz glass deposition furnace. The distance between the observation window and the furnace top is not greater than 40mm. The number of observation windows can be one, two, or more. This embodiment does not limit the number of observation windows, as long as they can be used to observe the morphological image of the billet inside the quartz glass deposition furnace using an industrial camera. Optionally, the material of the aforementioned observation window is any one of quartz glass, sapphire glass, or CaF2, and the material is a light-transmitting material with a transmittance of not less than 70% in the visible light band.

[0022] The aforementioned industrial camera is positioned outside the observation window. The distance between the industrial camera and the observation window is not limited to 100mm. The focal length of the industrial camera lens is not less than 6mm, and a global camera is preferred. The industrial camera is used to acquire morphological images of the billet after the deposition process has started.

[0023] The aforementioned moving mechanism may include a lifting mechanism and a rotating mechanism. The lifting mechanism is used to raise or lower the billet during the deposition process, and the rotating mechanism is used to rotate the billet during the deposition process. Both the lifting mechanism and the rotating mechanism are located on the central axis of the quartz glass deposition furnace.

[0024] It should be noted that the preform is the initial form of quartz glass. In the quartz glass manufacturing process, silicon-containing precursor materials undergo a chemical reaction in a high-temperature deposition furnace using an oxyhydrogen flame or plasma as a heat source to generate SiO2 particles. These particles are then stacked through deposition steps to form a solid structure with a preliminary morphology but not yet fully melted and vitrified. The initial morphology includes the deposition surface and the side surfaces.

[0025] Specifically, in the quartz glass preparation process, the quartz glass deposition furnace is first started, providing a high-temperature environment through a deposition heat source. Silicon-containing precursor material enters the deposition chamber of the furnace through the silicon-containing precursor interface, initiating the quartz glass deposition process. Simultaneously, an industrial camera is activated to monitor the morphology of the deposition surface and sides of the billet in real time through an observation window. The industrial camera can acquire images of the billet's morphology at preset time intervals through the observation window. This preset time interval can be customized according to actual needs, for example, it could be 10 minutes or 5 minutes.

[0026] Simultaneously, process parameters during the billet deposition process in the furnace are collected in real time using process parameter acquisition equipment. These process parameters may include deposition temperature, carrier gas flow rate, furnace differential pressure, exhaust volume, raw material flow rate, and descent rate. An infrared thermometer, installed at the furnace top, collects real-time temperature data of the deposition area to obtain the deposition temperature. A flow sensor, such as a carrier gas carrying silicon precursors, is installed on the carrier gas delivery pipeline to directly measure the real-time flow rate of the carrier gas. The carrier gas is a mass flow meter. A differential pressure gauge, connected to the interface at the bottom of the furnace, measures the pressure difference between the furnace and the external environment. An air volume sensor is installed on the exhaust pipe of the furnace to monitor the real-time air volume of the exhaust gas. A flow meter, such as a precision flow meter, is installed on the silicon precursor delivery pipeline to obtain the real-time feed flow rate of the precursor. A displacement sensor on the lifting mechanism along the central axis of the furnace measures the descent speed of the lifting platform during billet deposition, i.e., the longitudinal growth rate of the billet.

[0027] After obtaining the morphological images and process parameters of the billet in the quartz glass deposition furnace at multiple time points, they can be transmitted to a computer device so that the computer device can control the morphology of the quartz glass deposition based on the morphological images and process parameters.

[0028] On the other hand, this application embodiment provides a control system for the deposition morphology of quartz glass. The system includes: a computer device and a control device for the deposition morphology of quartz glass. The control device for the deposition morphology of quartz glass is used to: provide a high-temperature environment through a deposition heat source after the quartz glass deposition furnace is started, and the silicon-containing precursor enters the furnace through the silicon-containing precursor interface. After the quartz glass deposition process begins, the device collects morphology images and process parameters of the billet in the quartz glass deposition furnace at multiple time points through an industrial camera and transmits them to the computer device. The computer equipment is used to control the deposition morphology of quartz glass based on the morphology images and process parameters of the billet at multiple time points in the quartz glass deposition furnace.

[0029] Optionally, the aforementioned computer equipment may include a terminal, a server, and a data storage system. The terminal communicates with the server via a network. The data storage system can store morphological images and process parameters of the quartz glass deposition furnace billet at multiple time points acquired by the server. The data storage system can be set up independently, integrated into the server, or located in the cloud or on other servers.

[0030] The terminals can be, but are not limited to, various desktop computers, laptops, smartphones, tablets, IoT devices, and portable wearable devices. IoT devices can include smart speakers, smart TVs, smart air conditioners, and smart in-vehicle systems. Portable wearable devices can include smartwatches, smart bracelets, and head-mounted devices. Servers can be implemented using independent servers, server clusters composed of multiple servers, or cloud servers.

[0031] On another front, this application provides a method for controlling the deposition morphology of quartz glass. This method is executed by a computer device, specifically a terminal or server, or both. In this application, the method is described using the aforementioned server as an example. Please refer to [link to relevant documentation]. Figure 3 As shown, the procedure includes steps S101 to S106. Wherein: Step S101: Obtain morphological images and process parameters of the billet in the quartz glass deposition furnace at multiple time points; the morphological images are acquired by an industrial camera after the quartz glass deposition process is started.

[0032] The aforementioned morphological images may include the deposition surface, side surface, or other background area of ​​the billet. The side surface may include two sides. The process parameters refer to the parameters collected during the deposition process of the billet. The aforementioned multiple time points may be time points generated according to a preset time interval. For example, an industrial camera may collect the deposition process of the billet every ten minutes through an observation window to obtain multiple morphological images.

[0033] Optionally, the morphological images and process parameters of the billet at multiple time points can be obtained from a database or blockchain, imported from external devices, or collected in real time by industrial cameras and process parameter acquisition devices and transmitted to computer devices.

[0034] As an alternative approach, the quartz glass deposition furnace is started first. The internal deposition heat source provides a high-temperature environment. The silicon-containing precursor enters the deposition furnace through the interface and begins the deposition process. Then, an industrial camera is activated to monitor the billet inside the furnace in real time through the observation window, obtain morphological images, and collect process parameters in real time through various process parameter acquisition devices.

[0035] Step S102: Select the deposition surface area and side surface area of ​​the billet from the morphology image.

[0036] After acquiring multiple topographic images, for each image, the deposition surface region and side region of the billet are selected. The deposition surface region refers to the image area corresponding to the dynamic growth interface at the top of the billet formed by the deposition of silicon-containing precursor particles during the quartz glass deposition process. The side region is the image area corresponding to the longitudinally extending sidewall portion of the billet, reflecting the radial morphology of the billet.

[0037] As an alternative approach, during the quartz glass deposition process, an industrial camera is used to monitor the deposition surface wheel in real time, and morphological images of the deposition surface of the billet are acquired at multiple time points. Then, the deposition surface area is extracted from these morphological images.

[0038] As another alternative implementation, during the quartz glass deposition process, the side of the billet is monitored in real time by an industrial camera, and morphological images of the side of the billet at multiple time points are acquired separately. The side area is then extracted from the second morphological image.

[0039] As another alternative approach, during the quartz glass deposition process, an industrial camera is used to monitor the deposition wheel and the side of the billet in real time, and morphological images containing the deposition surface and side surface are acquired at multiple time points. The deposition surface area and side surface area are then extracted from these morphological images.

[0040] Step S103: The sedimentation surface region is fitted into a sedimentation surface fitting curve, and a first correspondence between the fitting coefficient of the sedimentation surface fitting curve and the process parameters is established; the side region is fitted into a side fitting curve, and a second correspondence between the fitting coefficient of the side fitting curve and the process parameters is established.

[0041] The aforementioned deposition surface fitting curve is used to reflect the morphology of the deposition interface at the top of the billet, and its shooting angle focuses on the core area of ​​the billet's longitudinal growth; the side fitting curve is used to reflect the morphology of the billet's circumferential radial direction, and its shooting angle focuses on the area of ​​the billet's lateral expansion.

[0042] In one embodiment, a specific implementation method is also provided for fitting the deposition surface region into a deposition surface fitting curve and establishing a first correspondence between the fitting coefficient of the deposition surface fitting curve and the process parameters. Please refer to [link to relevant documentation]. Figure 4 As shown, the method includes: Step S201: Extract the contour line of the sedimentary surface based on the sedimentary surface region.

[0043] Step S202: Perform polynomial fitting on the sedimentation surface contour line at least twice to obtain the sedimentation surface fitting curve.

[0044] Step S203: Through correlation analysis, establish the first correspondence between the fitting coefficient of the deposition surface fitting curve and the process parameters; the process parameters include: deposition temperature, feed gas flow rate, furnace pressure difference, exhaust gas volume, raw material flow rate, and drop rate.

[0045] For each topographic image, after acquiring the sedimentary surface region, its contour line is extracted. The corresponding sedimentary surface contour line is then extracted and subjected to at least two polynomial fitting processes to obtain the sedimentary surface fitting curve. Specifically, in the process of extracting the sedimentary surface contour line based on the sedimentary surface region, the sedimentary surface region can first be converted to grayscale to obtain a grayscale image; the grayscale image is then filtered to obtain a filtered image; finally, an edge contour line extraction algorithm is used to binarize the filtered image to extract the sedimentary surface contour line from the sedimentary surface region.

[0046] Optionally, the edge contour extraction algorithm described above can be, for example, Canny, Roberts, Sobel, Scharr, Prewitt, LoG, etc. The filtering algorithm described above can be, for example, mean filtering, Gaussian filtering, etc. This embodiment does not impose any limitations on the edge contour extraction algorithm or the filtering algorithm. The filtering algorithm only needs to be able to filter the image to obtain the filtered image, and the edge contour extraction algorithm only needs to be able to binarize the filtered image and extract the deposition surface contour.

[0047] Specifically, when the topographic image contains both a sedimentary surface region and a lateral region, the sedimentary surface region and the lateral region are processed separately. When the topographic image is a sedimentary surface image containing a sedimentary surface region, two sedimentary surface images can be acquired at a preset time interval, and each sedimentary surface image is processed separately. Since the sedimentary surface region selected in the original image is a color image, it needs to be converted to grayscale to eliminate color interference. Then, filtering is performed. An m×m matrix filtering function can be used to filter the grayscale image to remove stray noise, resulting in a filtered image with clearer and smoother contours, where m≥3. For the first sedimentary surface image, a corresponding schematic diagram can be found in [reference needed]. Figure 5 As shown, after filtering the first sedimentary surface image to obtain the filtered image, an edge contour extraction algorithm is used for edge extraction. For example, the Canny algorithm can be used to binarize the filtered image to extract the sedimentary surface contour lines. The corresponding sedimentary surface contour lines can be found in [reference needed]. Figure 6 As shown. The contour line of the deposition surface can be represented by multiple pixel coordinates, that is, including the pixel coordinates of the deposition surface contour line.

[0048] In obtaining the sedimentary surface fitting curve, a pre-set multi-function model is used to perform multiple fitting processes on the sedimentary surface contour line based on the pixel coordinates of the contour line, resulting in the current sedimentary surface fitting curve, and the goodness of fit is calculated. If the goodness of fit meets the standard, the current sedimentary surface fitting curve is determined as the initial sedimentary surface fitting curve. Based on the process parameters, the coefficients of the initial sedimentary surface fitting curve are corrected to obtain the final sedimentary surface fitting curve. The sedimentary surface fitting curve corresponding to the first sedimentary surface image can be found in [reference needed]. Figure 7 As shown.

[0049] In this embodiment, the aforementioned multi-function model can be represented by a multi-function expression. At least two fitting operations are required, for example, a four-fold fitting process. By performing at least two fitting operations, a more explicit correspondence can be formed with specific process parameters, improving the accuracy of morphological description and enhancing process correlation capabilities. Goodness of fit reflects the degree of fit between the fitted curve and the original deposition surface contour line, and is a core indicator for judging the reliability of the fitted curve. The multi-function expression can be represented by the following formula: L1=P m x m +P m-1 x m-1 +…+P2x 2 +P1x+P0; Among them, P0-P m The fitting coefficients can be in the range (-∞, +∞) without any restrictions.

[0050] For example, taking a quadratic function model as an example, the pixel coordinates (x, y) of the deposition surface contour line are used as the basic data, and the quadratic function L4=P4 is applied. x 4 +P3 x 3 +P2 x 2 +P1 A curve fitting operation is performed using x+P0, where P4 is the coefficient of the 4th-order term, P3 is the coefficient of the 3rd-order term, P2 is the coefficient of the 2nd-order term, P1 is the coefficient of the 1st-order term, and P0 is the coefficient of the constant term. This yields a fitted curve of the current deposition surface containing the fitting coefficients, which are then P0-P4. Simultaneously, the degree of fit between this curve and the original contour line (i.e., the goodness of fit R) is calculated. 2 When the goodness of fit is less than the preset threshold of 98%, it indicates that the goodness of fit is not up to standard. The polynomial degree is then readjusted or the fitting algorithm optimized, and the fitting is repeated until the goodness of fit meets the standard. When the goodness of fit is not less than the preset threshold of 98%, it indicates that the goodness of fit is up to standard, i.e., the R-value of the current fitted curve is [value missing]. 2 If this requirement is met, it means that the curve accurately reflects the actual shape of the sedimentary surface, and it is determined as the initial sedimentary surface fitting curve.

[0051] The initial deposition surface fitting curve is constructed solely based on contour pixel data and does not reflect the actual impact of process parameters. Therefore, it needs to be corrected by incorporating real-time acquired process parameters such as deposition temperature T and carrier gas flow rate Vo. The correction logic can utilize the correlation between the fitting coefficients established in previous experiments and the process parameters, substituting the real-time values ​​of each process parameter into a preset correction formula to adjust the fitting coefficients (P0-P...) of the initial curve. m The process is dynamically adjusted to obtain a sedimentary surface fitting curve that can simultaneously reflect the relationship between sedimentary surface morphology and process parameters, laying the foundation for subsequent inference of the direction of process parameter adjustment through coefficient changes.

[0052] Similarly, the second sedimentary surface image is processed in the same way as the first, sequentially performing grayscale conversion, filtering, and binarization to obtain the sedimentary surface contour line, and then fitted to obtain the sedimentary surface fitting curve. Please refer to [link to relevant documentation]. Figures 8-10 As shown, Figure 8 This is a schematic diagram of the second deposition surface image provided in an embodiment of this application. Figure 9 This is a schematic diagram of the sedimentary surface outline corresponding to the second sedimentary surface image. Figure 10 The diagram below shows the fitted curve of the sedimentary surface corresponding to the second sedimentary surface image, thus obtaining the fitting coefficients and corresponding goodness of fit of the two sedimentary surface images, as shown in Table 1 below: Table 1

[0053] Understandably, based on the mathematical meaning of quartic functions, the absolute value of the quartic coefficient P4 determines the width of the curve. A wider sedimentary surface profile corresponds to a smaller quartic coefficient. The size (diameter) of the sedimentary surface is significantly affected by the sedimentation temperature T; the higher the temperature, the larger the sedimentary surface, and the smaller the quartic coefficient P4 of the fitted profile. The cubic coefficient P3 affects the inflection point of the curve, especially when the sedimentary surface is abnormal, exhibiting irregular structures such as ^, V, and W. Early intervention can be made based on the cubic coefficient P3, primarily through adjusting the carrier gas flow rate Vo. The quadratic coefficient P2 affects the curve's... Curvature characterizes the degree to which the curve deviates from a straight line. The pressure difference Pa inside the furnace affects the fluidity of the molten glass and can change the curvature of the deposition surface. The coefficient of the first-order term P1 affects the slope of the curve. The larger the slope, the more curved the curve. Experiments show that the larger the exhaust gas volume, the sharper the deposition surface profile; the smaller the exhaust gas volume, the flatter the deposition surface. Therefore, the coefficient of the first-order term is related to the exhaust gas volume W. The constant term P0 represents the ordinate of the curve, which is the height of the deposition surface and is related to the growth rate V of the quartz glass. The relative growth rate is related to the raw material flow rate Vsi and the falling rate Vd. Therefore, the quartz function model needs to be modified. The formula of the modified quartz function model can be expressed as follows: L4=P14 T x 4 +P13 Vo x 3 +P12 Pa x 2 +P11 W x+(P101 Vsi+P102 Vd); Where T is the deposition temperature, Vo is the feed gas flow rate, Pa is the pressure difference, W is the exhaust gas volume, Vsi is the feed flow rate, Vd is the rate of decrease, and P14, P13, P12, P11, P101, and P102 are the coefficients of the quartic function correction model.

[0054] Simultaneously with acquiring the first and second sedimentary surface images, the corresponding process parameters are obtained, as shown in Table 2 below: Table 2

[0055] Since process parameters determine the deposition morphology, and the contour morphology is determined by the fitting coefficient, there is a proportional relationship between the process parameters and the fitting coefficient. That is, the fitting coefficient of the corrected model = proportionality coefficient. Process parameters. Based on the above process parameters, the coefficients of the quartic function correction model are obtained by calculating using the formula modified from the above model. The coefficients of the quartic function correction model are shown in Table 3 below: Table 3

[0056] In other words, by substituting the coefficients of the above modified model into the formula of the above quartic function modified model, the final sedimentary surface fitting curve is obtained. This sedimentary surface fitting curve is a fitting curve that can simultaneously reflect the relationship between sedimentary surface morphology and process parameters.

[0057] After obtaining the fitted curve of the deposition surface, multiple sets of synchronous data need to be accumulated. This is achieved by synchronously recording the real-time values ​​of the current process parameters (e.g., T=1700℃, Vo=5L / min, etc.) at fixed intervals during the deposition process, and obtaining the fitting coefficients through polynomial fitting of the corresponding deposition surface contour. By covering a reasonable adjustment range for each process parameter, such as a gradient change of T between 1600-1900℃, at least 30 sets of "process parameter-fitting coefficient" samples are obtained to ensure data comprehensiveness. First, single-factor correlation screening is performed, calculating the correlation coefficient between each fitting coefficient and a single process parameter, and selecting strongly correlated combinations (e.g., |correlation coefficient|≥0.8). For example, the correlation coefficient between the quadratic coefficient P2 and the furnace pressure difference Pa is found to be 0.91, indicating a significant correlation. The second step is multi-factor quantitative modeling. For strongly correlated combinations, a functional relationship is established through regression analysis, such as P2=k1×Pa+k2×T+c (k1 and k2 are weighting coefficients, and c is a constant), clarifying the specific rules governing the coefficient changes with the parameters (e.g., for every 10Pa increase in Pa, P2 increases by 0.02).

[0058] In this embodiment, by extracting the contour line from the deposition surface region and performing at least two polynomial fittings to obtain the deposition surface fitting curve, it is possible not only to accurately capture the complex and subtle morphology of the deposition surface, such as curvature and local concavity and convexity, avoiding the simplification and distortion of morphological features caused by low-order fitting, but also to quantify the differences in deposition surface morphology through fitting coefficients. This provides a precise mathematical basis for establishing the correlation with process parameters such as deposition temperature and carrier gas flow rate, effectively replacing the fuzzy control method of manual observation and marking, significantly improving the accuracy of longitudinal morphology control of the billet, reducing optical and mechanical performance defects caused by morphological inconsistencies, and thus improving the efficiency of quartz glass preparation. Furthermore, establishing the first correspondence between the fitting coefficients of the deposition surface fitting curve and the process parameters provides a good data foundation for accurately locating key process parameters and calculating adjustment amounts through coefficient changes, and achieving quantitative control of longitudinal consistency of the billet.

[0059] In one embodiment, a specific implementation method for fitting the lateral region into a lateral fitting curve is also provided, including: Based on the side region, extract the side contour line; the side contour line includes: the pixel coordinates of the side contour line. By using a preset exponential function model, the side contour line is fitted with at least one exponential function based on the pixel coordinates of the side contour line to obtain the side fitting curve.

[0060] Specifically, after obtaining the lateral region, the processing method is the same as that for the deposition surface region. First, the contour lines of the lateral region are extracted. This involves converting the lateral region to grayscale to obtain a grayscale image, filtering the grayscale image to obtain a filtered image, and then using the Canny algorithm to binarize the filtered image to extract the lateral contour lines. Finally, a preset exponential function model is used to fit at least one exponential function to the lateral contour lines, resulting in a lateral fitting curve with at least one exponential term. This preset exponential function model can be expressed by the following formula: Le=a m e bmx + a m-1 e bm-1x +…+ a1e b1x ; Among them, a1-a m b1-bm are the fitting coefficients. a1-a m is the amplitude coefficient, and b1-bm is the exponential growth coefficient.

[0061] In this embodiment, during the quartz glass deposition process, the side of the billet is monitored in real time using an industrial camera. A side image can be captured every 10 minutes or more. For example, the first side image can be found in [reference needed]. Figure 11 As shown, the image is converted to grayscale, then the side area of ​​the grayscale image is selected and cropped. This is done using 3... A filtering matrix of 3 is used to filter and remove stray points, making the contour lines smoother and clearer. The deposition surface is then binarized using the Canny algorithm to obtain the side contour lines. The side contour lines of this first side image can be found in [reference needed]. Figure 12 As shown. The side profile of the billet can be represented by multiple pixel coordinates, that is, it includes the pixel coordinate data of the side profile of the billet. Taking the preset exponential function model as a double exponential function model as an example, based on the pixel coordinate data of the side profile, and by using the double exponential function model for fitting, the corresponding side fitting curve is obtained. The side fitting curve of the first side image can be found in [reference needed]. Figure 13 As shown, the double exponential function model can be expressed by the following formula: Le2=a2e b2x +a1e b1x ; Where a1 and a2 are amplitude coefficients that determine the positive or negative amplitude, and b1 and b2 are exponential growth coefficients.

[0062] Similarly, the image obtained ten minutes after cropping the first side view image is used as the second side view image. The processing procedure is the same as for the first side view image: the side contour line is extracted from the second side view image. A schematic diagram of the side contour line of this second side view image can be found in [reference needed]. Figure 14 As shown, the side profile is fitted with at least one exponential function to obtain the side fitting curve. The side fitting curve of this second side image can be found in [reference needed]. Figure 15 As shown in Table 4 below, the fitting curves for the first and second side images are as follows: Table 4

[0063] According to the mathematical meaning of the double exponential function, a1 and a2 are amplitude coefficients, and b1 and b2 are exponential growth rates. The side profile of the quartz glass preform is a linear superposition of two exponential terms, and the growth behavior of the preform is dominated by the coefficients and exponents. The amplitude coefficients a1 and a2 determine the longitudinal change in the position of the preform's side profile, which is related to the material rate and fall rate in the deposition process. The exponential growth rates b1 and b2 determine the change in the side profile of the preform. When b1 and b2 > 0, it is a double exponential compound growth model, and the glass preform is in a process of synchronous radial and longitudinal growth, which is related to the deposition temperature T and the exhaust gas volume W. When b1 > 0 and b2 < 0, it is a model of coexistence of growth and decay, reflecting the process of radial growth and slowing longitudinal growth of the preform morphology, reflecting the transient response of adjusting process parameters to the glass. When b1 and b2 < 0, it is a double exponential decay, with rapid and slow decay, reflecting the decrease in the radial and longitudinal growth rates of the preform. Therefore, the double exponential function model needs to be modified. The formula for the modified double exponential function model can be expressed as follows: Le2=a 12 + a 11 ; Among them, a 12 a 11 b12 and b11 are the coefficients of the modified model.

[0064] While acquiring the first and second side images, the corresponding process parameters are also obtained. These process parameters are shown in Table 5 below. Table 5

[0065] In other words, by substituting the coefficients of the above-mentioned modified model into the formula of the above-mentioned double exponential function modified model, the final side fitting curve is obtained. This side fitting curve is a fitting curve that can simultaneously reflect the relationship between the side morphology and the process parameters. Among them, the deviation of the correction coefficient is less than 1%, indicating that the model can reflect the changes in the morphology of the billet.

[0066] During the deposition process, real-time process parameters, such as exhaust volume W=20m³, are recorded at fixed intervals. 3 The parameters included / h, deposition temperature T=1750℃, etc. A double exponential function was applied to the extracted side profile lines to obtain the corresponding fitting coefficients. By covering the typical adjustment range of each process parameter, ≥30 sets of data related to the side fitting coefficients of each process parameter were accumulated to ensure that the samples could reflect the side morphological characteristics under different working conditions. Furthermore, through single-factor correlation screening, the correlation coefficients between individual fitting coefficients and each process parameter were calculated, identifying strongly correlated combinations (e.g., |correlation coefficient|≥0.8). For example, the correlation coefficient between b2 and the process parameter reached 0.89, indicating a significant correlation. A functional relationship was established through multiple regression analysis, such as b2=k1×W+k2×Pa+c (k1, k2 are weighting coefficients, c is a constant), clarifying the law of coefficient change with parameters, such as W increasing by 2m... 3 / h, b2 is increased by 0.015, and an interaction term is introduced to correct the collaborative effect of multiple parameters and improve the correlation accuracy.

[0067] In this embodiment, by establishing a second correspondence between the fitting coefficient of the side fitting curve and the process parameters, it is possible to accurately locate key process parameters and calculate the adjustment amount through coefficient changes, effectively avoiding problems such as irregular cross-section and diameter deviation of the blank material caused by uneven side shape, and ensuring the dimensional accuracy and performance stability of the quartz glass finished product.

[0068] Step S104: Obtain camera parameters, and calculate the actual profile curve, growth rate and curvature of the billet based on the camera parameters.

[0069] The growth rates mentioned above include longitudinal growth rate and radial growth rate; curvature includes deposition surface curvature and lateral curvature; camera parameters include the tilt angle of the industrial camera relative to the horizontal plane.

[0070] It should be noted that the longitudinal growth rate mentioned above refers to the growth rate of the billet along the growth axis (height direction), while the radial growth rate refers to the growth rate of the billet along the circumference of the cross-section (horizontal direction); the deposition surface curvature is the degree of curvature of the deposition surface contour line, and the side curvature is the degree of curvature of the side contour line; the tilt angle of the industrial camera relative to the horizontal plane is the angle between the optical axis of the camera lens and the horizontal plane.

[0071] In the process of calculating the actual contour curve, growth rate, and curvature of the billet based on camera parameters, the pixel length and actual length of the billet can be obtained first; the image-to-object ratio coefficient of the billet can be calculated based on the tilt angle of the industrial camera relative to the horizontal plane, the pixel length, and the actual length; using the image-to-object ratio coefficient, the pixel coordinates of the deposition surface fitting curve can be converted into the actual size coordinates of the deposition surface, and the pixel coordinates of the side fitting curve can be converted into the actual size coordinates of the side; based on the actual size coordinates of the deposition surface and the actual size coordinates of the side, the actual contour curve of the billet can be obtained; based on the deposition surface fitting curve, the longitudinal growth rate and the curvature of the deposition surface can be calculated, and based on the side fitting curve, the radial growth rate and the side curvature can be calculated.

[0072] Specifically, the aforementioned pixel length can be the cumulative pixel count of the image outline, while the actual length can be obtained by measuring the actual billet. The tilt angle of the industrial camera relative to the horizontal plane is obtained through the angle sensor used when the camera is mounted. After obtaining the pixel length and actual length of the billet, the image-to-object ratio coefficient of the billet can be calculated using the following formula: r=L 像 cosθ / L 物 ; Where r is the image-object scale factor, L 像 L is the pixel length of the blank. 物 θ represents the actual length of the blank, and θ is the tilt angle of the industrial camera relative to the horizontal plane.

[0073] After obtaining the image-object scale factor, the pixel coordinates of the deposition surface fitting curve and the side profile fitting curve are converted into actual size coordinates using this scale factor, which can be expressed by the following formula: L 物 =L 像 cosθ / r, L 像 = r L 物 / cosθ; Based on the actual size coordinates of the deposition surface and the side surface, an actual contour curve reflecting the true shape of the billet is obtained, solving the problem of dimensional deviation caused by perspective distortion in the image. Based on the transformed actual size coordinates of the deposition surface, the longitudinal growth rate is obtained by calculating the displacement change of the deposition surface contour line along the longitudinal (height) direction per unit time. The curvature of the deposition surface is calculated using the ratio of the second derivative to the square of the sum of the squares of the first derivatives. Similarly, based on the actual size coordinates of the side surface, the radial growth rate is obtained by calculating the radial (horizontal) expansion of the side surface contour line per unit time. The side curvature is calculated using the same curvature formula. Ultimately, this achieves a quantitative characterization of the billet's growth dynamics and morphological curvature, providing a physically meaningful quantitative basis for subsequent consistency assessment and process parameter adjustment.

[0074] In this embodiment, camera parameters such as the tilt angle of the industrial camera relative to the horizontal plane are used, combined with the pixel length and the actual length to calculate the image-object ratio coefficient. The pixel coordinates of the fitted curves of the deposition surface and side surface are accurately converted into actual size coordinates, thereby accurately obtaining the actual contour curve of the billet. At the same time, the longitudinal / radial growth rate and the curvature of the deposition surface / side surface are quantitatively calculated, effectively eliminating image perspective distortion and the deviation between pixels and actual size. This achieves accurate quantitative characterization of the true shape, growth dynamics and bending degree of the billet, providing a reliable physical quantitative basis for subsequent billet consistency evaluation and precise adjustment of process parameters.

[0075] Step S105: Reconstruct the overall shape based on the fitting curve of the deposition surface and the fitting curve of the side surface to obtain the overall outline of the billet.

[0076] It is understandable that, due to the different shooting targets on the deposition surface and the side, the shooting angle and focus area of ​​the industrial camera are different, which causes the two fitted curves to not be continuously connected in the actual space and to have a physical discontinuity distance. This physical discontinuity distance is the actual blank area between the lower end of the deposition surface contour line and the upper end of the side contour line.

[0077] The aforementioned multiplicative function model is one representation of the deposition surface model, and the exponential function model is one representation of the radial contour model. After obtaining the deposition surface model and the radial contour model, a complete outline model of the billet is constructed to evaluate the deposition process of quartz glass. The complete outline of the billet consists of a deposition surface contour and two side radial contours. Since the deposition surface and the side views are not continuous, the complete outline constructed from these contours is discontinuous. The pixel discontinuity distance is determined based on the actual discontinuity distance, and the complete outline of the billet is reconstructed online.

[0078] Specifically, in obtaining the actual discontinuity distance, a high-precision distance sensor can be installed in the corresponding discontinuous area within the deposition furnace, or a calibration object of known actual size can be used to assist in the measurement. This allows for the direct acquisition of the true physical gap length between the lower edge of the deposition surface and the upper edge of the side surface, i.e., the actual discontinuity distance, ensuring that the data reflects the true spatial positional relationship. Then, based on the image-object ratio coefficient calculated above, the actual discontinuity distance is converted into pixel discontinuity distance. This can be achieved through the calculation logic of "pixel discontinuity distance = actual discontinuity distance × image-object ratio coefficient," accurately mapping the gap length in the physical space to the gap length in the image pixel coordinate system. This unifies the discontinuity distance from the physical dimension to the image dimension, providing a quantitative basis for adapting image data for contour stitching.

[0079] After obtaining the pixel discontinuity distance, contour stitching processing is performed based on the pixel discontinuity distance. It can be using the pixel discontinuity distance as the "connection scale" to supplement a smooth transition contour segment that conforms to the growth law of the blank (such as generating a transition curve according to the curvature trend of adjacent contours) between the lower edge of the deposition surface fitting curve and the upper edges of the side fitting curves on both sides, seamlessly connecting the originally scattered deposition surface fitting curve and the side fitting curves on both sides, and finally forming a complete blank contour line covering the "top deposition surface - sides on both sides". This complete contour line can be seen as shown in Figure 16 shown. The complete contour line can be expressed by the following formula: ; where, L 全 is the complete blank contour line, describing the overall shape of the blank. a1 - a2, b1 - b2 are the parameters of the radial contour line model. a1 - a2 is the amplitude coefficient of the exponential function, reflecting the longitudinal position of the side contour. b1 - b2 is the exponential growth coefficient of the exponential function, reflecting the radial growth rate of the side contour. P0 - P4 are the parameters of the deposition surface model, that is, the fitting coefficients of the polynomial function, corresponding to the quartic term, cubic term, quadratic term, linear term, and constant term respectively, jointly describing the complex shapes such as bending and concavity - convexity of the deposition surface. M, N, R, S refer to the interval demarcation points in the pixel coordinate system, used to divide the pixel ranges of the deposition surface contour and the side contours on both sides. M is the abscissa value of the end position of the left - hand radial contour line model. N, R are the abscissa values of the start position and end position of the deposition surface contour line. S is the abscissa value of the start position of the right - hand radial contour line model. x ≤ M corresponds to the left - hand side contour, N ≤ x ≤ R corresponds to the deposition surface contour, S < x corresponds to the right - hand side contour. The first discontinuity distance N - M ≥ 0, the second discontinuity distance S - R ≥ 0. In this embodiment, the first discontinuity distance is selected as 300, and the second discontinuity distance is selected as 80. The first discontinuity distance N - M is the interval distance between the right end of the left - hand radial contour line and the left end of the deposition surface contour line. The second discontinuity distance S - R is the interval distance between the left end of the right - hand radial contour line and the left end of the deposition surface contour line.

[0080] In this embodiment, by obtaining the actual discontinuity distance between the deposition surface and the side shooting positions and combining with the object - image ratio coefficient, the pixel discontinuity distance can be accurately determined. Based on this, the deposition surface and the side fitting curves are stitched to obtain the complete blank contour line, effectively solving the problem of contour discontinuity caused by the shooting angle difference, completely restoring the overall shape of the blank, avoiding misjudgment of the shape caused by shooting discontinuity, providing complete and accurate shape data support for the consistency evaluation of the blank, and improving the pertinence and effectiveness of subsequent process parameter adjustment.

[0081] Step S106, based on the growth rate, curvature, the first correspondence, the second correspondence, and the complete contour line, control and adjust the process parameters to meet the consistency evaluation standard.

[0082] The aforementioned consistency assessment criteria include longitudinal consistency criteria and radial consistency criteria. Longitudinal growth rate is used to characterize the growth rate of the billet along the height direction, deposition surface curvature is used to characterize the degree of curvature of the deposition surface profile, radial growth rate is used to characterize the growth rate of the billet along the circumferential direction, and lateral curvature is used to characterize the degree of curvature of the lateral profile.

[0083] Based on growth rate, curvature, first correspondence, second correspondence, and overall contour line, process parameters are controlled and adjusted to achieve consistency evaluation standards, including: Based on the overall outline of the billet, longitudinal growth rate, and curvature of the deposition surface, it is determined whether the billet meets the longitudinal consistency standard. Based on the overall outline of the billet, radial growth rate, and lateral curvature, it is determined whether the billet meets the radial consistency standard. When the billet does not meet the longitudinal consistency standard, the first process parameter to be adjusted is determined according to the first correspondence relationship and the first process parameter is adjusted until the longitudinal consistency standard is met. And / or, when the radial consistency standard is not met, the second process parameter to be adjusted is determined and the second process parameter is adjusted until the radial consistency standard is met.

[0084] Specifically, after determining the overall outline, longitudinal growth rate, and deposition surface curvature, the deposition state can be assessed by using the overall outline of the billet as a basic morphological reference, combined with the longitudinal growth rate and deposition surface curvature, to determine whether the billet meets the longitudinal consistency standard from three dimensions: overall morphology, growth dynamics, and local curvature. This deposition state can include a flat or gently sloping deposition surface, a curved deposition surface, or a steep deposition surface. A curved deposition surface refers to a circular arc shape, while a steep deposition surface refers to a conical shape. This longitudinal consistency standard can be customized according to actual needs.

[0085] When the billet does not meet the longitudinal consistency standard, based on the first correspondence between the deposition surface fitting coefficient and the process parameters established in the early stage, the key process parameters that cause the deviation are located, such as deposition temperature T and carrier gas flow rate Vo. The specific adjustment amount of the first process parameter is calculated according to the coefficient deviation, such as T needs to be increased by 5℃ and Vo needs to be decreased by 0.3L / min. By adjusting the first process parameter in real time, the deposition surface morphology is continuously optimized until the billet meets the longitudinal consistency standard, thus achieving closed-loop precise control of the billet morphology.

[0086] Similarly, by combining radial growth rate and lateral curvature, it is determined whether the billet meets the radial consistency standard. When it does not meet the radial consistency standard, the second process parameters are adjusted according to the second correspondence to make it meet the radial consistency standard. Furthermore, the deposition process of quartz glass is continuously monitored by an industrial camera, with data updated every 10 minutes, to achieve real-time closed-loop control of the deposition morphology. By combining the process parameters and the change law of billet morphology, and by accumulating deposition model coefficients and process parameters and storing them in a database, they can be used as training data for deep learning. This allows for the specification of appropriate process parameters, resulting in quartz glass with a transmittance ≥85%@180~2600nm, optical uniformity ≤1ppm, and stress birefringence ≤1nm / cm.

[0087] Please see Figure 17 As shown, during the quartz glass deposition process, real-time photos are acquired using an industrial camera and processed sequentially through contour line extraction, curve fitting, and model building. This process constructs a deposition surface model and a radial contour line model, obtaining model parameters and morphological data such as longitudinal dimensions, curvature, and radial dimensions. Combining these model parameters with deep learning, the longitudinal and radial growth rates are calculated and correlated with the billet morphology. The billet morphology is then evaluated in conjunction with process parameters, and adjustments are made to the process parameters based on consistency evaluation criteria to improve the optical and mechanical properties of the quartz glass, specifically enhancing its transmittance, optical uniformity, and stress birefringence. Ultimately, this achieves intelligent and precise control from data acquisition to process optimization and performance feedback, enabling fine-scale control of the billet morphology at the microscale. This allows for a deeper understanding of the quartz glass deposition and growth behavior, facilitating finer-grained control of the billet morphology.

[0088] The method for controlling the deposition morphology of quartz glass provided in this embodiment acquires images of the billet morphology at multiple time points using an industrial camera and simultaneously obtains process parameters. Compared with manual observation, it can accurately capture subtle changes in a short period of time, solving the problem that subtle changes are difficult to detect with the naked eye. The deposition surface region and the side region are selected from the morphology images and fitted separately to obtain the deposition surface fitting curve and the side fitting curve. The actual problem is transformed into a mathematical model, which facilitates the subsequent accurate establishment of the correspondence between the fitting coefficient and the process parameters. This replaces manual labeling control, avoids the errors of manual labeling, and improves the control accuracy of the billet morphology. By combining camera parameters to calculate the actual contour curve, growth rate, and curvature, a precise conversion from image to actual size was achieved, providing a quantitative basis for subsequent consistency evaluation of the billet. Furthermore, by reconstructing the overall shape, a complete contour line was obtained, which allowed for an intuitive judgment of the overall morphology of the billet, solving the problem of inconsistent morphology caused by manual observation. Finally, based on multi-dimensional data such as growth rate, curvature, correspondence, and overall contour line, process parameters were adjusted to ensure that the billet met the consistency standard, thereby improving the optical and mechanical properties of quartz glass, increasing the control precision of the billet morphology, and further improving the preparation efficiency of quartz glass.

[0089] On the other hand, embodiments of this application also provide a control system for the deposition morphology of quartz glass, the control system for the deposition morphology of quartz glass including: computer equipment and control device for the deposition morphology of quartz glass; The morphology control device for quartz glass deposition is used to: provide a high-temperature environment through the deposition heat source after the quartz glass deposition furnace is started, and the silicon-containing precursor enters the furnace through the silicon-containing precursor interface. After the quartz glass deposition process begins, the morphology images and process parameters of the billet in the quartz glass deposition furnace at multiple time points are collected by an industrial camera and transmitted to the computer equipment. The computer equipment is used to: acquire morphological images and process parameters of the billet at multiple time points in the quartz glass deposition furnace; select the deposition surface region or side region of the billet from the morphological images; fit the deposition surface region into a deposition surface fitting curve and establish a first correspondence between the fitting coefficient of the deposition surface fitting curve and the process parameters; fit the side region into a side fitting curve and establish a second correspondence between the fitting coefficient of the side fitting curve and the process parameters; acquire camera parameters and calculate the actual contour curve, longitudinal growth rate, and curvature of the billet based on the camera parameters; reconstruct the overall appearance based on the deposition surface fitting curve and the side fitting curve to obtain the overall outline of the billet; and control and adjust the process parameters based on the growth rate, curvature, first correspondence, second correspondence, and overall outline to achieve a consistency evaluation standard.

[0090] The quartz glass deposition morphology control system provided in this embodiment acquires billet morphology images at multiple time points using an industrial camera and simultaneously obtains process parameters. Compared to manual observation, this system can accurately capture subtle changes over a short period, solving the problem of subtle changes being difficult to detect with the naked eye. The system selects the deposition surface region and side region from the morphology images and performs fitting processing to obtain deposition surface fitting curves and side fitting curves, transforming the practical problem into a mathematical model. This facilitates the subsequent accurate establishment of the correspondence between fitting coefficients and process parameters, replacing manual labeling control, avoiding errors from manual labeling, and improving the control accuracy of the billet morphology. By combining camera parameters to calculate the actual contour curve, growth rate, and curvature, a precise conversion from image to actual size was achieved, providing a quantitative basis for subsequent consistency evaluation of the billet. Furthermore, by reconstructing the entire picture, a complete overall contour line was obtained, which can intuitively judge the overall shape of the billet and solve the problem of inconsistent morphology caused by manual observation. Finally, based on multi-dimensional data such as growth rate, curvature, correspondence, and overall contour line, the process parameters were adjusted to ensure that the billet meets the consistency standard, thereby improving the optical and mechanical properties of quartz glass, increasing the control precision of the billet morphology, and further improving the preparation efficiency of quartz glass.

[0091] In one exemplary embodiment, a computer device is provided, which may be a server or a terminal, and its internal structure diagram may be as follows. Figure 18 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The database stores video tag processing data. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communication with external terminals via a network connection. When executed by the processor, the computer program implements a method for controlling the deposition morphology of quartz glass.

[0092] Those skilled in the art will understand that Figure 18 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0093] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0094] In one exemplary embodiment, a computer-readable storage medium is provided storing a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0095] In one exemplary embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above-described method embodiments.

[0096] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0097] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium. When executed, the computer program can include the processes of the embodiments described above. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM).

[0098] The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0099] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0100] This document uses specific examples to illustrate the principles and implementation methods of this application. The descriptions of the above embodiments are only for the purpose of helping to understand the methods and core ideas of this application. Furthermore, those skilled in the art will recognize that, based on the ideas of this application, there will be changes in the specific implementation methods and application scope. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A method of controlling the deposition morphology of quartz glass, characterized by, The quartz glass deposition morphology control method comprises: Obtaining the morphology images and process parameters of the blank in the quartz glass deposition furnace at multiple time points; the morphology images are collected by an industrial camera after starting the quartz glass deposition process; Selecting the deposition surface area and side area of the blank from the morphology images; Fitting the deposition surface area to a deposition surface fitting curve, establishing a first correspondence relationship between the fitting coefficients of the deposition surface fitting curve and the process parameters, and fitting the side area to a side fitting curve, and establishing a second correspondence relationship between the fitting coefficients of the side fitting curve and the process parameters; Obtaining camera parameters, and calculating the actual profile curve, growth rate and curvature of the blank according to the camera parameters; Reconstructing the overall appearance according to the deposition surface fitting curve and the side fitting curve to obtain the overall profile curve of the blank; Based on the growth rate, curvature, first correspondence relationship, second correspondence relationship and overall profile curve, the process parameters are adjusted to achieve a consistency evaluation standard.

2. The method of controlling the deposition morphology of quartz glass according to claim 1, characterized by Fitting the deposition surface area to a deposition surface fitting curve, and establishing a first correspondence relationship between the fitting coefficients of the deposition surface fitting curve and the process parameters, comprises: Based on the deposition surface area, the deposition surface profile line is extracted; The deposition surface profile line is subjected to at least two times of polynomial fitting processing to obtain a deposition surface fitting curve; Through correlation analysis, a first correspondence relationship between the fitting coefficients of the deposition surface fitting curve and the process parameters is established; the process parameters include: deposition temperature, carrier gas flow, furnace pressure difference, tail gas exhaust amount, raw material flow and speed reduction.

3. The method of controlling the deposition morphology of quartz glass according to claim 2, characterized in that Based on the deposition surface area, the deposition surface profile line is extracted, comprising: The deposition surface area is subjected to grayscale processing to obtain a grayscale image; The grayscale image is subjected to filtering processing to obtain a filtered image; The filtered image is subjected to binaryzation processing by using an edge profile line extraction algorithm to extract the deposition surface profile line.

4. The method of controlling the deposition morphology of quartz glass according to claim 3, characterized by The deposition surface profile line comprises deposition surface profile line pixel coordinates, and the deposition surface profile line is subjected to at least two times of polynomial fitting processing to obtain a deposition surface fitting curve, comprising: Based on the deposition surface profile line pixel coordinates, the deposition surface profile line is subjected to multiple times of fitting processing by using a preset multiple function model to obtain a current deposition surface fitting curve, and the goodness of fit is calculated; If the goodness of fit meets the standard, the current deposition surface fitting curve is determined as an initial deposition surface fitting curve; According to the process parameters, the initial deposition surface fitting curve is subjected to coefficient correction to obtain the deposition surface fitting curve.

5. The method of controlling the deposition morphology of quartz glass according to claim 1, wherein Fitting the side area to a side fitting curve, comprising: Based on the side area, a side profile line is extracted; the side profile line comprises side profile line pixel coordinates; Based on the side profile line pixel coordinates, the side profile line is subjected to at least one time of exponential function fitting processing by using a preset exponential function model to obtain the side fitting curve.

6. The method of controlling the deposition morphology of quartz glass according to claim 1, wherein The growth speed includes longitudinal growth speed and radial growth speed; the curvature includes deposition surface curvature and side surface curvature; the camera parameter includes an inclination angle of the industrial camera relative to a horizontal plane; According to the camera parameter, an actual profile curve of the blank is calculated, and the growth speed and the curvature are included; A pixel length and an actual length of the blank are obtained; According to the inclination angle of the industrial camera relative to the horizontal plane, the pixel length and the actual length, a pixel-to-actual length conversion coefficient of the blank is calculated; By using the pixel-to-actual length conversion coefficient, pixel coordinates of the deposition surface fitting curve are converted into deposition surface actual size coordinates, and pixel coordinates of the side surface fitting curve are converted into side surface actual size coordinates; Based on the deposition surface actual size coordinates and the side surface actual size coordinates, the actual profile curve of the blank is obtained; Based on the deposition surface fitting curve, the longitudinal growth speed and the deposition surface curvature are calculated, and based on the side surface fitting curve, the radial growth speed and the side surface curvature are calculated.

7. The method of controlling the deposition morphology of quartz glass according to claim 6, characterized in that According to the deposition surface fitting curve and the side surface fitting curve, a full profile curve of the blank is obtained, including: An actual interval distance between a deposition surface and a side surface shooting position is obtained; According to the pixel-to-actual length conversion coefficient, the actual interval distance is converted into a pixel interval distance; Based on the pixel interval distance, the deposition surface fitting curve and the side surface fitting curve are spliced to obtain the full profile curve of the blank.

8. The method of controlling the deposition morphology of quartz glass according to claim 6, wherein The consistency evaluation standard includes longitudinal consistency standard and radial consistency standard; Based on the growth speed, the curvature, the first corresponding relationship, the second corresponding relationship and the full profile curve, the process parameters are adjusted to achieve the consistency evaluation standard, including: Based on the full profile curve of the blank, the longitudinal growth speed and the deposition surface curvature, it is judged whether the blank meets the longitudinal consistency standard, and based on the full profile curve of the blank, the radial growth speed and the side surface curvature, it is judged whether the blank meets the radial consistency standard; When the longitudinal consistency standard is not met, the first process parameter to be adjusted is determined according to the first corresponding relationship, and the first process parameter is adjusted until the longitudinal consistency standard is met; and / or, When the radial consistency standard is not met, the second process parameter to be adjusted is determined according to the second corresponding relationship, and the second process parameter is adjusted until the radial consistency standard is met.

9. An apparatus for controlling the deposition morphology of quartz glass, characterized by The quartz glass deposition morphology control device includes: A quartz glass deposition furnace, a deposition chamber is arranged in the quartz glass deposition furnace, the deposition chamber is provided with at least one observation window, a deposition heat source interface, a silicon-containing precursor interface and an exhaust port, the deposition heat source interface and the silicon-containing precursor interface are located at the top of the deposition chamber; the exhaust port is located at the bottom of the quartz glass deposition furnace; A deposition heat source, the deposition heat source is located in the deposition heat source interface; An industrial camera, the industrial camera is arranged outside the observation window; A movable mechanism, the movable mechanism is located on the central axis of the quartz glass deposition furnace, the movable mechanism includes a lifting frame and a rotating mechanism, and the blank is placed on the movable mechanism; The process parameter acquisition device comprises a differential pressure gauge and a temperature measuring instrument; the differential pressure gauge is located at one side of the exhaust port close to the top of the quartz glass deposition furnace, and is used for acquiring the pressure difference in the furnace; and the temperature measuring instrument is located at the top of the quartz glass deposition furnace, and is used for acquiring the deposition temperature.

10. A system for controlling the deposition morphology of quartz glass, characterized in that The quartz glass deposition morphology control system comprises a computer device and a quartz glass deposition morphology control device. The quartz glass deposition morphology control device is used for: after starting the quartz glass deposition furnace, providing a high-temperature environment through a deposition heat source, and after the quartz glass deposition process is started, acquiring the morphology images and process parameters of the blank in the quartz glass deposition furnace at multiple time points through an industrial camera and transmitting them to the computer device. The computer device is used for: acquiring the morphology images and process parameters of the blank in the quartz glass deposition furnace at multiple time points; selecting the deposition surface region or the side surface region of the blank from the morphology images; fitting the deposition surface region to a deposition surface fitting curve, establishing a first corresponding relationship between the fitting coefficients of the deposition surface fitting curve and the process parameters, and fitting the side surface region to a side surface fitting curve, and establishing a second corresponding relationship between the fitting coefficients of the side surface fitting curve and the process parameters; acquiring the camera parameters, and calculating the actual profile curve, the longitudinal growth speed and the curvature of the blank according to the camera parameters; reconstructing the overall morphology according to the deposition surface fitting curve and the side surface fitting curve to obtain the overall profile curve of the blank; and based on the growth speed, the curvature, the first corresponding relationship, the second corresponding relationship and the overall profile curve, adjusting the process parameters to achieve the consistency evaluation standard.

Images