Glass frit re-detection system and detection method

Abstract

The application relates to the technical field of automatic control of glass manufacturing, in particular to a glass material weight detection system and a detection method, which comprises a discharging mechanism used for extruding glass material liquid downwards; a material cutting mechanism arranged at the side of the discharging mechanism and used for cutting the glass material liquid after the glass material liquid flows for a preset time, so as to form falling material drops; a camera arranged at the side of the discharging mechanism and used for collecting images of the material drops; and a control analysis device in communication connection with the discharging mechanism, the material cutting mechanism and the camera respectively, and used for adjusting the size of the material drops dropped by the discharging mechanism according to the first image. The application has the effect of conveniently detecting the weight of the material drops and adjusting the extrusion speed of the discharging mechanism.

Description

Technical Field

[0001] This invention relates to the field of automated control technology in glass manufacturing, specifically to a glass material weight detection system and method. Background Technology

[0002] On automated glassware production lines, the weight and shape of the glass droplets determine the quality of the final product. Conventional droplet forming devices typically include a discharge mechanism, a plunger mechanism, and a shearing mechanism. Molten glass is extruded from the circular orifice of the discharge mechanism by the plunger mechanism, and then cut by the shearing mechanism to form the falling droplets.

[0003] However, due to factors such as ambient temperature and the temperature of the molten glass itself, the weight of each droplet is not exactly the same, affecting the subsequent production of glass products. Therefore, how to conveniently and accurately detect the weight of the droplets and adjust the extrusion speed of the discharging mechanism in real time has become a technical problem that urgently needs to be solved in this field. Summary of the Invention

[0004] To facilitate the detection of droplet weight in order to adjust the extrusion speed of the discharge mechanism, this application provides a glass droplet weight detection system and detection method.

[0005] The glass weight detection system and method provided in this application adopt the following technical solution: A glass material weight detection system, characterized in that it includes: The discharge mechanism is used to extrude molten glass downwards; A shearing mechanism is located to the side of the discharge mechanism and is used to cut the molten glass after a preset time has elapsed to form falling droplets. A camera is positioned to the side of the discharge mechanism to capture images of the material droplets. The control and analysis device is communicatively connected to the discharge mechanism, the shearing mechanism, and the camera, and adjusts the size of the droplets falling from the discharge mechanism based on the first image.

[0006] By adopting the above technical solution, a camera acquires real-time images of the droplets, and a control and analysis device calculates their weight, thus facilitating the measurement of droplet weight. The control and analysis device automatically adjusts the discharge speed of the discharge mechanism based on weight deviations, thereby ensuring uniform discharge and improving the quality stability of the final glass products.

[0007] Optionally, the camera has a built-in strobe light source, which is electrically connected to the control and analysis device; after the shearing mechanism performs the shearing action, the control and analysis device triggers the strobe light source to flash and simultaneously triggers the camera to acquire images.

[0008] By employing the above technical solution, the droplet falls rapidly, which easily results in motion blur and ghosting under ordinary light sources. The stroboscopic light source provides high-intensity illumination and is synchronously triggered with the camera, effectively freezing the high-speed moving droplet. This yields a high-quality image with clear outlines and distinct details, providing a reliable data foundation for subsequent accurate weight calculations.

[0009] Optionally, a laser rangefinder sensor is also included, which is disposed on one side of the discharge mechanism and is used to detect the time point when the falling molten glass blocks the laser.

[0010] By adopting the above technical solution, the camera employs two shooting modes: high-frequency and low-frequency. In low-frequency mode, the camera reduces its shooting frequency, extending its lifespan and reducing the system's data processing burden. The laser rangefinder, as a low-cost, high-response monitoring method, continuously monitors the droplet's falling speed in low-frequency mode. The control and analysis device controls the camera to switch between high-frequency and low-frequency modes based on time points, thereby achieving an optimized balance between detection efficiency and system cost.

[0011] Optionally, the laser rangefinder is fixedly mounted on the camera. The laser rangefinder is used to detect the distance between the camera and a preset reference object. The control and analysis device determines whether the camera is tilted based on the distance change detected by the laser rangefinder.

[0012] By adopting the above technical solution, when the camera tilts, the laser rangefinder will tilt along with the camera. The control and analysis device determines whether the camera is tilted based on the distance change detected by the laser rangefinder, allowing for timely notification to staff to adjust the camera.

[0013] A method for testing the weight of glass frit, the method comprising the following steps: The shearing mechanism is controlled to cut the molten glass into droplets, and the first image of the falling droplets is captured by a camera. The final weight of the droplet is obtained from the first image; The discharge speed of the discharge mechanism is adjusted based on the comparison between the final weight and the preset target weight.

[0014] By adopting the above technical solution, image acquisition, data processing and actuator control are organically combined, achieving full automation from detection to feedback adjustment, reducing manual intervention and improving production efficiency and product consistency.

[0015] Optionally, the specific process of obtaining the final weight of the droplet based on the first image includes: The first image is preprocessed and edge extracted to identify the shape of the material droplets; Extract the length, neck diameter, and droplet head contour feature parameters of the droplet shape; The contour feature parameters are input into a pre-established fluid volume prediction model to calculate the volume of the droplet; The volume of the droplet is converted into its final weight based on the preset density of the molten glass.

[0016] By employing the above technical solution, compared to directly estimating weight from the area of ​​a two-dimensional image, this method extracts multiple key contour feature parameters of the droplet and inputs them into the fluid volume prediction model, enabling a more accurate reconstruction of the droplet's three-dimensional morphology and calculation of its volume. Combined with the known glass density, this significantly improves the accuracy of the final weight calculation.

[0017] Optional, also includes: Calculate the average absolute value of the discharge speed adjustment range within the preset time window to obtain the average range; When the average amplitude is greater than the first amplitude threshold, the camera is controlled to enter the high-frequency mode. The high-frequency mode is when the camera takes pictures after the shearing mechanism cuts the molten glass each time. When the average amplitude is less than or equal to the second amplitude threshold, the camera is controlled to enter a low-frequency mode. The low-frequency mode is when the camera takes pictures after the shearing mechanism cuts the molten glass every N times, where N is an integer greater than 1. When the average amplitude is less than or equal to the first amplitude threshold and greater than the second amplitude threshold, the shooting mode remains unchanged.

[0018] By adopting the above technical solution, the magnitude of the output speed adjustment directly reflects the stability of the production process. When adjustments are frequent and large, it indicates system instability, and the system automatically switches to high-frequency mode for close monitoring; when adjustments are gradual, it indicates system stability, and the system switches to low-frequency mode to conserve resources.

[0019] Optionally, after entering low-frequency mode, the following may also be included: The time point at which the laser ranging sensor detects the falling glass molten metal blocking the laser beam is obtained. Calculate the time from the emission of the shearing signal to the detection time when the laser is blocked; Determine whether the detection time is within a preset time range. If yes, maintain the low-frequency mode; otherwise, determine that the material output speed is abnormal and control the camera to enter the high-frequency mode.

[0020] By adopting the above technical solution, an additional layer of safety monitoring is added to the low-frequency mode. By calculating the time it takes for the droplet to fall from the shearing point to the position of the laser rangefinder, the discharge speed of the discharging mechanism can be indirectly determined. If the time is abnormal, it indicates that the weight of the droplet may have changed significantly, and the camera switches from low-frequency mode to high-frequency mode, thereby ensuring that abnormal situations can be detected and handled in a timely manner.

[0021] Optional, also includes: Record the initial distance when the camera is horizontal and not tilted, as well as the detection distance detected during use; The tilt angle is obtained based on the initial distance and the detection distance; Based on the tilt angle and the target weight of the droplet, the corresponding compensation value is queried in the pre-established error correction model; When adjusting the discharge speed of the discharge mechanism based on the comparison between the final weight and the preset target weight, the compensation value is introduced for correction.

[0022] By adopting the above technical solution, the problem of measurement accuracy being affected by camera tilt due to equipment vibration and thermal expansion and contraction is addressed. The tilt angle is monitored in real time by a laser rangefinder, and compensation values ​​are retrieved from a pre-stored error correction model for correction. This allows the system to avoid immediate shutdown for physical calibration when the camera tilts slightly, ensuring production continuity while maintaining detection accuracy.

[0023] Optionally, the method for constructing the error correction model includes: The camera was adjusted to different tilt angles, and at each angle, the experimental weight of droplets with different target weights was measured. Weigh the material droplet to obtain the experimental weight; The compensation value is obtained based on the difference between the experimental weight and the target weight; An error correction model is formed by using tilt angle, experimental weight, and compensation value.

[0024] By adopting the above technical solution, an error correction model containing tilt angle, experimental weight, and compensation value can be constructed. Based on this error correction model, the compensation value obtained by query can be used to correct the discharge speed adjustment of the discharge mechanism when the camera tilts, thus ensuring production continuity and detection accuracy.

[0025] In summary, this application includes at least one of the following beneficial technical effects: The first image of the material droplet is captured by a camera. The control and analysis device calculates the weight of the material droplet non-contactly based on the first image and the fluid volume prediction model. The discharge speed is automatically adjusted according to the deviation between the final weight and the preset target weight, thereby ensuring the uniformity of the discharge. By switching between high-frequency and low-frequency modes of the camera and using the auxiliary monitoring of the laser rangefinder, the system processing load is reduced while ensuring detection accuracy, and abnormal situations can be captured in a timely manner. By monitoring the camera's tilt angle and compensating for it using a laser rangefinder, the system can be adjusted without immediately stopping production when the camera tilts slightly, thus improving the continuity and stability of production. Attached Figure Description

[0026] Figure 1 This is a schematic diagram of the overall structure and installation of Embodiment 1 of this application; Figure 2 This is a flowchart of steps S1-S3 of Embodiment 1 of this application; Figure 3 This is a flowchart of steps S21-S24 of Embodiment 1 of this application; Figure 4 This is a schematic diagram of the overall structure and installation of Embodiment 2 of this application; Figure 5 This is a flowchart of steps S4-S7 of Embodiment 2 of this application; Figure 6 This is a flowchart of steps S8-S12 of Embodiment 2 of this application; Figure 7 This is a flowchart of steps S13-S16 of Embodiment 2 of this application.

[0027] Explanation of reference numerals in the attached diagram: 1. Discharge mechanism; 2. Shearing mechanism; 3. Camera; 4. Control and analysis device; 5. Laser rangefinder sensor. Detailed Implementation

[0028] The following is in conjunction with the appendix Figure 1-7 This application will be described in further detail. Example

[0029] This application discloses a glass material weight detection system and detection method.

[0030] Reference Figure 1 The glass material weight detection system includes a material discharge mechanism 1, a material shearing mechanism 2, a camera 3, and a control and analysis device 4.

[0031] The discharge mechanism 1 is used to extrude molten glass downwards. The discharge mechanism 1 includes a discharge head and a plunger assembly. When the molten glass enters the discharge head, it is extruded from the discharge orifice by the plunger assembly. The combination of the discharge head and the plunger assembly allows the molten glass to be stably extruded downwards, providing a foundation for subsequent droplet formation.

[0032] The shearing mechanism 2 is located to the side and below the discharge mechanism 1. The shearing mechanism 2 includes scissors and a drive component. The scissors typically consist of two sharp blades. The drive component can be a cylinder, which is connected to the scissors via a piston rod. When the piston of the cylinder moves, it drives the piston rod to extend and retract, thus opening and closing the scissors. Another optional drive component is a motor, which transmits power to the scissors through a transmission mechanism, such as gears or chains, to complete the shearing action. After a preset time has elapsed, the drive component activates, causing the scissors to close rapidly, cutting the molten glass into falling droplets.

[0033] Camera 3 is also located to the side and below the discharge mechanism 1. Camera 3 is used to capture the first image of the material droplet formation and send it to the control and analysis device 4. The control and analysis device 4 is communicatively connected to the discharge mechanism 1, the shearing mechanism 2, and the camera 3. By acquiring the first image of the material droplet in real time through camera 3, the control and analysis device 4 calculates the weight of the material droplet based on the first image, and then automatically adjusts the discharge speed of the discharge mechanism 1 according to the difference between the weight of the material droplet and the required weight of the material droplet.

[0034] Camera 3 has a built-in strobe light source, which is electrically connected to the control and analysis device 4. After the shearing mechanism 2 performs the shearing action, the control and analysis device 4 triggers the strobe light source to flash, and simultaneously triggers camera 3 to capture the material droplet to form the first image.

[0035] The stroboscopic light source can be a xenon flash lamp, which features high luminous intensity and a short flash time. Its luminous intensity is sufficient to illuminate the falling droplet in a very short time, while the short flash time avoids image blurring caused by droplet movement. When the control and analysis device 4 receives the signal from the shearing mechanism 2 to perform the shearing action, it immediately sends a trigger signal to the stroboscopic light source, causing it to emit a high-intensity flash instantaneously. Simultaneously, the control and analysis device 4 also sends an image acquisition signal to the camera 3. The camera 3 acquires an image the instant the stroboscopic light source illuminates the droplet, thus obtaining a clear image of the droplet. Besides xenon flash lamps, LED stroboscopic lights can also be used as the stroboscopic light source, offering advantages such as low energy consumption and long lifespan.

[0036] Reference Figure 2 , Figure 3 The method for testing the weight of glass frit includes the following steps: Step S1: Control the shearing mechanism 2 to cut the molten glass into droplets, and capture the first image of the falling droplets through the camera 3. Then execute step S2.

[0037] Step S2: Obtain the final weight of the droplet based on the first image. Then proceed to step S3.

[0038] The specific process of step S2 is as follows: Step S21: Preprocess and extract edges from the first image to identify the shape of the droplets. Then proceed to step S22.

[0039] Step S22: Extract the length, neck diameter, and head contour feature parameters of the droplet shape. Then proceed to step S23.

[0040] Step S23: Input the profile feature parameters into the pre-established fluid volume prediction model to calculate the volume of the droplet. Then execute step S24.

[0041] Step S24: Based on the preset density of the molten glass, convert the volume to obtain the final weight. Then proceed to step S3.

[0042] The method for establishing a fluid volume prediction model includes: acquiring sample images of the liquid column morphology formed during the outflow of molten glass at different discharge velocities and temperatures, and using the measured outflow volume at the corresponding time as training labels; extracting the length, neck diameter, and head contour features of the liquid column in the sample images using image processing algorithms; and establishing a mapping function relationship between the liquid column morphology features and the outflow volume based on the fluid dynamics continuity equation or the least squares fitting algorithm, thereby obtaining the fluid volume prediction model.

[0043] Step S3: Adjust the discharge speed of the discharge mechanism 1 based on the comparison between the final weight and the preset target weight.

[0044] If the final weight is greater than the preset target weight, the control and analysis device 4 will reduce the discharge speed of the discharge head; if the final weight is less than the preset target weight, the control and analysis device 4 will increase the discharge speed of the discharge head, thereby achieving precise control of the discharge amount.

[0045] The implementation principle of the glass drop weight detection system and method in this application is as follows: by organically combining the discharging mechanism 1, the shearing mechanism 2, the camera 3, and the control and analysis device 4, real-time detection of the glass drop weight and automatic adjustment of the discharging speed are achieved. The discharging mechanism 1 stably extrudes the molten glass, the shearing mechanism 2 cuts the formed droplets at the appropriate time, the camera 3 captures droplet images, and the control and analysis device 4 processes and analyzes the first image, adjusting the discharging speed according to the weight deviation. Compared with traditional glass drop production methods, this system reduces the problem of inconsistent droplet weight caused by factors such as ambient temperature and molten glass temperature, ensures the uniformity of the output, improves the quality stability of the final glass products, and reduces production costs. Example

[0046] This application discloses a glass material weight detection system and detection method.

[0047] Reference Figure 4The glass weight detection system of this embodiment differs from that of Embodiment 1 in that it also includes a laser rangefinder sensor 5. The laser rangefinder sensor 5 is fixedly mounted on the camera 3, and the control and analysis device 4 is electrically connected to the laser rangefinder sensor 5. The control and analysis device 4 detects the time points when the molten glass blocks the laser rangefinder sensor 5. Then, when the laser rangefinder sensor 5 is not blocked by the molten glass, the control and analysis device 4 detects the change in distance between the camera 3 and a preset reference object using the laser rangefinder sensor 5.

[0048] Reference Figure 5 To balance detection accuracy and processing speed, this embodiment adopts the following solution: Step S4: Calculate the average absolute value of the discharge speed adjustment range within the preset time window to obtain the average range. When the average range is greater than the first range threshold, proceed to step S5. When the average range is less than or equal to the second range threshold, proceed to step S6. When the average range is less than or equal to the first range threshold and greater than the second range threshold, proceed to step S7.

[0049] Step S5: Control camera 3 to enter high-frequency mode. In high-frequency mode, camera 3 takes one picture after each time the shearing mechanism 2 cuts the molten glass.

[0050] The average amplitude is greater than the first amplitude threshold, indicating a larger adjustment range for each adjustment of the discharge speed. When camera 3 enters high-frequency mode, the control and analysis device 4 adjusts the discharge speed more frequently, thus facilitating timely adjustment of the discharge speed.

[0051] Step S6: Control the camera 3 to enter low-frequency mode. In low-frequency mode, the camera 3 takes a picture once after the shearing mechanism 2 cuts the molten glass every N times, where N is an integer greater than 1.

[0052] The average amplitude is less than or equal to the second amplitude threshold, indicating that the adjustment range of the discharge speed is small each time. When camera 3 enters low-frequency mode, the frequency of the control and analysis device 4 in adjusting the discharge speed decreases, thereby achieving both uniform discharge and energy saving.

[0053] Step S7: Keep the shooting mode unchanged.

[0054] If the average amplitude is less than or equal to the first amplitude threshold and greater than the second amplitude threshold, the camera 3 will maintain its current shooting mode. This will reduce the occurrence of the camera 3 repeatedly switching between high-frequency and low-frequency modes when the average amplitude fluctuates around the first or second amplitude threshold.

[0055] Reference Figure 6 After camera 3 enters low-frequency mode, monitoring is performed using laser rangefinder 5, specifically including the following steps: Step S8: Obtain the time point at which the falling molten glass blocks the laser, detected by laser rangefinder 5. Then proceed to step S9.

[0056] As the molten glass is extruded from the discharge head, the bottom of the molten glass gradually descends. When the bottom of the molten glass blocks the laser rangefinder 5, the control and analysis device 4 records the current time point.

[0057] Step S9: Calculate the time from the emission of the shearing signal to the time when the laser is blocked and detected. Then proceed to step S10.

[0058] The control and analysis device 4 calculates the time between the cutting of the molten glass by the shearing mechanism 2 and the time point to form the detection time.

[0059] Step S10: Determine whether the detection time is within the preset time range; if yes, proceed to step S11; if no, proceed to step S12.

[0060] Step S11: Maintain low frequency mode.

[0061] Step S12: Determine that the material discharge speed is abnormal, and control camera 3 to enter high-frequency mode.

[0062] The discharge speed varies due to factors such as ambient temperature, molten glass temperature, and blockage of the discharge head. When the discharge speed decreases, the time required for the molten glass to block the laser rangefinder 5 increases; conversely, when the discharge speed increases, the time required decreases. The laser rangefinder 5 assists in determining whether the discharge speed is abnormal. When an abnormal discharge speed occurs, the camera 3 enters high-frequency mode, facilitating timely adjustment of the discharge speed and ensuring uniform discharge.

[0063] Reference Figure 7 Furthermore, in actual production, camera 3 may vibrate or tilt, thereby increasing the error in detecting the weight of the droplet. This embodiment solves this problem in the following way: Step S13: Record the initial distance of camera 3 when it is horizontal and not tilted, as well as the detection distance during use. Then proceed to step S14.

[0064] Step S14: Obtain the tilt angle based on the initial distance and the detection distance. Then proceed to step S15.

[0065] The tilt angle is calculated using the inverse cosine function, which is the ratio of the initial distance to the detection distance.

[0066] Step S15: Based on the real-time detected tilt angle and the target weight of the droplet, query the corresponding compensation value in the pre-established error correction model. Then execute step S16.

[0067] Methods for establishing error correction models include: Camera 3 is adjusted to different tilt angles, and at each angle, the experimental weight of droplets with different target weights is measured. A compensation value is obtained based on the difference between the experimental weight and the target weight. An error correction model is formed using the tilt angle, experimental weight, and compensation value.

[0068] Step S16: When adjusting the discharge speed of the discharge mechanism 1 based on the comparison between the final weight and the preset target weight, a compensation value is introduced for correction.

[0069] When adjusting the discharge speed, a compensation value is added to the calculated final weight to reduce measurement errors caused by tilting, eliminating the need for immediate machine stoppage to adjust the level of camera 3. Before adjusting the level of camera 3, the laser rangefinder 5 will also tilt with camera 3, but the tilt angle of camera 3 is small, allowing the laser rangefinder 5 to continue being used for detection steps S8-S12.

[0070] In another embodiment, an error correction model can also be used to compensate for the time range or detection time in steps S8-S12.

[0071] In another embodiment, a laser rangefinder 5 and a tilt sensor can be used. The tilt sensor is fixedly mounted on the top side of the camera 3 for performing steps S13-S16. Another laser rangefinder 5 is separately fixed from the camera 3 for performing steps S8-S12.

[0072] At the same time, staff can promptly determine whether camera 3 is tilted based on the obtained tilt angle, which facilitates the staff to reasonably arrange maintenance time to level the position of camera 3.

[0073] The implementation principle of the glass material weight detection system and detection method in this application embodiment is as follows: by intelligently switching the shooting mode of camera 3 and compensating for the tilt of camera 3, the detection process is further optimized, the system cost is reduced, the detection accuracy is guaranteed, and manual intervention is reduced, thus comprehensively improving and upgrading the existing glass drop production technology.

[0074] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A glass material weight detection system, characterized in that, include: The discharge mechanism (1) is used to extrude molten glass downwards; The shearing mechanism (2) is located on the side of the discharge mechanism (1) and is used to shear the glass molten material after a preset time of discharge to form falling droplets. A camera (3) is set on the side of the discharge mechanism (1) for capturing images of the material droplets; The control and analysis device (4) is connected to the discharge mechanism (1), the shearing mechanism (2) and the camera (3) respectively, and adjusts the size of the material droplets falling from the discharge mechanism (1) according to the first image.

2. The glass charge weight detection system according to claim 1, characterized in that: The camera (3) has a built-in strobe light source, which is electrically connected to the control and analysis device (4). After the shearing mechanism (2) performs the shearing action, the control and analysis device (4) triggers the strobe light source to flash and simultaneously triggers the camera (3) to acquire images.

3. The glass charge weight detection system according to claim 1, characterized in that: It also includes a laser rangefinder (5), which is located on one side of the discharge mechanism (1) and is used to detect the time point when the molten glass falls and blocks the laser.

4. The glass charge weight detection system according to claim 3, characterized in that: The laser rangefinder (5) is fixedly installed on the camera (3). The laser rangefinder (5) is used to detect the distance between the camera and the preset reference object. The control and analysis device (4) determines whether the camera (3) is tilted based on the distance change detected by the laser rangefinder (5).

5. A method for detecting the weight of glass charge, the glass charge detection system according to any one of claims 1-4, characterized in that, The method includes the following steps: The shearing mechanism (2) cuts the molten glass into droplets, and the camera (3) captures the falling droplets to obtain the first image. The final weight of the droplet is obtained from the first image; The discharge speed of the discharge mechanism (1) is adjusted according to the comparison between the final weight and the preset target weight.

6. The method for detecting the weight of glass charge according to claim 5, characterized in that, The specific process of obtaining the final weight of the droplet based on the first image includes: The first image is preprocessed and edge extracted to identify the shape of the material droplets; Extract the length, neck diameter, and droplet head contour feature parameters of the droplet shape; The contour feature parameters are input into a pre-established fluid volume prediction model to calculate the volume of the droplet; The volume of the droplet is converted into its final weight based on the preset density of the molten glass.

7. The method for detecting the weight of glass charge according to claim 5, characterized in that, Also includes: Calculate the average absolute value of the discharge speed adjustment range within the preset time window to obtain the average range; When the average amplitude is greater than the first amplitude threshold, the camera (3) is controlled to enter the high-frequency mode. The high-frequency mode is when the camera (3) takes pictures after the shearing mechanism (2) cuts the molten glass each time. When the average amplitude is less than or equal to the second amplitude threshold, the camera (3) is controlled to enter the low frequency mode. The low frequency mode is when the camera (3) takes pictures after the shearing mechanism (2) cuts the glass melt every N times, where N is an integer greater than 1. When the average amplitude is less than or equal to the first amplitude threshold and greater than the second amplitude threshold, the shooting mode remains unchanged.

8. The method for detecting the weight of glass charge according to claim 7, characterized in that, After entering low-frequency mode, it also includes: The laser ranging sensor (5) detects the time point at which the falling glass melt blocks the laser. Calculate the time from the emission of the shearing signal to the detection time when the laser is blocked; Determine whether the detection time is within the preset time range. If yes, maintain the low frequency mode; otherwise, determine that the material output speed is abnormal and control the camera (3) to enter the high frequency mode.

9. The method for detecting the weight of glass charge according to claim 5, characterized in that, Also includes: Record the initial distance of the camera (3) when it is horizontal and not tilted, and the detection distance detected during use; The tilt angle is obtained based on the initial distance and the detection distance; Based on the tilt angle and the target weight of the droplet, the corresponding compensation value is queried in the pre-established error correction model; When adjusting the discharge speed of the discharge mechanism (1) based on the comparison between the final weight and the preset target weight, the compensation value is introduced for correction.

10. The method for detecting the weight of glass charge according to claim 9, characterized in that, The method for constructing the error correction model includes: The camera (3) was adjusted to different tilt angles, and the experimental weight of the droplets with different target weights was measured at each angle. The compensation value is obtained based on the difference between the experimental weight and the target weight; An error correction model is formed by using tilt angle, experimental weight, and compensation value.

Images