Method and system for removing hot inclusions in rock wool production

Abstract

The application discloses a method and system for removing sundries in the production process of rock wool. The method for removing thermal inclusions in the production process of rock wool comprises the following steps: step 1, collecting a rock wool image on a conveying mechanism, the rock wool image comprising initial cotton fibers and sundries, and identifying the positions of the sundries; step 2, acquiring position data of the sundries through a system sensor; and step 3, removing the thermal inclusions from the initial cotton fibers through fluid injection. The application positions the thermal inclusions through the addition of a visual identification system, and removes the thermal inclusions through the control of corresponding nozzle systems, thereby effectively reducing the negative influence of thermal inclusions on equipment or products in the production process.

Description

Technical Field

[0001] This invention belongs to the technical field of rock wool production equipment, specifically relating to a method and system for removing thermal inclusions during the rock wool production process. Background Technology

[0002] The statements herein provide only background information in relation to this invention and do not necessarily constitute prior art.

[0003] The production process of rock wool involves melting raw materials such as basalt, slag, and dolomite to form a molten material. This molten material is then transported to a fiber-forming unit, where a binder is sprayed to cool the fibrous strands, causing them to fall evenly onto a collecting mesh. During this process, high-temperature molten particles or binder gel-like substances become embedded in the rock wool's mesh structure. Because the surrounding rock wool acts as insulation, these particles or binder gel-like substances gradually cool and remain within the fiber network structure. This can negatively impact the conveying equipment, subsequent processing, and the final product, such as causing equipment jams or saw blade malfunctions, and negatively affecting the mechanical or thermal conductivity properties of the final product.

[0004] Patent CN209866842U discloses a processing device for rock wool boards, equipped with multiple rock wool board debris adsorption devices. This device, through the cooperation of the rock wool board debris adsorption devices mounted on a stable base and an exhaust fan, can promptly adsorb the edge wool on the surface of the rock wool board. Furthermore, the cooperation of a drive motor and rolling wheels ensures more comprehensive adsorption during the movement of the rock wool board. With the cooperation of a filtration device and an edge wool treatment device, the edge wool on the rock wool board can be processed promptly. However, it cannot effectively remove thermal inclusions embedded in the rock wool mesh structure. Summary of the Invention

[0005] To address the shortcomings of existing technologies and ensure the safety and quality of rock wool production, this invention proposes a method and system for removing impurities during rock wool production. This invention utilizes a visual recognition system to locate thermal inclusions and controls a corresponding nozzle system to remove them, effectively reducing the negative impact of thermal inclusions on equipment or products during production.

[0006] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for removing thermal inclusions during the production of rock wool, comprising the following steps: Step 1: Acquire images of rock wool on the conveying mechanism. The rock wool images include initial cotton fibers and thermal inclusions. Identify the thermal inclusions. Step 2: Obtain the location data of thermal inclusions through a visual recognition system; Step 3: The conveying mechanism moves the initial cotton fibers containing thermal inclusions to the impurity removal station, where the thermal inclusions are removed from the initial cotton fibers by positive pressure jetting fluid or negative pressure adsorption.

[0007] In some embodiments of the present invention, in step 1, the initial cotton fiber is the cotton fiber before the pendulum is formed.

[0008] In step 1, the identification method is as follows: images are acquired by a thermal imager, then the images are processed to form an image database, and different types of thermal inclusions are identified.

[0009] In step 2, the data acquisition method is as follows: (1) Image coordinate positioning of monitoring equipment in the visual recognition system: In the thermal image, the area with abnormal temperature is identified by image processing algorithms (such as threshold segmentation, region growth, contour search, etc.). The system can calculate the center point coordinates (x, y) of this area in the image pixel coordinate system. (2) Real space positioning (calibration required): Calibration is performed according to the actual position of the thermal imager set on the side. The image coordinates (x, y) can be converted into three-dimensional coordinates (X, Y, Z) through geometric calculation.

[0010] In step 3, fluid, including air and water, is injected through a nozzle system.

[0011] Secondly, the present invention provides a system for removing thermal inclusions during the production of rock wool, including a rock wool conveying module for continuously conveying rock wool products, wherein the rock wool products are cotton fibers (primary felt) before pendulum molding. A visual recognition module is used to identify thermal inclusions on cotton fibers and determine their location. The debris removal module is used to remove thermal inclusions from cotton fibers using a positive pressure jet fluid or a negative pressure adsorption system.

[0012] The rock wool conveying module includes a first belt conveyor, a second belt conveyor, and a third belt conveyor located behind the cotton collecting machine; The visual recognition module includes a visual recognition system located above the initial cotton fibers at the second belt conveyor. The debris removal module includes a nozzle system; It also includes a control module, which comprises a computer system and is signal-connected to the rock wool conveying module, the visual recognition module, and the debris removal module, respectively.

[0013] The visual recognition system includes an industrial camera and an infrared thermal imager, which acquire the characteristics of thermal inclusions and transmit signals to a computer system.

[0014] The thermal inclusion features include size data, location data, and heat data.

[0015] The nozzle system includes at least one row of spray guns arranged laterally between the second and third belt conveyors and above the initial cotton fibers, providing high-pressure air or high-pressure water flow to remove hot inclusions from the initial cotton fibers. It also includes a hot inclusion removal device located below the nozzle system, which removes hot inclusions that have detached from the initial cotton fibers and fallen under the drive of the nozzle system. In thick initial cotton fibers, when using a nozzle system to spray out impurities, the impurities are blocked by the cotton fibers, requiring increased airflow or water pressure.

[0016] As another optional implementation, the nozzle system is positioned above the second conveyor belt device and behind the scanning range of the vision recognition system. It uses 1~2MPa air at 20~25℃ to blow the initial cotton fibers on the second conveyor belt device, cooling the initial cotton fibers along with any possible hot inclusions. At the same time, it causes any possible hot inclusions to move to the lower part of the initial cotton fibers.

[0017] Furthermore, an air extraction pipe is installed in the gap between the second and third conveyor belt devices, below the initial cotton fibers. The other end of the air extraction pipe is sequentially connected to an air pump and a cyclone separator. The air extraction pipe provides a negative pressure of 1~5MPa to draw the hot inclusions into the cyclone separator.

[0018] Preferably, a visual recognition system is also provided on the side of the initial cotton fiber between the first belt conveyor and the second belt conveyor to monitor the position of the thermal inclusions on the side of the initial cotton fiber, thereby obtaining the depth of the thermal inclusions in the initial cotton fiber.

[0019] As an alternative, an ultrasonic probe is installed above and below the initial cotton fiber between the first and second belt conveyors, with the line connecting the two sets of ultrasonic probes perpendicular to the horizontal line to avoid mutual interference. The two sets of ultrasonic probes are equidistant from the initial cotton fiber.

[0020] The beneficial effects achieved by one or more embodiments of the present invention described above are as follows: The method and apparatus provided by this invention ensure the stability of rock wool products produced by cupola furnaces.

[0021] This invention uses a visual recognition system fixed diagonally above the conveyor belt to identify the location of thermal inclusions. The system's sensors acquire position data, and the nozzle system is then used to remove the bulky thermal inclusions from the initial cotton fibers. The visual recognition system uses an infrared or thermal imager as the identification sensor, sequentially scanning the continuously conveyed rock wool fibers line by line to generate a thermal image of the fibers. A computer then generates corresponding nozzle system control data.

[0022] High-pressure nozzles use air or water to blow away thermal inclusions. The nozzle system pressure is adjusted based on the detected size, heat, and depth of the thermal inclusions to control the pressure and remove them from the initial cotton fibers. The nozzles are effective only in a limited space and for a short time. The initial cotton fibers treated by the nozzle system retain their basic shape and do not affect the final product. After treatment, the cotton fibers are folded and spread in the downstream pendulum unit to form a primary felt, free of thermal inclusions, which can be evenly pushed into the curing oven. Attached Figure Description

[0023] The accompanying drawings, which form part of this invention, are used to provide a further understanding of the invention. The illustrative embodiments of the invention and their descriptions are used to explain the invention and do not constitute an improper limitation of the invention.

[0024] Figure 1 This is a schematic diagram of the device structure for removing thermal inclusions during the rock wool production process provided in Embodiment 1 of the present invention.

[0025] Figure 2 This is a top view of the second conveyor roller and belt, and the debris conveyor belt and roller.

[0026] Figure 3 This is a schematic diagram of the device for removing thermal inclusions during the rock wool production process provided in Embodiment 2 of the present invention.

[0027] Figure 4 This is a schematic diagram of the method for removing thermal inclusions during the production of rock wool provided by the present invention.

[0028] Figure 5 This is a schematic diagram of the device for removing thermal inclusions during the rock wool production process provided in Embodiment 3 of the present invention.

[0029] The components are: 1-Cotton collector, 2-Initial cotton fibers, 3-First conveyor roller and belt, 4-Hot inclusions, 5-Vision recognition system, 6-Nozzle system, 7-Impact conveyor belt, 8-Belt idler, 9-Hot inclusion movement direction, 10-Second conveyor roller and belt, 11-Third conveyor roller and belt, 12-Cotton felt conveyor support, 13-Connector, 14-Independent support, 15-Conveyor roller and direct connector, 16-Upper conveyor roller, 17-Lower conveyor roller, 18-Belt idler connector, 19-Transmission line equipment, 20-Computer system, 21-Control nozzle circuit, 22-Ultrasonic probe, 23-Extraction pipe, 24-Extraction pump, 25-Cyclone separator, 26-Settling chamber. Detailed Implementation

[0030] It should be noted that the following detailed description is illustrative and intended to provide further explanation of the invention. Unless otherwise specified, all technical and scientific terms used in this invention have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains.

[0031] The present invention will be further described below with reference to the embodiments.

[0032] Example 1: Device for removing thermal inclusions during rock wool production like Figure 1 As shown, the system includes a first belt conveyor 3, a second belt conveyor 10, and a third belt conveyor 11 located behind the cotton collector 1. A vision recognition system 5, a nozzle system 6, a hot inclusion removal device 7 located below the nozzle system 6, and a computer system 20 are also present above the initial cotton fibers 2 at the second belt conveyor 10. The vision recognition system 5 is connected to the computer system 20 via a transmission line device 19, and the nozzle system 6 is connected to the computer system 20 via a control nozzle line 21. The first belt conveyor 3, the second belt conveyor 10, and the third belt conveyor 11 transmit the belt speed signal to the computer system 20.

[0033] The visual recognition system 5 includes an industrial camera and an infrared thermal imager. The transmission line device 19 has a Camera Link receiving port to transmit images acquired by the industrial camera and the infrared thermal imager to the computer system 20.

[0034] The location data of thermal inclusions are obtained through monitoring equipment in the visual recognition system: the equipment is an infrared imager with an infrared resolution of 640×480 and an effective field of view of 1.5 meters. The monitoring equipment is located above the second belt conveyor device 10 after the cotton collector, at a height of 2 meters. There are two devices, which can ensure that the monitoring covers the entire conveyor belt.

[0035] In another embodiment, a visual recognition system 5 is also provided on the side of the initial cotton fiber 2 between the first belt conveyor 3 and the second belt conveyor 10 to monitor the position of the thermal inclusions on the side of the initial cotton fiber 2, thereby obtaining the depth of the thermal inclusions in the initial cotton fiber 2.

[0036] The nozzle system 6 includes at least one row of spray guns arranged laterally above the initial cotton fiber 2, located between the second belt conveyor 10 and the third belt conveyor 11, to provide high-pressure air or high-pressure water flow to remove thermal inclusions in the initial cotton fiber 2.

[0037] In another embodiment, the spray guns are arranged in 2-3 rows.

[0038] like Figure 2As shown, the first belt conveyor 3, the second belt conveyor 10, and the third belt conveyor 11 are respectively mounted on the cotton felt conveying support 12 via their respective connectors 13. Below the cotton felt conveying support 12, between the second belt conveyor 10 and the third belt conveyor 11, an independent support 14 perpendicular to the cotton felt conveying support 12 is provided. An upper conveying roller 16 and a lower conveying roller 17 are provided on the independent support 14. The upper conveying roller 16 and the lower conveying roller 17 are connected to the independent support 14 via a conveying roller connector 15. The bottom also includes a belt idler 8 connected via a belt idler connector 18.

[0039] The cotton felt conveying support 12 is placed horizontally, and the independent support 14 is below the cotton felt conveying support 12, with an angle of about 35° between the support and the horizontal of the cotton felt conveying support 12. The top of the upper conveying roller 16 is about 40cm away from the cotton felt conveying support 12, and the top of the lower conveying roller 17 is lower than the upper conveying roller 16.

[0040] The initial cotton fibers are collected by the cotton collector 1 and sent to the pendulum process via the first, second, and third belt conveyor devices (3, 10, 11). If there are thermal inclusions 4 during the cotton fiber collection process, the visual recognition system 5 located above the second conveyor belt device 10 identifies the thermal inclusions 4. The visual recognition system 5 scans the continuously conveyed rock wool fibers line by line to generate a thermal image of the cotton fibers. The computer system 20 generates the corresponding volume, position, and heat information of the thermal inclusions 4 as control data for the nozzle system. Based on the volume, position, and heat information of the thermal inclusions 4, the computer system 20 sets the pressure of the corresponding nozzle system 6 to blow the cotton fibers containing thermal inclusions. The removed thermal inclusions are received by the conveyor belt 7 and collected and processed by the rotation of the upper conveyor roller 16, the lower conveyor roller 17, and the belt idler roller 8 at the bottom of the conveyor belt 7.

[0041] Example 2 The device structure provided in Example 1 is further modified as follows: Figure 3 As shown: First, a set of ultrasonic probes 22 are installed above and below the initial cotton fiber between the first belt conveyor 3 and the second belt conveyor 10. The line connecting the two sets of ultrasonic probes is perpendicular to the horizontal line to avoid mutual interference. The two sets of ultrasonic probes are equidistant from the initial cotton fiber and are respectively connected to the computer system 20.

[0042] Each ultrasonic probe set includes a pair of matched signal transmitting and receiving probes. One probe transmits ultrasonic waves, and the other receives them. The location of thermal inclusions is determined based on the Time of Flight Diffraction (TOFD) method. When ultrasonic waves encounter defects such as inclusions, they emit weak diffraction signals from the defect tip in all directions. TOFD works by precisely capturing and analyzing these diffraction signals. Using two sets of ultrasonic probes, the diffraction signals at the upper and lower tips of the thermal inclusion are measured respectively. By calculating the time difference between the two measurements, it is determined whether the thermal inclusion is located in the upper or lower half of the initial cotton fiber.

[0043] When the time difference between the upper and lower ultrasonic probes is positive, it is determined that the thermal inclusions are located in the lower half of the initial cotton fiber. At this time, the computer system 20 controls the airflow pressure of the nozzle system 6 to 1.5-2 MPa to eject the thermal inclusions. When the time difference between the upper and lower ultrasonic probes is negative, it is determined that the thermal inclusions are located in the upper half of the initial cotton fiber. At this time, the computer system 20 adjusts the airflow pressure of the nozzle system 6 to 3-4 MPa to eject the thermal inclusions.

[0044] Example 3 The device structure provided in Example 1 is further modified as follows: Figure 5 As shown: The nozzle system 6 is positioned above the second conveyor belt device 10 and behind the scanning range of the vision recognition system 5. It uses 1~2MPa air at 20~25℃ to blow the initial cotton fibers on the second conveyor belt device 10, cooling the initial cotton fibers along with any possible hot inclusions. This can temporarily reduce the temperature of the hot inclusions to below 150℃.

[0045] An air extraction pipe 23 is installed in the space between the second conveyor belt device 10 and the third conveyor belt device 11, below the initial cotton fiber. The upper end of the air extraction pipe 23 is positioned 2-5 cm away from the lower surface of the initial cotton fiber. The other end of the air extraction pipe 23 is sequentially connected to an air pump 24, a cyclone separator 25, and a settling chamber 26. The air extraction pipe 23 provides a negative pressure of 1-5 MPa to draw the hot impurities after the air has been cooled to the cyclone separator 25. The air pump 24 is signal-connected to a computer system 20 and is started and stopped under the control of the latter.

[0046] The advantage of this structural design is that the initial cotton fibers, along with any possible thermal inclusions, are cooled by air on the second conveyor belt device 10. At the same time, any possible thermal inclusions are moved to the lower part of the initial cotton fibers, and the negative pressure adsorption system can remove the thermal inclusions with a small adsorption force.

[0047] Example 4 The method for removing thermal inclusions during rock wool production includes the following steps: Step 1: Acquire images of rock wool on the conveying mechanism. The rock wool images include initial cotton fibers and debris. Identify the location of the debris. Step 2: Acquire the location data of the debris using the system's sensors; Step 3: Remove thermal inclusions from the initial cotton fibers by jetting fluid.

[0048] In step 1, the initial cotton fiber is the cotton fiber (primary felt) before the pendulum is formed. This primary felt is relatively thin and contains a certain amount of moisture. Subsequently, the pendulum unit folds and spreads the primary felt, then applies pressure and cures it. During the curing stage, it is concentratedly evaporated and cured.

[0049] In some embodiments of the present invention, a visual recognition device is used to collect images of rock wool on the conveyor mechanism and identify the location of debris. The visual recognition system is fixed at a position diagonally above the conveyor belt.

[0050] In step 1, the identification method is as follows: images are acquired by a thermal imager, then the images are processed to form an image database, and different types of thermal inclusions are identified.

[0051] The thermal imager has a monitoring range of 0-600℃ and can collect and monitor large-sized foreign objects (thermal inclusions) that are above room temperature. The thermal imager identifies and monitors thermal inclusions with an accuracy rate of over 90%.

[0052] In step 2, the location identification methods are as follows: (1) Image coordinate positioning of monitoring equipment in the visual recognition system: Within the thermal image, the area with abnormal temperature is identified by image processing algorithms (such as threshold segmentation, region growing, contour finding, etc.). The system can calculate the center point coordinates (x, y) of this area in the image pixel coordinate system. (2) Real space positioning (calibration required): Calibration is performed based on the actual position of the thermal imager. The image coordinates (x, y) can be converted into real-world three-dimensional coordinates (X, Y, Z) through geometric calculation.

[0053] As an alternative, in the real-space positioning of step 2, an ultrasonic probe is used to detect the depth of thermal inclusions in the initial cotton fibers, and the image coordinates (x, y) are converted into real-world three-dimensional coordinates (X, Y, Z).

[0054] Meanwhile, after thermal inclusions are detected at the thermal imaging station, the computer system, based on the current location and the belt speed, infers the time it will take for the thermal inclusions to reach the cleaning station, and immediately controls the nozzle system to start for blowing treatment.

[0055] Image processing stage flow as follows Figure 4 As shown: (1) Image acquisition, preprocessing and enhancement: Due to the inherent characteristics of infrared detectors and environmental interference, the original image must be processed by field-scale rescaling, wavelet domain loss function and median / morphological filtering to improve image quality and prepare for subsequent analysis.

[0056] The specific process is as follows: First, acquire the original infrared image. Then, scale the image to a suitable scale (e.g., map temperature data to a 0-255 grayscale range) according to subsequent processing requirements. If necessary, use methods such as CLAHE to enhance local contrast. Use a 3×3 or 5×5 median filter window to quickly remove isolated impulse noise and salt-and-pepper noise. Select a structuring element larger than the target size and perform morphological opening to obtain background estimation. Subtract the background from the original image to obtain the target saliency map (morphological background suppression). Perform wavelet decomposition on the morphological background suppression result. Apply thresholding with wavelet domain loss function constraints or a deep learning model to the high-frequency subband to specifically remove complex non-uniform noise such as stripe noise. Finally, through inverse wavelet transform and necessary post-processing, output a high-quality, high signal-to-noise ratio infrared image for subsequent target detection, segmentation, and other tasks.

[0057] (2) Image processing: Traditional image processing methods or deep learning methods can be used to determine the target location. Traditional image processing methods include image segmentation, which divides the image into target and background. Methods include: a combination of Canny and Otsu algorithms, adaptive thresholding, and deep learning segmentation networks. As a typical example, a coarse target region is obtained through Otsu global thresholding, and a fine edge is obtained through Canny edge detection. Then, a fused segmentation result is obtained through logical operations (AND / OR). Subsequently, the target region features are compared with the template feature library to calculate the similarity and determine the best match. The location of the target in the image is obtained through feature matching.

[0058] Image processing, including image segmentation, can also be performed using deep learning methods. The segmentation results are used as region priors input to the detection network, thereby reducing the search space of the detection network and providing attention guidance, thus improving detection accuracy. Real-time detection is performed using RetinaNet or other network types to obtain bounding boxes and target category labels. Infrared imaging is achieved by combining temperature information fusion, thermal feature encoding, and other specially designed network techniques, outputting the target location and category after special design optimization.

[0059] (3) Detection and recognition, location and recognition of specific targets in the image: convert the coordinates detected in step (2) into physical world coordinates, and finally obtain the position coordinates and contour size of the thermal inclusions.

[0060] In step 3, fluid is injected through a nozzle system. The nozzle system can use either water or air as long as the pressure is sufficient to remove thermal inclusions, with air jetting being preferred. In some embodiments, the nozzle pressure is adjusted to a range of 1-5 MPa.

[0061] The nozzle system pressure is adjusted according to the size of the detected thermal inclusions. Based on experiments and experience, the system is set as follows: for blocky objects with a size of less than φ30mm, the pressure is set to 1-2MPa; for objects with a size of more than φ30mm, the pressure is set to 3-4MPa. If the thermal inclusions are located in the upper half of the cotton, set the pressure adjustment to 1.5-2MPa; if the thermal inclusions are located in the lower half of the cotton, set the pressure adjustment to 3-4MPa. The thermal data of thermal inclusions mainly comes from the detection of thermal imaging equipment. The higher the temperature, the better the water spray treatment. The system is set to a temperature range of 25-150℃, in which air spray treatment is preferred, and the temperature is greater than 150℃, in which case water spray treatment is preferred.

[0062] If any one of the above three conditions is met—that is, the size is φ30mm or larger, or located in the lower half—the pressure is set to 3-4MPa.

[0063] After the blown felt treatment, the felt may have slight depressions or voids, which can be adjusted in the subsequent pleating process. After the water-blown treatment, the subsequent curing process can adjust the curing temperature of the cotton fibers and improve the moisture evaporation effect to offset the above effects.

[0064] In the removal method of the negative pressure adsorption system provided in Example 3, step 3 includes the following process: the nozzle system 6 blows air onto the initial cotton fiber to drive the thermal inclusions to the bottom layer of the initial cotton fiber. When the initial cotton fiber containing thermal inclusions moves to the position of the negative pressure adsorption system, the computer system controls the air pump of the negative pressure adsorption system to operate and remove the thermal inclusions located at the bottom layer of the initial cotton fiber.

[0065] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A method for removing thermal inclusions during rock wool production, characterized in that, Includes the following steps: Step 1: Acquire images of rock wool on the conveying mechanism. The rock wool images include initial cotton fibers and impurities. Identify the location of thermal inclusions. Step 2: Obtain the location data of thermal inclusions through a visual recognition system; Step 3: Remove thermal inclusions from the initial cotton fibers by positive pressure jetting or negative pressure adsorption.

2. The method for removing thermal inclusions during rock wool production according to claim 1, characterized in that, In step 1, the identification method is as follows: images are acquired by a thermal imager, then the images are processed to form an image database, and different types of thermal inclusions are identified.

3. The method for removing thermal inclusions during rock wool production according to claim 1, characterized in that, In step 2, the identification data acquisition method includes: (1) image coordinate positioning of monitoring equipment in the visual recognition system: in the thermal image, the area with abnormal temperature is identified by the image processing algorithm, and the system calculates the center point coordinates (x, y) of this area, that is, the position in the image pixel coordinate system; (2) real space positioning: convert the image coordinates (x, y) into three-dimensional coordinates (X,Y, Z).

4. The method for removing thermal inclusions during rock wool production according to claim 1, characterized in that, In step 3, fluid is injected through a nozzle system, the fluid being selected from air or water.

5. A system for removing thermal inclusions during rock wool production, characterized in that, Includes a rock wool conveying module for continuously conveying rock wool products, wherein the rock wool products are cotton fibers before the pendulum is formed; A visual recognition module is used to identify thermal inclusions on cotton fibers and determine their location. The debris removal module is used to remove thermal inclusions from cotton fibers using a positive pressure jet fluid or a negative pressure adsorption system. The rock wool conveying module includes a first belt conveyor, a second belt conveyor, and a third belt conveyor located behind the cotton collecting machine; The visual recognition module includes a visual recognition system located above the initial cotton fibers at the second belt conveyor. The debris removal module includes a nozzle system; It also includes a control module, which comprises a computer system that is signal-connected to the rock wool conveying module, the vision recognition module, and the debris removal module, and controls the debris removal module to remove thermal inclusions.

6. The system for removing thermal inclusions during rock wool production according to claim 5, characterized in that, The visual recognition system includes an industrial camera and an infrared thermal imager, which acquire the characteristics of thermal inclusions and transmit signals to a computer system.

7. The system for removing thermal inclusions during rock wool production according to claim 6, characterized in that, The thermal inclusion features include size data, location data, and heat data.

8. The system for removing thermal inclusions during rock wool production according to claim 6, characterized in that, The nozzle system includes at least one row of spray guns arranged laterally above the second and third belt conveyors and above the initial cotton fibers, providing high-pressure air or high-pressure water flow to spray away thermal inclusions in the initial cotton fibers.

9. The system for removing thermal inclusions during rock wool production according to claim 6, characterized in that, Between the first belt conveyor and the second belt conveyor, a set of ultrasonic probes is set above and below the initial cotton fiber, and the line connecting the two sets of ultrasonic probes is perpendicular to the horizontal line.

10. The system for removing thermal inclusions during rock wool production according to claim 6, characterized in that, The nozzle system is located above the second belt conveyor and behind the scanning range of the vision recognition system, and uses 1~2MPa air at 20~25℃ to blow the initial cotton fibers on the second belt conveyor. An air extraction pipe is installed in the gap between the second and third conveyor belt devices and below the initial cotton fibers. The other end of the air extraction pipe is connected in sequence to an air pump and a cyclone separator. The air extraction pipe provides a negative pressure of 1~5MPa to draw the hot inclusions into the cyclone separator.

Images