KR20240028966A - Detection processing method and experimental system for the occurrence of agglomeration of fluidized bed fibril-like particles - Google Patents

Abstract

The present invention discloses a processing method for detecting agglomeration generated by a group of fluid-bed fibrillar particles, and accurately performs three-dimensional recognition of aggregation by adopting the pixel area method. It is used to capture the three-dimensional overlap in the dense phase region of soft fibrillar particles during the fluidized bed fibrillar phase experiment. The experimental device adopts three flow guide plates with angles of 75 to 80°, 60 to 65°, and 50 to 55°, respectively, to homogenize the particles. The image discrimination method mainly includes step 1 of synchronously photographing a fluidized bed shell frozen state experiment; Step 2 of loading image information onto the PC; Step 3 to delete background; Step 4 performing grayscale processing; Step 5, obtaining a binarization threshold; Step 6, performing first color inversion; Step 7 to find connecting domains, perform area calculation and staining; Step 8 to delete small areas; and step 9 of acquiring the target cohesive image and area. The system and method can analyze the entire process of changes in the formation, area size, and winding of agglomerates within the riser area of cut sheath agglomerates during the experimental process.

Description

Detection processing method and experimental system for the occurrence of agglomeration of fluidized bed fibril-like particles

The present invention relates to the field of cut filler drying process technology, and more specifically, to a detection processing method and experimental system for agglomeration generated in a group of fluid-bed fibrillar particles.

Fluidized bed dryer is widely used in many industrial processing processes such as pharmaceutical, agriculture, chemical and food engineering, etc. due to its high gas-solid contact efficiency and easy collection, etc. Its working principle is powder type, paste type, suspension and A fluid substance, such as a solution, is placed on an airflow distribution plate, such as a perforated plate, and is transferred from its bottom to a drying medium at a considerable speed. When the medium flow rate is relatively low, the gas flows between the material particles, and the entire material layer does not move. As the airflow speed is gradually increased, the layer of material begins to expand, increasing the spacing between particles. If the airflow speed is increased again, a significant portion of the material becomes suspended, forming a gas-solid mixed layer, or fluidized layer. Because the suspended material in the fluidized bed is very similar to a boiling liquid, it is also called a boiling layer. It also exhibits the properties of a fluid in many aspects, with the upper interface being clear and level. When the airflow speed is increased again, almost all particles are removed by the airflow, converted to gas, and transported.

The types of particle materials used are diverse, the particle sizes are large, and the physical properties and shapes of the particles are different. For example, in a fluidized bed dryer, grass, tobacco leaves and seaweed are cut into small pieces and dried. These special particles are soft and flexible in the form of long fibers and are commonly called soft filamentous particles. Soft filamentous particles are widely applied in industrial fluidization drying processes. The heterogeneity of the special particle flow has a significant impact on the hydrodynamic properties and equipment drying performance. This type of particles is easy to stick together and difficult to separate due to the heterogeneity of flexibility, filiform shape and moisture content, and then they can be easily associated with higher-order structures with slow quantitative changes (e.g. increasing particle quantity), and finally forming so-called agglomerates. In this process, once the agglomerates are formed, they do not easily disappear, and the formation and shape changes of the agglomerates are closely related to the kinetic state of the agglomerates. This has a significant impact on the hydrodynamic properties and drying performance of the equipment. For example, during the fluidization drying process of cut filler, due to the appearance of agglomeration, wet lumps are formed in the cut filler, which seriously affects the consistency of product quality. Currently, there are many studies on particle agglomeration, but the purpose of this is to study the causes of particle aggregation and the specific impact of the appearance of agglomeration on the course of fluidized bed experiments of fibrillar particles, and to make the formation of agglomerates possible through large-scale research. The goal is to find corresponding measures to prevent it. However, there is no published method for capturing the three-dimensional structure of particle condensation, and there is a lack of convenient processing methods when processing experimental images, causing some difficulty in data processing during experimental research.

In view of the above-mentioned problems, the purpose of the present invention is to provide a processing method for detecting agglomeration generated in a fluidized bed fibril-like particle group, thereby making up for the gap in the image method recognition of fibril-like particle aggregation among conventional image processing techniques. there is. Additionally, a detection and processing system thereof is provided.

The processing method for detecting agglomeration generated in a fluid bed fibrillar particle group includes the following steps.

Step 1: Carry out a fluidized bed shear freezing state experiment on the experimental system. Fibril particles are uniformly blown into the riser through the flow guide plate, and high-speed camera 1 and high-speed camera 2 are used to capture the YZ coordinate plane and XZ coordinate of the riser, respectively. The plane is photographed and the flow state of each second is recorded while supplementing light in the YZ coordinate plane and XZ coordinate plane.

Step 2: Load the image signals acquired from high-speed camera 1 and high-speed camera 2 into the PC, and the PC system uses the signal identification code for each frame in the large image database to identify the YZ images acquired by shooting from high-speed camera 1 within the same signal time. The image on the coordinate plane and the image on the

Step 3: Subtract the background image through morphological and subtraction operations on the 3D image obtained through simulation.

Step 4: Perform grayscale processing on the image with the subtracted background.

Step 5: Obtain the binarization threshold using an algorithm that automatically acquires the threshold, and perform binarization on the grayscale processed image.

Step 6: Perform first-order color inversion on the binary image.

Step 7: Find the connection domain, calculate the area and stain, use the pixel area method to calculate the pixel values of particles and aggregates at different thresholds to obtain the area of particles and aggregates in the image, and distinguish between particle and aggregate staining. do. where pixel values from 0 to 1500 are stained with color 1 and denoted as small particles, pixel values from 1500 to 3000 are stained with color 2 and denoted as large particles, and pixel values above 3000 are stained with color 3 and denoted as target aggregation. do.

Step 8: After staining, the portion smaller than a predetermined threshold pixel area, that is, the single particle portion, of the image of the particle in the YZ coordinate plane or XZ coordinate plane is deleted.

Step 9: Secondary color inversion is performed on the image from which the portion smaller than the threshold area is deleted, and finally the fibrillar particle aggregation image and aggregation area data are obtained, and the calculation is completed.

Furthermore, both high-speed camera 1 and high-speed camera 2 undergo parameter initialization processing before shooting, the initial transmission signals of high-speed camera 1 and high-speed camera 2 are all synchronized, and the shooting area of high-speed camera 1 and high-speed camera 2 is that of the experimental system. It is a dense area.

Furthermore, in step 8, the threshold area is set to 3000 to 3500 pixels, and if the particle area S in the YZ coordinate plane and the XZ coordinate plane are both greater than the set threshold, the coordinate plane with the larger recognition area is selected to perform the calculation. Perform. Most preferably, the cut sheath dimensions used in the experiments are between 7.6 mm x 0.9 mm and 15 mm x 0.9 mm.

Furthermore, in step 4 the grayscale value of the pixel point on the image is set to 0 or 255.

The experimental system using the above fluidized bed fibril-like particle cluster generation detection and processing method includes a fluidized bed, a surface light source lamp 1, a surface light source lamp 2, an auxiliary light lamp 1, an auxiliary light lamp 2, a high-speed camera 1, and a high-speed camera 2. , hub, and PC stage are further included, and surface light source light 1 and surface light source light 2 are installed on the XZ coordinate plane and YZ coordinate plane of the riser of the fluidized bed, respectively, and the It is installed on one side of the vertical pipe, and auxiliary light 1 is installed on one side of this, and the YZ coordinate plane facing the fluidized vertical pipe of high-speed camera 2 is installed on one side of the fluidized vertical pipe, and auxiliary light 2 is installed on one side of this, High-speed camera 1 and high-speed camera 2 are each connected to the PC terminal through a hub.

Furthermore, the bottom of the riser is connected to a flow guide pipe, and within the flow guide pipe are three arc-shaped low, middle, and high plates installed at an angle from bottom to top, and within the low plate and the flow guide pipe. The spacing between the bottom walls is 5 to 7 cm, the spacing between the middle and low plates is 12 to 14 cm, and the spacing between the high and middle plates is 7 to 9 cm. Most preferably, the arcuate angles of the low plate, middle plate, and high plate are 75 to 80°, 60 to 65°, and 50 to 55°, respectively.

Most preferably, the opposite sides of the low plate, middle plate, and high plate each engage with the two inner walls of the flow guide tube, and the engagement gap is 0.5 to 1.5 mm.

Furthermore, the fluidized bed further includes a fan, a storage bin, a drawing plate, a nut, an unloader, an unloading bin, a cyclone separator, a transport vessel, a reflux pipe, control valve 1 and control valve 2, and the other end of the flow guide pipe is ventilated. It is connected to the fan through a pipe, the top of the riser communicates with the cyclone separator through a bent pipe, the bottom of the cyclone separator communicates with the transport container, the top of the cyclone separator communicates with the side wall of the ventilation pipe through a reflux pipe, and the reflux pipe Control valve 1 and control valve 2 are installed separately from each other, the storage bin is installed above the unloading bin, the two are connected by a nut, a drawing plate is installed at the connection point, and the unloader is installed at the top of the unloading bin. It is mounted, and the bottom of the unloading bin communicates with the upper wall of the flow guide tube.

Compared with the prior art, the advantages of the present invention are as follows.

This method uniformly blows fibrillar particles into the riser through the action of a removable flow guide plate, and the arc of the three flow guide plates and the distance between the three significantly improve particle uniformity. Image collection and calculation aspect: After supplementing the light by using surface light source light 1, surface light source light 2, auxiliary light 1, and auxiliary light 2 at the cross-sectional point of the riser, the shaded area can be optimized, high-speed camera 1, high-speed camera 2 Images of the fluidized bed experiment process are captured and acquired for each YZ coordinate plane and XZ coordinate plane using The image in the YZ coordinate plane captured and acquired by high-speed camera 1 within the visual field and the image on the The following steps are performed on the three-dimensional image: background deletion, grayscale processing, automatic acquisition of threshold algorithm binarization, first-order color inversion, finding connection domains, deletion of parts smaller than the threshold area, and second-order color inversion to obtain the experimental image. Process. According to this method, running processing is performed on the experimental images, and finally, images of the entire process of fibrillar particles forming agglomerates in the fluidized bed riser area and target aggregation area data are captured during the experimental process. Since this process can be performed in as little as a few minutes, it is very convenient as it greatly shortens the experimental data processing cycle. The device also uses interlocking flow guide plates to improve particle uniformity and make subsequent replacement or separation more flexible. Here, the aggregation measurement principle is to adopt the pixel area method. Compared to gridded methods, the pixel area method more accurately measures the dimensions of agglomeration. In addition, the present invention is applied not only to drying cut fillers, but also to the study of flow characteristics in a fluidized bed of mostly fibrillar particles. In order to reduce the formation of agglomerates during the flow process, such as drying of fibrillar particles in a fluidized bed, it provides a very effective agglomeration capture method, and further contributes to some extent to experiments in fluidized beds of fibrillar particles such as cut filler.

Figure 1 is a schematic diagram of the experimental system of the present invention.
Figure 2 is a structural diagram of a flow guide plate.
Figure 3 is a flow chart of the method according to the invention.
Figure 4 is a schematic diagram of an embodiment.

Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings and specific examples. These examples are for illustrating the present invention and do not limit the scope of the present invention.

A method for detecting agglomeration of a fluidized bed fibril-like particle group includes the following steps, as shown in FIG. 3.

Step 1: Experiment on the frozen state of the fluidized bed cut sheath on the experimental system, the filamentous particles are uniformly blown into the riser through the flow guide plate, and high-speed camera 1 (11) and high-speed camera 2 (12) are used to check the flow state of the cut sheath in the riser area. ) to capture the YZ coordinate plane and the XZ coordinate plane. Before shooting, on the system, initialize the parameters of high-speed camera 1 (11) and high-speed camera 2 (12) and collect signals at the same time to ensure synchronization of signal collection, and surface light source light 1 (8) and surface light source light 2 (7) And auxiliary light 1 (10) and auxiliary light 2 (13) are used to partially supplement the light to the riser.

Step 2: The image signal is loaded into the PC stage (15) through the hub (14), and the system combines the image in the YZ coordinate plane obtained by shooting from high-speed camera 1 (11) within the same signal time in a large image database and high-speed The images on the

Step 3: Subtract the background image through morphological and subtraction operations on the 3D image obtained through simulation.

Step 4: Perform grayscale processing on the subtracted background image to improve computation speed.

Step 5: Obtain the binarization threshold using an algorithm that automatically acquires the threshold, perform binarization on the image, and binarization can convert the grayscale image into a binary image.

Step 6: Perform first-order color inversion on the binary image, and first-order color inversion can indirectly calculate the area of fibril particle aggregation.

Step 7: Find the connecting domain, calculate the area, and stain. According to the pixel number area of the connection domain, the image of the fibril-like particle dense phase area is divided into three types: small particles, large particles, and target aggregation connection domain, and color classification of three different colors is performed. The dyeing logic is as follows. That is, pixel values from 0 to 1500 are dyed with color 1, for example red, and displayed as small particles. Pixel values 1500 to 3000 are dyed color 2, for example yellow, and displayed as large particles. Pixel values above 3000 are stained with color 3, e.g. blue, and marked as target aggregation.

Step 8: Delete the part smaller than the threshold area. After staining, the portion of the particle image in the YZ coordinate plane or the When the number is greater than or equal to 1, the surface with the largest recognition area is selected for calculation).

Step 9: Perform secondary color inversion. A secondary color inversion is performed on the image in which a small area is deleted (Figure 4(d)), that is, the remaining image is the agglomerated portion, and the final image of the fibrillar particle aggregation as shown in FIG. 4(e) is obtained. Acquire and terminate the operation.

The image change process of the entire process is as shown in Figures 4(a) to (e).

The experimental system using the above fluidized bed fibril-like particle cluster generation detection processing method, as shown in Figures 1 and 2, includes a fluidized bed, surface light source lamp 1 (8), surface light source lamp 2 (7) , auxiliary light 1 (10), auxiliary light 2 (13), high-speed camera 1 (11), high-speed camera 2 (12), hub (14), and PC stage (15). Surface light source light 1 (8) and surface light source light 2 (7) are installed on the XZ coordinate plane and YZ coordinate plane of the riser of the fluidized bed, respectively, and the It is installed on one side of the straight pipe, and an auxiliary light 1 (10) is installed on one side of this, and the YZ coordinate plane facing the fluidized vertical pipe of high-speed camera 2 (12) is installed on one side of the fluidized vertical pipe, and an auxiliary light is installed on one side of this. 2 (13) is installed, and the signals of high-speed camera 1 (11) and high-speed camera 2 (12) are each connected to the PC terminal (15) through the hub (14).

The bottom of the riser is connected to a flow guide pipe (9), and within the flow guide pipe (9) there are three arc-shaped low plate (91), middle plate (92), and high plate ( 93) are included, and all three of these are flow guide plates, and the gap between the lower plate 91 and the bottom wall in the flow guide pipe 9 is 5 to 7 cm, and the gap between the middle plate 92 and the lower plate 91 is 12 to 14 cm, and the distance between the upper plate 93 and the middle plate 92 is 7 to 9 cm. The arc degrees of the low plate 91, middle plate 92, and high plate 93 are 75 to 80°, 60 to 65°, and 50 to 55°, respectively.

The fluidized bed consists of a fan (1), storage bin (2), drawing plate (3), nut (4), unloader (5), unloading bin (6), cyclone separator (16), transport vessel (17), and reflux. It further includes a pipe 18, a control valve 1 (19), and a control valve 2 (20), and the other end of the flow guide pipe (9) is connected to the fan (1) through a ventilation pipe, and the upper end of the riser is connected to the curved pipe. communicates with the cyclone separator 16 through the cyclone separator 16, the bottom of the cyclone separator 16 communicates with the transport container 17, and the top of the cyclone separator 16 communicates with the side wall of the ventilation pipe through the reflux pipe 18, and reflux Control valve 1 (19) and control valve 2 (20) are installed spaced apart on the pipe (18), and the storage bin (2) is installed above the unloading bin (6), and these two are connected by a nut (4). A drawing plate (3) is installed at the connection point, the unloader (5) is mounted at the upper part of the unloading bin (6), and the bottom of the unloading bin (6) is attached to the upper wall of the flow guide pipe (9). communicates with

The nut (4) is used for the connection between the storage bin outlet and the fluidized bed supply port and facilitates attachment and detachment of the unloader and storage bin. By using three different angles of flow guide plates to blow particles uniformly into the riser, the uniformity of particles being blown into the riser after passing through the flow guide plates can be enhanced. The connection method between the flow guide plate and the wall is engagement, the gap size is 0.5 to 1.5 mm, and the engagement method can facilitate subsequent replacement or disassembly.

The following preparations must be made in advance before conducting the experiment. That is, auxiliary light 1 (10) and auxiliary light 2 (13) are placed right behind high-speed camera 1 (11) and high-speed camera 2 (12), respectively, and supplement light in the riser YZ coordinate plane and XZ coordinate plane, respectively, and image shading In order to improve the effect of the part on the computation of the image, light is supplemented using surface light source lamp 1 (8) and surface light source lamp 2 (7) in the YZ coordinate plane and XZ coordinate plane, respectively. Before shooting, the system must perform parameter initialization settings on two high-speed cameras and simultaneously collect signals to ensure synchronization of signal collection.

Put the cut candle into the storage bin (2), start the fan (1) and control the wind speed to 5.3 m/s by adjusting the frequency. The wind speed can completely blow the sheath up to the threshold of the riser. Through the experimental system, the unloader 5 controls the material flow rate by controlling the motor and controls the material discharge from the drawing plate 3 to perform the experiment. Fibril-like particles are uniformly blown into the riser through a flow guide plate, and images of the entire process of particles forming agglomerates in the riser YZ coordinate plane and XZ coordinate plane are taken through high-speed camera 1 and high-speed camera 2, respectively, to capture fibril particles. It is used to analyze the agglomeration distribution situation. After the cut grass flows into the cyclone separator (16), it falls into the transport container (7) and is collected. The reflux pipe 18 controls the outlet point pressure through control valve 1 (19) and control valve 2 (20), respectively. During the fluidized bed experiment, sheath agglomerates generally appear in the dense phase region, so selecting and calculating images of the dense phase region is advantageous for capturing the sheath agglomerates. The image signals collected by the two sets of cameras are transmitted to the PC stage 15 through the hub 14, and the system uses images in the YZ coordinate plane obtained by shooting from high-speed camera 1 within the same signal time in a large image database and high-speed The images on the

Specifically, the background image subtraction subtracts the background image from the loaded image through morphological and subtraction operations. In other words, it distinguishes soft filamentous particle clusters from the background. Morphological operation is an image processing method developed based on mathematical morphological set theory methods for binary images. Typically, morphological image processing comes in the form of adjacency operations, with some specially defined adjacencies called "structuring elements". On each pixel location it performs a specific logical operation on the corresponding region of the binary image, and the result of the logical operation is the corresponding pixel in the output image. The effectiveness of morphological operations depends on the size and content of the structural elements and the nature of the logical operations.

Specifically, the grayscale processing can improve computation speed for image grayscaling. In other words, the RGB values of each pixel point can be integrated into the same value. The grayscaled image changes from three channels to a single channel, making processing data from a single channel much easier. The purpose of image grayscaling is to simplify the matrix and improve computational speed.

Specifically, the threshold automatic acquisition algorithm can binarize an image, and binarization can convert a grayscale image into a binary image. This is as shown in Figure 4 (b). Binarization is implemented by setting the pixel grayscale greater than a certain threshold grayscale value as the maximum grayscale value, and setting the pixel smaller than this value as the minimum grayscale value. That is, set the grayscale value of the pixel point on the image to 0 or 255. In other words, it creates a visual effect where the entire image is noticeably black and white.

Specifically, the first color inversion can indirectly calculate the area of fibril particle aggregation, and since the fibrillar particle itself is black "0" and the background is white "1", 0 cannot be calculated. , the area of the particle cannot be obtained. After color inversion, the fibrillar particles are white "1" and the background is black "0". At this time, the area of the fibril-shaped particle can be calculated.

Specifically, the linking domain finding and staining can distinguish single particles and aggregates of different value domains and stain them with different colors.

Specifically, deleting a portion smaller than the threshold area may delete a portion smaller than an area of 3000 to 3500 pixels. When the size of the cut sheath used in the experiment is 7.6 mm . To determine single particle, target aggregation, when the particle area (S) in the YZ coordinate plane or the This largest face is selected and the calculation is performed), that part is no longer a single particle and is considered a fibrillar particle agglomeration, and the entirety that remains and appears in the image is a target agglomerate larger than 3000 to 3500 pixels. Finally, after secondary color inversion, an image of fibrillar particle aggregation is obtained, as shown in Figure 4(e). The results of this experiment meet the researcher's experimental requirements, and the data on the aggregate area and area formed by fibrillar particles in the fluidized bed vertical pipe during the experiment are measured.

Claims (10)

In the method for detecting agglomeration of fluidized bed fibrillar particles,
It includes the following steps,
Step 1: Carry out a fluidized bed shear freezing state experiment on the experimental system. Fibril particles are uniformly blown into the riser through the flow guide plate, and high-speed camera 1 and high-speed camera 2 are used to capture the YZ coordinate plane and XZ coordinate of the riser, respectively. Take pictures of the plane, record the flow state of each second, and at the same time supplement the light in the YZ coordinate plane and XZ coordinate plane;
Step 2: Load the image signals acquired from high-speed camera 1 and high-speed camera 2 into the PC, and the PC system uses the signal identification code for each frame in the large image database to identify the YZ images acquired by shooting from high-speed camera 1 within the same signal time. Select the image on the coordinate plane and the image on the
Step 3: Subtract the background image through morphological and subtraction operations on the three-dimensional images obtained by simulation;
Step 4: Perform grayscale processing on the image with subtracted background;
Step 5: Obtain the binarization threshold using an algorithm for automatically obtaining the threshold, and perform binarization on the grayscale processed image;
Step 6: Perform first-order color inversion on the binary image;
Step 7: Find the connection domain, calculate the area and stain, use the pixel area method to calculate the pixel values of particles and aggregates at different thresholds to obtain the area of particles and aggregates in the image, and distinguish between particle and aggregate staining. where pixel values 0 to 1500 are stained color 1 and are indicated as small particles, 1500 to 3000 pixel values are stained color 2 and indicated large particles, and pixel values >3000 are stained color 3 and are indicated as target aggregation. It is displayed as;
Step 8: After staining, delete a portion of the particle image in the YZ coordinate plane or the XZ coordinate plane that is smaller than a predetermined threshold pixel area, that is, the single particle portion;
Step 9: Perform secondary color inversion on the image from which the portion smaller than the threshold area is deleted, and finally obtain the fibril-phase particle aggregation image and aggregation area data, and terminate the operation. Method for detecting and processing aggregation of phase particle swarms. According to paragraph 1,
High-speed camera 1 and high-speed camera 2 both undergo parameter initialization processing before shooting, the initial transmission signals of high-speed camera 1 and high-speed camera 2 are all synchronized, and the shooting areas of high-speed camera 1 and high-speed camera 2 are the dense image area of the experimental system. A method for detecting aggregation of a group of fluid-bed fibril particles, characterized in that: According to paragraph 1,
In step 8, the threshold area is set to 3000 to 3500 pixels, and if the particle area S in the YZ coordinate plane and the XZ coordinate plane are both greater than the set threshold, the coordinate plane with the larger recognition area is selected to perform the operation. A method for detecting and processing aggregation of a group of fluidized bed fibril particles, characterized in that: According to paragraph 1,
A method for detecting agglomeration of a group of fluidized bed fibrillar particles, characterized in that the sheath size used in the experiment is 7.6 mm × 0.9 mm to 15 mm × 0.9 mm. According to paragraph 1,
In step 4, the grayscale value of the pixel point on the image is set to 0 or 255. An experimental system using the method for detecting agglomeration of fluidized bed fibrillar particles according to any one of claims 1 to 5, comprising:
Contains a fluidized layer, surface light source lamp 1 (8), surface light source lamp 2 (7), auxiliary light lamp 1 (10), auxiliary light lamp 2 (13), high-speed camera 1 (11), high-speed camera 2 (12), hub (14), it further includes a PC stage (15), and surface light source lamp 1 (8) and surface light source lamp 2 (7) are installed on the XZ coordinate plane and YZ coordinate plane of the riser of the fluidized bed, respectively, and a high-speed camera 1 ( The It is installed on one side of the fluidization vertical pipe, and auxiliary light 2 (13) is installed on one side, and high-speed camera 1 (11) and high-speed camera 2 (12) transmit signals to the PC stage (15) through the hub (14), respectively. An experimental system using a method for detecting and processing aggregation of a cluster of fibril-like particles in a fluidized bed, characterized in that they are connected. According to clause 6,
The bottom of the riser is connected to a flow guide pipe (9), and within the flow guide pipe (9), there are three arc-shaped low plate (91), middle plate (92), and high plate (93) installed diagonally spaced from bottom to top. is included, the gap between the bottom wall of the low plate 91 and the flow guide pipe 9 is 5 to 7 cm, the gap between the middle plate 92 and the low plate 91 is 12 to 14 cm, and the high plate 93 ) and the intermediate plate 92 are 7 to 9 cm apart. In clause 7,
Detection of agglomeration of fluidized bed fibril-like particles, characterized in that the arcuate angles of the low plate 91, the middle plate 92, and the high plate 93 are 75 to 80°, 60 to 65°, and 50 to 55°, respectively. Experimental system using processing methods. In clause 7,
The opposing sides of the low plate 91, the middle plate 92, and the high plate 93 are each engaged with the two inner walls of the flow guide pipe 9, and the engagement gap is 0.5 to 1.5 mm. An experimental system using a method for detecting aggregation of fibrillar particles. In clause 7,
The fluidized bed consists of a fan (1), storage bin (2), drawing plate (3), nut (4), unloader (5), unloading bin (6), cyclone separator (16), transport vessel (17), and reflux. It further includes a pipe 18, a control valve 1 (19), and a control valve 2 (20), and the other end of the flow guide pipe (9) is connected to the fan (1) through a ventilation pipe, and the upper end of the riser is connected to the curved pipe. communicates with the cyclone separator 16 through the cyclone separator 16, the bottom of the cyclone separator 16 communicates with the transport container 17, and the top of the cyclone separator 16 communicates with the side wall of the ventilation pipe through the reflux pipe 18, and reflux Control valve 1 (19) and control valve 2 (20) are installed spaced apart on the pipe (18), and the storage bin (2) is installed above the unloading bin (6), and these two are connected by a nut (4). A drawing plate (3) is installed at the connection point, the unloader (5) is mounted at the upper part of the unloading bin (6), and the bottom of the unloading bin (6) is attached to the upper wall of the flow guide pipe (9). An experimental system using a method for detecting and processing aggregation of a cluster of fibrillar particles in a fluidized bed, characterized in that it communicates with.