Method for evaluating pleat shape and method for manufacturing pleated body

Abstract

To provide a method for evaluating a pleat shape, the method making it possible to objectively evaluate the pleat shape of a pleat body; and a method for manufacturing a pleat body.SOLUTION: This method for evaluating the pleat shape of a pleat body 20 obtained by processing a sheet-like material 30 into a pleat shape comprises: an extraction step for extracting a feature pertaining to the pleat shape from image data 56 of the pleat body 20; and an evaluation step for evaluating the pleat shape on the basis of the feature extracted in the extraction step.SELECTED DRAWING: Figure 13

Description

[Technical field]

[0001] The present disclosure relates to a method for evaluating a pleat shape and a method for manufacturing a pleated body. [Background technology]

[0002] 2. Description of the Related Art Conventionally, pleated bodies have been known in which a sheet-like material is processed into a pleated shape and used, such as the filter medium described in Patent Document 1 (JP Patent Publication No. 2016-026870).

[0003] It is desirable for such a pleated body to be processed into an intended pleat shape by a processing machine in order to achieve a desired function or for other purposes. DISCLOSURE OF THEINVENTION [Problem to be solved by the invention]

[0004] However, in order to process a sheet-like material into a pleated body of the intended shape, it may be necessary to adjust the manufacturing conditions depending on the material and physical properties of the sheet-like material, the settings of the processing machine, and the usage environment, and the current situation is that the ability to adjust the manufacturing conditions is dependent on the intuition of an experienced worker who visually inspects the pleat shape.

[0005] For this reason, it is desirable to be able to objectively evaluate the pleat shape of a pleated body. [Means for solving the problem]

[0006] The evaluation method according to a first aspect is a method for evaluating the pleat shape of a pleated body obtained by processing a sheet-like material into a pleat shape, and includes an extraction step and an evaluation step. In the extraction step, feature amounts related to the pleat shape are extracted from image data of the pleated body. In the evaluation step, the pleat shape is evaluated based on the feature amounts extracted in the extraction step.

[0007] According to this evaluation method, it becomes possible to objectively evaluate the pleat shape of the pleated body.

[0008] The evaluation method according to a second aspect is the evaluation method according to the first aspect, wherein the pleated body is a filter medium.

[0009] This evaluation method makes it possible to objectively evaluate the pleat shape of the filter medium, and therefore the performance of the filter medium can also be evaluated objectively.

[0010] An evaluation method according to a third aspect is the evaluation method according to the first or second aspect, wherein the evaluation step includes a step of comparing the feature amount extracted in the extraction step with a predetermined criterion.

[0011] This evaluation method makes it possible to grasp whether or not the pleat shape of the pleated body satisfies a predetermined standard.

[0012] The evaluation method according to a fourth aspect is any one of the evaluation methods according to the first aspect to the third aspect, in which the feature amount is any one of the pleat pitch, pleat height, length in the second direction of a connecting portion connecting adjacent peaks and valleys as viewed in the first direction, degree of curvature of the connecting portion as viewed in the first direction, symmetry between adjacent connecting portions as viewed in the first direction, length and width of a spacing portion that maintains the pleat spacing when the spacing portion is provided in the pleated body, and combinations thereof. Here, the first direction is the direction in which the fold lines extend. The second direction is the direction in which the pleat shape continues. The third direction is a direction perpendicular to both the first direction and the second direction.

[0013] This evaluation method makes it possible to improve the objectivity of the evaluation of the pleat shape.

[0014] An evaluation method according to a fifth aspect is the evaluation method according to any one of the first to fourth aspects, further comprising an evaluation model learning step. In the evaluation model learning step, an evaluation model is created by machine learning the relationship between the image data of the pleated body and the evaluated data using a plurality of pairs of image data of the pleated body and evaluated data associated with the image data of the pleated body. In the evaluation step, the feature amount extracted in the extraction step is evaluated based on the evaluation model created in the evaluation model learning step.

[0015] This evaluation method increases the objectivity of the evaluation results.

[0016] An evaluation method according to a sixth aspect is any one of the evaluation methods according to the first aspect to the fifth aspect, further comprising an input condition learning step. In the input condition learning step, a relationship between the input conditions and the evaluation result data is learned using a plurality of pairs of input conditions and evaluation result data. The input conditions include processing condition data input to a processing machine that processes a sheet-like material into a pleated shape, and physical property data related to the sheet-like material. The evaluation result data is data obtained by evaluating a pleated body obtained in response to the input conditions in the evaluation step.

[0017] This evaluation method further increases the objectivity of the evaluation results.

[0018] The evaluation method according to the seventh aspect is the evaluation method according to the sixth aspect, and the processing condition data includes the amount of gap between the folding blade and the folding plate of the processing machine, the conveying speed at which the processing machine conveys the material, the amount of resin applied to the material to form the spacing retaining portion, and any combination of these.

[0019] This evaluation method further increases the objectivity of the evaluation results.

[0020] An evaluation method according to an eighth aspect is the evaluation method according to the sixth or seventh aspect, in which the input conditions are output by an output device based on the model obtained in the input condition learning step.

[0021] According to this evaluation method, it is possible to grasp the input conditions based on an objective evaluation.

[0022] A manufacturing method according to a ninth aspect manufactures a pleated body using the input conditions output in the evaluation method according to the eighth aspect.

[0023] This manufacturing method makes it easy to realize the intended pleat shape in the manufactured pleated body. [Brief description of the drawings]

[0024] [Figure 1] FIG. 2 is a schematic external perspective view of an example of a pleated body. [Diagram 2] FIG. 13 is a schematic external perspective view of an example of a pleated body provided with a spacing portion. [Diagram 3] 4 is a diagram showing the shape of a pleat of a pleated body when viewed in a first direction. FIG. [Figure 4] FIG. 2 is a cross-sectional view of an example sheet of material. [Diagram 5] FIG. 2 is an external perspective view of the air filter unit. [Figure 6] FIG. 1 is a schematic diagram of an apparatus used for stretching a sheet-like material. [Figure 7] FIG. 2 is a diagram showing a schematic diagram of an apparatus (left half) used for stretching a sheet-like material in the width direction and an apparatus (right half) for laminating an air-permeable support material onto the sheet-like material. [Figure 8] FIG. 1 is a schematic diagram of a reciprocating folder. [Figure 9] FIG. 1 is a schematic diagram showing an apparatus used in the coating and repleating steps of the spacing portion. [Figure 10] FIG. 13 is a diagram showing a state in which the folding blade is retracted in the re-pleating process. [Figure 11] FIG. 13 is a diagram showing a state in which the folding blade has moved to a folding position in the re-pleating process. [Figure 12] FIG. 2 is a system configuration diagram of an example of a camera and an evaluation information processing device. [Figure 13] 13 is a flowchart of evaluation processing by the evaluation information processing device. [Figure 14] FIG. 13 is a system configuration diagram of another example of the camera and evaluation information processing device. [Figure 15] 13 is a flowchart of evaluation processing by the evaluation information processing device. [Figure 16] FIG. 2 is a system configuration diagram of an example of a processing machine, a camera, and an evaluation information processing device. [Figure 17] 4 is a flowchart of a manufacturing process of a pleated body. [Figure 18]13 is a side view photograph of the pleated body obtained when the gap between the folding blade and the folding plate is changed. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0025] (1) Evaluation method for pleat shape The pleat shape evaluation method evaluates the pleat shape of a pleated body obtained by processing a sheet-like material into a pleat shape. This evaluation method includes an extraction step and an evaluation step. In the extraction step, feature amounts related to the pleat shape are extracted from image data of the pleated body. In the evaluation step, the pleat shape is evaluated based on the feature amounts extracted in the extraction step. This makes it possible to objectively evaluate the pleat shape of the pleated body without relying on visual evaluation of the pleat shape by an expert.

[0026] The pleat shape of the pleated body may be, for example, a shape in which mountain folds and valley folds are repeated on a sheet-like material 30 so that a plurality of mountain fold lines X and a plurality of valley fold lines Y are aligned parallel to each other, as shown in Fig. 1, which is a schematic external perspective view of an example of a pleated body 20. Here, the straight lines perpendicular to each of the plurality of mountain fold lines X and the straight lines perpendicular to each of the plurality of valley fold lines Y may be parallel to each other.

[0027] The method of acquiring image data of the pleated body is not particularly limited, and may be, for example, an image captured from a direction in which the mountain fold lines and valley fold lines extend. The image data of the pleated body is not particularly limited, and may be, for example, framed so that the camera captures 3 to 30 pairs of peaks and valleys. When acquiring images of the pleated body with a camera from a predetermined position, it is preferable to capture images at a predetermined time interval such that the peak and valley pairs move a predetermined number of times as the pleated body is transported in the transport direction.

[0028] (2) Extraction process In the extraction step, features related to the pleat shape are extracted from the image data of the pleated body. The extraction process here is preferably performed using an information processing device such as a computer having a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit) and a memory such as a RAM or a ROM.

[0029] (2-1) Features of pleat shape The feature amount relating to the pleat shape may be, for example, a numerical representation of the shape of curved or straight edge portions seen when the pleat shape is viewed from the direction in which the mountain folds and valley folds extend, and may consist of only one feature amount or may consist of multiple feature amounts, thereby increasing the objectivity of the evaluation of the pleat shape.

[0030] More specifically, the characteristic amount is, for example, the pleat pitch, the pleat height, the length in the second direction of the connecting portion connecting adjacent peaks and valleys in the first direction, the degree of curvature of the connecting portion in the first direction, the symmetry between adjacent connecting portions in the first direction, the length and width of the spacing portion when the spacing portion for maintaining the pleat spacing is provided in the pleated body, and any combination thereof. Here, the first direction is the direction in which the multiple peak fold lines X and the multiple valley fold lines Y extend. The second direction is the direction in which the pleat shape continues, and can be said to be the direction in which the multiple peak fold lines X are arranged, or the direction in which the multiple valley fold lines Y are arranged. The third direction is a direction perpendicular to both the first direction and the second direction. In addition, the third direction can be said to be the direction in which the treated fluid passes through the pleated filter medium.

[0031] Here, as shown in the schematic external perspective view of FIG. 2, the pleated body may be provided with either or both of a peak spacing portion 26 for maintaining a spacing between adjacent peaks and a valley spacing portion 27 for maintaining a spacing between adjacent valleys. Spacing portions such as the peak spacing portion 26 and the valley spacing portion 27 are optional, and the positions, shapes, thicknesses, etc. of the pleated body may be optional. Note that FIG. 2 shows a state in which the folding gap of the pleated body 20 provided with spacing portions is slightly widened, but as shown in FIG. 3, which is a cross-sectional view seen from the folding line direction, the peak spacing portion 26 has a function of maintaining a spacing so that the opposing surfaces of adjacent peaks do not contact each other, and the valley spacing portion 27 has a function of maintaining a spacing so that the opposing surfaces of adjacent valleys do not contact each other. Note that in FIG. 3, the depth direction of the paper is the first direction, the left-right direction of the paper is the second direction, and the up-down direction of the paper is the third direction. In addition, in Figure 3, an example is given of the spacing retaining portion where the peak spacing retaining portion 26 is provided outside the peak fold line X in the third direction, and the valley spacing retaining portion 27 is provided outside the valley fold line Y in the third direction.

[0032] The parameters of the feature amounts will be described below with reference to FIG.

[0033] The pleat pitch a refers to the length between adjacent mountain fold lines X in the second direction, and can also be referred to as the length between multiple valley fold lines Y in the second direction. Here, the mountain fold lines X refer to the ends of the mountain portions in the third direction. Also, the valley fold lines Y refer to the ends of the valley portions in the third direction. Instead of the pleat pitch a, a value obtained by subtracting the thickness of the sheet from the pleat pitch a may be used as a parameter.

[0034] Pleat height b refers to the length in the third direction between an intersection Z of a line connecting adjacent mountain fold lines X and a line extending in the third direction from a valley fold line Y between the adjacent mountain fold lines X, and the valley fold line Y. In addition, pleat height b refers to the length in the third direction between an intersection Z of a line connecting adjacent valley fold lines Y and a line extending in the third direction from a mountain fold line X between the adjacent valley fold lines Y, and the mountain fold line X.

[0035] The length c in the second direction of the connecting portion 35 connecting adjacent peaks and valleys in the first direction refers to the length (c1) in the second direction of the rightmost undulating portion on the valley side and the leftmost undulating portion on the mountain side of the connecting portion 35a connecting the valley and the adjacent peak on the right side in Fig. 3. Also, the length c in the second direction of the connecting portion 35 connecting adjacent peaks and valleys in the first direction refers to the length (c2) in the second direction of the rightmost undulating portion on the mountain side and the leftmost undulating portion on the valley side of the connecting portion 35b connecting the peak and the adjacent valley on the right side in Fig. 3. Here, in Fig. 3, the right side can be referred to as one side in the second direction, and the left side can be referred to as the other side in the second direction.

[0036] The degree of curvature of the connecting portion as viewed in the first direction is not particularly limited, but can be, for example, the ratio (c / b) of the length c of the connecting portion in the second direction to the length d of the connecting portion in the third direction. The greater the value of this ratio, the greater the degree of curvature can be evaluated. For example, in the case where the pleated body is a filter medium, the greater the value of the ratio (c / b), the greater the degree of curvature, the narrower the flow path for passing a fluid, and the reduced effective surface area, which can be evaluated as increasing pressure loss.

[0037] The symmetry between adjacent connecting parts when viewed in the first direction is not particularly limited, and may be, for example, a difference in the degree of curvature between adjacent connecting parts. More specifically, it may be, for example, a difference in the ratio (c / b) between adjacent connecting parts.

[0038] When the pleated body is provided with a spacing maintaining portion for maintaining the pleat spacing, the length of the spacing maintaining portion may be the length of a portion extending in a direction intersecting the mountain fold line X or the valley fold line Y. Note that, instead of this length, for example, the ratio of the length of the spacing maintaining portion in the third direction to the pleat height b may be used.

[0039] When the pleated body is provided with spacing maintaining portions for maintaining the pleat spacing, the width of the spacing maintaining portions may be a length in a first direction that is a direction parallel to the mountain fold lines X or the valley fold lines Y.

[0040] Furthermore, in the case where spacing maintaining portions that maintain the pleat spacing are provided in the pleated body at predetermined intervals in the first direction, the spacing in the first direction may also be used as a feature amount.

[0041] (3) Evaluation process In the evaluation step, the pleat shape is evaluated based on the feature amounts extracted in the extraction step. The evaluation process here is preferably performed using an evaluation information processing device such as a computer having a processor such as a CPU (Central Processing Unit) or a GPU (Graphics Processing Unit) for performing various information processing, and memories such as a RAM and a ROM.

[0042] The evaluation step preferably includes, for example, a step of comparing the feature amount extracted in the extraction step with a predetermined criterion, which makes it possible to determine whether or not the pleat shape of the pleated body satisfies the predetermined criterion.

[0043] The predetermined criteria may be set for each of the above-mentioned features, or may be determined by a mathematical formula using at least two or more of the above-mentioned features.

[0044] Furthermore, the method for determining the predetermined standard is not particularly limited, and may be, for example, a standard for ensuring that the pleat shape of the obtained pleated body satisfies desired conditions.

[0045] Furthermore, for example, in the case where the pleated body is a filter medium, a predetermined standard may be set so that the pressure loss when the fluid to be treated is passed through the obtained pleated body satisfies a predetermined condition.

[0046] (4) Uses of the pleated body The material 30 in a sheet state processed into a pleated shape is not particularly limited, but examples thereof include a filter medium, an adsorbent, and the like. Such a filter medium or adsorbent may be configured with a plurality of layers. As shown in the cross-sectional view of FIG. 4, the filter medium may be, for example, a filter medium having a main collection layer 31, a first support layer 32 laminated on one side thereof, and a second support layer 33 laminated on the other side thereof. In addition, the filter medium may have a support layer laminated only on one side or the other side of the main collection layer 31 in the direction in which the treated fluid passes.

[0047] The filter material may be an air filter material in which the fluid to be treated is a gas, a filter material in which the fluid to be treated is a liquid, a filter material in which the fluid to be treated is a powder, or a mixture of at least two of the fluid to be treated, which are a gas, a liquid, and a powder.

[0048] Examples of the filter material include glass fiber filter material, organic fiber filter material, and filter material with nanofiber. Examples of the organic fiber filter material include fluororesin porous membranes such as porous membranes (hereinafter, sometimes referred to as PTFE) made of polytetrafluoroethylene (hereinafter, sometimes referred to as PTFE porous membrane). Such fluororesin porous membranes have higher dust collection efficiency than glass fiber filter material when compared at the same pressure loss, and are therefore particularly suitable for use in HEPA filters (High Efficiency Particulate Air Filters) and ULPA filters (Ultra Low Penetration Air Filters).

[0049] Here, when the pleated filter medium is used by passing the fluid to be treated in the direction intersecting the plane on which the multiple mountain fold lines X are located and the plane on which the multiple valley fold lines Y are located, the pleat shape is deformed by resistance, and the opposing surfaces on the downstream side of the filter medium come into contact with each other, which may cause an increase in the passage resistance and a decrease in the effective filtration area, resulting in a decrease in performance. This deformation of the pleat shape is likely to occur when the uniformity of the pleat shape is low, so it is required that the uniformity of the pleat shape of the obtained pleated body is high. In contrast, this evaluation method makes it possible to objectively evaluate the pleat shape of the filter medium, and therefore makes it possible to objectively evaluate the performance of the filter medium. In addition, it is also possible to estimate that the performance of the filter medium is good for the pleat shape of the pleated body that has been evaluated as good by this evaluation method.

[0050] (5) Pleated unit Next, a unit having a pleated body will be described with reference to FIG.

[0051] The unit 100 equipped with the pleated body includes the above-described pleated body 20 and a frame body 25 for housing the pleated body 20.

[0052] The frame 25 is made by combining plate materials such as resin and metal, and the gap between the pleated body 20 and the frame 25 is preferably sealed with a sealant. The sealant is for preventing leakage between the pleated body 20 and the frame 25, and is made of resin such as epoxy, acrylic, or urethane.

[0053] (6) Manufacturing method of filter media An example of a method for producing a filter medium, which is an example of a sheet-like material, is outlined in Figures 6 and 7. Note that, here, a PTFE porous membrane, which is an example of a fluororesin porous membrane, is taken as an example for explanation.

[0054] In Fig. 6, 1 indicates an unwinding roll, 2 indicates a winding roll, 3 to 5 indicate rolls, 6 and 7 indicate heat rolls, and 8 to 12 indicate rolls. In Fig. 7, 14 indicates an unwinding roll, 15 indicates a preheating zone, 16 indicates a stretching zone, 17 indicates a heat setting zone, 19 indicates a laminating roll, and 21 indicates a winding roll.

[0055] In the porous membrane preparation process, a PTFE porous membrane is prepared by biaxially stretching a PTFE unsintered film through the process shown in Fig. 6 and the process shown in the left half of Fig. 7. Since such a fluororesin porous membrane is thin and soft, it is supported by laminating a support material in the following process.

[0056] In the thermal lamination step, the breathable support material 18 made of nonwoven fabric is thermally laminated on both sides of the PTFE porous membrane by the step shown in the right half of FIG. 7 to obtain a sheet-like material 30 which is a filter medium.

[0057] In the winding process, the sheet-like material 30 which is the filter medium is wound around the winding roll 21 .

[0058] In addition, the filter medium is preferably one that can be used as a HEPA filter, a ULPA filter, or a medium-performance filter. These filter mediums using a fluororesin porous membrane have a very fine fiber diameter, a fine fiber structure, and a small membrane thickness. In addition, the membrane thickness of the fluororesin porous membrane may be, for example, 1 μm or more and 100 μm or less, and from the viewpoint of reducing pressure loss, it is preferably 1 μm or more and 50 μm or less, and more preferably 1 μm or more and 30 μm or less. In addition, the average fiber diameter of the fluororesin porous membrane may be, for example, 0.01 μm or more and 0.25 μm or less, and preferably 0.05 μm or more and 0.2 μm or less. The average fiber diameter may be calculated as a number average fiber diameter by randomly selecting 50 fibers from an image of a scanning electron microscope photograph.

[0059] (7) Manufacturing method of pleated body 8 and 10 show an outline of an example of a manufacturing method for the pleated body 20. In FIG.

[0060] The manufacturing method of the pleated body includes an unwinding step, a pleating step, an unfolding step, a spacing portion coating step, and a re-pleating step.

[0061] In the unwinding process, the sheet-like material 30 that was wound around the winding roll 21 in the winding process is unwound. Note that the material 30 wound around the winding roll 21 may have a curl, and the curl tends to become larger in areas with a smaller radius of the winding roll 21 because the material is wound with a larger curvature.

[0062] In the pleating process, as shown in FIG. 8, a pair of upper folding blades 41 and lower folding blades 42 of a processing machine are used to alternately fold back the sheet-like material 30 to a corrugated shape (pleats processing), forming corrugated fold lines. The processing machine folds the material 30 with the upper folding blade 41 and the lower folding blade 42 while feeding the sheet-like material 30 to a folding processing location between the upper folding blade 41 and the lower folding blade 42. Therefore, the material 30 is folded by the upper folding blade 41 and the lower folding blade 42 while a force is applied to the material 30 in the conveying direction. Here, the processing machine can adjust the conveying speed of the material 30 by accepting setting input, etc. After that, the material 30 processed into a corrugated shape is heated by the heater 22 to create a fold.

[0063] In the unfolding step, the pleated material 30 is unfolded into a sheet, as shown in FIG.

[0064] In the spacing-maintaining portion coating process, as shown in Fig. 9, a processing machine coats the spread material 30 with a material such as resin that forms spacing portions such as ridge spacing portions 26, 28 and valley spacing portions 27, 29. Coating of the spacing portions is performed by coating molten resin from nozzles arranged at predetermined intervals in a direction perpendicular to both the conveyance direction and thickness direction of the material 30. The processing machine can adjust the amount of coating of the spacing portions by receiving setting inputs, etc. Specifically, the amount of coating is adjusted by adjusting the discharge speed of the molten resin relative to the conveyance speed.

[0065] In the repleating process, the material 30 coated with the spacing portion is pleated again using a pair of upper folding blades 43 and lower folding blades 44 of the processing machine to obtain the pleated body 20. The upper folding blade 43 and the lower folding blade 44 are comb-shaped so as not to interfere with the spacing portion. As in the above-mentioned pleating process, the processing machine folds the material 30 with the upper folding blade 43 and the lower folding blade 44 while feeding the sheet-shaped material 30 to the folding processing location between the upper folding blade 43 and the lower folding blade 44. Therefore, the material 30 is folded by the upper folding blade 43 and the lower folding blade 44 while a force is applied to the material 30 in the conveying direction. Here, the processing machine can adjust the conveying speed of the material 30 by receiving a setting input, etc.

[0066] In the repleating process, as shown in FIG. 10, the upper folding blade 43 retracts upward and the lower folding blade 44 retracts downward, and the material 30 is conveyed in the conveying direction. As shown in FIG. 11, the upper folding blade 43 descends to a lower folding position and the lower folding blade 44 rises to an upper folding position, thereby processing the material 30 into a pleated shape. In the following, both the pleating process and the repleating process may be referred to as pleating. Here, as shown in FIG. 11, the pleating machine adjusts the vertical distance u between the upper folding plate 45 and the lower folding plate 46 by accepting setting input or the like so as to correspond to the pleat height b or a length that is a predetermined amount larger than the pleat height b. Furthermore, in the pleating machine, the vertical gap s when the upper surface of the lower folding plate 46 and the lower end of the upper folding blade 43 are closest to each other is adjusted by accepting setting input or the like. Similarly, in the pleating machine, the vertical gap t when the lower surface of the upper folding plate 45 and the upper end of the lower folding blade 44 are closest to each other is adjusted by receiving a setting input or the like.

[0067] As described above, in the pleating machine, the shape of the resulting pleated body can be changed by adjusting the conveying speed of the material 30, the gap s, the gap t, the amount of application of the material such as resin forming the spacing portion, and the like. For example, when the conveying speed of the material 30 changes, the pressure applied to the material 30 from the folding blades such as the upper folding blade 43 and the lower folding blade 44 changes, which may cause a change in the length c in the second direction of the connecting portion 35 connecting the adjacent peaks and valleys and a change in the pleat pitch a, and the shape of the resulting pleated body may change. In addition, when the application amount of the material such as resin forming the spacing portion changes, the pleat pitch a may change, and the shape of the resulting pleated body may change. Furthermore, when the gap s or the gap t changes, the pressure applied to the material 30 from the folding blades such as the upper folding blade 43 and the lower folding blade 44 changes, which changes the position where the material 30 is folded by the folding blades, which may cause a change in the length c in the second direction of the connecting portion 35 connecting the adjacent peaks and valleys and a change in the pleat pitch a, and the shape of the resulting pleated body may change. In addition, when the material 30 has a high rigidity or a large thickness, the shape of the resulting pleated body may change because the pleating process is less likely to cause folding.

[0068] The image data of the pleated body is obtained, for example, by using a camera 48 to capture an image of the pleated body obtained through the re-pleating process.

[0069] (8) An example of pleat shape evaluation Below, an example will be described in which the pleat shape of a pleated body used as an air filter medium is evaluated using an evaluation information processing device 50 and a camera 48, with the average value in one image of the length c in the second direction of the connecting portion connecting adjacent peaks and valleys when viewed in the first direction being used as an index.

[0070] Here, as shown in Fig. 9, the camera 48 is provided so as to photograph the pleated body in which the pleats have been formed after the re-pleating process, from the extension direction of the folds. Also, as shown in Fig. 12, the camera 48 records image data 49 of the photographed pleated body, which can be sent to an evaluation information processing device 50 via a communication line.

[0071] The evaluation information processing device 50 is an information processing device such as a computer having a processor 51 such as a CPU or a GPU (Graphics Processing Unit) for performing various information processing, a memory 53 including a RAM, a ROM, etc., and a display 52 capable of displaying and outputting data. The memory 53 stores image data 49 sent from the camera 48 as is as image data 56. The memory 53 also stores an indexing program 54 for indexing the image data 56, and reference value data 55.

[0072] In the above-described configuration, the evaluation process flow by the evaluation information processing device 50 is shown in FIG.

[0073] In step S11, the processor 51 stores the image data 49 transmitted from the camera 48 in the memory 53 as image data 56 as is.

[0074] In step S12, the processor 51 executes the indexing program 54 stored in the memory 53 on the image data 56 acquired in step S11 to calculate an index value, which is a feature amount corresponding to the image data 56. Specifically, the average value of the lengths c in the second direction of the multiple connecting parts 35 (see FIG. 3) included in one image data 56 is calculated as the index value. Specifically, the value obtained by dividing the sum of c1 and c2 shown in FIG. 3 in one image by the number of connecting parts 35 in one image is calculated as the index value.

[0075] In step S13, the processor 51 determines whether or not the index value calculated in step S12 is smaller than the reference value indicated by the reference value data stored in the memory 53. If it is determined that it is smaller, the process proceeds to step S14, and if it is determined that it is larger, the process proceeds to step S16.

[0076] In step S14, the processor 51 determines that the pleat shape is good with small undulations because the average value of the length c in the second direction of the connecting portions 35 of the pleated body 20 is smaller than the reference value.

[0077] In step S15, the processor 51 outputs a message to the display 52 indicating that the product is non-defective, and proceeds to the process of acquiring the next image data.

[0078] In step S16, the processor 51 determines that the pleat shape is poor because the average value of the length c of the connecting portions 35 of the pleated body 20 in the second direction is greater than the reference value.

[0079] In step S17, the processor 51 displays on the display 52 that the product is defective, and proceeds to the process of acquiring the next image data.

[0080] In the above process, the reference value data 55 may be determined based on the index value of a pleated body 20 that is visually recognized as having a good pleat shape among already obtained pleated bodies 20. The reference value data 55 may also be determined based on the index value of a pleated body 20 that is confirmed to have a measured value of a predetermined pressure loss lower than a predetermined value among already obtained pleated bodies 20.

[0081] (9) An example of machine learning for evaluation methods Hereinafter, a pleated body used as an air filter medium will be described by way of example in which the pleat shape is evaluated by machine learning using an evaluation information processing device 50a configured as shown in Fig. 14 and a camera 48, with the average value in one image of the length c in the second direction of the connecting portion connecting adjacent peaks and valleys in the first direction as viewed as an index. Note that the evaluation information processing device 50a differs from the evaluation information processing device 50 in that it stores an evaluation machine learning program 58 in the memory 53, accumulates evaluation correspondence data 57 in which the evaluation results and corresponding image data are associated in the memory 53, and stores an evaluation model 59 obtained by machine learning in the memory 53.

[0082] In the above-described configuration, the evaluation processing flow by the evaluation information processing device 50a is shown in Fig. 15. Note that steps S11 to S15 and S17 are omitted because they are the same as the evaluation processing flow by the evaluation information processing device 50. After the processing of step S15 or S17 is completed, the process proceeds to step S18, respectively.

[0083] In step S18, the processor 51 causes the memory 53 to store the evaluation results in step S15 in association with the corresponding image data as evaluation correspondence data 57. Similarly, the processor 51 causes the memory 53 to store the evaluation results in step S17 in association with the corresponding image data as evaluation correspondence data 57.

[0084] In step S19, the processor 51 judges whether or not a predetermined amount or more of the evaluation correspondence data 57 has been accumulated. If the accumulated amount of the evaluation correspondence data 57 is less than the predetermined amount, the process returns to step S11 to continue accumulating the evaluation correspondence data 57. If the accumulated amount of the evaluation correspondence data 57 is equal to or greater than the predetermined amount, the process proceeds to step S20.

[0085] In step S20, the processor 51 uses a plurality of sets of data of image data and evaluation results included in the evaluation correspondence data 57 to train a neural network by deep learning or the like to obtain an evaluation model 59. Note that the machine learning algorithm is not limited to deep learning, and any known algorithm such as a support vector machine, a Gaussian process, a decision tree, or a random forest may be used.

[0086] In step S21, the processor 51 stores the evaluation model 59 obtained in step S20 in the memory 53. As a result, in the indexing in step S12, indexing is performed using the evaluation model 59 stored in the memory 53, and the evaluation is continued.

[0087] As described above, by indexing the pleat shape of the pleated body using evaluation model 59 obtained by machine learning and evaluating it, it becomes possible to objectively evaluate the pleat shape of the pleated body.

[0088] The machine learning is not particularly limited, and may be, for example, a device for obtaining an evaluation model 59 that enables new indexing by extracting features common to image data of pleated bodies judged to be good products through machine learning. The reference value may be updated to correspond to the value indexed by the evaluation model 59. In this case, the evaluation of the pleat shape can be made more objective. The feature amount extracted from the image data may be any of those described in "(2-1) Feature amount of pleat shape" above, or a combination thereof.

[0089] (10) Information processing during manufacturing of pleated bodies Hereinafter, information processing during the production of a pleated body will be described with reference to the block diagram of Fig. 16 and the flow chart of Fig. 17, using the processing machine 60, the evaluation information processing device 50, and the camera 48 as examples. Note that here, the processing machine 60, the evaluation information processing device 50, and the camera 48 are connected to each other so that they can communicate with each other.

[0090] The processing machine 60 is equipped with a processor 61 such as a CPU (Central Processing Unit) or GPU (Graphics Processing Unit) that performs various types of information processing, a memory 70 such as a RAM or a ROM, a display 62, a temperature sensor 63, a humidity sensor 64, a receiving section 65, a conveying section 66, a folding blade section 67, an application section 68, a roll section 69, etc.

[0091] The temperature sensor 63 is a sensor for detecting the air temperature in the space where the pleating process is performed.

[0092] The humidity sensor 64 is a sensor for detecting the humidity in the space where the pleating process is carried out.

[0093] The receiving unit 65 accepts input of various information. Here, inputs such as physical property data of the sheet-like material 30 as the filter material to be processed, setting data of the vertical interval u between the upper folding plate 45 and the lower folding plate 46 according to the pleat height b of the pleated body, setting data of the vertical gap s when the upper surface of the lower folding plate 46 and the lower end of the upper folding blade 43 are closest to each other, setting data of the vertical gap t when the lower surface of the upper folding plate 45 and the upper end of the lower folding blade 44 are closest to each other, setting data of the coating amount of the spacing retaining portion, setting data of the rotation speed of each roller, etc. are accepted. For example, the physical property data of the sheet-like material 30 includes information indicating the rigidity of the material 30, information indicating the thickness of the material 30, information indicating the material constituting the material 30, information indicating a predetermined pressure loss of the material 30, information indicating a predetermined collection efficiency of the material 30, information indicating the average fiber diameter in the main collection layer of the material 30 such as a fluororesin porous membrane, etc.

[0094] The conveying section 66 adjusts the rotation speed of each roller during pleating.

[0095] The folding blade section 67 adjusts the vertical distance u between the upper folding plate 45 and the lower folding plate 46 during pleating, the vertical gap s when the upper surface of the lower folding plate 46 and the lower end of the upper folding blade 43 are closest to each other, and the vertical gap t when the lower surface of the upper folding plate 45 and the upper end of the lower folding blade 44 are closest to each other, based on the reception information of the reception section 65.

[0096] The applicator 68 adjusts the amount of application to the spacing portion during pleating based on the information received by the receiver 65.

[0097] The roll unit 69 obtains information indicating the degree of curl for each portion of the material 30 sent out from the winding roll 21 during pleating. For example, the roll unit 69 uses information on the radius of the winding roll 21 at which the material 30 sent out from the winding roll 21 was located, and determines that the curl is greater in a portion located at a smaller radius. The degree of the curl is determined in conjunction with the sending out from the winding roll 21. Note that the curl may be obtained from the relationship between the overall length and thickness of the material 30, rather than from the information on the radius of the winding roll 21.

[0098] The display 62 displays and outputs information during various controls.

[0099] The processor 61 performs pleating by driving and controlling each part of the processing machine 60 to realize setting data for the vertical spacing u between the upper folding plate 45 and the lower folding plate 46, which corresponds to the pleat height b of the pleated body, setting data for the vertical gap s when the upper surface of the lower folding plate 46 and the lower end of the upper folding blade 43 are closest to each other, setting data for the vertical gap t when the lower surface of the upper folding plate 45 and the upper end of the lower folding blade 44 are closest to each other, setting data for the amount of coating applied to the spacing maintaining section, and setting data for the rotational speed of each roller, which are received by the receiving unit 65.

[0100] During this pleating process, the manufacturing conditions such as the physical property data of the sheet-like material 30 to be processed, the information on the degree of curling grasped by the roll unit 69, the temperature information grasped by the temperature sensor 63, the humidity information grasped by the humidity sensor 64, the setting data of the vertical interval u between the upper folding plate 45 and the lower folding plate 46, the setting data of the vertical gap s when the upper surface of the lower folding plate 46 and the lower end of the upper folding blade 43 are closest to each other, the setting data of the vertical gap t when the lower surface of the upper folding plate 45 and the upper end of the lower folding blade 44 are closest to each other, the setting data of the coating amount of the interval holding unit, and the setting data of the rotation speed of each roller are grasped in real time, and the manufacturing conditions are stored in the memory 70 while corresponding to the pleated body obtained under the manufacturing conditions. Note that the identification of the correspondence is not particularly limited, and the processor 61 may appropriately calculate and obtain it from the conveying speed, conveying distance, processing time, etc.

[0101] The obtained pleated body is evaluated in the same manner as described above by the evaluation information processing device 50, which stores image data 49 captured by the camera 48 and transmitted as image data 56 in the memory 53 while indexing the data. The evaluation results are transmitted from the evaluation information processing device 50 to the processing machine 60 and are stored in the memory 70 as manufacturing condition correspondence data 71, which corresponds to the corresponding manufacturing conditions and indicates the relationship between the manufacturing conditions and the evaluation results.

[0102] Then, when a predetermined amount of manufacturing condition correspondence data 71 has been accumulated, the manufacturing condition correspondence data indicating the relationship between the manufacturing conditions and the evaluation results is used to train a neural network by deep learning or the like to obtain a manufacturing condition model 72, which is stored in the memory 70. Note that the machine learning algorithm is not limited to deep learning, and known algorithms such as a support vector machine, a Gaussian process, a decision tree, and a random forest may also be used.

[0103] After the manufacturing condition model 72 is obtained, the processor 61 performs pleating by controlling the drive of each part of the processing machine 60 so as to realize the processing machine control amount grasped by inputting the setting data of the vertical interval u between the upper folding plate 45 and the lower folding plate 46 corresponding to the pleat height b of the pleated body received by the reception unit 65, the physical property data of the sheet-like material 30 to be processed, the information on the degree of curling grasped by the roll unit 69, the temperature information grasped by the temperature sensor 63, and the humidity information grasped by the humidity sensor 64 into the manufacturing condition model 72. Here, the processing machine control amount includes the vertical gap s when the upper surface of the lower folding plate 46 and the lower end of the upper folding blade 43 are closest to each other, the vertical gap t when the lower surface of the upper folding plate 45 and the upper end of the lower folding blade 44 are closest to each other, the coating amount of the spacing portion, and the rotation speed of each roller.

[0104] The above process will be explained again with reference to the flowchart of FIG.

[0105] In step S31, the processor 61 of the processing machine 60 receives inputs of the physical property data of the sheet-like material 30, the setting data of the vertical interval u between the upper folding plate 45 and the lower folding plate 46 according to the pleat height b of the pleated body, the setting data of the vertical gap s when the upper surface of the lower folding plate 46 and the lower end of the upper folding blade 43 are closest to each other, the setting data of the vertical gap t when the lower surface of the upper folding plate 45 and the upper end of the lower folding blade 44 are closest to each other, the setting data of the coating amount of the spacing retaining unit, the setting data of the rotation speed of each roller, the physical property data of the sheet-like material 30 to be processed, and the like, via the receiving unit 65. The processor 61 also receives information on the degree of curling grasped by the roll unit 69, temperature information grasped by the temperature sensor 63, and humidity information grasped by the humidity sensor 64. From these, the processor 61 grasps the manufacturing conditions according to the setting data. The setting data is not particularly limited, and may be conditions empirically grasped to obtain a desired pleat shape.

[0106] In step S32, the processor 61 performs pleating in accordance with the manufacturing conditions grasped in step S32 to obtain a pleated body.

[0107] In step S33, the processor 51 of the evaluation information processing device 50 evaluates the pleat shape of the pleated body manufactured under the manufacturing conditions grasped in step S32.

[0108] In step S34, the processor 61 of the processing machine 60 associates the evaluation results in step S33 sent from the evaluation information processing device 50 with the corresponding manufacturing conditions, and stores them in the memory 70 as manufacturing condition correspondence data 71.

[0109] In step S35, the processor 61 determines whether or not a predetermined amount of data has been accumulated as the manufacturing condition correspondence relationship data 71. If a sufficient amount of data has been accumulated, the process proceeds to step S38. If the amount of data is insufficient, the process returns to step S31 and continues to acquire relationship data through manufacturing and evaluation.

[0110] In step S36, the processor 61 uses multiple sets of data of manufacturing conditions and evaluation results included in the manufacturing condition correspondence data 71 to train a neural network using deep learning or the like to obtain a manufacturing condition model 72.

[0111] In step S37, the processor 61 stores in the memory 70 the manufacturing condition model 72 obtained in step S36.

[0112] In step S38, the processor 61 grasps the manufacturing conditions other than the processing machine control amount. Specifically, the processor 61 grasps the vertical distance u between the upper folding plate 45 and the lower folding plate 46, the physical property data of the sheet-like material 30 to be processed, the information on the degree of curling grasped by the roll unit 69, the temperature information grasped by the temperature sensor 63, and the humidity information grasped by the humidity sensor 64.

[0113] In step S39, the processor 61 inputs the manufacturing conditions other than the processing machine control amount obtained in step S38 into the manufacturing condition model 72 obtained in step S36 to grasp the manufacturing conditions. Then, pleating is performed according to the obtained manufacturing conditions to obtain a pleated body. The manufacturing conditions identified here are displayed on the display 62.

[0114] In step S40, the processor 51 of the evaluation information processing device 50 evaluates the pleat shape of the pleated body manufactured in step S39.

[0115] In step S41, the processor 61 of the processing machine 60 matches the evaluation results in step S40 sent from the evaluation information processing device 50 with the corresponding manufacturing conditions, and stores them in the memory 70 as manufacturing condition correspondence data 71.

[0116] In step S42, the processor 61 determines whether or not a predetermined amount or more of data has been accumulated as the manufacturing condition correspondence data 71. If a sufficient amount of data has been accumulated, the process proceeds to step S36, where machine learning is performed to update the manufacturing condition model 72. If the amount of data is insufficient, the process returns to step S38, where the processor 61 continues to acquire the manufacturing condition correspondence data 71 through manufacturing and evaluation.

[0117] As described above, by manufacturing a pleated body using manufacturing conditions determined by the manufacturing condition model, it is possible to stably manufacture a pleated body having a pleat shape that can be objectively judged to be good while suppressing variation.

[0118] In addition, by obtaining a manufacturing condition model, even if manufacturing conditions other than the processing machine control variables, such as the vertical distance u between the upper folding plate 45 and the lower folding plate 46, the physical property data of the sheet-like material 30 being processed, the information on the degree of curling tendency grasped by the roll section 69, the temperature information grasped by the temperature sensor 63, and the humidity information grasped by the humidity sensor 64, change, it becomes possible to stably manufacture a pleated body having a pleat shape that can be objectively judged to be good while suppressing variation.

[0119] Among the above manufacturing conditions, the setting data for the vertical gap s when the upper surface of the lower folding plate 46 and the lower end of the upper folding blade 43 are closest to each other, and the setting data for the vertical gap t when the lower surface of the upper folding plate 45 and the upper end of the lower folding blade 44 are closest to each other were used as examples to confirm how the pleat shape of the resulting pleated body changes when these setting data are changed.

[0120] Here, the gap s and gap t under the manufacturing conditions for one pleated body were set to the same values, and the manufacturing conditions were changed so that the relationship between gap s and gap t was (a)<(b)<(c)<(d)<(e) to obtain pleated bodies (a), (b), (c), (d), and (e) shown in Figure 18.

[0121] As is clear from FIG. 18, it was confirmed that there is a correlation between the gap s, the gap t, and the pleat shape of the pleated body.

[0122] Although the embodiments of the present disclosure have been described above, it will be understood that various changes in form and details can be made without departing from the spirit and scope of the present disclosure described in the claims. [Explanation of symbols]

[0123] 20 Pleated 26, 27, 28, 29 Spacing section 30 Sheet materials 35 Contact section 56 Image data 59 Evaluation Model 62 Display (output device) a Pleats pitch b Pleats height c Length of the connecting part in the second direction [Prior art documents] [Patent documents]

[0124] [Patent Document 1] JP 2016-026870 A

Claims

1. A method for evaluating the pleat shape of a pleated body (20) obtained by processing a sheet-like material (30) into a pleat shape, comprising: an extraction step of extracting features relating to the pleat shape from image data (56) of the pleated body (20); an evaluation step of evaluating the pleat shape based on the feature amount extracted in the extraction step; Equipped with The pleated body is a filter material. Evaluation method.

2. The evaluation step includes a step of comparing the feature amount extracted in the extraction step with a predetermined criterion. The evaluation method according to claim 1 .

3. The feature amount is Pleats pitch (a), Pleats height (b), When the direction in which the fold line extends is defined as a first direction, the direction in which the pleat shape continues is defined as a second direction, and a direction perpendicular to both the first direction and the second direction is defined as a third direction, the length (c) in the second direction of a connecting portion (35) connecting adjacent peaks and valleys when viewed in the first direction, A degree of bending of the connection portion when viewed in the first direction; Symmetry between the adjacent connecting portions when viewed in the first direction; When a spacing maintaining portion (26, 27, 28, 29) for maintaining the pleat spacing is provided in the pleated body, the length and width of the spacing maintaining portion are and, Combinations of these Either The evaluation method according to claim 1 or 2.

4. A method for evaluating the pleat shape of a pleated body (20) obtained by processing a sheet-like material (30) into a pleated shape, comprising: an extraction step of extracting features relating to the pleat shape from image data (56) of the pleated body (20); an evaluation step of evaluating the pleat shape based on the feature amount extracted in the extraction step; an evaluation model learning process for creating an evaluation model (59) by machine learning the relationship between the image data of the pleated body and the evaluated data using a plurality of pairs of the image data of the pleated body and the evaluated data associated with the image data of the pleated body; Equipped with In the evaluation step, the feature amount extracted in the extraction step is evaluated based on the evaluation model created in the evaluation model learning step. Evaluation method.

5. A method for evaluating the pleat shape of a pleated body (20) obtained by processing a sheet-like material (30) into a pleated shape, comprising: an extraction step of extracting features relating to the pleat shape from image data (56) of the pleated body (20); an evaluation step of evaluating the pleat shape based on the feature amount extracted in the extraction step; an input condition learning process for learning a relationship between the input conditions and the evaluation result data by using a plurality of sets of input conditions including processing condition data input to a processing machine that processes the sheet-like material into a pleated shape and physical property data related to the sheet-like material, and evaluation result data obtained by evaluating the pleated body obtained in the evaluation process corresponding to the input conditions; Equipped with Evaluation method.

6. The processing condition data includes a gap between a folding blade and a folding plate of the processing machine, a conveying speed when the processing machine conveys the material, an application amount when a resin forming a spacing retention portion is applied to the material, and any combination thereof. The evaluation method according to claim 5.

7. outputting the input conditions to an output device (62) based on the model obtained in the input condition learning step; The evaluation method according to claim 5.

8. The pleated body is manufactured using the input conditions output in the evaluation method according to claim 7. Manufacturing method.

Images