Abnormality detecting method and abnormality detecting system

Abstract

To provide an abnormality detecting method and an abnormality detecting system capable of detecting abnormality during formation of lamination without relying on the shape of a formed component or the resolution of a photographic image.SOLUTION: An abnormality detecting method for detecting abnormality while performing formation of lamination by the repetition of laying a metallic powder material in a laminar form on a forming plate and solidifying the powder material, includes: a photographing step of photographing a video image from above a forming plate while formation of lamination is being performed; a descending step of descending the forming plate after laying the powder material on the forming plate or on the powder material solidified on the forming plate; an extraction step of extracting a still image when the forming plate is descended; and an abnormality determination step of performing abnormality determination during formation of lamination based on the still image.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present disclosure relates to an anomaly detection method and an anomaly detection system. [Background technology]

[0002] Patent Document 1 describes a method for detecting anomalies that appear during 3D additive manufacturing, which manufactures products by stacking metals. In this method, a visual image of the layer is obtained before or after stacking powder material in each layer, and the gray values ​​of the pixels in the visual image are compared with reference gray values ​​of pixels in a reference image when additive manufacturing is performed without anomalies, thereby determining whether or not there is an anomaly in the layer corresponding to the visual image. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Special Publication No. 2018-538167 Summary of the Invention [Problem to be solved by the invention]

[0004] In the method described in Patent Document 1, attention is paid to each layer when a molded part is additively molded on a mold plate, and the pixel reference gray value in the reference image is compared. However, since the location and state of occurrence of an anomaly differ depending on the shape of the molded part, and the image quality of the image captured by the camera differs depending on the installation position and resolution of the camera, it is sometimes difficult to determine the presence or absence of an anomaly from the captured image of each layer of the molded part.

[0005] In view of the above circumstances, at least one embodiment of the present disclosure aims to provide an anomaly detection method and an anomaly detection system capable of detecting anomalies during additive manufacturing regardless of the shape of the molded part or the resolution of the captured image. [Means for solving the problem]

[0006] In addition, the anomaly detection method disclosed herein is an anomaly detection method for detecting an abnormality during additive manufacturing, which is performed by repeatedly laying layers of metallic powder material on a building plate and solidifying the powder material, and includes an imaging step of capturing an image from above the building plate while the additive manufacturing is being performed, a lowering step of lowering the building plate after laying the powder material on the building plate or on the powder material solidified on the building plate, an extraction step of extracting from the image a still image of when the building plate is lowered, and an anomaly judgment step of judging an abnormality during the additive manufacturing based on the still image.

[0007] In order to achieve the above-mentioned object, the anomaly detection system of the present disclosure comprises an additive manufacturing device including a shaping plate on which additive manufacturing is performed by repeatedly laying metal powder material in layers and solidifying the powder material, and a lifting device for raising and lowering the shaping plate, a photographing device for capturing images from above the shaping plate, and a control device, and the control device comprises a control unit for controlling the operation of the additive manufacturing, a descent data receiving unit for receiving from the control unit descent data that the lifting device has lowered the shaping plate in response to a command from the control unit, an extraction unit for extracting a still image from the image captured by the photographing device when the descent data receiving unit receives the descent data, and an anomaly judgment unit for making an abnormality judgment during the additive manufacturing based on the still image extracted by the extraction unit. Effect of the Invention

[0008] According to the anomaly detection method and anomaly detection system disclosed herein, a still image of when the building plate descends can be extracted from the video captured from above the building plate, and an anomaly can be judged based on the extracted still image, thereby making it possible to detect anomalies during additive manufacturing regardless of the shape of the built part or the resolution of the captured image. [Brief description of the drawings]

[0009] [Figure 1] 1 is a schematic diagram illustrating a configuration of an anomaly detection system according to an embodiment of the present disclosure. [Diagram 2] FIG. 13 is a schematic diagram illustrating a configuration of a modified example of a control device of the anomaly detection system according to an embodiment of the present disclosure. [Diagram 3] 1 is a diagram showing an example of an image above a build plate captured in an anomaly detection method according to an embodiment of the present disclosure. FIG. [Figure 4] 1 is a diagram showing an example of a still image (in a state where an abnormality has occurred) extracted from an image above a modeling plate captured in an anomaly detection method according to an embodiment of the present disclosure. FIG. [Diagram 5] FIG. 5 is a diagram showing a binarized image of the still image in FIG. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, an anomaly detection method and an anomaly detection system according to an embodiment of the present disclosure will be described with reference to the drawings. The embodiment described below shows one aspect of the present disclosure, and does not limit the disclosure. The embodiment can be modified as desired within the scope of the technical idea of ​​the present disclosure.

[0011] <Configuration of anomaly detection system according to an embodiment of the present disclosure> As shown in FIG. 1, an anomaly detection system 20 according to an embodiment of the present disclosure is a system for detecting an anomaly during an operation of layer-by-layer manufacturing of a three-dimensional product P by irradiating a metal powder material 5 laid in layers with a beam B such as a laser beam to melt and solidify the powder material 5. The anomaly detection system 20 includes an additive manufacturing apparatus 1 including a modeling plate 2 on which the powder material 5 is laid and an elevating device 3 for raising and lowering the modeling plate 2, a camera 4 for taking an image from above the modeling plate 2, and a control device 10. The additive manufacturing apparatus 1 also includes a device for laying the powder material 5, such as a coater 6, an irradiation device 7 for irradiating the beam B, and a housing 8 for housing the modeling plate 2 and the elevating device 3. The camera 4 may be installed anywhere above the modeling plate 2, but is preferably installed above the end of the modeling plate 2 (the portion corresponding to the first straight line L1 in FIG. 5 described later).

[0012] The control device 10 is a device for controlling the operation of the additive manufacturing of the product P by the additive manufacturing device 1 and for detecting an abnormality during the additive manufacturing of the product P. The control device 10 is composed of, for example, a central processing unit (CPU), a random access memory (RAM), a read only memory (ROM), and a computer-readable storage medium. As an example, a series of processes for realizing various functions is stored in a storage medium or the like in the form of a program, and the CPU reads the program into the RAM or the like and executes information processing and arithmetic processing to realize various functions. The program may be installed in a ROM or other storage medium in advance, provided in a state stored in a computer-readable storage medium, or distributed via a wired or wireless communication means. The computer-readable storage medium may be a magnetic disk, a magneto-optical disk, a CD-ROM, a DVD-ROM, a semiconductor memory, or the like.

[0013] Specifically, the control device 10 includes a control unit 11 that controls the additive manufacturing operation of the product P by the additive manufacturing device 1, a descent data receiving unit 12 that receives from the control unit 11 descent data that the lifting device 3 has lowered the modeling plate 2 in response to a command from the control unit 11, an extraction unit 13 that extracts still images (to be described later) from the video captured by the imaging device 4, an abnormality determination unit 14 that performs an abnormality determination for the additive manufacturing based on the still images extracted by the extraction unit 13, a device 15 that notifies the abnormality when the abnormality determination unit 14 determines an abnormality in the additive manufacturing, such as a display that displays the abnormality, a speaker that notifies the abnormality by sound, voice, music, etc., and a storage unit 16 that stores a program for the control unit 11 to control the additive manufacturing operation of the product P and a program for the extraction unit 13 to extract a still image from the video captured by the imaging device 4. The storage unit 16 may be a hard disk or a memory, or may be a storage medium in which the program is stored and a reading device that reads the program from the storage medium.

[0014] 2, the control device 10 may be configured to include a first control device 10a and a second control device 10b, which are separate devices. The first control device 10a may include a control unit 11, a descent data receiving unit 12, an extraction unit 13, a device 15 for notifying an abnormality, and a storage unit 16, and the second control device 10b may include an abnormality determination unit 14. The first control device 10a may further include a data transmission / reception unit 17 that transmits data of the still image extracted by the extraction unit 13 to the abnormality determination unit 14 of the second control device 10b and receives data of the determination result in the abnormality determination unit 14.

[0015] For example, the first control device 10a may be a computer installed at the site where the additive manufacturing of the product P is performed, and the second control device 10b may be a server installed at a location away from the site where the additive manufacturing of the product P is performed and performing communication control within a factory including the site where the additive manufacturing of the product P is performed. The second control device 10b may be a computer installed outside the factory including the site where the additive manufacturing of the product P is performed, or may be a cloud server. When the control device 10 is configured to include the first control device 10a and the second control device 10b, it is necessary to transmit still image data from the first control device 10a to the second control device 10b, so the program for extracting still images from the video captured by the imaging device 4 may also include a program for transmitting still image data via wired or wireless communication.

[0016] <Operation of anomaly detection system according to an embodiment of the present disclosure (anomaly detection method)> Next, an operation (anomaly detection method) of the anomaly detection system 20 according to an embodiment of the present disclosure will be described. This operation is an operation for detecting an anomaly during additive manufacturing of the product P.

[0017] First, the operation of layered manufacturing of the product P will be described. This operation is controlled by the control unit 11 according to a program for controlling the operation of layered manufacturing of the product P. Based on a command from the control unit 11, the coater 6 lays the powder material 5 in layers on the modeling plate 2. Based on a command from the control unit 11, the lifting device 3 lowers the modeling plate 2 by a predetermined distance. Based on a command from the control unit 11, the irradiation device 7 irradiates the laid powder material 5 with the beam B. As a result, at least a part of the powder material 5 constituting the layer is melted and solidified. The melted and solidified part constitutes a part of the product P. After this, the process returns to the process of laying the powder material 5 on the modeling plate 2 by the coater 6, and the above-mentioned steps are repeated to layered manufacture the product P.

[0018] Next, an operation for detecting an abnormality during additive manufacturing of the product P (an abnormality detection method) will be described. During the above-mentioned additive manufacturing operation of the product P, the image capturing device 4 captures an image from above the shaping plate 2, particularly an image including the end of the shaping plate 2 (the portion corresponding to the first straight line L1 in FIG. 5 described later). The captured image data is transmitted to the extraction unit 13. Alternatively, the image data may be transmitted to the storage unit 16 and stored in the storage unit 16. When the image capturing device 4 is provided directly above the shaping plate 2, an image of the shape of the shaping plate 2, for example, if the shaping plate is square, an image of the square (corresponding to an image of at least one of the upper surface of the shaping plate 2, the powder material 5, or a part of the product P) is captured. When the camera 4 is provided outside the housing 8 and captures an image of the upper part of the shaping plate 2 through a window (not shown) provided in the housing 8, the camera 4 captures an image from diagonally above the shaping plate 2, and if the shaping plate 2 has a square shape, a trapezoidal image is captured as shown in Fig. 3. Note that the higher the position in Fig. 3, the farther the camera 4 is from the position. In this embodiment, the latter case will be described as an example.

[0019] As shown in FIG. 1, after the control unit 11 transmits a command to the lifting device 3 to lower the shaping plate 2, the control unit 11 transmits descent data to the descent data receiving unit 12. The descent data receiving unit 12, having received the descent data, transmits a command to the extraction unit 13 to extract a still image from the video data. The extraction unit 13 extracts a still image from the video transmitted from the imaging device 4 or the video stored in the memory unit 16 according to a program for extracting a still image. The extracted still image should include the end of the shaping plate 2 (the part corresponding to the first straight line L1 in FIG. 5 described later). The still image data extracted by the extraction unit 13 is transmitted to the abnormality determination unit 14.

[0020] The anomaly determination unit 14 performs an anomaly determination for the additive manufacturing of the product P based on the still image. Here, the "anomaly" determined by the anomaly determination unit 14 will be described. If the product P has a sharp part during additive manufacturing and the sharp part protrudes above the upper surface of the housing 8, when the powder material 5 is laid by the coater 6 (see FIG. 1), the powder material 5 from the coater 6 may come into contact with the sharp part protruding above the upper surface of the housing 8, causing the laying of the powder material 5 to be disturbed. Specifically, as shown in FIG. 4, streaks S1 and S2 are generated on the upper surface of the powder material 5 downstream of parts P1 and P2 of the sharp part from the upper surface of the powder material 5 in the direction D0 in which the coater 6 moves. If such streaks S1 and S2 are generated, there is a risk that each layer constituting the product P cannot be formed in the correct shape. In this embodiment, such a state is defined as "anomaly during additive manufacturing of the product P". If the laying of powder material 5 from coater 6 becomes disordered, then every time powder material 5 is laid thereafter, streaks will appear on the upper surface of laid powder material 5 parallel to the direction D0 in which coater 6 moves, even if a portion of the pointed portion of powder material 5 does not protrude from the upper surface of housing 8 above. For this reason, even if an abnormality cannot be detected immediately after the laying of powder material 5 from coater 6 becomes disordered, it can be detected somewhere in the subsequent process.

[0021] Next, the operation of abnormality determination will be described. As shown in FIG. 1, the still image extracted by the extraction unit 13 is an image of the state in which the shaping plate 2 has been lowered after the powder material 5 has been laid, so that a straight line corresponding to the step between the upper surface of the housing 8 and the upper surface of the powder material 5 laid on the shaping plate 2 is captured in the still image. The abnormality determination unit 14 first binarizes the still image. The above-mentioned straight lines are detected from the binarized image. When the shape of the shaping plate 2 is a square, as shown in FIG. 5, four straight lines can be detected, but it is set to detect one of two straight lines (first straight line L1) that are perpendicular to the direction D0 in which the coater 6 (see FIG. 1) moves. These four straight lines correspond to the above-mentioned step, so that the positions that appear in the still image can be specified within a certain range. Note that, since deviations may occur in the position or length of the first straight line L1 that appears in the still image due to differences in the shooting angle by the shooting device 4 (see FIG. 1), a step of correcting at least one of the position or length of the first straight line L1 before abnormality determination may be added. In order to correct the first straight line L1 in a still image, a reference still image is prepared in advance, and the range of deviation from the reference image is specified. By correcting the position and length of the first straight line L1 in advance in this way, it is possible to eliminate the effects on the still image caused by differences in the installation position and shooting angle of the image capture device 4.

[0022] When an abnormality as shown in FIG. 4 occurs (streaks S1 and S2 occur), second straight lines L2 and L3 corresponding to the streaks S1 and S2 are seen in FIG. 5. Considering the direction D0 in which the coater 6 (see FIG. 1) moves, the second straight lines L2 and L3 can be detected as straight lines extending in a direction intersecting the first straight line L1. In many cases, the direction D0 in which the coater 6 moves is perpendicular to the first straight line L1 of the printing plate, so the second straight lines L2 and L3 appear in a direction perpendicular to the first straight line L1. When such perpendicular second straight lines L2 and L3 (streaks S1 and S2) occur, it can be reliably determined that there is a printing abnormality, and the risk of erroneous detection can be reduced. Note that the perpendicular angle described here is not limited to an angle of 90 degrees, but includes a range from 85 degrees to 95 degrees. When the abnormality determination unit 14 detects such second straight lines L2 and L3, it determines that an abnormality has occurred. When the abnormality determination unit 14 determines that an abnormality has occurred, it displays an abnormality message on a display serving as an abnormality notification device 15, and / or emits a sound, voice, music, or the like to notify the abnormality from a speaker serving as an abnormality notification device 15. This allows the occurrence of an abnormality to be known.

[0023] In this way, a still image of the shaping plate 2 when it descends can be extracted from the video captured from above the shaping plate 2, and an abnormality can be determined based on the extracted still image. In other words, an abnormality during additive manufacturing can be detected not from an image of the shaping part, but from an image of the shaping plate generated in conjunction with the operation of the shaping device, so that an abnormality during additive manufacturing can be detected regardless of the shape of the shaping part or the resolution of the captured image.

[0024] In the above explanation, the number of sharp portions P1 and P2 was two, and therefore the number of lines S1 and S2 and the number of second straight lines L2 and L3 were also two, but depending on the number of sharp portions, the number of second straight lines may be one or three or more.

[0025] 2, when the control device 10 has a configuration including a first control device 10a and a second control device 10b, the still image data extracted by the extraction unit 13 is transmitted to the data transmission / reception unit 17, which transmits the still image data to the abnormality determination unit 14 of the second control device 10b. The abnormality determination unit 14 receives the still image data and performs an abnormality determination during additive manufacturing based on the still image by the same operation as described above. The abnormality determination unit 14 transmits the result of the abnormality determination to the data transmission / reception unit 17, which causes the device 15 that notifies an abnormality to display the result, as described above.

[0026] With such a configuration, the variation in usage patterns of the anomaly determination unit 14, which requires high computing power, can be expanded, and depending on the usage pattern, it is possible to reduce the cost of the anomaly detection system 20. In particular, when the second control device 10b is a cloud server, by using the anomaly determination unit 14 provided in the cloud server, it becomes unnecessary to provide an anomaly determination unit 14, which requires high computing power, for each anomaly detection system 20, and therefore it is possible to reduce the cost of the anomaly detection system 20.

[0027] The contents described in each of the above embodiments can be understood, for example, as follows.

[0028] [1] An anomaly detection method according to one aspect includes: 1. A method for detecting an abnormality during additive manufacturing, which is performed by repeatedly laying a layer of a powdered metal material (5) on a modeling plate (2) and solidifying the powdered metal material (5), comprising the steps of: a photographing step of photographing an image from above the modeling plate (2) while the additive manufacturing is being performed; a lowering step of lowering the shaping plate (2) after laying the powder material (5) on the shaping plate (2) or on the powder material (5) solidified on the shaping plate (2); an extraction step of extracting a still image when the shaping plate (2) is lowered from the image; an abnormality determination step of determining an abnormality during the additive manufacturing process based on the still image; Includes.

[0029] According to the anomaly detection method disclosed herein, a still image of when the building plate descends can be extracted from the video captured from above the building plate, and an abnormality can be judged based on the extracted still image, making it possible to detect abnormalities during additive manufacturing regardless of the shape of the built part or the resolution of the captured image.

[0030] [2] An anomaly detection method according to another aspect is the anomaly detection method according to [1], The abnormality determination step includes: binarizing the still image; A step of detecting a first straight line (L1) that is predicted to appear from the binarized still image; A step of detecting the presence or absence of at least one second straight line (L2, L3) that extends in a direction intersecting the first straight line (L1) from the binarized still image; determining that an abnormality has occurred when the second straight line (L2, L3) is detected; Includes.

[0031] According to this method, abnormalities can be judged based on still images taken from above the building plate, making it possible to detect abnormalities during additive manufacturing regardless of the shape of the part being built or the resolution of the captured image.

[0032] [3] A method for detecting an anomaly according to yet another aspect is the method for detecting an anomaly according to [2], The second straight line (L2, L3) extends in a direction perpendicular to the first straight line (L1).

[0033] According to this method, when the second straight line occurs, it can be reliably determined that the second straight line is a printing abnormality, and the risk of erroneous detection can be reduced.

[0034] [4] A method for detecting an anomaly according to yet another aspect is the method for detecting an anomaly according to [2] or [3], The abnormality determination step includes a step of correcting at least one of the position and the length of the first straight line (L1) before performing the abnormality determination.

[0035] According to this method, it is possible to eliminate the influence on a still image caused by differences in the installation position of the image capturing device and the image capturing angle.

[0036] [5] An anomaly detection system according to one aspect includes: An additive manufacturing device (1) including a modeling plate (2) on which additive manufacturing is performed by repeatedly laying a metal powder material (5) in layers and solidifying the powder material (5), and a lifting device (3) for lifting and lowering the modeling plate (2); a photographing device (4) for photographing an image from above the shaping plate (2); A control device (10) Equipped with The control device (10) A control unit (11) for controlling the operation of the additive manufacturing; a lowering data receiving unit (12) that receives, from the control unit (11), lowering data indicating that the lifting device (3) has lowered the shaping plate (2) in response to a command from the control unit (11); an extraction unit (13) that extracts a still image at the time when the descent data receiving unit (12) receives the descent data from the video captured by the imaging device (4); an abnormality determination unit (14) that determines an abnormality during the additive manufacturing process based on the still image extracted by the extraction unit (13); Equipped with.

[0037] According to the anomaly detection system disclosed herein, a still image of when the building plate descends can be extracted from the video captured from above the building plate, and an anomaly can be judged based on the extracted still image, making it possible to detect anomalies during additive manufacturing regardless of the shape of the built part or the resolution of the captured image.

[0038] [6] An anomaly detection system according to another aspect is the anomaly detection system according to [5], The control device (10) a first control device (10a) including the control unit (11), the descent data receiving unit (12), and the extraction unit (13); a second control device (10b) different from the first control device (10a), the second control device including the abnormality determination unit (14); Equipped with.

[0039] According to such a configuration, it is possible to expand the variety of usage patterns of the anomaly determination unit, which requires high computing power, and therefore it is possible to reduce the cost of the anomaly detection system depending on the usage pattern.

[0040] [7] A system for detecting an anomaly according to yet another aspect is the system for detecting an anomaly according to [6], The second control device (10b) is a cloud server.

[0041] According to this configuration, by utilizing an anomaly judgment unit provided in the cloud server, there is no need to provide an anomaly judgment unit that requires high computing power for each anomaly detection system, thereby making it possible to reduce the cost of the anomaly detection system. [Explanation of symbols]

[0042] 1 Additive manufacturing equipment 2 Modeling plate 3 Lifting device 4. Imaging Device 5 Powder material 10 Control device 10a First control device 10b Second control device 11 Control section 12 Descent data receiver 13 Extraction part 14 Abnormality judgment section 20 Anomaly Detection System L1 1st straight line L2,L3 second straight line

Claims

1. 1. A method for detecting an anomaly during additive manufacturing, which is performed by repeatedly laying a layer of powdered metal material on a modeling plate and solidifying the powdered metal material, comprising: An imaging step of capturing an image from above the building plate while the additive manufacturing is being performed; a lowering step of lowering the building plate after laying the powder material on the building plate or on the powder material solidified on the building plate; An extraction step of extracting a still image when the building plate is lowered from the video; an abnormality determination step of determining an abnormality during the additive manufacturing process based on the still image; The anomaly detection method includes:

2. The abnormality determination step includes: binarizing the still image; A step of detecting a first straight line, the first straight line being predicted to appear, from the binarized still image; detecting, from the binarized still image, the presence or absence of at least one second straight line extending in a direction intersecting the first straight line; determining that an abnormality has occurred when the second straight line is detected; The anomaly detection method according to claim 1 , comprising:

3. The anomaly detection method according to claim 2 , wherein the second straight line extends in a direction perpendicular to the first straight line.

4. 4. The anomaly detection method according to claim 2, wherein the anomaly determination step includes a step of correcting at least one of a position or a length of the first straight line before performing the anomaly determination.

5. an additive manufacturing device including a modeling plate on which additive manufacturing is performed by repeatedly laying a metal powder material in layers and solidifying the powder material, and a lifting device that lifts and lowers the modeling plate; A photographing device that photographs an image from above the shaping plate; Control device and Equipped with The control device includes: A control unit that controls the operation of the additive manufacturing method; a lowering data receiving unit that receives, from the control unit, lowering data indicating that the lifting device has lowered the modeling plate in response to a command from the control unit; an extracting unit that extracts a still image when the descent data receiving unit receives the descent data from the video captured by the imaging device; an anomaly determination unit that determines an anomaly during the additive manufacturing process based on the still image extracted by the extraction unit; An anomaly detection system comprising:

6. The control device includes: A first control device including the control unit, the descent data receiving unit, and the extraction unit; a second control device different from the first control device, the second control device including the abnormality determination unit; The anomaly detection system of claim 5 .

7. The anomaly detection system according to claim 6 , wherein the second control device is a cloud server.

Images