JP7051619B2 - Molded product take-out machine equipped with an image pickup device - Google Patents

Description

The present invention relates to a molded product take-out machine equipped with an image pickup device.

Japanese Patent Application Laid-Open No. 2003-136566 (Patent Document 1) discloses that an image pickup device as a human detector is provided on a main body frame on the release position side of a molded product.
Further, Japanese Patent Application Laid-Open No. 2007-326270 (Patent Document 2) is provided with an image pickup member that is fixedly attached to the release position side of the main body frame and images a chuck moving at the release position and outputs image quality data. Is disclosed.

Further, Japanese Patent Application Laid-Open No. 2004-306490 (Patent Document 3) discloses a molded product take-out machine in which an image pickup device for inspecting a molded product is attached to the tip of a mounting arm (traverse frame) for inspection of the molded product. Has been done.

Further, Japanese Patent Application Laid-Open No. 2006-334836 (Patent Document 4) discloses that an image pickup device is used as a distance measuring means for measuring a distance to an obstacle within a movable range of a drawing frame. This publication describes that the distance measuring means is provided at each portion of the tip of the transfer arm, the tip of the elevating arm, the tip of the runner elevating arm, and the tip of the take-out head in a desired transverse direction. There is.

Japanese Patent Application Laid-Open No. 2003-136566 Japanese Unexamined Patent Publication No. 2007-326270 Japanese Unexamined Patent Publication No. 2004-306490 Japanese Unexamined Patent Publication No. 2006-334836

The inventions shown in Patent Documents 1 to 3 utilize an image pickup device for the purpose of human intrusion, product inspection, and the like. Further, the invention shown in Patent Document 4 utilizes an image pickup device for the purpose of measuring the distance to an obstacle. The mounting position of the image pickup apparatus in the conventional invention is selected as the optimum position for achieving the object of those inventions. Since the image pickup device provided for a specific purpose is provided so as to limit the imaging target, it is difficult to obtain information other than the purpose.

In addition, when the supervisor or manager who is responsible for the entire factory remotely monitors the situation in the factory, he / she will see the images of many image pickup devices installed in various places in the factory. The large number of image pickup devices imposes a heavy burden on the viewer of the image.

An object of the present invention is to provide a molded product ejecting machine capable of obtaining a large amount of monitor information from a still image or a moving image from at least one monitor image pickup device.

Another object of the present invention is to provide a molded product ejecting machine capable of viewing monitor information and an image at the same time.

The object of the present invention is a molded product take-out machine installed on the molding machine. In the present invention, at least one monitor image pickup device is attached to a movable frame that moves during a take-out operation or a traveling body for moving the movable frame. At least one monitor image pickup device is a part of the molding die of the molding machine and a part thereof in a still image or a moving image taken at least when the molded product is taken out, when the molded product is opened, and when the movable frame is retracted. The mounting position, angle, viewing angle and focal length are defined so as to include objects present around the. The still image or moving image obtained by the present invention includes at least a part of the molding die of the molding machine and the surroundings of the part when the molded product is taken out, the molded product is opened, and the movable frame is retracted. included. Still images or moving images taken for a limited purpose include images suitable for that purpose, but are unlikely to contain images from which other information can be obtained. On the other hand, the still image or moving image obtained by the present invention is molded as long as it contains an image of a part of the molding die and its surroundings at least at the time of taking out, opening, and retracting the movable frame. Since it is an image of the central area of the operation of the machine and the molded product take-out machine, it is urgent if the rate of defective products increases, the cycle time for taking out the molded product is longer than the set value, or it is urgent. If the number of outages is increasing, or if the mold replacement work is taking a long time, it is likely that you can get a hint as to where the cause is. Therefore, according to the present invention, a large amount of monitor information can be obtained from a still image or a moving image from at least one monitor image pickup device.

The image pickup device for a monitor may be configured so that the angle, viewing angle and focal length can be adjusted. If the image pickup device for a monitor has such a function, the image range of the obtained still image or moving image can be adjusted, and the versatility is increased. Further, the image pickup device for a monitor may be equipped with, for example, an infrared camera or the like, regardless of the type, as long as the above-mentioned necessary information can be obtained.

Furthermore, if the monitor image pickup device is configured to automatically adjust the angle, viewing angle, and focal length according to the stop position or movement path of the movable frame or the traveling body, more information can be obtained. become.

When the movable frame moves laterally along the transverse frame extending laterally and the elevating frame that moves in the vertical direction and the front-rear direction orthogonal to the lateral direction is a pull-out frame that is movably mounted in the front-rear direction. When the elevating frame moves in the front-rear direction and is farthest from the traverse frame, it is defined as the maximum movement position on the extraction frame. It is preferably installed at the end of the drawing frame located on the tip side of the drawing frame. When the position of one monitor image pickup device is determined in this way, the still image or moving image obtained from this monitor image pickup device includes information on taking out the molded product, information on opening the molded product, and a molding die. It will contain most of the information when exchanging, information about objects approaching the perimeter of the part take-out machine, and information about the movement of the operator. Therefore, it is possible to obtain various information when the molded product ejecting machine is operating, based on a still image or a moving image from one monitor image pickup device.

Further, one monitor image pickup device may be installed at the bottom portion of the extraction frame located closer to the base side of the extraction frame than the maximum moving position. At this position, although the presence of the elevating frame is an obstacle, various information can be obtained when the molded product ejector is operating based on the still image or moving image from one monitor image pickup device. ..

Further, a monitor image pickup device may be installed on the first traveling body that moves the elevating frame in the front-rear direction. By doing so, it is possible to obtain information when the elevating frame is moving in the front-rear direction.

Further, a monitor image pickup device may be installed on the second traveling body that moves the extraction frame in the lateral direction. Also in this case, various information can be obtained when the molded product ejecting machine is operating, based on the still image or the moving image from the image pickup device for the monitor.

The monitor image pickup device is preferably arranged within the end of the extraction frame. By doing so, the existence of the image pickup device itself becomes inconspicuous, and the operator does the work without being aware of the image pickup device, so that the normal work state can be known from the image data.

It is preferable that the electric wire connected to the image pickup apparatus is housed inside the drawing frame. By doing so, there is no possibility that the electric wire will be cut, so that the reliability is greatly improved.

The movable frame on which the image pickup device for a monitor is mounted may be an elevating frame that moves laterally along a transverse frame extending in the lateral direction and moves in the vertical direction and in the front-rear direction orthogonal to the lateral direction. The size of the image obtained by mounting the image pickup device for a monitor on the elevating frame changes, but various information can be obtained even if the image changes in size.

The display mode of the information from one or more monitor image pickup devices is arbitrary. For example, the molded product ejector analyzes a display unit provided with a display screen, a detection unit that detects the operating status of one or more types of controlled devices included in the molded product ejector, and detection data detected by the detection unit. Operation status detection analysis storage unit consisting of an analysis unit and a data storage unit that stores one or more types of analysis data analyzed by the analysis unit as display information, and a data storage unit of the operation status detection analysis storage unit. An information display system including a display control unit for displaying the displayed information on the display screen may be further provided. In this case, the data storage unit stores a still image or a moving image as display information of image data. The display control unit preferably displays both the display information of the analysis data stored in the data storage unit and the display information of the image data in synchronization with the display screen. When such a synchronized display is performed, a person who has some doubts about the operation or work in the displayed information may be able to find the cause of the doubt by looking at the necessary image while checking the analysis information. It gets higher.

When one or more monitor image pickup devices are attached to a movable frame that moves during a take-out operation or a traveling body for moving the movable frame, the one or more monitor image pickup devices are at least a molded product. The mounting position, angle, and viewing angle so that the still image or moving image taken at the time of taking out the molded product and opening the molded product includes a part of the molding mold of the molding machine and an object existing around the part. And the focal length is preferably defined.

When a device that counts the cycle time of the molded product take-out machine is used as the detection unit of the operation status detection analysis storage unit, the analysis unit detects a plurality of time-series cycle times counted within a predetermined period. As a display information, the number of cycles including a stop time longer than necessary in one cycle and the time of occurrence may be analyzed. By doing so, it is possible to know how much the stop (chocolate stop) that has not been set occurs during the so-called one cycle. If it is possible to know the occurrence status of a stop (chocolate stop) that has not been set during this one cycle, it is possible to know a setting error or a device failure.

A machine learning device that learns the conditions associated with the occurrence of a defect in the molded product take-out machine may be further provided in order to know a setting error or a failure of the device. In this case, the detection unit further includes a monitor image pickup device that detects the operating state of the molded product take-out machine as an image. Then, the machine learning device observes a state variable including data on the number of cycles including a stop time longer than necessary, data on the occurrence time, data on the stop time, and image data of the operation status in one cycle detected by the detection unit. According to the training data set created based on the combination of the state observation unit, the judgment data acquisition unit that acquires the judgment data that determines the degree of failure of the molded product ejector, and the state variable and the determination data, the molded product ejector. It is equipped with a learning unit that learns the conditions associated with the occurrence of problems. If such a machine learning device is provided, the molded product unloading machine can actually set the conditions associated with the occurrence of a defect in the molded product unloading machine according to the training data set created based on the combination of the state variables and the judgment data. Since learning is performed while operating, accurate failure occurrence conditions according to actual usage conditions are learned.

It is a perspective view which shows the whole structure of the molded article take-out machine of embodiment of this invention. It is a perspective view which showed the tip of the drawing frame in an enlarged manner. (A) is a perspective view showing a state in which the elevating frame is located on the open mold of the molding machine, and (B) is the elevating frame on the pallet of the product stocker device arranged next to the molding machine. It is a perspective view which shows the state which is positioned. An example of a simulated image of an image taken by an image pickup device when a drawing frame is placed on a mold of a molding machine is shown. It is a simulated figure of a still image taken by a monitor image pickup device when the extraction frame is located on the pallet of the product stocker device and the take-out head is located above the pallet. It is a simulated figure of the still image taken by the image pickup apparatus for a monitor when the drawing frame is in the retracted position. It is a simulated view of the still image taken by the image pickup device for a monitor when the molding die was removed from the molding machine, the product stocker device was moved, and the extraction frame was in the retracted position for the replacement of the molding die. It is a figure which shows another example of the mounting position of the image pickup apparatus for a monitor. It is a perspective view which shows the other example of the reversing unit attached to the tip of the elevating frame in another embodiment of the molded article take-out machine of this invention. It is a block diagram which shows the structure of an example of the information display system which displays the information from one or more monitor image pickup apparatus. It is a figure which shows the display example of a display screen. It is a block diagram for demonstrating the structure of the machine learning apparatus.

Hereinafter, embodiments of the molded product take-out machine of the present invention will be described in detail with reference to the drawings. FIG. 1 is a perspective view of the overall configuration of the molded product take-out machine 1 of the present embodiment as viewed from different angles. The molded product take-out machine 1 is a traverse-type molded product take-out machine, and the base is supported by a fixed platen of a molding machine (not shown). The molded product take-out machine 1 shown in FIG. 1 includes a traversing frame 3, a control box 5, a drawing frame 7, a runner elevating unit 9, and a molded product adsorbing elevating unit 11. The traversing frame 3 has a cantilever beam structure extending in the X-axis direction horizontally orthogonal to the longitudinal direction of the molding machine (not shown). The control box 5 is supported by the traversing frame 3, and the traveling body 6 moves back and forth along the traversing frame 3 in the X-axis direction using an AC servomotor (not shown) included in the servo mechanism as a drive source. The drawing frame 7, which is a movable frame, has a base attached to the traveling body 6 and extends in the Y-axis direction parallel to the longitudinal direction of the molding machine (not shown). In the drawing frame 7, a runner elevating unit 9 and a molded product adsorbing elevating unit 11 are supported so as to be movable in the Y-axis direction using an AC servomotor (not shown) included in the servo mechanism as a drive source. A control box 8 is attached to the base of the drawing frame 7.

The elevating unit 9 for a runner has a structure in which a traveling body 15 movably supported by a pull-out frame 7 is provided with an elevating frame 17 that elevates and elevates in the Z-axis direction. The traveling body 15 is driven by an AC servomotor (not shown) and moves in the Y-axis direction. The elevating frame 17 is moved up and down in the vertical direction (Z-axis direction) by the AC servomotor 19. The elevating frame 17 includes a chuck 21 as an attachment for holding the runner to be discarded.

Further, the traveling body 23 included in the elevating unit 11 for adsorbing the molded product moves on the drawing frame 7 in the Y-axis direction by an AC servomotor (not shown). The elevating unit 11 for adsorbing molded products includes an elevating frame 25 as a movable frame that elevates and elevates in the vertical direction (Z-axis direction), and an inversion unit 27 that constitutes an attitude control device provided at the lower end of the elevating frame 25. .. The take-out head 28 shown by the broken line is attached to the reversing unit 27. Note that FIGS. 3 to 5 described later show the take-out head 28.

As shown in FIG. 1, the drawing frame 7 moves laterally along the transverse frame 3 extending in the lateral direction (X-axis direction). An elevating frame 17 and 25 that move in the vertical direction (Z-axis direction) and the front-rear direction (Y-axis direction) orthogonal to the lateral direction are mounted on the pull-out frame 7 so as to be movable in the front-rear direction.

In this embodiment, the drawing frame 7 has a hollow structure formed of an iron-based material, and the outer wall of the drawing frame 7 has one side wall 29 and the other side wall 31 facing the one side wall 29. The upper connecting plate 33 and the lower connecting plate 35 that connect one side wall 29 and the other side wall 31 on the upper and lower sides are welded to each other to form four side surfaces. The tip 7A , which is the free end of the drawing frame 7, is closed by the closing plate 37, and the control box 8 is attached to the rear end, which is the fixed end (base).

One side wall 29 of the drawing frame 7 has one flat side surface so as to allow the elevating frames 17 and 25 of the drawing frame 7 to move in the front-rear direction. Further, the other side surface of the other side wall 31 has an inclined surface 39 that is inclined so as to approach one side surface 30 of the one side wall 29 from the middle toward the tip 7A . A pair of guide rails 41, 41 for guiding the movement of the elevating frames 17, 25 are provided along the upper and lower edges of the flat side surface 30 of one side wall 29. Between the guide rails 41 and 41, a linear movement mechanism using a belt 43 for moving the elevating frames 17 and 25 along the pull-out frame 7 is arranged. The belt 43 is partially covered with the belt cover 45.

As shown in FIG. 2, a cap 47 that covers the tip of the guide rail 41 and the tip of the belt cover 45 is provided adjacent to the closing plate 37 at the tip of the drawing frame 7. In the present embodiment, when the elevating frames 17 and 25 move in the front-rear direction and are at the position farthest from the traverse frame 3, the maximum movement position on the extraction frame is defined as the maximum movement position. A monitor image pickup device 49 composed of a monocular camera is installed in a cap 47 (corresponding to the end of the extraction frame 7) located on the tip end side of the frame 7 downward. One monitor image pickup device 49 is a part of the molding die of the molding machine and a part thereof in a still image or a moving image taken at least when the molded product is taken out, when the molded product is opened, and when the drawing frame 7 is retracted. The mounting position, angle, viewing angle and focal length are defined so as to include objects present around the.

The still image or moving image obtained by one monitor image pickup device 49 of the present embodiment includes at least a part of the molding die of the molding machine and its surroundings at the time of taking out the molded product and opening the molded product. Includes those that exist in. Still images or moving images taken for a limited purpose include images suitable for that purpose, but are unlikely to contain images from which other information can be obtained. On the other hand, the still image or the moving image obtained by one monitor image pickup device 49 of the present embodiment is a part of the molding die and a part thereof at least at the time of taking out, opening, and retracting the movable frame. If the image includes the surrounding image, it is an image of the area that is the center of the operation and work of the molding machine and the molded product take-out machine. If the time is longer than the set value, or if the number of emergency stops is increasing, it is likely that you can get a hint as to where the cause is. Therefore, according to the present embodiment, a large amount of monitor information can be obtained from a still image or a moving image from one monitor image pickup device 49. Further, when the monitor image pickup device 49 is provided as in the present embodiment, the existence of the monitor image pickup device 49 itself becomes inconspicuous, and the operator can perform the work without being aware of the monitor image pickup device 49. Therefore, it is possible to know the normal working state from the image data. Further, in the present embodiment, the electric wire 51 connected to the monitor image pickup device 49 is housed inside the extraction frame 7. By doing so, there is no possibility that the electric wire 51 will be cut, so that the reliability is greatly improved.

3A and 3B are perspective views showing a state in which the molded product take-out machine 1 is installed on the platen 56 of the molding machine 53. Note that in FIGS. 3A and 3B, the shape of the drawing frame 7 is simply shown, so that the shape is different from the shape of the drawing frame shown in FIGS. 1 and 2. FIG. 3A shows a state in which the elevating frames 17 and 25 are positioned (taken out state) on the molding die 55 in which the molding machine 53 is open, and FIG. 3B shows the side of the molding machine 53. It shows a state (open state) in which the elevating frames 17 and 25 are located on the pallet 59 of the product stocker device 57 arranged in. As described above, the mounting position, angle (direction in which the lens is directed), viewing angle and focal length of the monitor image pickup device 49 are taken at least when the molded product is taken out, when the molded product is opened and when the drawer frame 7 is retracted. It is defined that the still image or the moving image includes a part of the molding die 55 of the molding machine 53 and an object (product stocker device 57) existing around the part.

FIG. 4 simulates an example of a still image taken by a monitor image pickup device 49 when the extraction frame 7 is on the molding die 55 of the molding machine 53 of FIG. 3 (A) (at the time of taking out). .. As can be seen from FIG. 4, from this still image, it is possible to know not only the information on the status of the molding die 55 of the molding machine 53 but also the information on the status on both sides of the molding machine 53.

Further, FIG. 5 shows a monitor image pickup device 49 when the extraction frame 7 is located on the pallet 59 of the product stocker device 57 (when opened) and when the take-out head 28 is located above the pallet 59. The simulated figure of the still image taken in is shown. From this still image, not only the information on the status of the molding die 55 but also the information on the status of the product stocker device 57 can be known.

FIG. 6 shows a simulated still image taken by the monitor image pickup device 49 when the extraction frame is in the retracted position. From this still image, it is possible to know information on the status of the molding die 55, information on the status of the product stocker device 57, and information on the status around the product stocker device 57. FIG. 7 shows a still image taken by the monitor image pickup device 49 when the molding die is removed from the molding machine 53, the product stocker device is moved, and the extraction frame is in the retracted position for the replacement of the molding die. A simulated diagram is shown. From this still image, it is possible to know information on the surrounding situation when the molding die 55 of the molding machine 53 is installed.

When the position and angle of the monitor image pickup device 49 are determined as in the present embodiment, the image data obtained from the monitor image pickup device 49 includes information when the molded product is taken out from the molding die 55 of the molding machine 53. It contains most of the information about releasing the taken-out molded product in the open position, the information when the extraction frame is in the retracted state, and the information about obstacles approaching the periphery of the molded product take-out machine. Therefore, according to the present embodiment, various information can be obtained when the molded product take-out machine is moving, based on the image data from one monitor image pickup device 49.

In the above embodiment, the monitor image pickup device 49 is arranged inside the cap 47, but it is needless to say that the monitor image pickup device 49 may be arranged outside the cap 47. Further, in the above embodiment, the monocular camera is used as the image pickup device 49 for the monitor, but a compound eye camera may be used, or a camera equipped with a zoom lens may be used. Further, the image pickup device 49 for a monitor may be a device for shooting a moving image.

(Position of image pickup device for monitor)
FIG. 8 is a diagram showing another example of the mounting position of the image pickup device for a monitor. FIG. 8 shows an example in which three monitor image pickup devices 49A to 49C are installed in addition to the monitor image pickup device 49 of the above embodiment.

The monitor image pickup device 49A is installed at the bottom portion of the extraction frame 7 located on the base side of the extraction frame 7 with respect to the maximum movement position of the elevating frame 25. At this position, although the presence of the elevating frame 25 is an obstacle, various information can be obtained when the molded product take-out machine is operating based on the still image or the moving image from the monitor image pickup device 49A.

Further, the monitor image pickup device 49B is installed on the first traveling body 23 that moves the elevating frame 25 in the front-rear direction. By doing so, it is possible to obtain information when the elevating frame 25 is moving in the front-rear direction.

Further, the monitor image pickup device 49C is installed in the second traveling body 6 that moves the extraction frame 7 in the lateral direction. Also in this case, various information can be obtained when the molded product ejector is operating, based on the still image or the moving image from the monitor image pickup device 49C. Even when one monitor image pickup device is provided among the four monitor image pickup devices 49, 49A to 49C shown in FIG. 8, the monitor image pickup device is at least when the molded product is taken out and when the molded product is opened. The mounting position, angle, viewing angle, and focal length so that the still image or moving image taken when the movable frame is retracted includes a part of the molding die of the molding machine and an object existing around the part. Should be determined.

FIG. 9 shows another example of the reversing unit 27 attached to the tip of the elevating frame 25 in another embodiment of the molded product take-out machine of the present invention. In the present embodiment, the reversing unit 27 is provided with a monitor image pickup device 49D having a lens 49Da in the opening / closing direction of the molding die 55. In this embodiment, the monitor image pickup device 49D is installed in the component storage space 34C formed between the two air cylinders 34A and 34B in which the air cylinders serving as the drive source 34 of the reversing unit 27 are arranged at intervals. It is stored. Of course, the monitor image pickup device may be installed directly on the elevating frame 25 instead of the reversing unit 27. The image obtained by mounting the monitor image pickup device on the elevating frame 25 changes in size, but various information can be obtained even if the image changes in size.

(Information display system)
FIG. 10 is a block diagram showing a configuration of an example of an information display system 60 that displays information from one or more monitor image pickup devices. The information display system 60 includes operation status detection analysis including monitoring image pickup devices 49, 49A and 49B constituting a part of the detection unit, a counting unit 62 forming a part of the detection unit, and a display control unit 65. It includes a storage unit 61, a controller 66 as a display unit including a display screen 67, a data communication device 68, a server 69, and a communication terminal device 70. The monitor image pickup devices 49, 49A and 49B, which form a part of the detection unit, detect the operating status of the molded product ejector as an image. The counting unit 62 counts the cycle time of the molded product take-out machine. The analysis unit 63 analyzes the detection data detected by the detection unit. Then, the data storage unit 64 stores one or more types of analysis data analyzed by the analysis unit 63 as display information. The display information stored in the data storage unit 64 of the operation status detection analysis storage unit 61 is displayed on the display screen of the controller 66 as the display control unit in which the display control unit 65 has the display screen 67. In this case, the data storage unit 64 stores image data of a still image or a moving image as display information. Then, as shown in FIG. 11, the display control unit 65 displays both the display information of the analysis data stored in the data storage unit 64 and the display information of the image data in synchronization with the display screen. When such a synchronized display is performed, a person who has some doubts about the operation or work in the displayed information may be able to find the cause of the doubt by looking at the necessary image while checking the analysis information. It gets higher.

It should be noted that at least a part of the molding die of the molding machine and its surroundings are present in the still image or the moving image taken at the time of taking out the molded product and opening the molded product of the image pickup devices 49, 49A and 49B for the monitor. It is preferable that the mounting position, angle, viewing angle and focal length are determined so as to include objects.

In the present embodiment, as the detection unit of the operation status detection analysis storage unit 61, a counting unit 62 for counting the cycle time of the molded product take-out machine is provided, so that the analysis unit 63 counts within a predetermined period. It has a function to analyze a plurality of cycle times of a series as detection data and analyze the number of cycles including a stop time longer than necessary in one cycle and the occurrence time as display information. When the analysis unit 63 has such a function, it is possible to know to what extent a so-called stop (chocolate stop) that has not been set occurs during one cycle. If it is possible to know the occurrence status of a stop (chocolate stop) that has not been set during this one cycle, it is possible to know a setting error or a device failure.

Further, in this example, the display unit is also attached to the communication terminal device 70 located at a position away from the molded product ejecting machine. Of course, the data storage unit 64 in the operation status detection analysis storage unit 61 may be provided in the server device located at a position away from the molded product ejecting machine. In this example, a data communication device 68 that enables data communication between the display control unit 65, the communication terminal device 70, and the server device is further provided. As a result, the manager outside the factory can confirm the data and the image of the analysis result through the display screen 71 of the communication terminal device 70.

(Machine learning device)
FIG. 12 is a block diagram showing an example of a defect occurrence prediction system. The defect occurrence prediction system 101 can learn the conditions associated with the occurrence of defects in the molded product take-out machine (hereinafter referred to as “defect occurrence conditions”) by using the machine learning device 103 having a machine learning function. Further, the defect occurrence prediction system 101 can create defect information according to the state of the molded product ejecting machine based on the result learned by the machine learning device 103.

In the present specification, the "defect of the molded product ejecting machine" includes a defect of the occurrence of a molded product defect due to a setting error of the molded article ejecting machine and an operation error. That is, the "malfunction of the molded product ejecting machine" is not limited to the state in which the intended function of the molded product ejecting machine cannot be executed, and includes, for example, the state in which the normal operation cannot be reproduced temporarily or permanently. ..

The "defect information" created by the defect occurrence prediction system 101 includes information indicating the presence or absence of a defect in the molded product ejecting machine or information indicating the "degree of defect". The "defect information" may include information indicating that the molded product ejecting machine is in a normal state. "Degree of defect" means the seriousness of the defect. The "degree of defect" may be limited to either the maximum value or the minimum value. The "degree of defect" may be a continuous quantity or a discrete quantity. The operator can determine whether the target component should be replaced or repaired immediately or at the next maintenance work, depending on the "degree of defect".

The molded product take-out machine 1 of FIG. 12 is a molded product take-out machine described with reference to FIGS. 1 to 11. The molded product take-out machine 1 operates according to a command from the control device 105. The control device 105 creates a command for the molded product take-out machine 1 according to the control program. The control device 105 includes a defect determination unit 106. The defect determination unit 106 determines the defect of the molded product take-out machine 1 by using a known defect diagnosis method. The defect determination unit 106 determines the presence or absence of a defect or the degree of the defect of the molded product ejecting machine 1 independently of the defect information created by the defect occurrence prediction system 101. For example, when the disturbance torque detected by the torque sensor or the vibration amplitude of the output data of the sensor exceeds a predetermined threshold value, or when the existence of an originally nonexistent object is detected from the image data, it is more than necessary in one cycle. When the data of the number of cycles including the stop time of the above exceeds the threshold value, when the occurrence time exceeds the threshold value, and when the stop time exceeds the threshold value, the defect determination unit 106 determines that a defect has occurred. judge. In this way, the defect determination unit 106 determines a defect based on various factors. The determination result by the defect determination unit 106 is input to the determination data acquisition unit 103A of the machine learning device 103.

The defect occurrence prediction system 101 includes a detection unit 107 that counts the cycle time of the molded product ejecting machine and detects an operation image. The detection unit 107 uses an image pickup sensor, a counting sensor, a force sensor, a torque sensor, a vibration sensor, a sound collection sensor, a distance sensor, a temperature sensor, a humidity sensor, a flow rate sensor, a light amount sensor, a pressure sensor, and the like provided with an image pickup device as sensors. It may be included. The data output from the detection unit 107 is input to the state observation unit 103B of the machine learning device 103.

The machine learning device 103 learns the condition for generating a defect in the molded product take-out machine 1. In this embodiment, the machine learning device 103 is built in the control device 105. Therefore, the machine learning device 103 executes machine learning by using the processor of the control device 105. In yet another embodiment, the machine learning device 103 may exist on the cloud server.

The determination data acquisition unit 103A acquires determination data from the defect determination unit 106. The determination data is input from the determination data acquisition unit 103A to the learning unit 103C, and is used when the machine learning device 103 learns the defect occurrence condition. The determination data is data for determining the presence or absence of a defect or the degree of the defect.

The state observation unit 103B observes a state variable as an input value for machine learning while the molded product take-out machine 1 is operating or stationary.

The state variable includes data on the number of cycles including a stop time longer than necessary in one cycle detected by the detection unit 107, data on the occurrence time and data on the stop time, and image data of the operation status. Further, the state variable may include the internal data of the control software that controls the molded product take-out machine 1. Internal data may include at least one of torque, position, velocity, acceleration, acceleration, current, voltage and estimated disturbance value. State variables may include computational data obtained based on output data or internal data.

The learning unit 103C learns the failure occurrence condition according to the training data set created based on the combination of the state variable output from the state observation unit 103B and the determination data output from the determination data acquisition unit 103A. The training data set is data in which state variables and judgment data are associated with each other.

An example of the learning process in the machine learning device 103 will be described. When the learning is started, the state observation unit 103B acquires the state variable. Next, the determination data acquisition unit 103A acquires the determination data based on the determination result by the defect determination unit 106. The learning unit 103C learns the failure occurrence condition according to the training data set created based on the combination of the acquired state variable and the acquired determination data. This process is repeatedly executed until the machine learning device 103 sufficiently learns the failure occurrence condition.

The learning unit 103C of the machine learning device 103 may learn the defect occurrence condition according to the neural network model, or may learn the defect occurrence condition by the AI chip.

The still image or moving image obtained by the present invention is a molding machine and a molded product as long as the image includes an image of a part of the molding die and its surroundings at least at the time of taking out, opening, and retracting the movable frame. Since it is an image of the area that is the center of the operation of the take-out machine, the number of emergency stops is high when the incidence of defective products is high, the cycle time for taking out molded products is longer than the set value, and so on. If the number is increasing, or if the time to replace the mold is long, it is likely that you can get a hint as to where the cause is. Therefore, according to the present invention, a large amount of monitor information can be obtained from a still image or a moving image from at least one monitor image pickup device.

1 Molded product take-out machine 3 Traverse frame 6 Traveling body 7 Extraction frame 9 Elevating unit for runner 11 Elevating unit for adsorbing molded product 13 AC servo motor 15 Traveling body 17 Elevating frame 19 AC servomotor 21 Chuck 23 Traveling body 25 Elevating frame 27 Inversion Unit 28 Extraction head 29 One side wall 31 The other side wall 33 Upper connecting plate 34 Drive source 35 Lower connecting plate 37 Closing plate 39 Inclined surface 41 Guide rail 47 Cap 49, 49A to 49C, 49D Monitor imager 53 Molding machine 55 Molded type 57 Product stocker device 59 Pallet 60 Information display system 61 Operation status detection analysis storage unit 62 Counting unit 63 Analysis unit 64 Data storage unit 65 Display control unit 66 Controller 68 Data communication device 69 server 70 Communication terminal device 101 Failure occurrence prediction system 103 Machine learning device 103A Judgment data acquisition unit 103B State observation unit 103C Learning unit 105 Control device 106 Defect judgment unit 107 Detection unit

Claims (7)

It is a molded product take-out machine installed on the molding machine.
At least one monitor image pickup device is attached to the movable frame that moves during the take-out operation.
The at least one monitor image pickup device includes a part of the molding die of the molding machine and a part of the molding machine in a still image or a moving image taken at least when the molded product is taken out, when the molded product is opened, and when the movable frame is retracted. The mounting position, angle, viewing angle and focal length are defined so that objects existing around the part are included.
The movable frame is a pull-out frame in which an elevating frame that moves laterally along a transverse frame extending laterally and moves in a vertical direction and a front-rear direction orthogonal to the lateral direction is movably mounted in the front-rear direction. It's a frame
When the elevating frame moves in the front-rear direction and is at the position farthest from the traverse frame is defined as the maximum movement position on the extraction frame, one monitor image pickup device moves the maximum movement. A molded product take-out machine characterized in that it is installed at the end of the withdrawal frame located on the tip end side of the withdrawal frame with respect to the position. The molded product take-out machine according to claim 1, wherein the image pickup device for a monitor is configured such that the angle, the viewing angle, and the focal length can be adjusted. The monitor image pickup device is configured so that the angle, the viewing angle, and the focal length are automatically adjusted according to the stop position or the movement path of the movable frame or the traveling body for moving the movable frame. The molded product take-out machine according to claim 1. The molded product take-out machine according to claim 1, wherein an electric wire connected to the monitor image pickup device is housed inside the drawing frame. A display unit with a display screen and
A detection unit that detects the operating status of one or more types of controlled equipment included in the molded product take-out machine, an analysis unit that analyzes the detection data detected by the detection unit, and one or more types of analysis analyzed by the analysis unit. An operation status detection analysis storage unit consisting of a data storage unit that stores data as display information,
Further, an information display system including a display control unit for displaying the display information stored in the data storage unit of the operation status detection analysis storage unit on the display screen is further provided.
The data storage unit stores the still image or moving image as display information of image data.
The first aspect of claim 1, wherein the display control unit displays both the display information of the analysis data and the display information of the image data stored in the data storage unit in synchronization with the display screen. Molded product take-out machine. A part of the detection unit of the operation status detection analysis storage unit counts the cycle time of the molded product ejecting machine.
The analysis unit analyzes a plurality of time-series cycle times counted within a predetermined period as the detection data, and analyzes the number of cycles including a stop time longer than necessary in one cycle and the occurrence time as the display information. The molded product take-out machine according to claim 5. The detection unit further includes the monitor image pickup device that detects the operating state of the molded product ejector as an image.
Further equipped with a machine learning device for learning the conditions associated with the occurrence of a defect in the molded product take-out machine.
The machine learning device is a state variable including data on the number of cycles including a stop time longer than necessary in the one cycle detected by the detection unit, data on the occurrence time and data on the stop time, and image data on the operation status. The state observation unit that observes
A determination data acquisition unit that acquires determination data for determining the degree of failure of the molded product ejecting machine, and a determination data acquisition unit.
The sixth aspect of claim 6 is characterized in that it includes a learning unit that learns conditions associated with the occurrence of a defect in the molded product ejecting machine according to a training data set created based on a combination of the state variables and the determination data. The described molded product take-out machine.

Images