Molded product removal machine

Abstract

To provide a molded product take-out machine capable of determining an occurrence of an adsorption error of a molded product and / or a possibility of the occurrence of the adsorption error with high precision, and knowing a cause thereof.SOLUTION: A pressure data storage part 73A of an adsorption error monitoring part 73 stores pressure data obtained from an output of a pressure detector 35 in a time series. An adsorption error cause determination part 73B determines an occurrence of an adsorption error of a molded product and / or a possibility of the occurrence of the adsorption error by an adsorbing member 31 based on comparison results of the time series obtained by comparing the time series of the pressure data stored in the pressure data storage part 73A with the predetermined time series of determination criteria data. When determining the occurrence of the adsorption error and / or the possibility of the occurrence of the adsorption error, the adsorption error cause determination part 73B determines a cause of the occurrence of the adsorption error and countermeasures from a change pattern of the time series.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a molded product remover having a function for determining whether a molded product has been picked up by a suction member without error. [Background technology]

[0002] Japanese Patent Application Publication No. 2007-319960 (Patent Document 1) discloses a molded product removal machine equipped with a vacuum abnormality determination means that determines whether a vacuum abnormality has occurred due to the state of the piping or suction member before suction, based on the degree of vacuum detected by a vacuum pressure detector before the molded product is suctioned by the suction member.

[0003] Furthermore, Japanese Patent Application Laid-Open No. 1-86088 (Patent Document 2) discloses a technology in which a vacuum pressure detection unit distinguishes between multiple levels of vacuum pressure in the nozzle when a component is being sucked, and distinguishes between good and bad component suction status as differences in vacuum pressure levels. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] Japanese Patent Application Laid-Open No. 2007-319960 [Patent Document 2] Japanese Patent Application Publication No. 1-86088 Summary of the Invention [Problem to be solved by the invention]

[0005] The invention described in Patent Document 1 detects the possibility of pipe damage or clogging by comparing the dry suction pressure state before suction with a specified level range, making it possible to detect the presence or absence of vacuum abnormalities before a workpiece is suctioned by the suction member and prevent suction errors and workpieces falling during transport. However, this conventional invention does not know whether suction errors occur or their causes after suction has begun. Furthermore, this invention does not know the possibility of a suction error occurring even if one has not actually occurred.

[0006] Furthermore, when distinguishing between multiple stages of vacuum pressure inside the nozzle during component pickup to determine whether a component is good or bad and the pickup state, as in the conventional invention shown in Patent Document 2, in equipment in which the head is changed frequently, such as a molded product extractor, the dry suction pressure changes each time the head is changed, making it difficult to set the vacuum pressure stage differences that serve as the basis for judgment. Furthermore, the invention described in this publication does not allow for highly accurate detection of the cause of pickup errors, much less for detection of the possibility of a pickup error occurring.

[0007] An object of the present invention is to provide a molded product remover that can determine with high accuracy the cause of a molded product suction error and / or the cause of the possibility of a suction error occurring.

[0008] In addition to the above object, another object of the present invention is to provide a molded product remover that can identify the cause of a suction error and / or the possible cause of a suction error and how to deal with it. [Means for solving the problem]

[0009] The molded product removal machine of the present invention includes a suction member that suctions a molded product, a vacuum suction device that includes a vacuum generator connected to the suction member via a pipe and applying vacuum pressure to the suction member, a pressure valve that controls the vacuum pressure applied to the suction member from the vacuum generator, and a pressure detector that detects the pressure acting on the suction member, a movement mechanism that moves the suction member from the molding machine's mold, a control device that controls the operation of the vacuum generator and the pressure valve and the drive of the movement mechanism, a pressure data storage unit that stores changes in pressure detected by the pressure detector as time-series pressure data, and a suction failure cause determination unit that determines the occurrence and cause of a failure to suction the molded product by the suction member. The suction failure cause determination unit compares the time-series pressure data with predetermined time-series determination reference data and determines the cause of the failure to suction the molded product by the suction member and / or the cause of the possibility of the failure (i.e., at least one cause of the failure to suction the molded product and the possibility of the failure) based on the time-series comparison results obtained.

[0010] The time-series pressure data includes all information about the suction status during the process leading up to and after suction. Even if the pressure does not increase or temporarily fluctuates for some reason during the suction operation, these conditions will be reflected in the time-series pressure data. Therefore, as in the present invention, by comparing the time-series pressure data with predetermined time-series judgment reference data and obtaining a time-series comparison result, even if the pressure temporarily exceeds the threshold, it will not be erroneously determined that no suction error has occurred. Furthermore, the pattern of change in the time-series pressure data can also detect the occurrence of some cause that may have caused a suction error during the operation. By understanding the characteristics of pressure changes that appear in the time-series pressure data when a suction error occurs due to a known cause, it is possible to determine the cause of the suction error. Furthermore, by understanding the characteristics of pressure changes that appear in the time-series pressure data during multiple removal processes before the suction error occurs, it is possible to determine the cause of the suction error. Therefore, according to the present invention, it is possible to determine with higher accuracy than before whether a pickup error has occurred and / or whether a pickup error is likely to occur, and it is also possible to know the cause of the pickup error and / or the cause of the possibility of a pickup error. Therefore, the present invention can be used not only to detect the occurrence of a pickup error, but also in predictive maintenance of a molded product take-out machine.

[0011] The time-series pressure data includes the dry suction pressure, which is the pressure in the pipe before the suction member sucks the molded product, the suction operation start pressure, which indicates that the suction member has started to suck the molded product, the post-removal rise pressure, which is the pressure after the suction member has risen to a predetermined rise position after the suction operation start pressure is detected, and the peak pressure, which is the maximum pressure in the pipe between the detection of the dry suction pressure and the detection of the post-removal rise pressure. Note that although these pressures are characteristic pressures, not all of them are clearly included in the time-series pressure data, and depending on the situation, they may or may not appear in the time-series pressure data. The time-series judgment criteria data is determined on the assumption that the various pressures described above are included in the pressure data.

[0012] The time-series judgment reference data includes a reference pressure value determined by prior testing and a plurality of pieces of time-series pressure change pattern information related to the causes of pickup errors and / or causes linked to the possibility of a pickup error, which have been collected in advance. The pickup error cause judgment unit judges that a pickup error has not occurred when the dry suction pressure is lower than the reference pressure value and the post-take-out rise pressure and the peak pressure are also lower than the reference pressure values, and judges that a pickup error has occurred or is likely to occur when these conditions are not met. If it is determined that a pickup error has occurred or is likely to occur, it can be configured to output information on the cause of the pickup error based on the comparison result and the plurality of pieces of time-series pressure change pattern information. When the time-series comparison result is used, the tendency appearing in the comparison result does not change significantly even if the pickup head used is changed, thereby preventing erroneous judgment of the causes of pickup errors and / or the causes linked to the possibility of a pickup error.

[0013] It is preferable that the reference pressure value is changeable, which allows fine adjustment and thus improves versatility.

[0014] The reference pressure values ​​may include a first reference pressure value used to determine whether a suction error has occurred and a second reference pressure value used to determine the possibility of a suction error occurring, thereby further improving the accuracy of the determination.

[0015] The pressure data storage unit stores a time-series change pattern of pressure for each take-out cycle, and the pickup error cause determination unit preferably outputs a countermeasure for the cause from the time-series change pattern when it determines that a pickup error has occurred or is likely to occur. If a countermeasure is output, the cause can be corrected immediately after determining that a pickup error has occurred, and the cause of a possible pickup error can be corrected before the pickup error actually occurs, thereby preventing a significant decrease in production efficiency.

[0016] The pickup error cause determination unit may acquire the cause of a pickup error or the possibility of a pickup error by inputting the time-series pressure data into a trained learning model that has undergone machine learning using previously collected multiple time-series pressure data sets when a pickup error occurred or when there was a possibility of a pickup error and the causes of the pickup error as training data. That is, the pickup error cause determination unit may be configured using so-called artificial intelligence (AI). Furthermore, the pickup error cause determination unit may be configured to acquire the cause of a pickup error and / or the cause of the possibility of a pickup error and the countermeasures for the pickup error by inputting the time-series pressure data into a trained learning model that has undergone machine learning using previously collected multiple time-series pressure data sets when a pickup error occurred and the causes of the pickup error and / or the causes of the possibility of a pickup error and the countermeasures for the pickup error as training data. In this way, using a machine learning model for pattern-based determination enables the cause of a pickup error and the countermeasures for the pickup error to be determined with even higher accuracy.

[0017] The cause of the pickup error and how to deal with it may be displayed on the screen of a display device or output as audio, allowing the operator to quickly take action to address the cause. [Brief explanation of the drawings]

[0018] [Figure 1] 1 is a block diagram showing the main components of a configuration of a molded product removal machine according to an embodiment of the present invention. [Figure 2] 10A and 10B are diagrams used to explain a movement path of a take-out head caused by a movement mechanism. [Figure 3] FIG. 10 is a diagram illustrating an example of configuring part of the function of the pickup error cause determination unit using a learning model. [Figure 4] (A) is a typical time-series change pattern of the suction pressure when no suction error has occurred, and (B) to (E) are examples of time-series change patterns of the suction pressure when a suction error has occurred or when there is a high possibility of a suction error occurring. [Figure 5]5 is a table summarizing examples of events, causes, and items to be checked (measures) in the case of the time-series change patterns of FIGS. 4(B) to 4(E). DETAILED DESCRIPTION OF THE INVENTION

[0019] An embodiment of the molded product removal machine of the present invention will be described in detail below with reference to the drawings. FIG. 1 is a block diagram showing the main components of a molded product removal machine 1 according to this embodiment of the present invention, which removes molded products from a molding machine 11. The molded product removal machine 1 includes a vacuum suction device 3, a movement mechanism 5, and a control device 7. The configuration in FIG. 1 illustrates the arrangement of a pressure valve 34 used when a vacuum pump is used as the vacuum suction device 3. When a vacuum ejector is used as the vacuum generator 33, the pressure valve 34 is disposed in front of the vacuum generator 33. Therefore, the present invention is not limited to the configuration shown in FIG. 1. The vacuum suction device 3 includes a suction member 31 consisting of a suction nozzle that suctions the molded product, a vacuum generator 33 connected to the suction member 31 via a pipe 32 and applying vacuum pressure to the suction member 31, a pressure valve 34 that opens or closes communication between the vacuum generator 33 and the suction member 31, and a pressure detector 35 that detects the pressure in the pipe 32 between the pressure valve 34 and the suction member 31. The suction member 31 is attached to a take-out head 8 that is moved by a moving mechanism 5.

[0020] When the movement mechanism 5 is a well-known three-axis movement mechanism, i.e., an XYZ-type movement mechanism, during the removal operation of a molded product as shown in Fig. 2, the removal head 8 equipped with the suction member 31 moves along the Z-axis and Y-axis in the following order: standby position (lowering start position) → lowered position → removal position → extraction position → raised position. At the removal position, the movement mechanism 5 uses the suction member 31 to suck and remove a molded product extruded from the molding mold 12 of the molding machine 11 by the protruding action of the ejector pins driven by the ejector pin drive mechanism 13. The movement mechanism 5 then moves the removal head 8 equipped with the suction member 31 from the raised position to an open position (not shown). At the open position, suction by the suction member 31 is released, and the molded product MP is released at the open position.

[0021] The control device 7 controls the operation of the vacuum generator 33 and the pressure valve 34 and the drive of the moving mechanism 5. As shown in Fig. 1, in this embodiment, the control device 7 at least includes a movement mechanism control unit 71 that controls the movement mechanism 5, a vacuum generator control unit 72 that controls the vacuum generator 33, a control device 7 that controls the operation of the vacuum generator 33 and the pressure valve 34 and the drive of the moving mechanism 5, and a suction failure monitoring unit 73 that monitors the suction failure of the suction member 31 to suction the molded product based on the output of the pressure detector 35. Of course, the control device 7 may include other components if necessary to reduce air consumption.

[0022] The suction failure monitoring unit 73 includes a pressure data storage unit 73A and a suction failure cause determination unit 73B. The pressure data storage unit 73A stores pressure changes output by the pressure detector 35 as time-series pressure data. The time-series pressure data includes all information about the suction status during the process leading up to and after suction. Therefore, even if the pressure does not increase or the pressure temporarily fluctuates for some reason during the suction operation, these conditions are reflected in the time-series pressure data. The time-series pressure data includes, in time series, the pressure in the pipe 32 before the suction member 31 suctions the molded product (dry suction pressure), the suction operation start pressure Pb indicating that the suction member 31 has started suctioning the molded product, the post-takeout rise pressure Pa after the suction member 31 has risen to a predetermined rise position after the suction operation start pressure is detected, and the peak pressure Pp, which is the maximum pressure in the pipe 32 between the detection of the dry suction pressure and the detection of the post-takeout rise pressure Pa. Although these pressures are characteristic pressures of the adsorption operation, not all of them are clearly included in the time-series pressure data, and they may or may not appear in the pressure data depending on the situation. Based on information such as the presence or absence of these pressures, the magnitude of these pressures, and the period during which these pressures appear (time-series pressure change pattern information), time-series judgment criteria data is created.

[0023] The time-series determination criterion data includes a reference pressure value determined through a prior test and a plurality of time-series pressure change pattern information related to causes of pickup errors and causes linked to the possibility of a pickup error. The time-series determination criterion data used by the pickup error cause determination unit 73B includes time-series determination criterion data for determining, for example, that a pickup error does not occur when the post-takeout rising pressure Pa and the peak pressure Pp are lower than the reference pressure value Pr, and determining that a pickup error has occurred or is likely to occur when this condition is not met. The pickup error cause determination unit 73B is configured to output information on the cause of the pickup error and / or the cause of the possibility of a pickup error based on the comparison result and the plurality of time-series pressure change pattern information. When the time-series comparison result is used, the tendency appearing in the comparison result does not change significantly even if the pickup head used is changed, thereby preventing erroneous determination of the cause of a pickup error and the cause of the possibility of a pickup error.

[0024] The suction error cause determination unit 73B compares the time-series pressure data stored in the pressure data storage unit 73A with predetermined criterion data to obtain a time-series comparison result, and determines and outputs the occurrence of a failure to suction the molded product by the suction member 31, its cause, and the cause of the possibility of a suction error occurring, as well as countermeasures for those causes. Specifically, the suction error cause determination unit 73B used in this embodiment includes a data storage unit 73Ba that stores information such as the reference pressure value and criterion data, and a comparison unit 73Bb and a search unit 73Bc that are constituted by processors.

[0025] The time-series determination criteria data stored in the data storage unit 73Ba in the pickup error cause determination unit 73B includes determination criteria data that serve as determination criteria determined in advance based on a plurality of representative time-series change patterns 1 to 4 in which pickup errors occur or are likely to occur, as shown in Fig. 4, the pressure change trends of these change patterns, and time-series pressure data when a pickup error actually occurs. In addition, the data storage unit 73Ba also stores reference pressure values ​​determined in advance by tests, previously collected information on causes of pickup errors and possible causes of pickup errors, and countermeasures for those causes, in order to determine the cause and how to deal with them.

[0026] The pickup error cause determination unit 73B uses the comparison unit 73Bb to compare the time-series pressure data stored in the pressure data storage unit 73A with the time-series reference data. The comparison unit 73Bb determines whether a pickup error has occurred and whether there is a possibility of a pickup error occurring based on the comparison result. The search unit 73Bc then searches for and outputs the cause of the pickup error, the cause of the possibility of a pickup error occurring, and a solution to the problem based on the comparison result of the comparison unit 73Bb. When the comparison unit 73Bb of the pickup error cause determination unit 73B determines that a pickup error has occurred or there is a possibility of a pickup error occurring, the result is output as an alarm to the movement mechanism control unit 71. The cause and solution found by the search unit 73Bc are displayed on the display device 75 in the control device 7. The display on the display device 75 may be an image or an audible signal.

[0027] For example, as shown in FIGS. 4A to 4E, the suction error cause determining unit 73B may use a reference pressure value Pr determined through a prior test as an example of the determination criteria data. In this case, the suction error cause determining unit 73B determines, as a first step of determination, that a suction error does not occur if the dry suction pressure is lower than the reference pressure value Pr and the post-takeout rising pressure Pa and peak pressure Pp (described later) are higher than the reference pressure value Pr. If these conditions are not met, the suction error cause determining unit 73B determines that a suction error has occurred. Using the time-series comparison results in this way prevents a false determination that a suction error has occurred due to momentary pressure fluctuations, even if the takeout head being used is changed. Here, the reference pressure value Pr is a pressure that is recognized from past test results as the pressure reached in most cases where a suction error has not occurred. It is known that the reference pressure value Pr varies depending on conditions such as the structure of the suction member 31, the length of the piping 32, and the type of vacuum generator 33. Therefore, it is preferable that the reference pressure value Pr be changeable. As this reference pressure value Pr, a first reference pressure value used to determine whether a suction error has occurred and a second reference pressure value used to determine whether a suction error may occur may be separately prepared. In this case, the second reference pressure value is set to a value smaller than the first reference pressure value. It is preferable to adjust the reference pressure value Pr so that it decreases as the length of the piping 32 increases.

[0028] In this embodiment, information such as judgment criteria data is stored in the data storage unit 73Ba of the pickup error cause judgment unit 73B for the second-stage judgment. As described above, the time-series judgment criteria data includes a reference pressure value determined through a preliminary test and multiple time-series pressure change pattern information related to the causes of pickup errors and the causes linked to the possibility of a pickup error, which have been collected in advance. The time-series judgment criteria data for comparison by the comparison unit 73Bb includes, for example, time-series judgment criteria data for determining that a pickup error has not occurred when the post-takeout rising pressure Pa and the peak pressure Pp are lower than the reference pressure value Pr, and for determining that a pickup error has occurred or is likely to occur when this condition is not met. The data storage unit 73Ba also includes multiple other time-series judgment criteria data for comparison. Based on the comparison results of the comparison unit 73Bb, the search unit 73Bc searches for causes of pickup errors or possible causes of pickup errors and countermeasures, which are pre-stored in the data storage unit 73Ba, and outputs the results to the display device 75.

[0029] The pressure data storage unit 73A stores time-series pressure data for each take-out cycle. As described above, when the pickup error cause determination unit 73B determines that a pickup error has occurred or that a pickup error may occur, it determines the cause of the pickup error or the cause of the possibility of a pickup error from the time-series change pattern and outputs a countermeasure.

[0030] To further improve accuracy, as shown in FIG. 3 , a trained learning model 74 constructed by machine learning may be used in the pick-up error cause determination unit 73B. In this case, the time-series pressure data may be input to the trained learning model 74, which has undergone machine learning using previously collected time-series pressure data when a pick-up error occurs, the cause of the pick-up error, and time-series pressure data before the pick-up error occurs, the cause of the pick-up error, and countermeasures for the error as training data, to acquire the cause of the pick-up error or the possible cause of the pick-up error. By using such a trained learning model 74, inputting the time-series pressure data into the learning model 74 makes it possible to acquire whether or not a pick-up error has occurred, whether or not a pick-up error is likely to occur, the cause of the pick-up error, and countermeasures for the error. Using the learning model 74 to determine the cause of the pick-up error or the possible cause of the pick-up error and the countermeasures for the error based on the change pattern of the time-series pressure data in this way allows for even more accurate determination of the cause of the pick-up error and the countermeasures for the error. The cause of the pick-up error, the possible cause of the pick-up error, and the countermeasures for the error may be displayed on the screen of the display device 75, or may be output as audio from the display device 75. This allows the worker to quickly address the cause of the problem.

[0031] FIG. 4A shows a typical change pattern of the suction pressure over time when no suction error has occurred. In contrast, FIGS. 4B to 4E show representative examples of the suction pressure over time when a suction error has occurred or is highly likely to occur. In the case of time-series change pattern 1, which is also used as the judgment criterion data in FIG. 4B, the suction pressure over time shows almost no change even when the suction timing arrives after the pressure reaches the dry suction pressure. In such a case, the suction member 31 may not be in contact with the molded product. When the comparison unit 73Bb outputs as a comparison result that such time-series pressure data has been input to the suction error cause determination unit 73B, the search unit 73Bc determines that the cause is either a teaching error or the installation condition of the suction member 31, and suggests measures such as redoing the teaching or checking the condition of the suction member.

[0032] 4(C), the time-series change pattern 2 serving as the time-series judgment criterion data shows that the suction pressure waveform rises once after reaching the dry suction pressure, but then drops before reaching the suction timing. In other words, the peak pressure Pp occurs before reaching the post-takeout rising pressure Pa. If the input time-series pressure data corresponds to this judgment criterion data, there is a possibility that the suction member 31 became detached from the molded product after the peak pressure Pp occurred. Therefore, when the comparison unit 73Bb outputs as a comparison result that such time-series pressure data has been input to the suction error cause judgment unit 73B, the search unit 73Bc determines that the cause is either a teaching error, a problem with the vacuum generator 33, or the installation of the suction member 31, and suggests the following countermeasures: redoing the teaching, checking the setting conditions of the vacuum generator 33, or checking the condition of the suction member.

[0033] In the case of time-series change pattern 3, which serves as the reference data for the time-series determination in FIG. 4(D), the suction pressure waveform rises after reaching the dry suction pressure, and its shape is similar to the normal pattern in FIG. 4(A). However, in this time-series change pattern, the pressure at the suction timing does not reach the reference pressure value Pr. In such a case, a suction error has occurred or is likely to occur in the near future. Therefore, when the comparison unit 73Bb outputs as a comparison result that such time-series pressure data has been input to the suction error cause determination unit 73B, the search unit 73Bc determines that the cause is incorrect teaching, a problem with the vacuum generator 33, a problem with the attachment of the suction member 31, or a problem with the setting of the reference pressure value Pr, and suggests the following countermeasures: redoing the teaching, checking the setting conditions of the vacuum generator 33, checking the condition of the suction member, or checking for an error in the setting of the reference pressure value Pr.

[0034] Furthermore, in the case of time-series change pattern 4, which serves as the time-series judgment reference data in Figure 4(E), the suction pressure waveform shows that the pressure rises after reaching the dry suction pressure, but continues to rise even after the suction timing is reached. The pressure at the suction timing does not reach the reference pressure value Pr, and then exceeds the reference pressure value Pr and reaches saturation. In such cases, the molded product may be suctioned by the suction member 31 but may not be sufficiently suctioned. In other words, when such time-series pressure data is input to the suction error cause determination unit 73B, the comparison unit 73Bb determines that a suction error may occur. The display device 75 then displays a message indicating that the suction timing setting is incorrect, and the search unit 73Bc suggests redoing the teaching as a remedy. It is also effective to suggest that the vacuum generator 33 or the suction member 31 needs to be checked as a remedy.

[0035] Figure 5 shows examples of events, causes, and countermeasures to be checked in the case of time series change patterns 1 to 4, which are the reference data for the time series shown in Figures 4(B) to 4(E). The learning model 74 shown in Figure 3 is constructed using these data as training data.

[0036] In this embodiment, the pickup error cause determination unit 73B displays the cause and a solution on the display screen of the display device or notifies the worker by voice, based on the determination result, which has the advantage that it becomes easier for the worker to take action after discovering a pickup error.

[0037] For example, to improve accuracy, as shown in FIG. 3 , a trained learning model 74 constructed by machine learning may be used in the pick-up error cause determination unit 73B. In this case, the cause of a pick-up error may be acquired by inputting the time-series pressure data into the trained learning model 74, which has undergone machine learning using previously collected multiple time-series pressure data when a pick-up error occurred, the cause of the pick-up error, and previously collected multiple time-series pressure data when there was a possibility of a pick-up error occurring, as training data. In this case, by using the trained learning model 74, which has undergone machine learning using previously collected multiple time-series pressure data when a pick-up error occurred, the cause of the pick-up error, and countermeasures, as training data, inputting the time-series pressure data into the learning model 74, it is possible to acquire whether a pick-up error occurred, whether there is a possibility of a pick-up error occurring, the cause of the occurrence, and the countermeasures. In this way, using the learning model 74 to determine the cause of a pick-up error or the possible cause of a pick-up error and the countermeasures, based on the change pattern of the time-series pressure data, allows for even more accurate determination of the cause of a pick-up error and the countermeasures. The cause of the pickup error, the possible cause of the pickup error, and the countermeasures may be displayed on the screen of the display device 75, or may be output as audio from the display device 75. In this way, the operator can quickly take measures to address the cause.

[0038] The time-series judgment criteria data and data on causes of pickup errors and how to deal with them shown in Figures 4 and 5 are part of the training data when creating the learning model 74 shown in Figure 3. When creating an actual learning model, however, as much data as possible from the use of multiple types of take-out heads and multiple types of pickup members is collected and used as training data. [Industrial Applicability]

[0039] According to the present invention, by comparing time-series pressure data with predetermined time-series judgment reference data and obtaining the time-series comparison results, even if the threshold value is temporarily exceeded, it is not possible to erroneously determine that a pickup error has not occurred simply because of that. Furthermore, the pattern of change in the time-series pressure data makes it possible to detect the occurrence of some cause that may have caused a pickup error during the operation. By understanding the characteristics of the pressure change that appears in the time-series pressure data when a pickup error occurs due to a known cause, it is also possible to determine the cause of the pickup error or the possibility of a pickup error occurring. Therefore, according to the present invention, it is possible to determine with higher accuracy than conventional methods the occurrence of a pickup error and / or the possibility of a pickup error occurring with a molded product, and also to know the cause of the occurrence. [Explanation of symbols]

[0040] 1 Molded product removal machine 3 Vacuum suction device 5 Moving mechanism 7 Control Device 8 Extraction head 11 Molding machine 12 Molding mold 14 Ejector pin movement mechanism 31 Adsorption member 32 Piping 33 Vacuum Generator 34 Pressure valve 35 Pressure detector 7 Control Device 71 Movement mechanism control unit 72 Vacuum generator control unit 73 Adsorption error monitoring unit 73A Pressure data storage unit 73B Pickup error cause determination unit

Claims

1. a vacuum suction device including a suction member that suctions a molded product, a vacuum generator connected to the suction member via a pipe and applying vacuum pressure to the suction member, a pressure valve that controls the vacuum pressure applied from the vacuum generator to the suction member, and a pressure detector that detects the pressure acting on the suction member; a moving mechanism that moves the suction member to remove the molded product from the molding die of the molding machine; a control device that controls the operation of the vacuum generating device and the pressure valve and the drive of the moving mechanism; a pressure data storage unit that stores the change in pressure detected by the pressure detector as time-series pressure data; a suction error cause determination unit that determines whether a suction error of the molded product by the suction member has occurred and / or whether a suction error may occur, and the cause thereof, The suction error cause determination unit determines, based on a time-series comparison result obtained by comparing the time-series pressure data with predetermined time-series judgment reference data, that a suction error has occurred in the suction member to suction the molded product and / or that a suction error may occur, as well as the cause of the suction error.

2. 2. The molded product removal machine according to claim 1, wherein the time-series pressure data includes an air suction pressure, which is the pressure in the piping before the suction member suctions the molded product, a suction operation start pressure which indicates that the suction member has started suctioning the molded product, a post-removal ascent pressure which is the pressure in the piping after the suction member has risen to a predetermined ascent position after the suction operation start pressure is detected, and a peak pressure which is the maximum pressure in the piping between the detection of the air suction pressure and the detection of the post-removal ascent pressure.

3. the time-series judgment reference data includes a reference pressure value determined by a previous test and a plurality of time-series pressure change pattern information related to causes of suction errors collected in advance, 3. The molded product take-out machine according to claim 2, wherein the suction error cause determination unit determines that there is no suction error under conditions that the dry suction pressure is lower than the reference pressure value and that the post-take-out rise pressure and the peak pressure are lower than the reference pressure values, and determines that a suction error has occurred or is likely to occur when these conditions are not satisfied, and when it determines that a suction error has occurred or is likely to occur, outputs information on the cause of the suction error and / or the cause of the possibility of the suction error based on the comparison result and the plurality of pieces of time-series pressure change pattern information.

4. 4. The molded product remover according to claim 3, wherein the reference pressure value is variable.

5. 5. The molded product removal machine according to claim 4, wherein the reference pressure values ​​include a first reference pressure value used to determine whether the suction error has occurred and a second reference pressure value used to determine whether the suction error may occur.

6. the pressure data storage unit stores the pressure data for each take-out cycle; 6. The molded product remover according to claim 1, wherein the suction error cause determination unit outputs a countermeasure for the cause when it determines that the suction error has occurred or that the suction error may occur and the cause of the suction error.

7. The molded product removal machine according to claim 4, wherein the suction error cause determination unit inputs the time-series pressure data into a trained learning model that has undergone machine learning using as training data a plurality of time-series pressure data collected in advance when a suction error has occurred and / or when there is a possibility of a suction error occurring and the causes of the suction errors, and thereby obtains from the learning model that a suction error has occurred and / or when there is a possibility of a suction error occurring, and the causes thereof.

8. The molded product removal machine according to claim 4, wherein the suction error cause determination unit inputs the time-series pressure data into a trained learning model that has undergone machine learning using as training data a plurality of time-series pressure data collected in advance when a suction error has occurred and / or when it has been determined that there is a possibility of a suction error, and causes and countermeasures for the occurrence and / or possibility of the occurrence of a suction error, and acquires the cause of the suction error and countermeasures from the learning model.

9. 7. The molded product remover according to claim 6, wherein the cause of the suction error or the possible cause of the suction error and a remedy for the suction error are displayed on a screen of a display device.

10. The molded product removing machine according to claim 6, wherein the cause of the suction error or the possible cause of the suction error and a solution to the problem are output by voice from a display device.

Images