Machine tool monitoring device and program

Abstract

To streamline the process of identifying the cause of malfunctions in machine tools. [Solution] The monitoring device for monitoring the operation of a machine tool comprises multiple cameras that photograph the machine tool from multiple directions, multiple sensors that detect the operation of the machine tool, a video file storage unit 104 that repeatedly stores multiple images taken by each of the multiple cameras as multiple video files, a sensor signal storage unit 106 that constantly stores multiple sensor signals detected by the multiple sensors, an abnormality detection unit 107 that detects the occurrence of an abnormality in the machine tool based on the multiple sensor signals, an abnormality video file storage unit 109 that, when an abnormality is detected, stores multiple abnormality video files corresponding to all of the multiple cameras, covering the period before and after the time the abnormality was detected, and a playback control unit 110 that synchronously plays back the multiple abnormality video files.

Description

Technical Field

[0001] The present invention relates to a monitoring device and a program for a machine tool.

Background Art

[0002] Conventionally, in a machine tool such as an injection molding machine, operation data such as screw torque, cylinder temperature, in-mold pressure, injection speed, in-mold resin temperature, and mold surface temperature is acquired by sensors, and the occurrence of operation abnormalities is monitored. In many injection molding machines, an alarm is issued when the sensor value indicates an abnormal value. Here, the abnormal value is a value closer to the normal value than the value (failure value) that is determined to be a failure and automatically stops, and is a value indicating a sign of failure. When an abnormal value is detected, the operation of the injection molding machine continues, while the administrator starts investigating the cause of the occurrence of the abnormality.

[0003] In a certain type of monitoring system, together with the operation data, a video of the injection molding machine taken by a camera is recorded as a video file. The administrator checks the occurrence of an abnormality with the operation data and checks the operation state of the machine tool at that time on the video, and speculates on the cause of the occurrence of the abnormality.

[0004] This operation requires alternately performing a playback operation of the operation data and a playback operation of the video, which is very complicated. Also, in some cases, the cause location may not be shown in a single camera. In that case, it is assumed that various locations of the machine tool are photographed with a plurality of cameras, but the operation of playing back a plurality of videos in cooperation becomes even more complicated.

Summary of the Invention

Problems to be Solved by the Invention

[0005] An object is to achieve labor saving in the work of identifying the cause of an abnormality in a machine tool.

Means for Solving the Problems

[0006] A monitoring device for monitoring the operation of a machine tool according to an embodiment of the present invention comprises: a plurality of cameras that photograph the machine tool from a plurality of directions; a plurality of sensors that detect the operation of the machine tool; a video file storage unit that repeatedly stores a plurality of images taken by each of the plurality of cameras as a plurality of video files; a sensor signal storage unit that constantly stores a plurality of sensor signals detected by the plurality of sensors; an abnormality detection unit that detects the occurrence of an abnormality in the machine tool based on the plurality of sensor signals; an abnormality video file storage unit that, when the occurrence of the abnormality is detected, stores a plurality of abnormality video files from the plurality of video files that correspond to all of the plurality of cameras over a period before and after the time the occurrence of the abnormality was detected; and a playback unit that synchronously plays back the plurality of abnormality video files. [Brief explanation of the drawing]

[0007] [Figure 1] A diagram showing the physical configuration of a monitoring device for a machine tool according to an embodiment of the present invention. [Figure 2] A perspective view showing an example of the arrangement of the sensor and camera in Figure 1 on the injection molding machine. [Figure 3] Figure 1 shows the functional configuration of the information processing device. [Figure 4] A flowchart showing the monitoring procedure of the monitoring device in Figure 1. [Figure 5] A conceptual diagram illustrating how abnormal video files are generated from video files in this embodiment. [Figure 6] Figure 2 shows an example of a display screen showing a list of the date and time of an anomaly and the type of anomaly, as displayed by the playback control unit. [Figure 7] Figure 6 shows an example of a playback screen for three abnormal video files corresponding to the date and time of the abnormality and the type of abnormality selected from the list. [Figure 8] This diagram shows the screen transitions associated with specifying the zoom command in Figure 7. [Modes for carrying out the invention]

[0008] Embodiments of the present invention will be described below. Machine tools include cutting machines such as lathes, milling machines, machining centers, or grinding machines used for cutting metals, plastic processing machines such as injection molding machines, electrical discharge machining machines such as wire cutting machines, and surface polishing machines. Here, an injection molding machine will be used as an example.

[0009] As shown in Figure 1, the monitoring device 1 according to this embodiment has an information processing device 10. The information processing device 10 is connected to a first camera 21, a second camera 22, a third camera 23, as well as a vibration sensor 31, a temperature sensor 32, and a microphone 33 as an acoustic sensor.

[0010] The information processing device 10 continuously records video captured by the three cameras 21, 22, and 23, and monitors for the occurrence of anomalies based on sensor signals from sensors 31, 32, and 33. When an anomaly is detected, it extracts and stores three partial video clips (corresponding to the anomaly video file described later) corresponding to the three cameras 21, 22, and 23, covering the same period before and after the time of the anomaly. The information processing device 10 also arranges the three partial video clips from the time of the anomaly specified by the user on a single screen and plays them back synchronously, that is, aligning the time axis. Furthermore, the information processing device 10 displays at least one of the vibration time waveform, temperature time waveform, and acoustic time waveform (referred to as a sensor graph) on the same screen as the three partial video clips. A seek bar indicating the playback position on the time axis is superimposed on the sensor graph, and the user can manually change the position of the time bar on the time axis to play back from any point in time. Details regarding the physical and functional configuration of the information processing device 10, as well as its operating procedure, will be described later.

[0011] The first camera 21, the second camera 22, and the third camera 23 are positioned via open windows so that both the fixed and movable molds, which are important moving parts of the injection molding machine 4, are within approximately the center of the field of view, as illustrated in Figure 2, and continuously photograph the injection molding machine 4 from multiple directions, typically from the front, back, and above, during the injection molding operation. The vibration sensor 31 is in contact with the fixed or movable mold to directly detect vibrations generated in the mold. The temperature sensor 32 is in contact with the fixed or movable mold to directly detect the surface temperature of the mold. The microphone 33 is positioned via open windows relative to the fixed and movable molds to detect operating sounds generated in the mold itself, the injection unit, and the clamping unit around it.

[0012] Returning to Figure 1, the information processing device 10 has a processor 11. Connected to the processor 11 via a data / control bus 19 are an interface (I / F) 17 for connecting to a storage device 14 such as RAM (Random Access Memory) 12, ROM (Read Only Memory) 13, HDD (Hard Disk Drive) or SSD (Solid State Drive), an input device 15 such as a keyboard or mouse, a display 16 such as an LCD, cameras 21-23 and sensors 31-33, and a communication unit 18 for connecting to an external cloud storage device 2 via an internet line 3.

[0013] The processor 11 consists of a CPU (Central Processing Unit) and a GPU (Graphics Processing Unit). The processor 11 executes a monitoring program loaded into RAM 12 from the storage device 14 or ROM 13, and performs the video recording and playback processes described later. RAM 12 functions as the main memory, work area, etc., of the processor 11. ROM 13 or storage device 14 stores the BIOS (Basic Input Output System), operating system program (OS), monitoring program according to this embodiment, and various data required for these processes, which are executed by the processor 11.

[0014] Figure 3 shows the functional configuration of the information processing device 10. The information processing device 10 has a control unit 101 as the control center and includes a camera signal input unit 102, a video file generation unit 103, a video file storage unit 104, a sensor signal input unit 105, a sensor signal storage unit 106, an anomaly detection unit 107, an anomaly video file generation unit 108, an anomaly video file storage unit 109, a video file transmission unit 113, a playback control unit 110, a playback operation input unit 111, and a display unit 112.

[0015] The video file generation unit 103 repeatedly generates video files of a predetermined duration, for example, 5 minutes in length, from the video signals of cameras 21, 22, and 23 input via the camera signal input unit 102. The video files are set with attribute data including a camera identification code to identify cameras 21, 22, and 23, and the date and time when the anomaly was detected. The video file storage unit 104 stores the video files. As a result, video from three directions is continuously recorded.

[0016] The sensor signal storage unit 106 continuously stores multiple sensor signals detected by sensors 31, 32, and 33 that are input via the sensor signal input unit 105.

[0017] The abnormality detection unit 107 detects the occurrence of an abnormality in the injection molding machine based on multiple sensor signals. Typically, with respect to vibration, abnormal vibration is detected when the amplitude or vibration acceleration exceeds a threshold. With respect to operating noise, abnormal noise is detected when the amplitude or decibel threshold is exceeded. With respect to temperature, an appropriate range for the amplitude or decibel generated during steady operation is preset, and abnormal temperature is detected when it deviates from this appropriate range.

[0018] When an abnormality (abnormal vibration, abnormal sound, abnormal temperature) is detected, the abnormal video file generation unit 108 generates three abnormal video files from multiple video files, corresponding to the three cameras 21, 22, and 23, covering the period before and after the detection of the abnormality. The duration of the abnormal video files, i.e., the length of the abnormal video files, can be arbitrarily set by the user, independently of the length of the video files. The abnormal video files are set with attributes related to a camera identification code that identifies cameras 21, 22, and 23, the date and time the abnormality was detected, and an abnormality code that identifies the type of abnormality. The abnormal video file storage unit 109 stores the generated abnormal video files.

[0019] Furthermore, the video files generated by the video file generation unit 103, the sensor signals detected by sensors 31, 32, and 33, and the abnormal video files generated by the abnormal video file generation unit 108 are stored in the storage units 104, 106, and 109, which correspond to the internal storage device 14, and are also transmitted to the cloud storage device 2 by the video file transmission unit 113 for storage.

[0020] The playback control unit 110 creates a list regarding the date and time when an abnormality was detected and the type of abnormality based on the attributes of the stored abnormal video files according to a user command related to abnormal video playback input from the playback operation input unit 111, and causes the display unit 112 to display it. Further, the playback control unit 110 synchronously plays back three abnormal video files of the three cameras 21, 22, and 23 corresponding to the date and time when an abnormality was detected and the abnormal code selected according to a user instruction via the playback operation input unit 111 on one screen of the display unit 112. The playback control unit 110 creates a sensor graph representing the temporal variation of the sensor signals during the same period as the period of the abnormal video file, and causes the sensor graph to be displayed on the display unit 112 together with the three abnormal video files. A seek bar indicating the playback positions of the three abnormal video files is superimposed on the sensor graph. The position of the seek bar on the time axis is arbitrarily moved according to a user instruction via the playback operation input unit 111. The three abnormal video files start playing back synchronously from the position of the seek bar operated by the user.

[0021] Fig. 4 shows the procedure of the monitoring operation. In conjunction with the main switch of the injection molding machine 4 being turned on and the start of operation of the injection molding machine 4 (step S11 “Yes”), the cameras 21, 22, and 23 are activated (S12), and the sensors 31, 32, and 33 are also activated (S13). Of course, it is not limited to the interlocking activation, and the cameras 21, 22, 23 and the sensors 31, 32, 33 may be activated by turning on the main switch of the monitoring device 1.

[0022] Video signals from the cameras 21, 22, and 23 are input via the camera signal input unit 102, and as shown in Fig. 5, video files MI(CA), MI(CB), and MI(CC) corresponding to the cameras 21, 22, and 23 with a predetermined period (time length) FLC, for example, 5 minutes long, are repeatedly generated every 5 minutes by the video file generation unit 103 (S14). Camera identification codes for identifying the cameras 21, 22, and 23 and the file generation date and time are set as attribute data in the video files MI(CA), MI(CB), and MI(CC).

[0023] Video files MI(CA), MI(CB), and MI(CC) are stored in the video file storage unit 104 (S15), and are transmitted to and stored in the cloud storage device 2 (S15).

[0024] Next, in step S17, it is determined whether the main switch is turned off. When the result is "No", that is, when the injection molding operation is continuing, the abnormality occurrence detection unit 107 individually determines whether abnormal vibration has occurred (S18), whether an abnormal temperature has occurred (S19), and whether an abnormal sound has occurred (S20).

[0025] When the occurrence of any one of abnormal vibration, abnormal temperature, and abnormal sound is detected, as shown in FIG. 5, the abnormal video file generation unit 108 generates an abnormal video file pMI(CA) corresponding to the first camera 21 over a predetermined period before and after the time point when the abnormal vibration, abnormal temperature, or abnormal sound occurred, from a plurality of video files MI(CA) of the camera 21. Typically, an abnormal video file pMI(CA) over a predetermined period (time length) FLA centered on the time point when the abnormal vibration, abnormal temperature, or abnormal sound occurred is generated (S21). For example, when the occurrence of an abnormal temperature is detected at time t13, an abnormal video file pMI(CA)t13 over the period FLA centered on time t13 is generated.

[0026] The period FLA corresponding to the abnormal video file pMI(CA) is set by the administrator to an arbitrary time, for example, 8 minutes, regardless of the period FLC of the video file pMI(CA), and can be changed at an arbitrary timing.

[0027] Similarly, from multiple video files MI(CB) of the second camera 22, an abnormal video file pMI(CB)t13 corresponding to the second camera 22 is generated, covering a predetermined period FLA before and after the time t13 when abnormal vibration, abnormal temperature, or abnormal sound occurred. In addition, from multiple video files MI(CC) of the third camera 23, an abnormal video file pMI(CC)t13 corresponding to the third camera 23 is generated, covering a predetermined period FLA before and after the time t13 when abnormal vibration, abnormal temperature, or abnormal sound occurred.

[0028] The abnormal video files pMI(CA), pMI(CB), and pMI(CC) are set with attributes related to the camera identification code that identifies cameras 21, 22, and 23, the date and time when the abnormality (abnormal vibration, abnormal temperature, or abnormal sound) occurred, and the abnormality code that identifies the type of abnormality.

[0029] Furthermore, the abnormal video file generation unit 108 generates abnormal video files pMI(CA), pMI(CB), and pMI(CC) separately for the occurrence of abnormal vibration, abnormal temperature, and abnormal sound. Therefore, as illustrated in Figure 5, when the occurrence of abnormal vibration is detected at time t37, abnormal video files pMI(CA)t37, pMI(CB)t37, and pMI(CC)t37 are generated for a period FLA centered around time t37. Similarly, when the occurrence of abnormal sound is detected at time t38, abnormal video files pMI(CA)t38, pMI(CB)t38, and pMI(CC)t38 are generated for a period FLA centered around time t38, and their periods may overlap.

[0030] The abnormal video files pMI(CA), pMI(CB), and pMI(CC) generated over the same period triggered by the detection of the same anomaly are stored in the video file storage unit 104 (S22) and also sent to the cloud storage device 2 for storage (S23). Then the process returns to step S14, and the next video files MI(CA), MI(CB), and MI(CC) are generated.

[0031] If no abnormal vibration, abnormal temperature, or abnormal sound is detected ("No" in S20, S21, and S22), the process returns to step S14.

[0032] Steps S14-S23 are repeated until it is determined that the main switch was turned off in step S17 ("Yes"), that is, while the injection molding operation is continuing.

[0033] When it is determined in process S17 that the main switch has been turned off ("Yes"), cameras 21, 22, 23 and sensors 31, 32, 33 are stopped, and the generation and storage of video files MI(CA), MI(CB), and MI(CC) is terminated. Of course, the generation and storage of abnormal video files pMI(CA), pMI(CB), and pMI(CC) is also terminated.

[0034] When an anomaly is detected in this manner, it triggers the generation of multiple corresponding anomaly video files for all cameras, each with the same start time, end time, and duration. This reduces the operational burden of video playback operations required to identify the cause of the anomaly afterward.

[0035] Figure 6 shows a screen displaying a list of the date and time an anomaly was detected and the type of anomaly. Although not shown, when an administrator inputs a command to play an anomaly video via the playback operation input unit 111, the playback control unit 110 generates a list of the date and time and type of anomaly and displays it on the display unit 112. The playback control unit 110 accesses the anomaly video file storage unit 109 and receives attribute data of anomaly video files triggered, for example, by the most recent predetermined number of anomalies. A list of the date and time an anomaly was detected and the type of anomaly is created according to the anomaly code that identifies the type of anomaly (abnormal vibration, abnormal temperature, or abnormal sound) from the attribute data.

[0036] The list is displayed on the display unit 112, and the administrator selects an arbitrary anomaly detection date and time and an anomaly type via the playback operation input unit 111. Upon this selection, the playback control unit 110 reads three anomaly video files pMI(CA), pMI(CB), and pMI(CC) corresponding to the selected anomaly detection date and time and anomaly type from the anomaly video file storage unit 109. The start date and end date and time of the video are identified from the attribute data of these anomaly video files, and multiple sensor signals related to vibration, temperature, and operating sound detected by sensors 31, 32, and 33 during that period are read from the sensor signal storage unit 106. From the read sensor signals, the playback control unit 110 creates graphs representing the temporal fluctuations of each.

[0037] As shown in Figure 7, the playback control unit 110 plays the three abnormal video files pMI(CA), pMI(CB), and pMI(CC) in synchronous playback, that is, with their timelines aligned, and displays them simultaneously in three areas CS1, CS2, and CS3, which are sections of the display unit 120 screen. The playback control unit 110 also displays a graph representing the temporal fluctuations of one to three sensor signals over the same period as the abnormal video in area SS of the same screen as the three abnormal video files pMI(CA), pMI(CB), and pMI(CC). In Figure 7, a temperature graph GT and a vibration graph GS are displayed, but these are merely examples, and one to three graphs arbitrarily specified by the administrator will be displayed. The screen displays a set of playback commands related to playback, pause, stop, and fast forward.

[0038] The graph is overlaid with seek bars that display the playback positions of the three abnormal video files. The seek bars can be moved to any position on the time axis via the playback operation input unit 111. The playback control unit 110 plays the three abnormal video files pMI(CA), pMI(CB), and pMI(CC) synchronously from the position of the moved seek bar.

[0039] Each of the three regions CS1, CS2, and CS3 that display abnormal videos has a magnifying glass-shaped icon corresponding to the zoom command. By selecting any icon, the playback control unit 110 zooms in on the abnormal video displayed in one of the regions CS1, CS2, or CS3 where the selected icon is located, and displays it across the entire screen, as shown in Figure 8. When the back button is selected, the playback control unit 110 returns to the display shown in Figure 7 through its display control.

[0040] As described above, according to this embodiment, multiple video files corresponding to each of the multiple cameras that photograph the injection molding machine from multiple directions are stored in both an internal storage device and an external cloud storage device, thereby increasing the reliability of video storage. Furthermore, the movement of the injection molding machine can be observed with fewer blind spots.

[0041] Furthermore, when an abnormality is detected, the system triggers the recording of multiple video files corresponding to the abnormality from all cameras that film the injection molding machine from multiple directions, ensuring that the start time, end time, and duration are the same. Vibration, temperature, and operating noise are also recorded simultaneously. Since these multiple abnormality video files related to the same abnormality can be played back synchronously by selecting them all at once, the operational burden of video playback for retrospectively identifying the cause of the abnormality can be reduced.

[0042] Furthermore, the movement of the injection molding machine can be observed from multiple directions in a synchronized manner, making it easy and highly accurate to identify the cause of any malfunctions. In addition, graphs of vibration, temperature, and operating noise are displayed on the same screen as the video, along with a seek bar, allowing simultaneous confirmation of the vibration, temperature, and operating noise levels at the time of video playback.

[0043] Thus, according to this embodiment, it is possible to reduce the labor involved in identifying the cause of malfunctions in machine tools such as injection molding machines, and moreover, to improve the accuracy of the process.

[0044] It should be noted that the embodiments are not limited to those described above, and various modifications can be made during implementation without departing from the gist of the invention. Furthermore, each embodiment may be combined as appropriate, and in that case, the combined effects can be obtained. Moreover, the embodiments described above include various inventions, and various inventions can be extracted by selecting combinations from the multiple constituent elements disclosed. For example, if the problem can be solved and effects obtained even if some constituent elements are deleted from all the constituent elements shown in the embodiment, then the configuration with these deleted constituent elements can be extracted as an invention. [Explanation of Symbols]

[0045] 1... Monitoring device, 10... Information processing device, 21... First camera, 22... Second camera, 23... Third camera, 31... Vibration sensor, 32... Temperature sensor, 33... Microphone, 101... Control unit, 102... Camera signal input unit, 103... Video file generation unit, 104... Video file storage unit, 105... Sensor signal input unit, 106... Sensor signal storage unit, 107... Anomaly detection unit, 108... Anomaly video file generation unit, 109... Anomaly video file storage unit, 110... Playback control unit, 111... Playback operation input unit, 120... Display unit.

Claims

1. A monitoring device for monitoring the operation of a machine tool, Multiple cameras that photograph the aforementioned machine tool from multiple directions, Multiple sensors for detecting the operation of the machine tool, A video file storage unit that repeatedly stores multiple videos captured by the multiple cameras as multiple video files, A sensor signal storage unit that continuously stores multiple sensor signals detected by the aforementioned multiple sensors, An abnormality detection unit that detects the occurrence of an abnormality in the machine tool based on the plurality of sensor signals, When the occurrence of the aforementioned anomaly is detected, the abnormal video file storage unit stores multiple abnormal video files from the multiple video files, corresponding to each of the multiple cameras, covering the period before and after the time the occurrence of the aforementioned anomaly was detected. A monitoring device comprising a playback unit that synchronizes and plays back the aforementioned multiple abnormal video files.

2. The monitoring device according to claim 1, wherein the abnormal video file is configured with attributes relating to a camera identification code that identifies the camera, the date and time the occurrence of the abnormality was detected, and an abnormality code that identifies the type of abnormality.

3. The monitoring device according to claim 1, further comprising a list display unit that displays a list of the date and time when the occurrence of the abnormality was detected and the type of the abnormality.

4. The monitoring device according to claim 1, wherein the playback unit displays a graph showing the temporal fluctuation of the sensor signal over the same period as the period of the abnormal video file, together with the plurality of abnormal video files.

5. A seek bar displaying the playback position of the multiple abnormal video files is superimposed on the graph representing the temporal fluctuation of the sensor signal. The monitoring device according to claim 3, wherein the playback unit starts playback of the plurality of abnormal video files in a synchronized manner from the position of the seek bar operated by the user.

6. The monitoring device according to claim 1, further comprising a transmission unit that transmits the aforementioned video file and the aforementioned abnormal video file to a cloud storage device via an internet connection.

7. The multiple sensors mentioned above: A vibration sensor that converts the mechanical vibrations of the machine tool into electrical vibration signals, A temperature sensor that converts the temperature of the machine tool into an electrical temperature signal, The monitoring device according to claim 1, further comprising a microphone that converts the operating sound of the machine tool into an electrical sound signal.

8. The monitoring device according to claim 6, wherein the abnormal video file storage unit generates the abnormal video file separately for the occurrence of abnormalities related to vibration, temperature, and operating noise.

9. A computer as a monitoring device to monitor the operation of machine tools, A means for repeatedly storing multiple images of the machine tool, captured from multiple directions by multiple cameras, as multiple video files in a storage unit, Means for storing multiple sensor signals detected by multiple sensors that detect the operation of the machine tool in the storage unit, A means for detecting the occurrence of an abnormality in the machine tool based on the plurality of sensor signals, When the occurrence of the aforementioned anomaly is detected, means for storing in the storage unit a plurality of anomaly video files corresponding to each of the plurality of cameras, covering the period before and after the time the occurrence of the anomaly was detected, A program that functions as a means for synchronously playing the aforementioned multiple abnormal video files.

Images