Performance inspection system

Abstract

To suppress influence of environmental sound in the periphery for a performance inspection system.SOLUTION: A microphone 54 acquires sound to be issued from an inspection device 60. An image conversion section 51A converts acoustic data acquired from the microphone 54 into a spectrogram image. An image extraction section 51B extracts an image from a predetermined start point time to a predetermined end point time from the spectrogram image generated with time. An accuracy calculation section 51E outputs accuracy with respect to an abnormality of a product based on the extracted image by the image extraction section 51B. A missing-product determination section 51F determines presence / absence of the abnormality of the product based on the accuracy. The accuracy calculation section 51E includes an image recognition neural network. The neural network outputs accuracy with respect to the abnormality with the extracted image as input data. The missing-product determination section 51F determines that the abnormality exists in the product when the accuracy exceeds a predetermined threshold.SELECTED DRAWING: Figure 11

Description

[Technical field]

[0001] This specification discloses a performance inspection system that judges the performance of a product based on acoustic data obtained during a performance inspection of the product.

[0002] For example, in Patent Documents 1 to 3, the presence or absence of an abnormality in the device or work content is determined based on acoustic data during the operation of the product. In these documents, the acoustic data is converted into a spectrogram image. Then, the presence or absence of an abnormality is determined based on the spectrogram image. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-22311 [Patent Document 2] JP 2022-112143 A [Patent Document 3] International Publication No. 2020 / 009210 Summary of the Invention [Problem to be solved by the invention]

[0004] However, when the presence or absence of an abnormality is judged based on the operating sound of a product, the judgment may be affected by surrounding environmental sounds. For example, when a test run for a performance test is being conducted on a product, a transport robot passes by the side of the product. The transport robot may output a melody from a speaker to call attention while moving. In such a case, the melody is included in the sound data. As a result, an erroneous judgment may be output, such as judging a normal product as having an abnormality.

[0005] Therefore, this specification discloses a performance inspection system that can suppress the influence of surrounding environmental sounds. [Means for solving the problem]

[0006] This specification discloses a performance inspection system. The system includes a performance inspection device, a microphone, an image conversion unit, an extraction unit, an accuracy calculation unit, and a judgment unit. The performance inspection device performs performance inspection of a product. The microphone acquires sound emitted from the performance inspection device. The image conversion unit converts acoustic data acquired from the microphone into a spectrogram image. The extraction unit extracts images from a specified start time to a specified end time from a spectrogram image generated over time. The accuracy calculation unit outputs an accuracy of an abnormality in the product based on the image extracted by the extraction unit. The judgment unit judges whether or not there is an abnormality in the product based on the accuracy. The accuracy calculation unit includes a neural network for image recognition. The neural network uses the extracted image as input data and outputs an accuracy of the abnormality. The judgment unit judges that there is an abnormality in the product if the accuracy exceeds a specified threshold.

[0007] According to the above configuration, the degree of certainty of an anomaly is output. In other words, the presence or absence of an anomaly is determined based on the similarity with the sound data at the time of an anomaly. According to this determination method, even if environmental sounds are mixed in during a performance inspection of a normal product, the sound data is dissimilar to the sound data at the time of an anomaly. Therefore, erroneous determination of an anomaly is suppressed.

[0008] In the above configuration, the performance inspection device may perform a test run on the product. In this case, the operating noise of a defective product during the test run is louder than the operating noise of a normal product during the test run.

[0009] According to the above configuration, the contribution of the environmental sound to the determination of whether a product is abnormal is relatively low. In other words, erroneous determination due to the inclusion of the environmental sound, that is, erroneous determination of a product that is abnormal as normal, is suppressed.

[0010] In the above configuration, the performance inspection device may include a movable arm that holds down the product. In this case, the extraction unit identifies the time of occurrence of an operating sound when the arm holds down the product as the start time of the extracted image.

[0011] According to the above configuration, it is possible to identify, as the starting point of the extracted image, a sound that is characteristic of the performance inspection and has a constant frequency characteristic regardless of whether the product is normal or abnormal.

[0012] In the above configuration, the product may be an actuator. In this case, the performance inspection device includes a workpiece transported by the actuator and a stopper that stops the movement of the workpiece. The extraction unit identifies the occurrence time of a collision sound when the workpiece hits the stopper as the end time of the extracted image.

[0013] According to the above configuration, it is possible to identify, as the end point of the extracted image, a sound that is characteristic of the performance inspection and has a constant frequency characteristic regardless of whether the product is normal or abnormal. Effect of the Invention

[0014] The performance inspection system disclosed in this specification can suppress the influence of surrounding environmental sounds. [Brief description of the drawings]

[0015] [Figure 1] FIG. 2 is a perspective view illustrating an actuator that is a product subject to a performance test. [Diagram 2] FIG. 13 is an exploded view of the actuator, illustrating a perspective view of the tray side structure. [Diagram 3] FIG. 2 is an exploded view of the actuator, illustrating the structure of the cover side. [Figure 4] FIG. 4 is a perspective view illustrating a connection portion (in a normal state) between a motor and a worm shaft of the actuator. [Diagram 5] FIG. 13 is a perspective view illustrating a connection portion (abnormal state) between a motor and a worm shaft of the actuator. [Figure 6] 1 is a perspective view illustrating a performance inspection system according to an embodiment of the present invention. [Figure 7] FIG. 1 is a perspective view illustrating the performance inspection process (1 / 3). [Figure 8]FIG. 13 is a perspective view illustrating the performance inspection process (2 / 3). [Figure 9] FIG. 13 is a perspective view illustrating the performance inspection process (3 / 3). [Figure 10] FIG. 2 is a diagram illustrating an example of a hardware configuration of a controller. [Figure 11] FIG. 2 is a diagram illustrating an example of functional blocks of a controller. [Figure 12] FIG. 1 is a diagram illustrating a process of converting sound during a performance test into a spectrogram image. [Figure 13] FIG. 13 is a diagram illustrating an example of a neural network for identifying a start time of an extracted image. [Figure 14] FIG. 13 is a diagram illustrating an example of a neural network for outputting a probability of an anomaly. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0016] <Product> 1 shows an example of a product to be subjected to a performance inspection. This product is, for example, an actuator 10. The actuator includes a guide 44 and a holder 46. The actuator 10 also includes a tray 12 that houses a drive mechanism, and a cover 14. The tray 12 is provided with an inlet 16 that is an inlet for receiving power.

[0017] An opening is provided in the cover 14. The holder 46 is exposed from the opening. The guide 44 extends linearly. The holder 46 moves linearly along the guide 44.

[0018] 2 and 3 show an example of the internal structure of the actuator 10. FIG. 2 shows an example of the internal structure on the tray 12 side. FIG. 3 shows an example of the internal structure on the cover 14 side. Referring to FIG. 2, the tray 12 accommodates a motor 20 and a worm shaft 30. The motor 20 receives power from an inlet 16 and is driven to rotate. The driving force of the motor 20 is transmitted to the worm shaft 30 via a motor shaft 22 and a gear shaft 32.

[0019] 1 and 3, the cover 14 accommodates a worm wheel 40, a final gear 42, a guide 44, and a holder 46 (a base portion thereof). Rotation of the worm shaft 30 (see FIG. 2) is transmitted to the worm wheel 40. Furthermore, the final gear 42, which meshes with the worm wheel 40, rotates together with the worm wheel 40. As the final gear 42 rotates, the holder 46 moves linearly along the guide 44.

[0020] The motor 20 is capable of rotating in both forward and reverse directions, so that the holder 46 is also capable of moving forward and backward along the guide 44.

[0021] <Product normal / abnormal> 4 shows an enlarged view of the connection between the motor 20 and the worm shaft 30. A U-shaped hammer 24 is provided at the tip of the motor shaft 22. A U-shaped anvil 34 is provided at the tip of the gear shaft 32. The opening widths of the hammer 24 and the anvil 34 are, for example, the same. Furthermore, the motor shaft 22 and the gear shaft 32 are arranged coaxially.

[0022] A damper 36 is attached between the hammer 24 and the anvil 34. The damper 36 is, for example, an elastic part, and has a function of a coupler that connects the hammer 24 and the anvil 34. By connecting the hammer 24 and the anvil 34 via the damper 36, the hammer 24 and the anvil 34 rotate synchronously.

[0023] 5 illustrates the actuator 10 in a so-called missing state, in which the damper 36 is not attached. In a state without the damper 36, the hammer 24 strikes the anvil 34 like an impact driver. The hammer 24 then elastically deforms and overcomes the anvil 34. The hammer 24 then makes another half revolution and strikes the anvil 34.

[0024] In this way, the actuator 10 in a missing state, in which the damper 36 is not attached, will make a louder operating noise than a normal actuator 10 in which the damper 36 is attached. As will be described below, the actuator 10 in a missing state, in which the damper 36 is not attached, is determined to be an abnormal actuator 10. Furthermore, the performance inspection system according to this embodiment determines whether or not the actuator 10 is abnormal, that is, whether or not the damper 36 is missing.

[0025] <Performance inspection system> A performance inspection system according to this embodiment is illustrated in Fig. 6. The performance inspection system includes an inspection device 60 (performance inspection device), a controller 50, a display unit 52, and a microphone .

[0026] The inspection device 60 performs a performance inspection on the assembled actuator 10 (product). In the performance inspection, a trial run is performed on the actuator 10. The details of the performance inspection will be described later.

[0027] The controller 50 is, for example, a computer. The controller 50 acquires acoustic data from a microphone 54. The controller 50 also causes a display unit 52 to display the inspection result (normal / abnormal) for the actuator 10 to be inspected. The controller 50 also controls the operation of an inspection device 60.

[0028] The microphone 54 picks up the sound emitted from the inspection device 60. The microphone 54 may be, for example, directional. In this case, the microphone 54 is directed toward a main stage 68 of the inspection device 60.

[0029] The display unit 52 is, for example, a display device. The display unit 52 is disposed, for example, opposite a connector stage 65 of the inspection device 60. In a performance inspection, an operator stands on the connector stage 65 side. In other words, the display unit 52 faces the operator.

[0030] <Performance inspection equipment> 6 to 9 show an orthogonal coordinate system to explain the position and arrangement of each component of the inspection device 60. The operator places the actuator 10 on the main stage 68 from the connector stage 65 side. Based on these working conditions, the FR axis is defined with the forward direction as seen by the operator. The RW axis is defined as the width direction axis of the main stage 68. The FR-RW plane is a horizontal plane. The UP axis is set perpendicular to the FR and RW axes. The UP axis is a vertical axis.

[0031] 6, an inspection device 60 (performance inspection device) includes rails 61, a work 62, stoppers 63A, 63B, a connector 64, a connector stage 65, guides 66, arms 67, a main stage 68, and a holder 69.

[0032] An insertion hole 68A is drilled in the main stage 68. The holder 46 of the actuator 10 (see FIG. 1) is inserted into the insertion hole 68A. A pair of guides 66, 66 are provided on the main stage 68. The distance between the guides 66, 66 is equal to the width W1 of the actuator 10 (see FIG. 1). That is, the actuator 10 is positioned by the guides 66, 66.

[0033] A pair of arms 67, 67 are provided on the main stage 68. The arms 67, 67 are disposed between the guides 66, 66 and the insertion hole 68A. The arms 67, 67 are movable and can be rotated between an open position (see FIG. 6) and a pressing position (see FIG. 7) by a servo motor (not shown).

[0034] The connector stage 65 is connected to the rear end of the main stage 68. The connector stage 65 is provided with, for example, a single-axis stage mechanism, and is capable of advancing and retreating relative to the main stage 68.

[0035] A connector 64 is disposed on a connector stage 65. As illustrated in Figures 7 and 8, the connector 64 can be inserted into the inlet 16 of the actuator 10. The connector 64 receives power from a power source (not shown).

[0036] A pair of rails 61, 61 are laid below a main stage 68. The rails 61, 61 extend in the front-rear direction along the FR axis. A workpiece 62 is placed on the rails 61, 61. The workpiece 62 can advance and retreat on the rails 61, 61.

[0037] Stoppers 63A, 63B are also arranged on the rails 61, 61. The stoppers 63A, 63B are fixed to the rails 61, 61. The separation distance between the stoppers 63A, 63B is determined to exceed the front-to-rear dimension of the workpiece 62. In other words, the workpiece 62 advances and retreats between the stoppers 63A, 63B.

[0038] A holder 69 is disposed above the main stage 68. The holder 69 is movable up and down by a Z-stage mechanism (not shown). A movable stamp is provided on the bottom surface of the holder 69. As will be described later, when a normal judgment is given to the actuator 10, a product ID and the like are printed on the upper surface of the actuator 10 by the stamp.

[0039] 7 to 9 show an example of the performance inspection process. In this performance inspection, a trial run is performed on the actuator 10. First, the actuator 10 is positioned along the guides 66, 66. The actuator 10 is turned (turned upside down) opposite to the orientation shown in FIG. 1, and the holder 46 is inserted into the insertion hole 68A. The holder 46 then holds the workpiece 62.

[0040] When the actuator 10 is positioned on the main stage 68, the arms 67, 67 are rotated to press the front of the actuator 10. Also, the holder 69 descends to press the actuator 10 from above.

[0041] 8, the connector stage 65 advances. This causes the connector 64 to be inserted into the inlet 16 (see FIG. 7). This causes power to be supplied to the actuator 10 from a power source (not shown).

[0042] Electric power is supplied to the motor 20 (see FIG. 2) of the actuator 10, and the motor 20 is rotated. This causes the holder 46 (see FIG. 1) to move forward. The workpiece 62 held by the holder 46 also moves forward and is stopped by the stopper 63B.

[0043] Next, the motor 20 is rotated in the reverse direction. This causes the holder 46 and the workpiece 62 to move backward. The workpiece 62 is stopped by the stopper 63A. Thereafter, a stamp (not shown) on the holder 69 prints a product number and the like on the top surface of the actuator 10. Thereafter, the connector 64 is removed from the inlet 16. The holder 69 further rises, and the arms 67, 67 rotate to the open position (see FIG. 6). The actuator 10 to be inspected is removed by the operator, and the next actuator 10 is placed on the main stage 68.

[0044] As described above, the holder 46 advances and retreats both in a normal state where the actuator 10 has the damper 36 (see FIG. 4) and in an abnormal state where the damper 36 is not present (missing) as shown in FIG. 5. However, as described above, in an abnormal state, the operating sound of the actuator 10 is louder than in a normal state. As will be described below, the presence or absence of an abnormality in the actuator 10 is determined by utilizing the difference in the operating sound.

[0045] <Controller> 10 illustrates an example of a hardware configuration of the controller 50. The controller 50 includes a CPU 50A as a calculation device, and a system memory 50C and a storage 50E as storage devices. The storage 50E may be a non-transient storage device such as a hard disk drive (HDD) or a solid state drive (SSD). The controller 50 also includes an input unit 50G as an input device such as a keyboard or a mouse. The controller 50 also includes an input / output controller 50I that manages input and output of information to and from external devices such as a microphone 54, a display unit 52, and an inspection device 60.

[0046] The controller 50 further includes a GPU 50B (Graphics Processing Unit) as an image processing unit for processing the spectrogram image generated by the image conversion unit 51A (see FIG. 11), a frame memory 50D, a RAMDAC 50F (Random Access Memory Digital-to-Analog Converter), and a display control unit 50H.

[0047] The GPU 50B is a calculation device for image processing, and is mainly operated when determining whether or not there is an abnormality in the actuator 10. The frame memory 50D is a storage device that stores the acoustic data acquired by the microphone 54 and the spectrogram image generated by the image conversion unit 51A. The RAMDAC 50F converts the spectrogram image data stored in the frame memory 50D into an analog signal for the display unit 52, which is an analog display. The display control unit 50H causes the display unit 52 to display the determination result by the missing part determination unit 51F.

[0048] 11 illustrates an example of a functional block of the controller 50. This functional block diagram is configured by the CPU 50A executing a program. The program is stored in, for example, the storage 50E, or in a computer-readable non-transitory storage medium such as a DVD.

[0049] The controller 50 includes, as functional blocks, an image conversion unit 51A, an image extraction unit 51B, a probability calculation unit 51E, a missing item determination unit 51F, a device control unit 51G, and a warning unit 51H.

[0050] The image conversion unit 51A converts the acoustic data acquired from the microphone 54 into a spectrogram image. For example, the acoustic data is illustrated in the upper part of FIG. 12. This acoustic data has time on the horizontal axis and amplitude on the vertical axis. This acoustic data is converted into a spectrogram image shown in the lower part of FIG. 12. In the spectrogram image, time is represented on the horizontal axis and frequency is represented on the vertical axis. Furthermore, a color according to the amplitude is displayed. For example, the greater the amplitude, the brighter the color displayed.

[0051] The process of generating a spectrogram image is known, so it will be explained briefly here. The time series of acoustic data is divided into small frames. A fast Fourier transform (FFT) is then performed within each frame. The images resulting from the Fourier transform are arranged in a time series to obtain a spectrogram image.

[0052] For example, from the start time to the end time of operation of a factory in which the inspection device 60 is installed, acoustic data is continuously transmitted from the microphone 54 to the controller 50. Therefore, spectrogram images are continuously generated from the start time to the end time of operation.

[0053] The image extraction unit 51B includes a start point identification unit 51C and an end point identification unit 51D. The start point identification unit 51C determines an extraction start point of an image (extracted image) used for abnormality determination from the spectrogram image. A neural network is implemented in the start point identification unit 51C to determine a start point time that is the extraction start point.

[0054] 13 illustrates a neural network for identifying an extraction start time. For example, the start point identification unit 51C includes a neural network for image recognition that includes an input layer, a hidden layer, and an output layer. This neural network is, for example, a convolutional neural network (CNN).

[0055] The spectrogram image data is input to the input layer. The input image data may be, for example, an image in frame units used when the fast Fourier transform is performed as described above. For example, the number of nodes in the input layer is equal to the number of pixels of the input image.

[0056] The hidden layer is provided between the input layer and the output layer, and may include, for example, a convolutional layer and a pooling layer.

[0057] The output layer is provided with, for example, only one node. This output node outputs the accuracy of the sound when the product is held down as shown in Fig. 7. As shown in Fig. 7, during the test run, the arms 67, 67 rotate and hit the front surface of the actuator 10. Also, the holder 69 moves down and hits the top surface of the actuator 10.

[0058] 12, the sound of the arms 67, 67 hitting the actuator 10 (product holder (front)) and the sound of the holder 69 hitting the actuator 10 (product holder (upper)) have prominent visual features compared to other times. Furthermore, there is no characteristic difference in the product holder sounds when the actuator 10 is normal and when it is abnormal.

[0059] In this way, a sound that is characteristic of the performance test and has no difference in frequency characteristics between when the product is abnormal and when it is normal is determined as the starting point for extracting images. For example, the neural network in FIG. 13 learns from images of frames including the time when the product is pressed (front) as training data. Alternatively, this neural network learns from images of frames including the time when the product is pressed (upper) as training data.

[0060] When the probability output from the output node exceeds a predetermined threshold (e.g., 80%), the origin identification unit 51C identifies the representative time of the input data image (e.g., the time when it is half-value in the frame) as the predetermined extraction origin time. That is, the occurrence time of the collision sound when the arm 67, 67 hits the actuator 10 or when the holder 69 hits the actuator 10 is identified as the origin time of the extraction image.

[0061] The end point identification unit 51D obtains an extraction end point of an extraction image from the spectrogram image. A neural network for obtaining an end point time that is an extraction end point is implemented in the end point identification unit 51D. This neural network is, for example, a convolutional neural network (CNN) similar to that shown in FIG. 13.

[0062] However, the neural network of the end point identification unit 51D learns, as teacher data, spectrogram image data including the impact sound when the workpiece 62 hits the stopper 63B, with reference to Fig. 12. Alternatively, as teacher data, spectrogram image data including the impact sound when the workpiece 62 hits the stopper 63A is used.

[0063] By such learning, the output node outputs the accuracy of the sound when the workpiece 62 collides with the stopper 63B as shown in FIG. 9. Alternatively, the accuracy of the sound when the workpiece 62 collides with the stopper 63A is output. When the accuracy exceeds a predetermined threshold (e.g., 80%), the end point identification unit 51D identifies the representative time of the input data image (e.g., the time when it becomes half-value in the frame) as the predetermined extraction end point time. In other words, the occurrence time of the collision sound when the workpiece 62 hits the stopper 63A is identified as the end point time of the extraction image.

[0064] 12, the sound of the workpiece 62 colliding with the stoppers 63A, 63B has a distinctive image compared to other times. Furthermore, there is no characteristic difference in the sound of this collision when the actuator 10 is normal and when it is abnormal. In this way, a sound that is characteristic in the performance inspection and has no difference in frequency characteristics between when the product is abnormal and when it is normal is determined as the end point and start point of the image.

[0065] Image extraction unit 51B extracts an image from the start time to the end time from the spectrogram image. This extracted image is an image of a time period during which motor 20 (see FIG. 2) of actuator 10 is rotating. In other words, the operating sound during this time period differs between an abnormal actuator 10 and a normal actuator 10.

[0066] The probability calculation unit 51E outputs the probability of an abnormality in the actuator 10 based on the extracted image. A neural network for determining a missing part is implemented in the probability calculation unit 51E as shown in Fig. 14. This neural network is a convolutional neural network (CNN) similar to that in Fig. 13.

[0067] The extracted image data is input to the input layer of this neural network. For example, the number of nodes in the input layer is equal to the number of pixels in the extracted image. The hidden layer is provided between the input layer and the output layer. For example, the hidden layer includes a convolution layer and a pooling layer. The output layer includes, for example, only one node. This output node outputs the probability of a missing part (abnormality) of the damper 36 (see FIG. 2), which is a component of the actuator 10.

[0068] 14, operation sound data of a test run when the actuator 10 is abnormal, i.e., when the damper 36 is missing, is used as teacher data. More specifically, spectrogram image data generated based on the operation sound data is used as teacher data.

[0069] As described above, the operating sound of the actuator 10 when the damper 36 is missing (abnormal operating sound) is louder than the operating sound when the damper 36 is properly attached (normal operating sound). For this reason, the influence of sounds (environmental sounds) from outside the inspection device 60 is smaller for abnormal operating sounds than for normal operating sounds. In this way, in the performance inspection system according to this embodiment, acoustic data that is relatively less influenced by surrounding environmental sounds is used as the criterion for determining whether or not there is an abnormality.

[0070] 11, the stockout determination unit 51F acquires the probability of the abnormality from the probability calculation unit 51E. The stockout determination unit 51F determines whether or not there is an abnormality in the actuator 10 based on the acquired probability. If the probability exceeds a predetermined threshold (e.g., 80%), the stockout determination unit 51F determines that there is an abnormality in the actuator 10 during the trial run (abnormality determination). If the probability of the abnormality is equal to or less than the threshold, the stockout determination unit 51F determines that the actuator 10 during the trial run is normal (normality determination).

[0071] The result of the determination by stockout determination unit 51F is transmitted to warning unit 51H and device control unit 51G. Upon receiving the abnormality determination, warning unit 51H displays a warning message on display unit 52. Furthermore, upon receiving the abnormality determination, device control unit 51G suspends the test run by inspection device 60. Referring to FIG. 12, the end point of the extracted image (stopper collision sound) is before printing by holder 69 (see FIG. 6), and therefore the test run can be suspended before the serial number or the like is printed on actuator 10. This makes it possible to distinguish between normal actuators 10 and actuators 10 with abnormalities based on the presence or absence of printing.

[0072] In the above embodiment, the absence of the damper 36 in the actuator 10 is determined as an abnormal state, but other aspects may be determined as abnormal. For example, if a foreign object such as a bolt gets into the actuator 10, an abnormal noise will occur during the test run of the actuator 10. For example, a rattling sound will be heard inside the actuator 10.

[0073] Therefore, the acoustic data during the test run that contains this abnormal sound may be used as training data to train the neural network shown in FIG. [Explanation of symbols]

[0074] 10 actuator, 20 motor, 24 hammer, 34 anvil, 36 damper, 50 controller, 51A image conversion unit, 51B image extraction unit (extraction unit), 51C starting point identification unit, 51D end point identification unit, 51E accuracy calculation unit, 51F missing part determination unit (determination unit), 51G device control unit, 51H warning unit, 54 microphone, 60 inspection device (performance inspection device), 62 work, 63A, 63B stoppers, 67 arm.

Claims

1. A performance inspection device for inspecting the performance of the product; A microphone for acquiring a sound emitted from the performance inspection device; an image conversion unit that converts the acoustic data acquired from the microphone into a spectrogram image; an extracting unit that extracts an image from a predetermined start time to a predetermined end time from the spectrogram images generated over time; a probability calculation unit that outputs a probability of an abnormality in the product based on the image extracted by the extraction unit; a determination unit that determines whether or not the product is abnormal based on the accuracy; and Equipped with the certainty calculation unit includes a neural network for image recognition that receives the extracted image as input data and outputs the certainty of an anomaly, The determination unit determines that the product has an abnormality when the certainty exceeds a predetermined threshold value. Performance testing system.

2. 2. The performance inspection system according to claim 1, The performance inspection device performs a test run on the product, The operating noise of the abnormal product during the test run is louder than the operating noise of the normal product during the test run. Performance testing system.

3. 3. The performance inspection system according to claim 2, The performance inspection device includes a movable arm that holds the product, The extraction unit identifies a time when an operation sound occurs when the arm holds the product as the start time of the extracted image.

4. 4. The performance inspection system according to claim 3, The article is an actuator, the performance inspection device includes a workpiece that is transported by the actuator and a stopper that stops the movement of the workpiece; A performance inspection system, wherein the extraction unit identifies the occurrence time of a collision sound when the workpiece hits the stopper as the end time of the extracted image.

Images