Defect detection device, defect detection system, defect detection method, and program

Abstract

This invention provides a defect detection device, a defect detection system, a defect detection method, and a program that can increase the degree of freedom in detecting defects in screw tightening processes. [Solution] One embodiment of the defect detection device of the present disclosure includes: a waveform data acquisition unit that acquires waveform data showing the waveform of a physical quantity generated in a power source during a screw tightening process; a feature quantity calculation unit that uses the waveform data to calculate a plurality of feature quantities that characterize each of a plurality of divided portions obtained by dividing the waveform into a plurality of parts; and a defect detection unit that uses the plurality of feature quantities to determine whether the screw tightening process is defective.

Description

[Technical Field]

[0001] The present disclosure relates to a defect determination apparatus, a defect determination system, a defect determination method, and a program. [Background Art]

[0002] In existing automatic screw tightening apparatuses that automatically perform screw tightening processing, a technology is known that automatically determines defects in screw tightening processing through threshold determination using physical quantities such as torque and rotational speed transmitted from a motor to a driver (see, for example, Patent Documents 1 to 3).

[0003] For example, Patent Document 1 discloses performing defect determination for screw tightening processing through threshold determination on a two-dimensional plane using a value specified as two-dimensional coordinates from composite feature amounts F1 and F2 calculated from a plurality of feature amounts related to the shape similarity between a reference torque waveform and a tightening torque waveform. According to Patent Document 1, defects such as oblique tightening and cam-out can be determined.

[0004] Further, for example, Patent Document 2 discloses performing defect determination for screw tightening processing through threshold determination using a difference value between the rotation angle corresponding to a first set current value and the rotation angle corresponding to a second set current value in waveform data represented by a current value (torque) and a rotation angle during screw tightening processing. According to Patent Document 2, defects such as oblique tightening can be determined.

[0005] Further, for example, Patent Document 3 discloses performing defect determination for screw tightening processing through threshold determination using the rotation speed of a first motor that causes rotational movement around the axis of a driver for tightening screws. According to Patent Document 3, defects such as cam-out can be determined. [Prior Art Documents] [Patent Documents]

[0006] [Patent Document 1] Japanese Unexamined Patent Publication No. 2020-6448 [Patent Document 2] Japanese Patent Publication No. 2019-150922 [Patent Document 3] Patent No. 7070164 [Overview of the Initiative] [Problems that the invention aims to solve]

[0007] However, all of the existing technologies mentioned above have room for improvement in terms of increasing the degree of freedom in determining defects in the screw tightening process.

[0008] This disclosure is made in view of the above circumstances and contributes to providing a defect detection device, defect detection system, defect detection method, and program that can increase the degree of freedom in determining defects in screw tightening processes. [Means for solving the problem]

[0009] One embodiment of the defect detection device of the present disclosure includes: a waveform data acquisition unit that acquires waveform data showing the waveform of a physical quantity generated in a power source during a screw tightening process; a feature quantity calculation unit that uses the waveform data to calculate a plurality of feature quantities that characterize each of a plurality of divided portions obtained by dividing the waveform into a plurality of parts; and a defect detection unit that uses the plurality of feature quantities to determine whether the screw tightening process is defective.

[0010] One embodiment of the defect detection device of this disclosure comprises: a waveform data acquisition unit that acquires waveform data showing the waveform of a physical quantity generated in a power source during a screw tightening process; a feature quantity calculation unit that calculates a feature quantity characterized by a part of the waveform using the waveform data; and a defect detection unit that performs defect detection of the screw tightening process using the feature quantity, wherein the feature quantity is the time, angle, or absolute value or percentage of the total work taken until the torque, which is the physical quantity, reaches the value of a first parameter; or the time, angle, or absolute value or percentage of the total work taken from the time the torque, which is the physical quantity, reaches the value of a first parameter until the fastening is completed. The parameters include, at least one of the following values: time, angle, absolute value or proportion of work, specified from the region in which the physical quantity torque is above or below the value of the second parameter; time, angle, absolute value or proportion of work, specified from the region in which the physical quantity rotational speed is above or below the value of the third parameter; and time, angle, absolute value or proportion of work, specified from the region in which the magnitude of the time slope of the torque, the angular slope of the torque, the time slope of the angle, the angular slope of the angle, the time slope of work, or the angular slope of work is within the range of the fourth parameter.

[0011] One embodiment of the defect detection device of the present disclosure includes: a waveform data acquisition unit that acquires waveform data showing the waveform of a physical quantity generated in a power source during a screw tightening process; a feature quantity calculation unit that calculates a feature quantity characterized by a part of the waveform using the waveform data; and a defect detection unit that performs defect detection of the screw tightening process using the feature quantity, wherein the plurality of feature quantities include at least one of the following values: the time, angle, absolute value of the work, or percentage of the total, or the number of times the magnitude of the time slope or angular slope of the physical quantity torque continues at or above the value of the fifth parameter; and the time, angle, absolute value of the work, or percentage of the total, or the number of times the magnitude of the time slope or angular slope of the physical quantity rotational speed or the absolute value of the rotational speed continues at or above the value of the sixth parameter.

[0012] One aspect of the defect detection system of the present disclosure comprises: a control unit for controlling a power source used in a screw tightening process; a waveform data acquisition unit for acquiring waveform data showing the waveform of a physical quantity generated in the power source during the screw tightening process; a feature calculation unit for calculating a plurality of feature quantities that characterize each of a plurality of divisions obtained by dividing the waveform into a plurality of parts using the waveform data; and a defect detection unit for performing a defect detection of the screw tightening process using the plurality of feature quantities, wherein each of the plurality of feature quantities is a feature quantity for distinguishing between different degrees of defects in the screw tightening process; the defect detection unit further determines the degree of defect in the screw tightening process; and the control unit performs feedback control according to the determination result of the degree of defect in the screw tightening process.

[0013] One aspect of the defect determination method of this disclosure includes: a waveform data acquisition step in which a waveform data acquisition unit acquires waveform data showing the waveform of a physical quantity generated in a power source during a screw tightening process; a feature calculation step in which a feature calculation unit calculates a plurality of feature quantities that characterize each of a plurality of divisions obtained by dividing the waveform into a plurality of parts using the waveform data; and a defect determination step in which a defect determination unit performs a defect determination of the screw tightening process using the plurality of feature quantities.

[0014] One aspect of the program of this disclosure is a method for causing a computer to perform the following steps: a waveform data acquisition step of acquiring waveform data showing the waveform of a physical quantity generated in a power source during a screw tightening process; a feature calculation step of calculating a plurality of feature quantities that characterize each of a plurality of divisions obtained by dividing the waveform into a plurality of parts using the waveform data; and a defect determination step of performing a defect determination of the screw tightening process using the plurality of feature quantities. [Effects of the Invention]

[0015] According to this disclosure, it is possible to provide a defect detection device, a defect detection system, a defect detection method, and a program that can increase the degree of freedom in determining defects in screw tightening processes. [Brief explanation of the drawing]

[0016] [Figure 1] FIG. 1 is a schematic diagram showing an example configuration of the screw fastening device according to a first embodiment. [Figure 2] FIG. 2 is a diagram showing an example of a normal torque waveform generated in a motor from the start to the end of screw fastening processing by the screw fastening device according to the first embodiment. [Figure 3] FIG. 3 is a diagram showing an example of a torque waveform when mild oblique tightening occurs in the screw fastening processing by the screw fastening device according to the first embodiment. [Figure 4] FIG. 4 is a diagram showing an example of a torque waveform when moderate oblique tightening occurs in the screw fastening processing by the screw fastening device according to the first embodiment. [Figure 5] FIG. 5 is a diagram showing an example of a torque waveform when severe oblique tightening occurs in the screw fastening processing by the screw fastening device according to the first embodiment. [Figure 6] FIG. 6 is a diagram showing an example of a torque waveform when severe oblique tightening occurs in the screw fastening processing by the screw fastening device according to the first embodiment. [Figure 7] FIG. 7 is a diagram showing an example of a torque waveform when mild cam-out occurs in the screw fastening processing by the screw fastening device according to the first embodiment. [Figure 8] FIG. 8 is a diagram showing an example of a torque waveform when moderate cam-out occurs in the screw fastening processing by the screw fastening device according to the first embodiment. [Figure 9] FIG. 9 is a diagram showing an example of a torque waveform when severe cam-out occurs in the screw fastening processing by the screw fastening device according to the first embodiment. [Figure 10] FIG. 10 is a block diagram showing an example of the hardware configuration of the defect determination device according to the first embodiment. [Figure 11] FIG. 11 is a block diagram showing an example of the functional configuration of the defect determination device according to the first embodiment. [Figure 12] FIG. 12 is a diagram showing an example of feature quantities according to the first embodiment. [Figure 13] FIG. 13 is a diagram showing an example of feature quantities according to the first embodiment. [Figure 14] Fig. 14 is a diagram showing an example of feature amounts according to the first embodiment. [Figure 15] Fig. 15 is a diagram showing an example of feature amounts according to the first embodiment. [Figure 16] Fig. 16 is a diagram showing an example of feature amounts according to the first embodiment. [Figure 17] Fig. 17 is a diagram showing an example of feature amounts according to the first embodiment. [Figure 18] Fig. 18 is a diagram showing an example of feature amounts according to the first embodiment. [Figure 19A] Fig. 19A is an explanatory diagram showing an example of the degree of defect and a defect determination method performed by the defect determination unit of the first embodiment. [Figure 19B] Fig. 19B is an explanatory diagram showing an example of the degree of defect and a defect determination method performed by the defect determination unit of the first embodiment. [Figure 20] Fig. 20 is a flowchart showing an example of screw tightening processing performed by the screw tightening device of the first embodiment. [Figure 21] Fig. 21 is a flowchart showing an example of defect determination processing performed by the defect determination device of the first embodiment. [Figure 22] Fig. 22 is a block diagram showing an example of the functional configuration of a defect determination device according to the second embodiment. [Figure 23] Fig. 23 is a block diagram showing an example of the functional configuration of a defect determination device according to the third embodiment. [Figure 24] Fig. 24 is a block diagram showing an example of the functional configuration of a defect determination device according to the fourth embodiment. [Figure 25] Fig. 25 is a block diagram showing an example of the functional configuration of a defect determination device according to the fifth embodiment. [Figure 26] Fig. 26 is an explanatory diagram showing an example of a preprocessing method for a target physical quantity performed by the preprocessing unit of the fifth embodiment. [Figure 27] Fig. 27 is a flowchart showing an example of advance preparation processing for performing preprocessing by the defect determination device of the fifth embodiment. [Figure 28]Figure 28 is a flowchart showing an example of preprocessing by the defect detection device of the fifth embodiment. [Figure 29] Figure 29 is a block diagram showing an example of the functional configuration of the defect detection device according to the sixth embodiment. [Figure 30] Figure 30 shows an example of a screen displayed by the defect detection device of the sixth embodiment. [Figure 31] Figure 31 is a block diagram showing an example of the functional configuration of the defect detection device and main control device according to the seventh embodiment. [Figure 32] Figure 32 is a block diagram showing an example of the functional configuration of the defect detection device according to the eighth embodiment. [Modes for carrying out the invention]

[0017] The embodiments of this disclosure (hereinafter simply referred to as "these embodiments") will be described in detail below with reference to the drawings. However, this disclosure is not limited to the following embodiments. Furthermore, the following embodiments and modifications can be combined as appropriate.

[0018] (First Embodiment) First, let me explain the overview of the screw tightening device.

[0019] Figure 1 is a schematic diagram showing an example of the configuration of the screw tightening device 10 of the first embodiment. The screw tightening device 10 is a device that automatically tightens a screw 5 into a screw hole 2 formed in a workpiece 1. The screw hole 2 corresponds to a female screw, and the screw 5 corresponds to a male screw.

[0020] As shown in Figure 1, the screw tightening device 10 comprises a driver bit 20, a motor 30, a stage 40, a motor drive unit 50, a stage drive unit 60, a main control unit 70, and a defect detection device 100.

[0021] The driver bit 20 is a rotary tool used to fasten a screw 5 into a screw hole 2. The driver bit 20 is mounted on the rotating shaft (not shown) of the motor 30 and rotates around its own axis in accordance with the rotation of the rotating shaft. By rotating around its axis with the screw 5 held in place by suction at its tip, the driver bit 20 transmits the torque generated in the motor 30 to the screw 5, thereby fastening the screw 5 into the screw hole 2.

[0022] Motor 30 is the power source for the screw tightening process. Motor 30 may be, but is not limited to, a servo motor. Motor 30 generates torque for fastening the screw 5 into the screw hole 2 by rotating its shaft according to a drive signal output from the motor drive unit 50. The drive signal may be, but is not limited to, a PWM (Pulse Width Modulation) signal. Motor 30 is also equipped with a sensor, such as an encoder (not shown), which detects the torque and rotation angle of the motor 30's shaft as analog values ​​and outputs them to the motor drive unit 50 as a feedback signal. Note that torque detection may be performed by detecting the load current value of the motor 30 instead of using an encoder.

[0023] Stage 40 is a stage that can move in the vertical, horizontal, and depth directions, and is connected to a motor 30 and a driver bit 20 via the motor 30. Stage 40 moves according to the drive signal output from the stage drive unit 60, thereby moving the driver bit 20 to a predetermined position. Stage 40 is equipped with a motor (not shown), and moves by rotating the motor according to the drive signal output from the stage drive unit 60. The motor (not shown) of Stage 40 is equipped with an encoder (not shown), and outputs the detection result of this encoder as a feedback signal to the stage drive unit 60. In the first embodiment, the case in which Stage 40 can move in three axes (vertical, horizontal, and depth) is described as an example, but it is not limited to this. For example, if the workpiece 1 can be moved in the horizontal and depth directions, Stage 40 may be a stage that can move in one axis direction (vertical).

[0024] The motor drive unit 50 rotates the rotating shaft of the motor 30. The motor drive unit 50 is not limited to, but includes, for example, a drive circuit such as a motor driver or a motor amplifier. The motor drive unit 50 receives motor control signals from the main control unit 70 that instruct the rotation angle, rotation speed, and torque of the motor 30. The motor drive unit 50 generates a drive signal that rotates the rotating shaft of the motor 30 according to the control content instructed by the received motor control signals and outputs it to the motor 30.

[0025] The motor drive unit 50 also receives the aforementioned feedback signal from the motor 30. Based on the received feedback signal, the motor drive unit 50 generates waveform data showing the waveform of the torque actually generated in the motor 30 and the waveform of the actual rotational speed of the motor 30, and transmits it to the main control unit 70. For example, the motor drive unit 50 generates waveform data by sampling the received feedback signal at a fixed sampling interval (e.g., 2 ms) and quantizing it.

[0026] The stage drive unit 60 moves the stage 40. The stage drive unit 60 may include, but is not limited to, a drive circuit such as a motor driver or a motor amplifier. The stage drive unit 60 receives a stage control signal from the main control unit 70 that instructs the movement position of the stage 40, generates a drive signal to move the stage 40 to the instructed movement position, and outputs it to the stage 40. The stage drive unit 60 also receives the aforementioned feedback signal from the stage 40 and transmits it to the main control unit 70.

[0027] The main control unit 70 (an example of a control unit) provides overall control of the screw tightening device 10. The main control unit 70 may be implemented, for example, by a PLC (Programmable Logic Controller), or by software using a general-purpose processor. The main control unit 70 generates the motor control signal and stage control signal described above according to a pre-registered control flow. The main control unit 70 controls the screw tightening process by the driver bit 20 by outputting the generated motor control signal to the motor drive unit 50. For example, the main control unit 70 controls the motor 30, which is the power source used for the screw tightening process. The main control unit 70 also controls the movement of the driver bit 20 by outputting the generated stage control signal to the stage drive unit 60. By controlling the driver bit 20 in this way, the main control unit 70 automates the screw tightening process by the screw tightening device 10.

[0028] Furthermore, the main control device 70 outputs waveform data of the torque waveform and rotational speed waveform of the motor 30 received from the motor drive device 50 to the failure detection device 100, and receives the failure detection result using this waveform data from the failure detection device 100.

[0029] The defect detection device 100 performs defect detection on the screw tightening process by the screw tightening device 10. The defect detection device 100 may be implemented as software using, for example, a general-purpose processor, or it may be implemented using a PLC. Furthermore, the defect detection device 100 may be implemented as the same device as the main control device 70, or it may be implemented as a separate device from the main control device 70.

[0030] The defect detection device 100 uses the torque and rotational speed waveform data of the motor 30 received from the main control device 70 to determine if the screw tightening process by the screw tightening device 10 is defective, and transmits the defect detection result to the main control device 70. In the first embodiment, the case in which the defect to be detected is oblique tightening or cam-out will be used as an example to explain the case, but the defects to be detected are not limited to these. The defects to be detected may be, for example, foreign matter jamming, foreign matter in the female thread, screw stripping, screw breakage, incorrect screw, free rotation at the opening, galling (seizing), and bolt elongation.

[0031] Diagonal tightening is a defect in which a screw is fastened to an object while being inserted at an angle into the screw hole. Cam-out is a defect in which the tip of the screwdriver bit lifts up and comes out of the screw hole. Foreign object jamming is a defect in which a screw is fastened to an object while debris or other foreign objects are caught in the screw hole. Female screw foreign object jamming is a defect in which a screw is fastened to an object while debris or other foreign objects are caught in the screw hole. Incorrect screw is a defect in which a screw of a different size or shape than specified is used for fastening. Free rotation at the screw head is a defect in which a screw is rotated without being able to enter the screw hole. Galling (seizing) is a defect in which a screw seizes up. Bolt stretching is a defect in which a screw stretches.

[0032] Next, as an example of waveform data, we will describe the waveform of the torque generated in the motor 30.

[0033] Figure 2 shows an example of a normal torque waveform generated in the motor 30 from the start to the end of the screw tightening process by the screw tightening device 10 of the first embodiment. In the example shown in Figure 2, the vertical axis represents torque and the horizontal axis represents time. The torque waveform shown in Figure 2 is a normal waveform that indicates the screw tightening was completed successfully without any defects.

[0034] As shown in Figure 2, the screw tightening process by the screw tightening device 10 consists of four steps (processes): alignment, screwing, final tightening, and torque maintenance.

[0035] The alignment process involves moving the screw 5, which is held by suction on the driver bit 20, to a position where it can be screwed into the screw hole 2. For example, the screw 5, held by suction on the driver bit 20, is moved to a position directly above the screw hole 2, and then moved directly below, thereby moving it to a position where screwing the screw 5 into the screw hole 2 can begin. In the first embodiment, the alignment is assumed to be performed while rotating the driver bit 20, but it is not limited to this, and may be performed without rotating the driver bit 20. During the alignment process, the screw 5 is held by suction on the driver bit 20, and no tightening into the screw hole 2 has been performed. Therefore, the torque is maintained at a relatively low value.

[0036] The screwing process involves rotating the driver bit 20 while screwing the screw 5 into the screw hole 2 until the seating surface of the screw head of the held screw 5 contacts the workpiece 1, and the screw 5 is seated in the workpiece 1. During the screwing process, the screw 5 is screwed into the screw hole 2, which is a female thread, and no tightening is performed on the screw hole 2. Therefore, the torque is maintained at a relatively low value at this stage and only begins to rise after the seating surface of the screw head of the screw 5 contacts the workpiece 1.

[0037] Final tightening is the process of tightening the screw 5, which is seated in the workpiece 1, into the screw hole 2. In the final tightening process, friction increases significantly between the seating surface of the screw 5 and the workpiece 1, and between the screw 5 and the screw hole 2, as the screw 5 is tightened into the screw hole 2. As a result, a torque exceeding this friction is used, and the torque value increases rapidly.

[0038] Torque maintenance is a process that maintains the torque at the time of final tightening for a certain period of time after the final tightening is complete. Therefore, the torque is maintained at a relatively high value. Although not shown in the diagram, once torque maintenance is complete, the torque is released, and the torque drops sharply.

[0039] Thus, the torque waveform during the screw tightening process by the screw tightening device 10 reflects the characteristics of each step in the screw tightening process. Therefore, there exists an ideal torque waveform for the screw tightening process flow in the screw tightening process by the screw tightening device 10. Since the time and torque required for the screw tightening process differ depending on the size of the screw, such as its length and diameter, there is an ideal torque waveform for each screw size. Therefore, when the screw tightening process for a screw of a certain size is completed successfully, the torque waveform will be a normal torque waveform that is close to the ideal torque waveform for that size. On the other hand, in a screw tightening process where some kind of defect occurs, the effect of this defect will be reflected in the torque waveform, resulting in an abnormal torque waveform that deviates from the normal torque waveform.

[0040] In the example shown in Figure 2, the screw tightening process by the screw tightening device 10 is described as consisting of four steps: alignment, screwing, final tightening, and torque maintenance. However, the screw tightening process is not limited to these four steps. For example, the process from when the seating surface of the screw head of the screw 5 contacts the workpiece 1 until the screw 5 is seated on the workpiece 1 in the screwing process described above may be called a temporary seating step, and the screw tightening process may consist of five steps.

[0041] Next, referring to Figures 3 to 6, we will explain several types of waveform patterns that may correspond to "diagonal tightening" as examples of abnormal torque waveforms.

[0042] Figure 3 shows an example of the torque waveform when slight oblique tightening occurs during screw tightening by the screw tightening device 10 of the first embodiment. In the example shown in Figure 3, as in Figure 2, the vertical axis represents torque and the horizontal axis represents time. In the screw tightening process shown in the torque waveform of Figure 3, oblique tightening occurs because the screw 5 (male screw) is inserted horizontally offset from the center of the screw hole 2 (female screw) when screwing begins. When oblique tightening occurs, the screw 5 becomes difficult to rotate, and as indicated by reference numeral 81, the torque value increases sharply. However, in the screw tightening process shown in the torque waveform of Figure 3, after oblique tightening occurs, the screw 5 is corrected to the correct position and returns to a normal screw-in state where the screw 5 (male screw) and the screw hole 2 (female screw) are properly engaged. Therefore, as indicated by reference numeral 81, the torque value that increased sharply returns to normal. The explanation from this point onward is the same as the normal waveform explained in Figure 2, so the explanation is omitted.

[0043] In the diagonal tightening shown in Figure 3, although diagonal tightening occurs, the duration is very short, and the screw tightening process is completed with almost no damage to the threaded portion of screw 5 (male thread) and screw hole 2 (female thread). For this reason, diagonal tightening with a waveform pattern like that shown in Figure 3 is classified as a mild form of diagonal tightening.

[0044] Figure 4 shows an example of the torque waveform when moderate oblique tightening occurs during the screw tightening process using the screw tightening device 10 of the first embodiment. In the example shown in Figure 4, as in Figure 2, the vertical axis represents torque and the horizontal axis represents time. In the screw tightening process shown in the torque waveform of Figure 4, when screwing begins, the screw 5 (male screw) is inserted horizontally offset from the center of the screw hole 2 (female screw), causing oblique tightening and a sharp increase in torque value. However, in the screw tightening process shown in the torque waveform of Figure 4, even after oblique tightening occurs, the screw 5 is not corrected to the correct position, and the screwing process is carried out while the threaded portions of the screw 5 (male screw) and the screw hole 2 (female screw) grind against each other, so the torque is maintained at a relatively high value. After screwing continues for a certain period of time, the screw 5 (male screw) and the screw hole 2 (female screw) return to a normal screwing state where they properly mesh through the destruction of the threaded portions of the screw 5 (male screw) and the screw hole 2 (female screw), so the torque value drops sharply, as indicated by reference numeral 83. From this point onward, the waveform is the same as the normal waveform explained in Figure 2, so the explanation will be omitted.

[0045] In the diagonal tightening shown in Figure 4, although the diagonal tightening continues for a certain period and damage occurs to the threaded portion of screw 5 (male thread) and screw hole 2 (female thread), the screw tightening process is completed. For this reason, diagonal tightening with a waveform pattern like that shown in Figure 4 is classified as a moderately defective diagonal tightening.

[0046] Figure 5 shows an example of the torque waveform when severe oblique tightening occurs during the screw tightening process using the screw tightening device 10 of the first embodiment. In the example shown in Figure 5, as in Figure 2, the vertical axis represents torque and the horizontal axis represents time. In the screw tightening process shown in the torque waveform of Figure 5, when screwing begins, the screw 5 (male screw) is inserted horizontally offset from the center of the screw hole 2 (female screw), causing oblique tightening and a rapid increase in torque value. In the screw tightening process shown in the torque waveform of Figure 5, even after oblique tightening occurs, the screw 5 (male screw) and the threaded portion of the screw hole 2 (female screw) continue to grind against each other during screwing, and the torque reaches the target torque without the screw 5 being corrected to the correct position.

[0047] As shown in Figure 5, in the screw tightening process with the torque waveform, the required screwing (rotation at low torque) is not performed, and the screw jams prematurely. As a result, the screw tightening process is not completed properly, the screw 5 is not fully tightened, and the seating surface of the screw head of the screw 5 does not contact the workpiece 1, resulting in a floating screw state. For this reason, oblique tightening with a waveform pattern like that shown in Figure 5 is classified as a severely defective oblique tightening.

[0048] Figure 6 shows an example of the torque waveform when severe oblique tightening occurs during the screw tightening process using the screw tightening device 10 of the first embodiment. In the example shown in Figure 6, as in Figure 2, the vertical axis represents torque and the horizontal axis represents time. In the screw tightening process shown in the torque waveform of Figure 6, when screwing begins, the screw 5 (male screw) is inserted horizontally offset from the center of the screw hole 2 (female screw), causing oblique tightening and a rapid increase in torque value. In the screw tightening process shown in the torque waveform of Figure 6, because the degree of obliqueness of the screw 5 is large, the screw 5 stops rotating after oblique tightening occurs, and the torque reaches the target torque with a smaller amount of rotation compared to the case shown in Figure 5.

[0049] Thus, even with the torque waveform shown in Figure 6, the screw tightening process does not proceed as intended (rotation at low torque), and jams occur prematurely. As a result, the screw tightening process is not completed properly, the screw 5 is not fully tightened, and the seating surface of the screw head of the screw 5 does not contact the workpiece 1, resulting in a state of screw lifting. For this reason, the oblique tightening with the waveform pattern shown in Figure 5 is also classified as a severely defective oblique tightening.

[0050] As explained in Figures 3 to 6, even taking "angled tightening" as an example of a defect, the degree of defect varies. For example, if a screw is initially inserted at an angle but then returns to the correct position, mild angled tightening (intermediate phenomenon) as shown in Figure 3 or moderate angled tightening (intermediate phenomenon) as shown in Figure 4 can occur, which can be classified as either angled tightening (defective) or normal. In existing technology, it is difficult to distinguish between mild angled tightening as shown in Figure 3 or moderate angled tightening as shown in Figure 4, so mild and moderate angled tightening are either classified as normal or grouped together with severe angled tightening and classified as defective. In contrast, the first embodiment described below increases the degree of freedom in determining defects in screw tightening by making it possible to determine the degree of defect in the screw tightening process. For example, according to the first embodiment described below, mild and moderate angled tightening can also be distinguished, making it possible to determine defects that were difficult with existing technology, such as classifying the cases shown in Figures 2 to 3 as normal (good product) and the cases shown in Figures 4 to 6 as defective (angled tightening). Furthermore, according to the first embodiment described below, it becomes possible to freely design the case to be normal (good product) as shown in Figure 2 and defective (diagonally fastened) as shown in Figures 3 to 6, or to be normal (good product) as shown in Figures 2 to 3 and defective (diagonally fastened) as shown in Figures 4 to 6, or to be normal (good product) as shown in Figures 2 to 4 and defective (diagonally fastened) as shown in Figures 5 to 6, and so on.

[0051] The details of the method for determining the degree of defects, including intermediate phenomena, will be described later, but for example, even if an abnormal load occurs before seating, the degree of defects, including intermediate phenomena, can be determined in oblique tightening by characterizing tendencies such as a small rotation angle in the tightening region (the region of 80% or more of the target torque) as a feature quantity. More specifically, in the case of normal fastening, screwing proceeds smoothly and the torque increases sharply after seating, but in oblique tightening, if the threaded part is crushed, there is a tendency for abnormal torque to occur from the time of screwing, or for the torque not to increase sharply even before or after seating. Therefore, by characterizing (quantifying) these tendencies as feature quantities, the degree of defects, including intermediate phenomena, can be determined in oblique tightening.

[0052] Similarly, referring to Figures 7 to 9, we will describe several types of waveform patterns that may correspond to "cam-out" as examples of abnormal torque waveforms.

[0053] Figure 7 shows an example of a torque waveform when slight cam-out occurs during screw tightening using the screw tightening device 10 of the first embodiment. In the example shown in Figure 7, as in Figure 2, the vertical axis represents torque and the horizontal axis represents time. In the screw tightening process shown in the torque waveform of Figure 7, cam-out occurs during final tightening when the engagement between the tool hole (Phillips head) of the screw head of the screw 5 (male screw) and the driver bit 20 is disengaged. When cam-out occurs, the screw 5 spins freely and the load is removed, so the torque value drops sharply, as indicated by the reference numeral 91. However, in the screw tightening process shown in the torque waveform of Figure 7, after the screw 5 spins freely for 90 degrees, the tool hole (Phillips head) of the screw head of the screw 5 (male screw) and the driver bit 20 engage again. Therefore, the free spinning of the screw 5 is eliminated and the load returns, so the torque value rises sharply, as indicated by the reference numeral 91. The subsequent steps are the same as the normal waveform explained in Figure 2, so the explanation is omitted.

[0054] In the cam-out shown in Figure 7, although cam-out occurs, the duration is very short, and the screw tightening process is completed with almost no wear on the tool hole (Phillips head) of screw 5 (male screw). For this reason, cam-out with a waveform pattern like that shown in Figure 7 is classified as a mild case of cam-out.

[0055] Figure 8 shows an example of the torque waveform when moderate cam-out occurs during the screw tightening process using the screw tightening device 10 of the first embodiment. In the example shown in Figure 8, as in Figure 2, the vertical axis represents torque and the horizontal axis represents time. In the screw tightening process shown in the torque waveform of Figure 8, cam-out occurs repeatedly (twice) during the final tightening. Therefore, as indicated by reference numeral 93, the torque value repeatedly (twice) drops sharply and then rises sharply. Note that the waveform after the repeated (twice) cam-out is the same as the normal waveform described in Figure 2, so the explanation is omitted.

[0056] In the cam-out shown in Figure 8, multiple cam-outs occur, and some wear is observed in the tool hole (Phillips head) of screw 5 (male screw), but the screw tightening process is completed. Therefore, cam-out with a waveform pattern like that shown in Figure 8 is classified as a moderately defective cam-out.

[0057] Figure 9 shows an example of a torque waveform when severe cam-out occurs during screw tightening using the screw tightening device 10 of the first embodiment. In the example shown in Figure 9, as in Figure 2, the vertical axis represents torque and the horizontal axis represents time. In the screw tightening process shown in the torque waveform of Figure 9, cam-out occurs during final tightening, and the tool hole (Phillips head) of the screw head of the screw 5 (male screw) is completely worn away. As a result, the screw 5 continues to spin freely (the driver bit 20 does not fit into the screw 5), and the load does not return, so the cam-out continues, and as indicated by reference numeral 95, the torque value drops sharply while periodic torque changes are repeated.

[0058] Thus, in the screw tightening process with the torque waveform shown in Figure 9, the final tightening process is not completed and the torque maintenance process is not performed, resulting in the screw 5 not being fully tightened because the screw tightening process is not completed properly. For this reason, cam-out with a waveform pattern like that shown in Figure 9 is classified as a severe cam-out.

[0059] As explained in Figures 7 to 9, even with a defect type like "cam-out" as an example, the degree of defect varies. For example, if the tool hole (Phillips head) of screw 5 (male screw) and the driver bit 20 momentarily disengage, but the Phillips head is not damaged and the screw then re-engages, a mild cam-out (intermediate phenomenon) as shown in Figure 7 or a moderate cam-out as shown in Figure 8 may occur, which can be classified as either a cam-out (defect) or a normal cam-out. In existing technology, it is difficult to distinguish between the mild cam-out shown in Figure 7 and the moderate cam-out shown in Figure 8, so mild and moderate cam-outs were either classified as normal or grouped together with severe cam-outs and classified as defects. In contrast, the first embodiment described below increases the degree of freedom in determining defects in the screw-tightening process by making it possible to determine the degree of defect in the screw-tightening process. For example, according to the first embodiment described below, mild and moderate cam-out can also be distinguished, making it possible to determine defects that were difficult with existing technologies, such as classifying the cases shown in Figures 2 and 7 as normal (good products) and the cases shown in Figures 8 to 9 as defective (cam-out). Furthermore, according to the first embodiment described below, for example, it becomes possible to freely design whether the case shown in Figure 2 is normal (good products) and the cases shown in Figures 7 to 9 are defective (cam-out), or whether the cases shown in Figures 2 and 7 are normal (good products) and the cases shown in Figures 8 to 9 are defective (diagonal tightening), or whether the cases shown in Figures 2 and 7 to 8 are normal (good products) and the case shown in Figure 9 is defective (cam-out). Details of the method for determining the degree of defect, including intermediate phenomena, will be described later, but for example, by characterizing the fact that the target torque is reached and that a rapid torque decrease and torque increase occur during the tightening process as feature quantities, it is possible to determine the degree of defect, including intermediate phenomena, in cam-out.

[0060] Next, the configuration of the defect detection device 100 will be described.

[0061] Figure 10 is a block diagram showing an example of the hardware configuration of the defect detection device 100 of the first embodiment. As shown in Figure 10, the defect detection device 100 includes a processor 101, a memory 103, an auxiliary storage device 105, an input / output interface 107, a communication interface 109, and various buses 111.

[0062] The processor 101, memory 103, auxiliary storage device 105, input / output interface 107, and communication interface 109 are connected via various buses 111. Thus, in the first embodiment, the failure detection device 100 is described as an existing hardware configuration using an existing computer, but it is not limited to this. As mentioned above, the failure detection device 100 may be implemented by a PLC or may have a dedicated hardware configuration.

[0063] The processor 101 controls the overall operation of the failure detection device 100. The processor 101 may be, for example, a CPU (Central Processing Unit), but is not limited to this. The CPU can be one or more cores, and can be single-core or multi-core.

[0064] Examples of memory 103 include, but are not limited to, ROM (Read Only Memory) and RAM (Random Access Memory). ROM stores various programs, such as control programs for controlling the failure detection device 100. RAM is used as a workspace for the processor 101 to perform various controls based on the programs stored in ROM.

[0065] The auxiliary storage device 105 stores the various programs and data described above. The various programs described above only need to be stored in at least one of the memory 103 and the auxiliary storage device 105. The auxiliary storage device 105 may be, but is not limited to, at least one of existing storage devices capable of magnetic, electrical, or optical storage, such as an HDD (Hard Disk Drive), SSD (Solid State Drive), and DVD (Digital Versatile Disc). The auxiliary storage device 105 may be built into the fault detection device 100 or externally connected to the fault detection device 100 via an interface such as USB (Universal Serial Bus). Furthermore, the auxiliary storage device 105 may be a NAS (Network Attached Storage) connected via a network such as a LAN (Local Area Network) or WAN (Wide Area Network).

[0066] The input / output interface 107 is an interface between the fault detection device 100 and various input devices and various display devices used for fault detection. Examples of input devices include, but are not limited to, keyboards, mice, and touch panels. Examples of display devices include, but are not limited to, liquid crystal displays, organic electro-luminescence (OLED) displays, and touch panel displays.

[0067] The communication interface 109 may be, for example, a communication interface for a wired LAN or a wireless communication interface for a wireless LAN, but is not limited to these. The communication interface 109 may be used to acquire the various programs and data mentioned above from an external source, or it may be used to output the failure detection results from the failure detection device 100 to an external source.

[0068] Figure 11 is a block diagram showing an example of the functional configuration of the defect detection device 100 of the first embodiment. As shown in Figure 11, the defect detection device 100 includes a waveform data acquisition unit 121, a feature quantity calculation unit 123, a defect detection unit 125, and a display control unit 127. The waveform data acquisition unit 121, the feature quantity calculation unit 123, the defect detection unit 125, and the display control unit 127 can be realized, for example, by the processor 101, memory 103, and communication interface 109 described in Figure 10.

[0069] For example, the processor 101 reads a control program stored in memory 103 (ROM) or auxiliary storage device 105, or acquired from an external source via the network through the communication interface 109, and loads it into memory 103 (RAM). The processor 101 then executes various processes according to the loaded control program, thereby realizing each of the above-mentioned functional units as software. Here, the example of realizing each of the above-mentioned functional units as software has been used, but at least a part of each of the above-mentioned functional units may be realized as hardware. In this case, the functional units to be realized as hardware can be realized using hardwired circuits such as ICs (Integrated Circuits), ASICs (Application Specific Integrated Circuits), and FPGAs (Field-Programmable Gate Arrays). Alternatively, any of the above-mentioned functional units may be realized through the cooperation of software and hardware.

[0070] The waveform data acquisition unit 121 acquires waveform data showing the waveforms of physical quantities generated in the power source during the screw tightening process. Examples of physical quantities include, but are not limited to, the torque, rotational speed, rotational angle, and thrust of the power source. In the first embodiment, the power source is a motor 30 and the physical quantity is the torque of the motor 30, and this will be explained as an example, but is not limited to this.

[0071] The feature calculation unit 123 uses the waveform data acquired by the waveform data acquisition unit 121 to calculate feature quantities that characterize a part of the waveform. In this embodiment, the feature calculation unit 123 uses the waveform data acquired by the waveform data acquisition unit 121 to divide the waveform into multiple parts and calculates multiple feature quantities that characterize each of the multiple divided parts.

[0072] For example, a control program (algorithm) defines rules for identifying each part of the waveform to be characterized, and parameters used to identify each part of the waveform are stored in an auxiliary storage device 105 or the like. The feature calculation unit 123 retrieves the parameters from the auxiliary storage device 105 and, according to the rules and parameters described above, identifies and divides multiple segments from the waveform data obtained by the waveform data acquisition unit 121. The feature calculation unit 123 then characterizes (quantifies) each of the divided segments and calculates multiple features.

[0073] The multiple features are each used to distinguish between different degrees of defects in the screw tightening process. For example, taking "angled tightening" as an example, the features mentioned above could be the time, angle, or absolute value or percentage of the work done until the torque (a physical quantity) reaches the value of a parameter (an example of the first parameter). However, the use of these features is not limited to distinguishing the degree of angled tightening; they can also be used to distinguish other defects.

[0074] A concrete example of this feature will be explained using Figure 12. Figure 12 is a diagram showing an example of a feature of the first embodiment, and specifically shows the absolute value 141 of the time it takes for the torque to reach 20% of the target torque (an example of the first parameter). The torque waveform shown in Figure 12 is a normal torque waveform, similar to that in Figure 2. The absolute value 141, which is a feature shown in Figure 12, can be used, for example, to determine whether the waveform shown in the waveform data acquired by the waveform data acquisition unit 121 is a normal waveform that does not correspond to diagonal tightening. This is because, as is clear from the diagonal tightening waveforms shown in Figures 3 to 6 and the normal waveform shown in Figure 12, in the case of a normal waveform, the screw rotates smoothly during the screwing process, so the time it takes for the torque to reach 20% of the target torque tends to be longer compared to when diagonal tightening occurs.

[0075] When the feature calculation unit 123 calculates features used to determine whether or not a waveform is normal, it uses, for example, a rule to identify a part of the waveform, such as "the absolute value of the time it took for the torque to reach X1% of the target torque," and a first parameter, "20." Using this rule and first parameter, the feature calculation unit 123 calculates an absolute value 141, which is a feature shown in Figure 12 used to determine whether or not a waveform is normal, from the waveform data acquired by the waveform data acquisition unit 121.

[0076] Similarly, taking "diagonal tightening" as an example, other examples of the feature quantities mentioned above include the time, angle, or absolute value of the work done or its proportion to the whole, from when the torque reaches the parameter value until the tightening is completed. However, the use of this feature quantity is not limited to determining the degree of diagonal tightening; it can also be used to identify other defects.

[0077] A concrete example of this feature will be explained using Figure 12. Figure 12 further shows the absolute value 142 of the time taken from when the torque reaches 20% of the target torque (an example of the first parameter) until fastening is completed. The absolute value 142, a feature shown in Figure 12, can be used in conjunction with the absolute value 141 to determine, for example, whether the waveform shown in the waveform data acquired by the waveform data acquisition unit 121 corresponds to a waveform of mild angled tightening (for example, the waveform shown in Figure 3). This is because, as is clear from the angled tightening waveforms shown in Figures 3 to 6 and the normal waveform shown in Figure 12, in the case of normal waveforms and mild angled tightening waveforms, the time taken from when the torque reaches 20% of the target torque until fastening is completed tends to be shorter compared to when moderate or greater angled tightening occurs. Furthermore, by using the absolute value 141, it is possible to determine whether it is a normal waveform or a waveform of mild angled tightening, as described above.

[0078] When the feature calculation unit 123 calculates a feature used to determine whether or not the waveform is one of slight angled tightening, it uses, for example, a rule to identify a part of the waveform, such as "the absolute value of the time taken from when the torque reaches XA% of the target torque until the tightening is completed," and a parameter "20." Using this rule and parameter, the feature calculation unit 123 calculates an absolute value 142, which is a feature shown in Figure 12 used to determine whether or not the waveform is one of slight angled tightening, from the waveform data acquired by the waveform data acquisition unit 121.

[0079] Similarly, taking "angled tightening" as an example, other examples of the features mentioned above include at least one of the following values: time, angle, or the absolute value or percentage of work done, which can be identified from the region where the physical quantity torque exceeds or falls below the value of a parameter (an example of a second parameter). However, the use of this feature is not limited to determining the degree of angled tightening; it can also be used to identify other defects.

[0080] A concrete example of this feature will be explained using Figure 13. Figure 13 is a diagram showing an example of the feature of the first embodiment, and specifically shows the absolute value 144 of time identified from the region 143 where the torque is below 20% of the target torque (an example of the second parameter). The torque waveform shown in Figure 13 is a normal torque waveform, similar to that in Figure 2. The absolute value 144, which is the feature shown in Figure 13, can be used to determine the degree of angled tightening, although details will be omitted.

[0081] When calculating the features shown in Figure 13, the feature calculation unit 123 uses, for example, a rule for identifying a part of the waveform, such as "the absolute value of the time identified from the region where the torque is below X2% of the target torque," and a second parameter, "20." Using this rule and second parameter, the feature calculation unit 123 calculates the absolute value 144, which is the feature shown in Figure 13, from the waveform data acquired by the waveform data acquisition unit 121.

[0082] Similarly, taking "angled tightening" as an example, other examples of the features mentioned above include at least one of the following values: time, angle, or the absolute value or percentage of work done, which can be identified from the region where the rotational speed, a physical quantity, exceeds or falls below the value of a parameter (an example of a third parameter). However, the use of this feature is not limited to determining the degree of angled tightening; it can also be used to identify other defects.

[0083] A concrete example of this feature will be explained using Figure 14. Figure 14 is a diagram showing an example of the feature of the first embodiment, and specifically shows the absolute value 147 of time identified from regions 145 and 146 where the rotation speed exceeds 80% of the set rpm (an example of the third parameter). The absolute value 147 is the sum of the absolute value of time identified from region 145 and the absolute value of time identified from region 146. The absolute value 147, which is the feature shown in Figure 14, can be used to determine the degree of diagonal tightening, although details will be omitted.

[0084] When calculating the features shown in Figure 14, the feature calculation unit 123 uses, for example, a rule for identifying a part of the waveform, such as "the absolute value of time identified from the region where the rotation speed exceeds X3% of the set rpm," and a third parameter, "80." By using this rule and third parameter, the feature calculation unit 123 calculates the absolute value 147, which is the feature shown in Figure 14, from the waveform data acquired by the waveform data acquisition unit 121.

[0085] Similarly, taking "diagonal tightening" as an example, other examples of the features mentioned above include the time, angle, or work rate within the range of a parameter (an example of a fourth parameter), or the absolute value of the work rate or its proportion to the whole, which are physical quantities such as the time slope of torque, the angular slope of torque, the time slope of an angle, the angular slope of an angle, the time slope of work, or the angular slope of work. However, the use of these features is not limited to determining the degree of diagonal tightening; they can also be used to identify other defects.

[0086] A concrete example of this feature will be explained using Figure 15. Figure 15 shows an example of a feature in the first embodiment, specifically the absolute value 148 of the time when the magnitude of the time slope of the torque angle is in the range of 80-90%. The absolute value 148, which is the feature shown in Figure 15, can be used to determine the degree of angled tightening, although the details will be omitted.

[0087] When calculating the features shown in Figure 15, the feature calculation unit 123 uses, for example, a rule to identify a part of the waveform, such as "the absolute value of the time when the magnitude of the time slope of the torque angle is in the range of X4-1 to X4-2%", and a fourth parameter, "80 to 90". By using this rule and fourth parameter, the feature calculation unit 123 calculates the absolute value 148, which is the feature shown in Figure 15, from the waveform data acquired by the waveform data acquisition unit 121.

[0088] For example, taking "cam-out" as an example, the aforementioned feature quantities could be at least one of the following: the value of the final torque, the average value of the torque around the completion of fastening, or the variation in the torque around the completion of fastening. However, the use of these feature quantities is not limited to determining the degree of cam-out; they can also be used to identify other defects.

[0089] A concrete example of this feature will be explained using Figure 16. Figure 16 is a diagram showing an example of the feature of the first embodiment, and specifically shows the torque variation 151 during the last 20% of the time from the start to the end of the screw tightening process. The torque waveform shown in Figure 16 is the torque waveform when severe cam-out occurs, similar to that in Figure 9. The variation 151, which is the feature shown in Figure 16, can be used, for example, to determine whether the waveform shown in the waveform data acquired by the waveform data acquisition unit 121 is the waveform when severe cam-out occurs. This is because, as is clear from the normal waveform shown in Figure 2 and the cam-out waveforms shown in Figures 7 to 9, in the case of severe cam-out, the value of the torque variation 151 in the torque around the completion of fastening (for example, the last 20% of the time of the screw tightening process) tends to be larger than in other cases.

[0090] When the feature calculation unit 123 calculates a feature used to determine whether or not the torque waveform is one in the event of severe cam-out, it uses, for example, a rule to identify a part of the waveform, such as "the torque variation during the last XB% of the time from the start to the end of the screw tightening process," and a parameter "20." Using this rule and parameter, the feature calculation unit 123 calculates the variation 151, a feature shown in Figure 16, from the waveform data acquired by the waveform data acquisition unit 121, which is used to determine whether or not the waveform is one in the event of severe cam-out.

[0091] Similarly, taking "cam-out" as an example, another example of the feature quantities mentioned above is the absolute value of the time, angle, or work done, or its proportion to the whole, or the number of times the magnitude of the time slope or angular slope of the physical quantity torque remains above the value of a parameter (an example of a fifth parameter), or the number of times such a slope occurs. However, the use of this feature quantity is not limited to determining the degree of cam-out; it can also be used to determine other defects.

[0092] A concrete example of this feature will be explained using Figure 17. Figure 17 is a diagram showing an example of a feature of the first embodiment, and specifically shows the absolute value of time 153 when the magnitude of the torque time slope remains at or above the value of -3 (an example of the fifth parameter). The torque waveform shown in Figure 17 is the torque waveform when mild cam-out occurs, similar to Figure 7. The absolute value 153, which is a feature shown in Figure 17, can be used in combination with the variation 151 described above to determine, for example, whether the waveform shown in the waveform data acquired by the waveform data acquisition unit 121 is the waveform when mild cam-out occurs. Furthermore, the absolute value 153 can be used to determine, for example, whether the waveform shown in the waveform data acquired by the waveform data acquisition unit 121 is the waveform when moderate cam-out occurs (for example, the waveform shown in Figure 8). This is because, as is clear from the normal waveform shown in Figure 2 and the cam-out waveforms shown in Figures 7 to 9, the value of the absolute value 153 tends to increase as the number of cam-out occurrences increases. Furthermore, since the magnitude of the absolute value of 153 does not differ significantly between the waveforms of mild and severe cam-out, it is possible to distinguish between the waveform of mild and severe cam-out by using the aforementioned variation of 151.

[0093] When calculating the features shown in Figure 17, the feature calculation unit 123 uses, for example, a rule to identify a part of the waveform, such as "the absolute value of the time when the magnitude of the torque time slope remains at or above the value of X5," and the fifth parameter, "-3." By using this rule and the fifth parameter, the feature calculation unit 123 calculates the absolute value 153, which is the feature shown in Figure 17, from the waveform data acquired by the waveform data acquisition unit 121.

[0094] Similarly, taking "cam-out" as an example, another example of the feature quantities mentioned above is the time, angle, absolute value of work, or percentage of the total work performed when the magnitude of the time or angular slope of rotational speed, or the absolute value of said rotational speed, remains at or above the value of the sixth parameter, or at least one of the number of times such a period occurs. However, the use of this feature quantity is not limited to determining the degree of cam-out; it can also be used to determine other defects.

[0095] A concrete example of this feature will be explained using Figure 18. Figure 18 is a diagram showing an example of a feature in the first embodiment, and specifically shows the absolute value of time 155 when the absolute value of the rotational speed remains at or above 80% of the target rotational speed. The absolute value 155, which is the feature shown in Figure 18, can be used to determine the degree of cam-out, although the details will be omitted.

[0096] When calculating the features shown in Figure 18, the feature calculation unit 123 uses, for example, a rule to identify a part of the waveform, such as "the absolute value of the time when the absolute value of the rotation speed remains at or above X6% of the target rotation speed," and the sixth parameter, "80." Using this rule and the sixth parameter, the feature calculation unit 123 calculates the absolute value 155, which is the feature shown in Figure 18, from the waveform data acquired by the waveform data acquisition unit 121.

[0097] The defect determination unit 125 uses the feature quantities calculated by the feature quantity calculation unit 123 to determine if the screw tightening process is defective. In this embodiment, the defect determination unit 125 uses multiple feature quantities calculated by the feature quantity calculation unit 123 to determine if the screw tightening process is defective. In this embodiment, the defect determination unit 125 uses multiple feature quantities calculated by the feature quantity calculation unit 123 to further determine the degree of defect in the screw tightening process.

[0098] For example, for each feature calculated by the feature calculation unit 123, a threshold value for determining the degree of defect of that feature is stored in the auxiliary storage device 105 or the like. The defect determination unit 125 obtains the corresponding threshold value for each feature calculated by the feature calculation unit 123 and performs threshold determination to determine the degree of defect of the waveform shown in the waveform data acquired by the waveform data acquisition unit 121, and makes a defect determination of whether it is a good product or a defective product. The threshold determination by the defect determination unit 125 may be performed as sequential processing, where the threshold determination of each feature is processed sequentially, or as parallel processing, where the threshold determination of each feature is processed in parallel. In addition, the threshold determination by the defect determination unit 125 may perform different threshold determinations for the same feature as sequential or parallel processing.

[0099] Furthermore, the defect detection unit 125 may perform defect detection and determination of the degree of defect on specific defects such as skewed tightening or cam-out, or it may perform detection on multiple defects. When targeting specific defects, for example, as a preprocessing step, it is sufficient to roughly determine whether or not the waveform data corresponds to the specific defect based on the overall shape of the waveform.

[0100] Figures 19A and 19B are explanatory diagrams showing an example of the degree of defect and defect determination method by the defect determination unit 125 of the first embodiment. Specifically, they are explanatory diagrams of a method for determining the degree of defect of diagonal tightening and whether or not it is diagonal tightening. In the example shown in Figures 19A and 19B, the normal waveform shown in Figure 2 is shown as waveform 161, the waveform of mild diagonal tightening shown in Figure 3 is shown as waveform 163, the waveform of moderate diagonal tightening shown in Figure 4 is shown as waveform 165, the waveform of severe diagonal tightening shown in Figure 5 is shown as waveform 167, and the waveform of severe diagonal tightening shown in Figure 6 is shown as waveform 169. In addition, in the example shown in Figures 19A and 19B, the determination boundary between defective (diagonal tightening) and good product is set between mild diagonal tightening and moderate diagonal tightening. For example, in the example shown in Figure 19A, the defect detection unit 125 determines that a product is good if it matches the normal waveform 161, and that it is defective if it matches the waveform of moderate angled tightening 165 or the waveforms of severe angled tightening 167 and 169. In the example shown in Figure 19B, the defect detection unit 125 determines that a product is good if it matches the waveform of light angled tightening 163, and that it is defective if it matches the waveform of moderate angled tightening 165 or the waveforms of severe angled tightening 167 and 169. Furthermore, although the examples shown in Figures 19A and 19B describe an example in which defect detection (threshold detection) is performed by sequential processing, as described above, defect detection (threshold detection) may also be performed by parallel processing.

[0101] In the example shown in Figure 19A, the defect determination unit 125, as a first step, determines whether the type of waveform shown in the waveform data acquired by the waveform data acquisition unit 121 is normal, for example, whether it is a waveform that does not correspond to diagonal tightening. In this case, the defect determination unit 125 makes the determination using the absolute value 141, for example, the feature quantity explained in Figure 12, which is among the feature quantities calculated by the feature quantity calculation unit 123. The absolute value 141 is the absolute value of the time it takes for the torque to reach 20% of the target torque. As mentioned above, in the case of a normal waveform 161, the time it takes for the torque to reach 20% of the target torque tends to be longer than when diagonal tightening occurs. Therefore, the threshold for the absolute value 141 should be set to a value that can distinguish between the normal waveform 161 and the diagonal tightening waveforms 163, 165, 167, and 169 based on the above time.

[0102] The defect detection unit 125 performs a threshold determination using the absolute value 141. If the absolute value 141 is greater than or equal to the threshold, it determines that the type of waveform shown in the waveform data is normal (waveform 161). If the absolute value 141 is less than the threshold, it determines that the type of waveform shown in the waveform data is abnormal (a waveform other than 161). In this way, the first stage of defect determination makes it possible to determine whether the waveform shown in the waveform data is normal or slanted. However, the first stage of defect determination does not determine the degree of slanted tightening. Therefore, in the second stage and beyond, other features are used to determine the degree of slanted tightening.

[0103] In the example shown in Figure 19B, the defect determination unit 125, in the second stage, determines whether the waveform type shown in the waveform data acquired by the waveform data acquisition unit 121 is the waveform of mild angled tightening 163 or the waveforms of moderate or greater angled tightening 165, 167, 169. In this case, the defect determination unit 125 makes the determination using the absolute value 142, for example, the feature quantity explained in Figure 12, from among the feature quantities calculated by the feature quantity calculation unit 123. The absolute value 142 is the absolute value of the time taken from when the torque reaches 20% of the target torque until the fastening is completed. As mentioned above, in the case of a normal waveform 161 and a waveform of mild angled tightening 163, the time taken from when the torque reaches 20% of the target torque until the fastening is completed tends to be shorter compared to when moderate or greater angled tightening occurs. Therefore, the threshold for the absolute value 142 only needs to be set to a value that can distinguish between the waveform of mild angled tightening 163 and the waveforms of moderate or greater angled tightening 165, 167, 169 based on the above time.

[0104] The defect determination unit 125 performs a threshold determination using the absolute value 142. If the absolute value 142 is less than the threshold, it determines that the type of waveform shown in the waveform data is mild slanted tightening (waveform 163). If the absolute value 142 is equal to or greater than the threshold, it determines that the type of waveform shown in the waveform data is not mild slanted tightening (waveforms other than 161 and 163). In this way, the second stage of defect determination makes it possible to determine whether the waveform shown in the waveform data is mild slanted tightening or moderate to severe slanted tightening.

[0105] Based on the determination of the degree of defect in the first and second stages, it is possible to determine whether the type of waveform shown in the waveform data acquired by the waveform data acquisition unit 121 is normal or slightly angled tightening. For example, the defect determination unit 125 can determine that if the type of waveform shown in the waveform data acquired by the waveform data acquisition unit 121 is normal or slightly angled tightening, the screw tightening process of that waveform is good (normal), and otherwise it is defective.

[0106] Furthermore, if the determination of the degree of defect of a waveform that has been determined to be defective is to be continued, the defect determination unit 125 should, as a third step, determine whether the type of waveform shown in the waveform data acquired by the waveform data acquisition unit 121 is a moderately angled tightening waveform 165 or a severely angled tightening waveform 167, 169. Examples of feature quantities used in this determination include the tightening time.

[0107] Alternatively, the second stage of judgment can be performed without conducting the first stage, allowing for a defect determination without assessing the degree of defect. This approach allows for a determination of whether a product is good or defective without distinguishing between normal and slightly crooked tightening, thus improving the speed of defect detection.

[0108] The display control unit 127 displays the defect judgment result of the screw tightening process on the input / output interface 107 (display device). The display control unit 127 also displays the judgment result of the degree of defect in the screw tightening process on the input / output interface 107 (display device). For example, the display control unit 127 displays the waveform data acquired by the waveform data acquisition unit 121, along with the defect judgment result (whether it is defective or good) and the judgment result of the degree of defect from the defect judgment unit 125.

[0109] Next, we will explain the processing flow of the screw tightening device 10.

[0110] Figure 20 is a flowchart showing an example of a screw tightening process using the screw tightening device 10 of the first embodiment.

[0111] First, the main control device 70 moves the driver bit 20 to a predetermined screw tightening position so that the screw 5, which is held by suction on the driver bit 20, is positioned directly above the screw hole 2 (step S101).

[0112] Next, the main control device 70 lowers the driver bit 20 and presses the suction-held screw 5 into the screw hole 2 (step S103).

[0113] Next, the main control device 70 operates the motor 30 to rotate the driver bit 20 (step S105).

[0114] Furthermore, steps S101 to S105 may be performed by the main control device 70 operating the motor 30 and rotating the driver bit 20.

[0115] Next, the main control device 70 determines whether the torque generated by the motor 30 has reached the target torque (step S107).

[0116] If the target torque has not been reached (No in step S107), the main control device 70 determines whether or not the maximum screw tightening time has been reached (step S109).

[0117] If the maximum screw tightening time has not been reached (No in step S109), the process returns to step S107. On the other hand, if the maximum screw tightening time has been reached (Yes in step S109), the process ends because some kind of defect has occurred and the screw tightening process was not completed successfully.

[0118] On the other hand, if the target torque is reached in step S107 (Yes in step S107), the failure detection device 100 performs a failure detection process (step S111).

[0119] Figure 21 is a flowchart showing an example of the defect detection process by the defect detection device 100 of the first embodiment.

[0120] First, the waveform data acquisition unit 121 acquires waveform data showing the waveform of physical quantities generated in the power source during the screw tightening process. The feature calculation unit 123 then uses this waveform data to calculate n feature quantities, each of which of the n (where n is a natural number greater than or equal to 2) types of waveform the waveform corresponds to (step S201).

[0121] Next, the defect determination unit 125 sets the variable i to 1 (step S203) and determines whether the feature quantity of type i calculated by the feature quantity calculation unit 123 satisfies the threshold for type i (step S205). If the answer in step S205 is No, the defect determination unit 125 increments the variable i (step S203) and continues the process from steps S203 to S205 until the feature quantity of type i calculated by the feature quantity calculation unit 123 satisfies the threshold for type i.

[0122] If the answer in step S205 is Yes, the defect determination unit 125 determines that the degree of defect in the screw tightening process indicated by the waveform data acquired by the waveform data acquisition unit 121 is of type i (step S207).

[0123] Next, the defect determination unit 125 determines that if the degree of defect for type i is poor (Yes in step S209), the waveform data acquired by the waveform data acquisition unit 121 indicates that the screw tightening process is defective (step S211).

[0124] On the other hand, if the defect determination unit 125 determines that the screw tightening process shown in the waveform data acquired by the waveform data acquisition unit 121 is a good product (step S213), then the defect determination unit 125 determines that the degree of defect of type i is not a defect (No in step S209).

[0125] As described above, according to the first embodiment, by performing defect determination using feature quantities capable of distinguishing between minor and moderate defects (intermediate phenomena), it becomes possible to determine defects at determination boundaries that were difficult with existing technologies, such as using the boundary between minor and moderate defects as the determination boundary, thereby increasing the degree of freedom in determining defects in screw tightening processes.

[0126] Furthermore, according to the first embodiment, by using multiple feature quantities to distinguish between different degrees of defects in the screw tightening process, it becomes possible to set various boundaries of the degree of defect as judgment boundaries and perform defect judgment, thereby increasing the degree of freedom in defect judgment in the screw tightening process.

[0127] For example, according to the first embodiment, it becomes possible to determine defects by classifying the slight angled tightening shown in Figure 3 as a good product and the moderate angled tightening shown in Figure 4 as a defective product, which was difficult with existing technology. This allows for solutions to situations where both slight and moderate angled tightening are incorrectly classified as defective, resulting in an excessively high defect rate.

[0128] (Second Embodiment) In the second embodiment, an example is described in which parameters used to characterize each part of the waveform are adjusted using information about the material used in the screw tightening process. Below, the differences from the first embodiment will be mainly explained, and components having the same function as in the first embodiment will be given the same names and reference numerals as in the first embodiment, and their descriptions will be omitted.

[0129] Figure 22 is a block diagram showing an example of the functional configuration of the defect detection device 200 of the second embodiment. As shown in Figure 22, the defect detection device 200 differs from the defect detection device 100 of the first embodiment in that it includes a material information acquisition unit 231 and a parameter adjustment unit 233, as well as a feature quantity calculation unit 223.

[0130] The material information acquisition unit 231 acquires material information of the materials used in the screw tightening process. The material information is information that indicates the materials used in the screw tightening process. Examples of materials used in the screw tightening process include, but are not limited to, the types and sizes of screws and bolts.

[0131] The parameter adjustment unit 233 uses the material information acquired by the material information acquisition unit 231 to adjust the parameters used to identify multiple divisions of the waveform shown in the waveform data acquired by the waveform data acquisition unit 121. Examples of parameters include the first to sixth parameters described in the first embodiment.

[0132] The feature calculation unit 223 identifies multiple divisions from the waveform data acquired by the waveform data acquisition unit 121 according to the parameters adjusted by the parameter adjustment unit 233, features each of the identified divisions, and calculates multiple features.

[0133] As described above, according to the second embodiment, differences in the type and size of materials used in the screw tightening process can be absorbed by parameter adjustment, so the number of rules for identifying each segment of the characterized waveform can be kept low, and the complexity of the algorithm can be suppressed. In addition, it eliminates the need for retraining and other procedures that occur with the statistical method of Patent Document 1.

[0134] (Third embodiment) In the third embodiment, an example is described in which the threshold used for defect detection is adjusted using information about the material used in the screw tightening process. Below, the differences from the first embodiment will be mainly explained, and components having the same function as in the first embodiment will be given the same names and reference numerals as in the first embodiment, and their explanations will be omitted.

[0135] Figure 23 is a block diagram showing an example of the functional configuration of the defect detection device 300 of the third embodiment. As shown in Figure 23, the defect detection device 300 differs from the defect detection device 100 of the first embodiment in that it includes a material information acquisition unit 231 and a threshold adjustment unit 335, and the defect detection unit 325 is also different.

[0136] The material information acquisition unit 231 is the same as described in the second embodiment, so a detailed explanation will be omitted.

[0137] The threshold adjustment unit 335 adjusts a plurality of thresholds used for defect determination using material information acquired by the material information acquisition unit 231. Examples of thresholds include the various thresholds described in the first embodiment.

[0138] The defect determination unit 325 uses multiple feature quantities calculated by the feature quantity calculation unit 123 and multiple threshold values ​​adjusted by the threshold adjustment unit 335 to determine defects in the screw tightening process. For example, the defect determination unit 125 obtains the corresponding threshold value adjusted by the threshold adjustment unit 335 for each feature quantity calculated by the feature quantity calculation unit 123, and performs threshold determination to determine the degree of defects in the waveform data obtained by the waveform data acquisition unit 121, thereby determining whether the product is good or defective.

[0139] As described above, according to the third embodiment, differences in the type and size of materials used in the screw tightening process can be absorbed by threshold adjustment, thus keeping the number of thresholds low and suppressing an increase in the amount of data. Furthermore, it eliminates the need for retraining and other processes that occur with the statistical method of Patent Document 1.

[0140] (Fourth embodiment) In the fourth embodiment, an example is described in which parameters used to characterize each part of the waveform and thresholds used for defect detection are adjusted using information about the material used in the screw tightening process. Below, the differences from the first embodiment will be mainly explained, and components having the same functions as in the first embodiment will be given the same names and reference numerals as in the first embodiment, and their explanations will be omitted.

[0141] Figure 24 is a block diagram showing an example of the functional configuration of the defect detection device 400 of the fourth embodiment. As shown in Figure 24, the defect detection device 400 differs from the defect detection device 100 of the first embodiment in that it includes a material information acquisition unit 231, a parameter adjustment unit 233, and a threshold adjustment unit 335, and the feature quantity calculation unit 223 and defect detection unit 325 are also included.

[0142] However, the fourth embodiment is a combination of the second and third embodiments, and the material information acquisition unit 231, parameter adjustment unit 233, and feature quantity calculation unit 223 are the same as in the second embodiment, and the threshold adjustment unit 335 and defect determination unit 325 are the same as in the third embodiment, so a detailed explanation is omitted.

[0143] As described above, the fourth embodiment can achieve the effects of the second and third embodiments.

[0144] (Fifth embodiment) In the fifth embodiment, in order to accommodate various material sizes, an example is described in which the waveform data acquired by the waveform data acquisition unit 121 is preprocessed before calculating the feature quantities. Below, the differences from the first embodiment will be mainly explained, and components having the same functions as in the first embodiment will be given the same names and reference numerals as in the first embodiment, and their explanations will be omitted.

[0145] Figure 25 is a block diagram showing an example of the functional configuration of the defect detection device 500 of the fifth embodiment. As shown in Figure 25, the defect detection device 500 differs from the defect detection device 100 of the first embodiment in that it includes a material information acquisition unit 231 and a preprocessing unit 537.

[0146] The material information acquisition unit 231 is the same as described in the second embodiment, so a detailed explanation will be omitted.

[0147] If the material information obtained by the material information acquisition unit 231 is not of a predetermined size, the preprocessing unit 537 normalizes at least a portion of the waveform data obtained by the waveform data acquisition unit 121 so that the waveform data matches the waveform of the physical quantities generated in the power source when a screw tightening process is performed using a material of a predetermined size. At least a portion of the waveform subject to normalization is, for example, a physical quantity from screw alignment to seating.

[0148] Figure 26 is an explanatory diagram showing an example of a preprocessing method for a target physical quantity by the preprocessing unit 537 of the fifth embodiment. In the example shown in Figure 26, an example of the torque waveform generated in the motor 30 from the start to the end of the screw tightening process by the screw tightening device 10 is shown. Waveforms 551n and 551a show the torque waveforms during the screw tightening process with a screw of a predetermined length. Waveform 551n is the waveform of normal torque, and waveform 551a is the waveform of abnormal torque. In the example shown in Figure 26, rules, parameters, and thresholds are defined for identifying each part of the waveform characterized based on waveforms 551n and 551a.

[0149] Furthermore, waveforms 553n and 553a show the torque waveforms during screw tightening with screws shorter than a predetermined length. Waveform 553n is a normal torque waveform, and waveform 553a is an abnormal torque waveform. Waveform 555n shows the torque waveform during screw tightening with screws longer than a predetermined length. In the fifth embodiment, even if the waveform data acquired by the waveform data acquisition unit 121 shows the torque waveforms during screw tightening with screws of a different length than the predetermined length screws such as waveforms 553n, 553a, and 555n, the preprocessing unit 537 normalizes these waveforms so that it can handle defect detection and determination of the degree of defect.

[0150] Here, the maximum torque value does not change with respect to the length of the screw, but the time from the start of the screw tightening process (alignment) to seating changes proportionally to the length of the screw. Therefore, as shown in Figure 26, even if the length of the screw is different, the shape of the waveform from seating to the end of the screw tightening process (torque maintenance) does not change much, but the shape of the waveform from the start of the screw tightening process to seating compresses or expands in the time direction depending on the size of the screw.

[0151] Therefore, in the fifth embodiment, the pre-processing unit 537 normalizes the torque waveforms during screw tightening with screws shorter than a predetermined length, such as waveforms 553n and 553a, so that the waveform from the start of the screw tightening process to seating becomes longer in the time direction. Specifically, the pre-processing unit 537 normalizes the first sample number, which is the number of samples from the start of the screw tightening process to seating for waveforms 553n and 553a, to a specified sample number, which is the number of samples from the start of the screw tightening process to seating for waveforms 551n and 551a.

[0152] Similarly, for torque waveforms during screw tightening with screws longer than a predetermined length, such as waveform 555n, the preprocessor 537 normalizes the waveform from the start of the screw tightening process to seating so that it is shortened in the time direction. Specifically, the preprocessor 537 normalizes the second sample count, which is the number of samples from the start of the screw tightening process to seating for waveform 555n, so that it becomes the specified number of samples from the start of the screw tightening process to seating for waveforms 551n and 551a.

[0153] This makes it possible to determine whether a screw is defective or how severe it is defective, even when the torque waveform of a screw with a length different from a predetermined length is input.

[0154] Furthermore, for waveforms from seating to the end of the screw tightening process, if the difference in the number of samples is within the margin of error, time normalization is not required.

[0155] Figure 27 is a flowchart showing an example of a pre-preparation process for pre-processing by the defect detection device 500 of the fifth embodiment.

[0156] First, the waveform data acquisition unit 121 inputs t normal torque waveforms during the screw tightening process using a screw of a new length, in order to accommodate a screw of a new length that is different from a screw of a predetermined length (step S401). Next, the pre-processing unit 537 detects the seating point for each of the t normal torque waveforms that have been input (step S403). Next, the pre-processing unit 537 calculates the average seating time, which is the average seating time from the start of the screw tightening process (alignment) to seating, for each of the t normal torque waveforms that have been input (step S405).

[0157] Figure 28 is a flowchart showing an example of preprocessing by the defect detection device 500 of the fifth embodiment.

[0158] First, the pre-processing unit 537 sets the average seating time calculated in the pre-preparation process and the specified number of samples from the start of the screw tightening process (alignment) with a predetermined length of screw to seating (step S501).

[0159] Next, the waveform data acquisition unit 121 inputs the torque waveform during a new screw tightening process using a screw of a new length as the torque waveform to be predicted (step S503).

[0160] Next, the pre-processing unit 537 divides the input torque waveform to be predicted into two waveforms using the average seating time: one from the start of the screw tightening process (alignment) to before seating, and another from after seating to the end of the screw tightening process (torque maintenance) (step S505).

[0161] Next, if the number of waveform samples from the start of the screw tightening process (alignment) to before seating is less than the specified number of samples (Yes in step S507), the pre-processing unit 537 interpolates the samples so that they become equal to the specified number of samples (step S509).

[0162] On the other hand, if the number of waveform samples from the start of the screw tightening process (alignment) to before seating is greater than the specified number of samples (No in step S507), the pre-processing unit 537 reduces the number of sample points to be equal to the specified number of samples (step S511).

[0163] Next, if the number of waveform samples from seating to the end of the screw tightening process (torque maintenance) is less than the specified number of samples (Yes in step S513), the pre-processing unit 537 interpolates the samples so that the number of samples becomes equal to the specified number (step S515).

[0164] On the other hand, if the number of waveform samples from seating to the end of the screw tightening process (torque maintenance) is greater than the specified number of samples (No in step S513), the pre-processing unit 537 reduces the number of sample points to equal the specified number of samples (step S517).

[0165] As described above, in the fifth embodiment, since the change in torque waveform differs before and after seating when the screw length is different, the normalization is also divided into before and after seating, and normalization is performed to a different degree for each. For this reason, according to the fifth embodiment, defect detection and determination of the degree of defect can be performed in response to a wide variety of screw sizes.

[0166] (Sixth Embodiment) In the sixth embodiment, an example of adjusting the threshold used for defect detection based on statistical analysis of feature quantities will be described. Below, the differences from the first embodiment will be mainly explained, and components having the same functions as in the first embodiment will be given the same names and reference numerals as in the first embodiment, and their descriptions will be omitted.

[0167] Figure 29 is a block diagram showing an example of the functional configuration of the defect detection device 600 of the sixth embodiment. As shown in Figure 29, the defect detection device 600 differs from the defect detection device 100 of the first embodiment in that it includes a threshold adjustment unit 635, and the feature quantity calculation unit 623, defect detection unit 625, and display control unit 627 are different.

[0168] The feature calculation unit 623 further calculates statistical values ​​for a predetermined type of feature from among the multiple calculated features. The predetermined type of feature may be any of the multiple features. Alternatively, each of the multiple features may be a predetermined type of feature. For example, the feature calculation unit 623 obtains the predetermined type of feature for which statistical values ​​are to be calculated for an arbitrary period (e.g., one month) at regular intervals (e.g., daily) and calculates statistical values. Examples of statistical values ​​include, but are not limited to, the mean and variance.

[0169] The threshold adjustment unit 635 adjusts the threshold of a predetermined type of feature used for defect detection according to the progression from previously calculated statistical values ​​to the latest statistical values ​​by the feature calculation unit 623. For example, the threshold adjustment unit 635 adjusts the threshold by reflecting the progression from previously calculated statistical values ​​to the latest statistical values ​​(the difference between previously calculated statistical values ​​and the latest statistical values) in the threshold of a predetermined type of feature used for defect detection.

[0170] The defect detection unit 625 uses predetermined types of feature quantities and adjusted thresholds to determine defects in the screw tightening process and the degree of defects.

[0171] The display control unit 627 displays a screen on the input / output interface 107 (display device) that shows the relationship between a predetermined type of feature quantity and an adjusted threshold. For example, the display control unit 127 displays a screen on the input / output interface 107 (display device) as shown in Figure 30. Figure 30 is a diagram showing an example of a screen displayed by the defect detection device 600 of the sixth embodiment. The screen shown in Figure 30 has a predetermined type of feature quantity on the vertical axis and time progression on the horizontal axis. In the screen shown in Figure 30, up to time T, the threshold range for the predetermined type of feature quantity is TH1 to TH2, but after time T, it is shifted upward by the difference D of the statistical value, and the threshold range for the predetermined type of feature quantity is TH3 to TH4. In the screen shown in Figure 30, hatched dots indicate the predetermined type of feature quantity, and if they are within the threshold range, it indicates a good product, and if they are outside the threshold range, it indicates a defect. In the screen shown in Figure 30, all feature quantities except 651 are judged to be good products, which shows that the threshold adjustment after time T is functioning well.

[0172] As described above, according to the sixth embodiment, since threshold adjustment for absorbing changes in waveform shape caused by users such as workers performing screw tightening operations, and changes in waveform shape caused by wear of the driver bit 20, etc., is automated, convenience can be greatly improved.

[0173] Furthermore, according to the sixth embodiment, it becomes possible to perform a full inspection that reveals the patterns of the screw tightening process, allowing for observation of the progression of each pattern and enabling the implementation of countermeasures against the causes.

[0174] (Seventh Embodiment) In the seventh embodiment, an example is described in which the determination result of the degree of defects in the screw tightening process is used for feedback control. Below, the differences from the first embodiment will be mainly explained, and components having the same functions as in the first embodiment will be given the same names and reference numerals as in the first embodiment, and their explanations will be omitted.

[0175] Figure 31 is a block diagram showing an example of the functional configuration of the defect detection device 100 and the main control device 770 of the seventh embodiment. As shown in Figure 31, the main control device 770 in the seventh embodiment differs from that of the first embodiment.

[0176] The main control unit 770 performs feedback control according to the results of the screw tightening process defect determination, which is performed in real time by the defect determination device 100. For example, if the main control unit 770 determines that the screw tightening process defect is minor, it performs feedback control such as reversing the rotation of the motor 30 and repeating the screw tightening process.

[0177] (Eighth embodiment) In the eighth embodiment, an example is described in which feature quantities used for defect detection are identified using information about the material used in the screw tightening process. Below, the differences from the first embodiment will be mainly explained, and components having the same functions as in the first embodiment will be given the same names and reference numerals as in the first embodiment, and their explanations will be omitted.

[0178] Figure 32 is a block diagram showing an example of the functional configuration of the defect detection device 800 of the eighth embodiment. As shown in Figure 32, the defect detection device 800 differs from the defect detection device 100 of the first embodiment in that it includes a material information acquisition unit 231 and a feature quantity identification unit 839, and a feature quantity calculation unit 823.

[0179] The material information acquisition unit 231 is the same as described in the second embodiment, so a detailed explanation will be omitted.

[0180] The feature identification unit 839 uses the material information acquired by the material information acquisition unit 231 to identify feature quantities to be used for defect detection.

[0181] The feature calculation unit 823 uses the waveform data acquired by the waveform data acquisition unit 121 to calculate the features identified by the feature identification unit 839 from among multiple features.

[0182] As described above, according to the eighth embodiment, feature quantities can be calculated taking into account that the degree of defects differs depending on the material used in the screw tightening process. For example, if the strength of the male screw is stronger than that of the female screw, the female screw tends to be crushed first and return to normal more easily, so moderate angled tightening is less likely to occur. Therefore, it becomes possible to take efficient measures such as omitting the calculation of feature quantities for discriminating moderate angled tightening.

[0183] (Variation 1) In the embodiments described above, defects such as crooked tightening and cam-out were mainly explained, but the methods of each embodiment can also be applied to defects other than these.

[0184] For example, in the case of foreign object jamming, depending on the height and softness of the foreign object such as dirt, mild or moderate foreign object jamming (intermediate phenomenon) can occur, which can be classified as either foreign object jamming (defective) or normal. Here, in the case of soft foreign objects, the torque waveform tends to rise slowly, while in the case of foreign objects that are tall or hard, the cumulative rotation angle tends to decrease. Therefore, by characterizing these tendencies as feature quantities, the degree of defect, including intermediate phenomena, can be determined for foreign object jamming, similar to the embodiments described above.

[0185] Furthermore, in the case of foreign matter in female threads, for example, depending on the size of the foreign matter such as debris, the screw may remain stuck and fastening may be completed without the screw coming loose. This can result in mild or moderate female thread foreign matter (intermediate phenomenon) that can be classified as either defective or normal. Therefore, even if an abnormal load occurs before seating, by characterizing the tendency for a small rotation angle in the tightening region (the region of 80% or more of the target torque, etc.) as a feature quantity, the degree of defect, including intermediate phenomena, can be determined for female thread foreign matter as well, similar to the embodiments described above.

[0186] For example, in the case of a screw error, depending on how much the screw length and diameter deviate from the design value, minor or moderate screw errors (intermediate phenomena) can occur that can be classified as either screw errors (defects) or normal. Here, regarding screw length, by characterizing the cumulative rotation angle as a feature, the degree of defect, including intermediate phenomena, can be determined for screw errors caused by screw length, similar to the embodiments described above. Regarding screw diameter, when a screw slightly thicker than the design is used, the torque during fastening tends to be larger, and when a screw slightly thinner than the design is used, the torque during fastening tends to be smaller. Therefore, by characterizing these tendencies as feature quantities, the degree of defect, including intermediate phenomena, can be determined for screw errors caused by screw diameter, similar to the embodiments described above.

[0187] Furthermore, for example, in the case of free rotation at the opening, depending on the degree of horizontal misalignment, the screw may initially free rotate at the opening of the female thread, but then engage with the female thread and complete fastening. This can result in mild or moderate free rotation at the opening (intermediate phenomenon) that can be classified as either free rotation at the opening (defective) or normal. Therefore, by characterizing the rotation angle and torque of the free-rotating portion as features, and whether or not fastening is completed when the target torque is reached, the degree of defect, including intermediate phenomena, can be determined for free rotation at the opening, similar to the embodiments described above.

[0188] Furthermore, in the case of galling (seizing), for example, the male and female threads may begin to bond but not completely, and then the screws rotate to complete the fastening. This can result in mild or moderate galling (intermediate phenomena) that can be classified as either galling (defective) or normal. Therefore, even if an abnormal load occurs before seating, by characterizing the tendency for a small rotation angle in the tightening region (the region of 80% or more of the target torque, etc.) as a feature quantity, the degree of defect, including intermediate phenomena, can be determined for galling as well, similar to the embodiments described above.

[0189] For example, in the case of bolt elongation, depending on the degree to which friction is low and the screw rotates too much, mild or moderate bolt elongation (intermediate phenomenon) can occur, which can be classified as either bolt elongation (defective) or normal. Therefore, by characterizing the trend, such as how much greater the rotation angle from seating to completion of fastening (reaching the target torque) is compared to normal, the degree of defect, including intermediate phenomena, can be determined for bolt elongation as well, similar to the embodiments described above.

[0190] (Modification 2) In the embodiments described above, the torque waveform was used as an example to explain the physical quantity generated in the motor 30, but it is not limited to this. The rotational speed waveform of the motor 30 can also be used, similar to the torque waveform, as it reflects the characteristics of each step in the screw tightening process. In the case of the rotational speed waveform of the motor 30, the rotational speed is maintained at a relatively high value until the tightening of the screw 5 into the screw hole 2 begins. Then, when the screw is temporarily seated, the rotational speed begins to fall from its relatively high value, and when the screw is fully tightened, the rotational speed decreases rapidly and is maintained at a relatively low value. Furthermore, in the embodiments described above, the physical quantity generated in the motor 30 may be a combination of the characteristic quantities of the torque waveform and the characteristic quantities of the rotational speed waveform.

[0191] (Variation 3) In the embodiments described above, a screw tightening device that automatically performs screw tightening based on control by a main control device was used as an example, but the invention is not limited to this and can also be applied to electric screwdrivers that are operated manually by users such as workers.

[0192] (Modification 4) In the fifth embodiment described above, the waveform of torque was used as an example to explain the physical quantity generated in the motor 30. However, when the length of the screw is different, a similar trend occurs in the waveform of the rotational speed of the motor 30. For this reason, the waveform of the rotational speed of the motor 30 can also be used in the same way as the waveform of torque.

[0193] (Variation 5) In the fifth embodiment described above, pretreatment for cases where the screw length is different was explained, but the method described in the fifth embodiment can also be applied when the screw diameter is different. However, when the screw diameter is different, the time required to complete the entire screw tightening process, not just until seating, changes. Also, when the screw diameter is different, the torque required to complete the entire screw tightening process changes. For this reason, when the screw diameter is different, normalization in the time direction and normalization in the torque direction should be considered.

[0194] (program) The programs executed by the failure detection devices of each of the above embodiments and modifications are provided as installable or executable files stored on a computer-readable storage medium such as a CD-ROM, CD-R, memory card, DVD, or flexible disk (FD).

[0195] Furthermore, the programs executed by the defect detection devices of each of the above embodiments and modifications may be stored on a computer connected to a network such as the Internet and provided by allowing download via the network. Alternatively, the programs executed by the defect detection devices of each of the above embodiments and modifications may be provided or distributed via a network such as the Internet. Furthermore, the programs executed by the defect detection devices of each of the above embodiments and modifications may be pre-installed in ROM or the like and provided.

[0196] The programs executed in the defect detection devices of each of the above embodiments and modifications are configured as modules for realizing the above-described parts on a computer. In actual hardware, for example, the CPU reads the learning program from the HDD into RAM and executes it, thereby realizing the above-described parts on the computer.

[0197] As described above, according to each of the embodiments and modifications described above, it is possible to increase the degree of freedom in determining defects in the screw tightening process.

[0198] The embodiments and modifications described above are merely examples of how this disclosure may be implemented, and they do not restrict the technical scope of this disclosure. Therefore, this disclosure can be implemented in various ways without departing from its essence or its main features. For example, the embodiments and modifications described above may be combined as appropriate on a component basis. Also, for example, some components may be removed from the total components in the embodiments and modifications described above.

[0199] In the above explanation, the notation "...part" used for each component may be replaced with other notations such as "...assembly," "...circuitry," "...device," "...unit," or "...module."

[0200] This disclosure includes the following aspects:

[0201] (1) A waveform data acquisition unit that acquires waveform data showing the waveform of physical quantities generated in the power source during screw tightening, A feature calculation unit that uses the waveform data to calculate multiple feature quantities that characterize each of the multiple divisions obtained by dividing the waveform into multiple parts, A defect determination unit that uses the above-mentioned multiple feature quantities to determine if the screw tightening process is defective, A defect detection device equipped with the following features.

[0202] (2) A material information acquisition unit that acquires material information of the material used in the above process, The system further includes a parameter adjustment unit that adjusts parameters used to identify the plurality of divided portions using the material information, The feature calculation unit identifies the plurality of divisions according to the adjusted parameters, characterizes each of the identified plurality of divisions, and calculates the plurality of feature quantities. (1) The defect detection device described above.

[0203] (3) A material information acquisition unit that acquires material information of the material used in the screw tightening process, The system further includes a threshold adjustment unit that adjusts a plurality of thresholds used for defect determination using the material information, The defect detection unit uses the plurality of feature quantities and the adjusted plurality of thresholds to determine if the screw tightening process is defective. (1) The defect detection device described above.

[0204] (4) A material information acquisition unit that acquires material information of the material used in the screw tightening process, A parameter adjustment unit that adjusts the parameters used to identify the plurality of divided parts using the material information, The system further includes a threshold adjustment unit that adjusts a plurality of thresholds used for defect determination using the material information, The feature calculation unit identifies the plurality of divisions according to the adjusted parameters, characterizes each of the identified plurality of divisions, and calculates the plurality of feature quantities. The defect detection unit uses the plurality of feature quantities and the adjusted plurality of thresholds to determine if the screw tightening process is defective. (1) The defect detection device described above.

[0205] (5) A material information acquisition unit that acquires material information of the material used in the screw tightening process, If the material indicated by the material information is not of a predetermined size, the system further includes a preprocessing unit that normalizes at least a portion of the waveform indicated by the waveform data so that the waveform indicated by the waveform data matches the waveform of a physical quantity generated in the power source when a screw tightening process is performed using the material of the predetermined size. (1) The defect detection device described above.

[0206] (6) The feature calculation unit further calculates statistical values ​​of a predetermined type of feature from among the plurality of features, The system further includes a threshold adjustment unit that adjusts the threshold of a predetermined type of feature quantity used for the defect determination, according to the trend from previously calculated statistical values ​​to the latest statistical values. The defect determination unit determines whether the screw tightening process is defective using the predetermined type of feature quantity and the adjusted threshold. (1) The defect detection device described above.

[0207] (7) The system further includes a display control unit that displays a screen showing the relationship between the predetermined type of feature quantity and the adjusted threshold. (6) The defect detection device described above.

[0208] (8) Each of the aforementioned feature quantities is a feature quantity for determining the degree of defects in the screw tightening process, The defect determination unit further determines the degree of defect in the screw tightening process. (1) The defect detection device described above.

[0209] (9) The system further includes a display control unit that displays the result of determining the degree of defects in the screw tightening process. (8) The defect detection device described above.

[0210] (10) The defect determination unit determines the degree of defect during the screw tightening process, The result of determining the degree of defects in the screw tightening process is used in feedback control. (8) The defect detection device described above.

[0211] (11) A material information acquisition unit that acquires material information of the material used in the screw tightening process, The system further comprises a feature quantity identification unit that identifies feature quantities used for defect determination using the material information, The feature calculation unit uses the waveform data to calculate the specified feature from among the plurality of feature quantities. (8) The defect detection device described above.

[0212] (12) The plurality of feature quantities include at least one of the following values: the time, angle, or absolute value or proportion of the total work taken for the physical quantity torque to reach the value of the first parameter; the time, angle, or absolute value or proportion of the total work taken from the time the physical quantity torque reached the value of the first parameter until fastening was completed; the time, angle, or absolute value or proportion of the total work specified from the region in which the physical quantity torque is above or below the value of the second parameter; the time, angle, or absolute value or proportion of the total work specified from the region in which the physical quantity rotational speed is above or below the value of the third parameter; and the time, angle, or proportion of the total work taken when the magnitude of the time slope of the torque, the angular slope of the torque, the time slope of the angle, the angular slope of the angle, the time slope of the work, or the angular slope of the work is within the range of the fourth parameter. (1) The defect detection device described above.

[0213] (13) The plurality of features include the time, angle, absolute value of work, or proportion of the total, or the number of times the magnitude of the time slope or angular slope of the physical quantity torque continues at or above the value of the fifth parameter, and at least one of the time, angle, or absolute value of work, or proportion of the total, or the number of times the magnitude of the time slope or angular slope of the physical quantity rotational speed or the absolute value of the rotational speed continues at or above the value of the sixth parameter, (1) The defect detection device described above.

[0214] (14) A waveform data acquisition unit that acquires waveform data showing the waveform of physical quantities generated in the power source during screw tightening, A feature calculation unit calculates a feature that characterizes a part of the waveform using the waveform data, A defect determination unit that uses the aforementioned feature quantities to determine if the screw tightening process is defective, Equipped with, The aforementioned feature quantity includes at least one of the following values: the time, angle, or absolute value or percentage of the total work taken for the physical quantity torque to reach the value of the first parameter; the time, angle, or absolute value or percentage of the total work taken from the time the physical quantity torque reached the value of the first parameter until fastening was completed; the time, angle, or absolute value or percentage of the total work specified from the region in which the physical quantity torque is above or below the value of the second parameter; the time, angle, or absolute value or percentage of the total work specified from the region in which the physical quantity rotational speed is above or below the value of the third parameter; and the time, angle, or absolute value or percentage of the total work taken when the magnitude of the time slope of the torque, the angular slope of the torque, the time slope of the angle, the angular slope of the angle, the time slope of the work, or the angular slope of the work is within the range of the fourth parameter. Defective judgment device.

[0215] (15) A waveform data acquisition unit that acquires waveform data showing the waveform of physical quantities generated in the power source during screw tightening, A feature calculation unit calculates a feature that characterizes a part of the waveform using the waveform data, A defect determination unit that uses the aforementioned feature quantities to determine if the screw tightening process is defective, Equipped with, The aforementioned multiple feature quantities include at least one of the following values: the time, angle, absolute value of the work done, or its proportion to the total, or the number of times the magnitude of the time slope or angular slope of the physical quantity torque continues at or above the value of the fifth parameter; and the time, angle, absolute value of the work done, or its proportion to the total, or the number of times the magnitude of the time slope or angular slope of the physical quantity rotational speed, or the absolute value of the rotational speed, continues at or above the value of the sixth parameter. Defective judgment device.

[0216] (16) A control unit that controls the power source used during the screw tightening process, A waveform data acquisition unit that acquires waveform data showing the waveform of a physical quantity generated in the power source during the screw tightening process, A feature calculation unit that uses the waveform data to calculate multiple feature quantities that characterize each of the multiple divisions obtained by dividing the waveform into multiple parts, The system includes a defect determination unit that uses the aforementioned multiple feature quantities to determine if the screw tightening process is defective, Each of the aforementioned feature quantities is a feature quantity used to distinguish between different degrees of defects in the screw tightening process. The defect determination unit further determines the degree of defect in the screw tightening process, The control unit performs feedback control according to the determination result of the degree of defect in the screw tightening process. Defect detection system.

[0217] (17) A waveform data acquisition step in which the waveform data acquisition unit acquires waveform data showing the waveform of a physical quantity generated in the power source during the screw tightening process, A feature calculation unit performs a feature calculation step in which it calculates multiple feature quantities that characterize each of the multiple divisions obtained by dividing the waveform into multiple parts using the waveform data, A defect determination step in which the defect determination unit uses the plurality of feature quantities to determine if the screw tightening process is defective, A defect detection method that includes the following.

[0218] (18) A waveform data acquisition step in which waveform data is obtained that shows the waveform of a physical quantity generated in the power source during the screw tightening process, A feature calculation step of calculating multiple feature quantities that characterize each of the multiple divisions obtained by dividing the waveform into multiple parts using the waveform data, A defect determination step that uses the above-mentioned multiple feature quantities to determine if the screw tightening process is defective, A program that causes a computer to execute something. [Explanation of Symbols]

[0219] 1 Work 2 screw holes 5 screws 10 Screw tightening device 20 Driver Bits 30 motors 40 stages 50 Motor drive unit 60 Stage drive unit 70, 770 Main control unit 100, 200, 300, 400, 500, 600, 800 Defective judgment device 101 Processors 103 memory 105 Auxiliary storage device 107 Input / Output Interfaces 109 Communication Interface 111 Various buses 121 Waveform data acquisition unit 123, 223, 623, 823 Feature calculation unit 125, 325, 625 Defective judgment part 127, 627 Display Control Unit 231 Material Information Acquisition Department 233 Parameter adjustment section 335, 635 Threshold adjustment section 537 Pre-processing section 839 Feature Extraction Unit

Claims

1. A waveform data acquisition unit acquires waveform data showing the waveform of physical quantities generated in the power source during screw tightening, A feature calculation unit that uses the waveform data to calculate multiple feature quantities that characterize each of the multiple divisions obtained by dividing the waveform into multiple parts, A defect determination unit that uses the above-mentioned multiple feature quantities to determine if the screw tightening process is defective, A defect detection device equipped with the following features.

2. A material information acquisition unit that acquires material information of the material used in the screw tightening process, The system further includes a parameter adjustment unit that adjusts parameters used to identify the plurality of divided portions using the material information, The feature calculation unit identifies the plurality of divisions according to the adjusted parameters, characterizes each of the identified plurality of divisions, and calculates the plurality of feature quantities. The defect detection device according to claim 1.

3. A material information acquisition unit that acquires material information of the material used in the screw tightening process, The system further includes a threshold adjustment unit that adjusts a plurality of thresholds used for defect determination using the material information, The defect detection unit uses the plurality of feature quantities and the adjusted plurality of thresholds to determine if the screw tightening process is defective. The defect detection device according to claim 1.

4. A material information acquisition unit that acquires material information of the material used in the screw tightening process, A parameter adjustment unit that adjusts the parameters used to identify the plurality of divided parts using the material information, The system further includes a threshold adjustment unit that adjusts a plurality of thresholds used for defect determination using the material information, The feature calculation unit identifies the plurality of divisions according to the adjusted parameters, characterizes each of the identified plurality of divisions, and calculates the plurality of feature quantities. The defect detection unit uses the plurality of feature quantities and the adjusted plurality of thresholds to determine if the screw tightening process is defective. The defect detection device according to claim 1.

5. A material information acquisition unit that acquires material information of the material used in the screw tightening process, If the material indicated by the material information is not of a predetermined size, the system further includes a preprocessing unit that normalizes at least a portion of the waveform indicated by the waveform data so that the waveform indicated by the waveform data matches the waveform of a physical quantity generated in the power source when a screw tightening process is performed using the material of the predetermined size. The defect detection device according to claim 1.

6. The feature calculation unit further calculates statistical values ​​of a predetermined type of feature from among the plurality of features, The system further includes a threshold adjustment unit that adjusts the threshold of a predetermined type of feature quantity used for the defect determination, according to the trend from previously calculated statistical values ​​to the latest statistical values. The defect determination unit determines whether the screw tightening process is defective using the predetermined type of feature quantity and the adjusted threshold. The defect detection device according to claim 1.

7. The system further includes a display control unit that displays a screen showing the relationship between the predetermined type of feature quantity and the adjusted threshold. The defect detection device according to claim 6.

8. Each of the aforementioned feature quantities is a feature quantity used to distinguish between different degrees of defects in the screw tightening process. The defect determination unit further determines the degree of defect in the screw tightening process. The defect detection device according to claim 1.

9. The system further includes a display control unit that displays the result of determining the degree of defects in the screw tightening process. The defect detection device according to claim 8.

10. The defect determination unit determines the degree of defect during the screw tightening process, The result of determining the degree of defects in the screw tightening process is used in feedback control. The defect detection device according to claim 8.

11. A material information acquisition unit that acquires material information of the material used in the screw tightening process, The system further comprises a feature quantity identification unit that identifies feature quantities used for defect determination using the material information, The feature calculation unit uses the waveform data to calculate the specified feature from among the plurality of feature quantities. The defect detection device according to claim 8.

12. The aforementioned plurality of feature quantities include at least one of the following values: the time, angle, or absolute value or percentage of the total work taken for the physical quantity torque to reach the value of the first parameter; the time, angle, or absolute value or percentage of the total work taken from the time the physical quantity torque reached the value of the first parameter until fastening was completed; the time, angle, or absolute value or percentage of the total work specified from the region in which the physical quantity torque is above or below the value of the second parameter; the time, angle, or absolute value or percentage of the total work specified from the region in which the physical quantity rotational speed is above or below the value of the third parameter; and the time, angle, or absolute value or percentage of the total work taken when the magnitude of the time slope of the torque, the angular slope of the torque, the time slope of the angle, the angular slope of the angle, the time slope of the work, or the angular slope of the work is within the range of the fourth parameter. The defect detection device according to claim 1.

13. The aforementioned multiple feature quantities include at least one of the following values: the time, angle, absolute value of the work done, or its proportion to the total, or the number of times the magnitude of the time slope or angular slope of the physical quantity torque continues at or above the value of the fifth parameter; and the time, angle, absolute value of the work done, or its proportion to the total, or the number of times the magnitude of the time slope or angular slope of the physical quantity rotational speed, or the absolute value of the rotational speed, continues at or above the value of the sixth parameter. The defect detection device according to claim 1.

14. A waveform data acquisition unit acquires waveform data showing the waveform of physical quantities generated in the power source during screw tightening, A feature calculation unit calculates a feature that characterizes a part of the waveform using the waveform data, A defect determination unit that uses the aforementioned feature quantities to determine if the screw tightening process is defective, Equipped with, The aforementioned characteristic quantity includes at least one of the following values: the time, angle, or absolute value or percentage of the total work taken for the physical quantity torque to reach the value of the first parameter; the time, angle, or absolute value or percentage of the total work taken from the time the physical quantity torque reached the value of the first parameter until fastening was completed; the time, angle, or absolute value or percentage of the total work specified from the region in which the physical quantity torque is above or below the value of the second parameter; the time, angle, or absolute value or percentage of the total work specified from the region in which the physical quantity rotational speed is above or below the value of the third parameter; and the time, angle, or absolute value or percentage of the total work taken when the magnitude of the time slope of the torque, the angular slope of the torque, the time slope of the angle, the angular slope of the angle, the time slope of the work, or the angular slope of the work is within the range of the fourth parameter. Defective judgment device.

15. A waveform data acquisition unit acquires waveform data showing the waveform of physical quantities generated in the power source during screw tightening, A feature calculation unit calculates a feature that characterizes a part of the waveform using the waveform data, A defect determination unit that uses the aforementioned feature quantities to determine if the screw tightening process is defective, Equipped with, The aforementioned multiple feature quantities include at least one of the following values: the time, angle, absolute value of the work done, or its proportion to the total, or the number of times the magnitude of the time slope or angular slope of the physical quantity torque continues at or above the value of the fifth parameter; and the time, angle, absolute value of the work done, or its proportion to the total, or the number of times the magnitude of the time slope or angular slope of the physical quantity rotational speed, or the absolute value of the rotational speed, continues at or above the value of the sixth parameter. Defective judgment device.

16. A control unit that controls the power source used in the screw tightening process, A waveform data acquisition unit that acquires waveform data showing the waveform of a physical quantity generated in the power source during the screw tightening process, A feature calculation unit that uses the waveform data to calculate multiple feature quantities that characterize each of the multiple divisions obtained by dividing the waveform into multiple parts, The system includes a defect determination unit that uses the aforementioned multiple feature quantities to determine if the screw tightening process is defective, Each of the aforementioned feature quantities is a feature quantity used to distinguish between different degrees of defects in the screw tightening process. The defect determination unit further determines the degree of defect in the screw tightening process, The control unit performs feedback control according to the determination result of the degree of defect in the screw tightening process. Defect detection system.

17. The waveform data acquisition unit acquires waveform data showing the waveform of physical quantities generated in the power source during the screw tightening process. A feature calculation unit performs a feature calculation step in which it calculates multiple feature quantities that characterize each of the multiple divisions obtained by dividing the waveform into multiple parts using the waveform data, A defect determination step in which the defect determination unit uses the plurality of feature quantities to determine if the screw tightening process is defective, A defect detection method that includes the following.

18. A waveform data acquisition step that acquires waveform data showing the waveform of physical quantities generated in the power source during screw tightening, A feature calculation step of calculating multiple feature quantities that characterize each of the multiple divisions obtained by dividing the waveform into multiple parts using the waveform data, A defect determination step that uses the above-mentioned multiple feature quantities to determine if the screw tightening process is defective, A program that causes a computer to execute something.

Images