Inspection system
阅读说明:本技术 检查系统 (Inspection system ) 是由 松冈惇 粟田阳 A·鲁丁 于 2019-07-02 设计创作,主要内容包括:本发明提供一种能够在抑制不良的漏检的同时向作业者即时地报知判定结果的检查系统。检查系统检查螺栓的紧固状态,具备:服务器装置,其基于紧固波形的形状相对于基准波形的形状的相似度,判定螺栓的紧固状态是否良好,该基准波形表示紧固螺栓时的理想的转矩的历时变化,该紧固波形表示作业者使用螺母紧固器紧固了螺栓时的转矩的历时变化;报知装置,其向作业者报知通过服务器装置判定出的螺栓的紧固状态是否良好。(The invention provides an inspection system capable of promptly notifying a judgment result to an operator while suppressing defective omission. The inspection system inspects the fastening state of the bolt, and comprises: a server device that determines whether or not a fastening state of the bolt is good based on a similarity of a shape of a fastening waveform to a shape of a reference waveform, the reference waveform indicating a temporal change in an ideal torque when the bolt is fastened, the fastening waveform indicating a temporal change in a torque when an operator fastens the bolt using a nut fastener; and a notification device that notifies an operator whether the fastening state of the bolt determined by the server device is good.)
1. An inspection system for inspecting a fastening state of a screw, comprising:
a determination device that determines whether or not a fastening state of the screw is good based on a similarity of a shape of a fastening waveform to a shape of a reference waveform, the reference waveform indicating a change with time of an ideal torque when the screw is fastened, the fastening waveform indicating a change with time of a torque when a worker fastens the screw using a fastening tool; and
and a notification device that notifies an operator whether the fastening state of the screw determined by the determination device is good.
2. The inspection system of claim 1,
the reference waveform is calculated based on a plurality of past tightening waveforms,
the inspection system is configured such that the plurality of tightening waveforms are classified into a plurality of groups based on the similarity, a group having the largest number of tightening waveforms among the plurality of groups is set as a normal group, and an average of the tightening waveforms of the normal group is set as the reference waveform.
3. The inspection system of claim 1 or 2,
the determination device is configured to calculate a plurality of feature quantities related to similarity between the shape of the reference waveform and the shape of the tightening waveform, and determine whether the tightening state of the screw is good or not based on the plurality of feature quantities.
4. An inspection system according to any one of claims 1 to 3,
the determination device is configured to include a database storing tightening waveforms, and delete a tightening waveform determined to be normal from among the tightening waveforms stored in the database.
5. An inspection system according to any one of claims 1 to 4,
the notification device is configured to be able to input a determination made by an operator as to whether or not the fastening state of the screw is good.
6. An inspection system according to any one of claims 1 to 5,
the determination device is configured to determine whether or not the fastening state of the screw is good based on a similarity between a shape of a fastening waveform indicating temporal changes in a rotation angle when the screw is fastened by an operator using the fastening tool and a shape of a reference waveform indicating temporal changes in an ideal rotation angle when the screw is fastened.
Technical Field
The present invention relates to inspection systems.
Background
Conventionally, an inspection system for inspecting a fastening state of a screw is known (for example, see patent document 1).
The inspection system of patent document 1 includes a fastening machine that fastens a bolt, and an external device connected to the fastening machine. T-theta waveform data indicating a relationship between torque and a rotation angle at the time of fastening a bolt is recorded in an external device. The tightening machine is configured to continuously collect torque and a rotation angle when tightening a bolt, and output the torque and the rotation angle as a T- θ actual tightening curve to an external device. The external device is configured to compare the tilt change point of the T- θ waveform data with the tilt change point of the actual T- θ fastening curve to determine whether or not fastening is performed normally.
Disclosure of Invention
Technical problem to be solved by the invention
However, when the determination as to whether the fastening state of the bolt is good is made based on the final result (torque, angle) at the time of fastening completion as in the conventional inspection system, there is a possibility that a failure is missed (although the failure is a failure, the erroneous determination is a normal). That is, even when the final result at the time of fastening completion falls within the range of normal conditions, it is considered that breakage of the screw thread or the like has occurred. For example, if an existing screw thread is broken by inserting a bolt obliquely, and the screw thread is tightened while newly forming the screw thread, defective missing detection may occur.
Here, when the operator visually checks the shape of the tightening waveform indicating the change with time of the torque at the time of tightening the bolt, it is possible to appropriately determine whether the bolt is good or not.
The present invention has been made to solve the above-described problems, and an object of the present invention is to provide an inspection system capable of immediately notifying a determination result to an operator while suppressing defective omission.
Means for solving the problems
The inspection system according to the present invention inspects a fastening state of a screw, and includes: a determination device that determines whether or not a fastening state of the screw is good based on a similarity of a shape of a fastening waveform to a shape of a reference waveform, the reference waveform indicating a change with time of an ideal torque when the screw is fastened, the fastening waveform indicating a change with time of a torque when a worker fastens the screw using a fastening tool; and a notification device that notifies the operator whether the fastening state of the screw determined by the determination device is good.
With this configuration, it is possible to determine whether the fastening state of the screw is good or not based on the similarity of the shape of the fastening waveform with respect to the shape of the reference waveform, thereby suppressing defective omission. The operator can be notified of the result of the determination by the determination device in real time by the notification device.
In the inspection system, the reference waveform may be calculated based on a plurality of past tightening waveforms, the plurality of tightening waveforms may be classified into a plurality of groups based on the similarity, a group having the largest number of tightening waveforms among the plurality of groups may be set as a normal group, and an average of the tightening waveforms of the normal group may be set as the reference waveform.
With this configuration, the normal group tightening waveform has a high probability of being a normal tightening waveform, and therefore, an appropriate reference waveform can be calculated.
In the inspection system, the determination device may calculate a plurality of feature amounts regarding similarity between the shape of the reference waveform and the shape of the tightening waveform, and determine whether the tightening state of the screw is good or not based on the plurality of feature amounts.
With this configuration, the similarity between the shape of the reference waveform and the shape of the tightening waveform can be quantified to determine whether the waveform is good or not. That is, since the similarity between the shape of the reference waveform and the shape of the tightening waveform can be expressed numerically as a plurality of feature values, it is possible to perform the determination of the adequacy using the plurality of feature values.
In the inspection system, the determination device may include a database storing tightening waveforms, and delete a tightening waveform determined to be normal from among the tightening waveforms stored in the database.
With this configuration, an increase in the data amount of the tightening waveform stored in the database can be suppressed.
In the inspection system, the notification device may be configured to be capable of inputting a determination made by an operator as to whether or not the fastening state of the screw is good.
With this configuration, even when the determination device determines that the failure is present, the determination result of the determination device can be corrected when the worker confirms that the failure is present.
In the inspection system, the determination device may be configured to determine whether or not the fastening state of the screw is good based on a similarity between a shape of a fastening waveform indicating temporal changes in a rotation angle when the screw is fastened by the operator using the fastening tool and a shape of a reference waveform indicating temporal changes in an ideal rotation angle when the screw is fastened.
With this configuration, whether the fastening state of the screw is good or not is determined based on the similarity of the shape of the fastening waveform with respect to the rotation angle of the screw to the shape of the reference waveform, and thus, it is possible to further suppress defective omission.
ADVANTAGEOUS EFFECTS OF INVENTION
According to the inspection system of the present invention, the determination result can be immediately notified to the operator while suppressing the omission of the failure.
Drawings
Fig. 1 is a block diagram showing a configuration of an inspection system according to the present embodiment.
Fig. 2 is a graph showing an example of a reference waveform stored in the server device of the inspection system of fig. 1.
Fig. 3 is a flowchart for explaining an operation example of the inspection system of fig. 1.
FIG. 4 shows a composite feature quantity F1As the vertical axis, the composite characteristic quantity F2A scattergram, which is an example of a large number of examination objects, is plotted on two-dimensional coordinates on the horizontal axis.
Description of the reference symbols
1 nut fastener (fastening tool)
2 Server device (decision device)
3 informing device
22 waveform database (database)
100 inspection system
Detailed Description
Hereinafter, an embodiment of the present invention will be described with reference to the drawings.
First, the configuration of the
The
The nut tightener 1 is used when an operator tightens a bolt, and is disposed in a predetermined step of a production line. The nut tightener 1 includes a nut tightener body 11, and a
The nut tightening body 11 includes a socket (not shown) into which a bolt head is engaged, a
The
The
The
The
The
The
The
Working example of the inspection System
Next, an operation example of the
First, in the nut tightener 1 (see fig. 1), when the bolt is tightened by the nut tightener main body 11, a tightening waveform of the torque and a tightening waveform of the rotation angle are generated by the
Then, in the
Next, in step S2, it is determined whether or not the tightening waveform received from the nut tightener 1 is the inspection target based on the information received from the nut tightener 1. For example, when the identification information of the nut tightener 1 is registered in the
Next, in step S3, noise is removed from the tightening waveform of the torque. For example, a spike (spike-like noise) of the fastening waveform is detected, and the spike is removed from the fastening waveform. The spike is caused by the roughness of the sliding surface between the bolt and the workpiece, and is removed because it is not a feature to be noted in the present embodiment.
The tightening pattern of the nut tightener 1 is a pattern in which tightening and stopping of the tightening are repeated three times from the start of tightening to the completion of tightening, and the tightening waveform of the torque is separated in step S4. Specifically, the fastening waveform is divided into three fastening waveforms of fastening from the first time to the stop thereof, fastening from the second time to the stop thereof, and fastening from the third time to the stop thereof. And, the three divided fastening waveforms are resampled, respectively.
Next, in step S5, a plurality of feature quantities relating to the similarity between the shape of the reference waveform and the shape of the tightening waveform are calculated. This can quantify the similarity between the shape of the reference waveform and the shape of the tightening waveform. The reference waveform (see fig. 2) is stored in the
(A) Standard deviation of
(B) Cross correlation coefficient
(C) Coefficient of quadratic regression
(D) Frequency analysis
(E) Second order differential coefficient of difference
The standard deviation is an index indicating a deviation of the entire fastening waveform from the reference waveform. The cross correlation coefficient is an index indicating whether the overall tendencies of the fastening waveform and the reference waveform are parallel or orthogonal. The quadratic regression coefficient is an index indicating whether the entirety of the tightening waveform and the reference waveform is convex upward or convex downward, and is a coefficient of the 1 st term of the quadratic approximation curve. The frequency analysis is an index indicating the degree of fine irregularities, and is a ratio (low frequency/high frequency) of a spectrum of a difference between a tightening waveform and a reference waveform. The second order differential coefficient of the difference is a detection index of local ripples (undulations), and is the maximum value of the absolute value of the second order differential coefficient of the difference between the tightening waveform and the reference waveform.
Next, in step S6, the criterion is determined based on the five feature amountsThe similarity of the shape of the wave form to the shape of the fastening wave form. For example, for the similarity determination, the composite feature quantity F is calculated from five feature quantities by the following formula (1)1And F2Using the composite characteristic quantity F1And F2This determination is made. Thus, the five feature quantities are converted into two composite feature quantities F1And F2Combining the composite characteristic quantity F1And F2The threshold value for determination can be graphically observed and set by plotting the results on a scatter diagram. Composite characteristic quantity F1For example, the composite feature quantity F is an index representing the similarity of the entire waveform2For example, an index indicating the similarity of the details of the waveform.
In the formula (1), a is a feature amount of a standard deviation, B is a feature amount of a cross correlation coefficient, C is a feature amount of a quadratic regression coefficient, D is a feature amount of a frequency analysis, and E is a feature amount of a differential second order differential coefficient. m is1~m5And n1~n5Is a coefficient determined by an experiment or the like.
And, the composite characteristic quantity F1And F2The results were plotted on a scattergram, and similarity was determined. In the present embodiment, the result of determination of the degree of similarity in multiple stages (degree of poor reproducibility) is output.
Here, fig. 4 shows a composite feature quantity F1As the vertical axis, the composite characteristic quantity F2A scattergram, which is an example of a large number of examination objects, is plotted on two-dimensional coordinates on the horizontal axis. In the scattergram, threshold curves TC1 to TC3 are set, and the regions are partitioned by the threshold curves TC1 to
Therefore, the calculation result (the calculated composite feature quantity F)1And F2) When the result is plotted in the region R1, it is determined to be normal, and when the result is plotted in the region R4, it is determined to be poor. When the calculation result is plotted in the region R2, it is determined that the poor performance is low, and when the calculation result is plotted in the region R3, it is determined that the poor performance is high. Further, whether the waveform is good or not is determined for each of the three divided tightening waveforms.
Next, in step S7, it is determined whether or not a warning is required for the operator. For example, when a failure is determined in at least one of the three tightening waveforms, it is determined that a warning is necessary, and when the deterioration of the failure is high in at least one of the three tightening waveforms, it is determined that a warning is necessary. That is, when it is determined that the three tightening waveforms are normal or poor in performance, it is determined that a warning is not necessary. If it is determined that a warning is necessary, the process proceeds to step S8. On the other hand, if it is determined that the warning is not necessary, the process proceeds to step S10.
Next, in step S8, the determination result is transmitted to the
Then, in step S11, the
Next, in step S12, a warning is displayed on the display unit 32 (see fig. 1). The warning may include, for example, a level of a failure (determined to be poor or determined to be poor with high reliability), or may include a display for urging the operator to perform a handling operation. The reference waveform and the tightening waveform (defective waveform) may be displayed on the
When the worker removes the bolt, confirms the bolt fastening point, and causes a fastening failure such as seizing, information on the situation (fastening failure) is input to the input unit 33 (see fig. 1) by the worker. In this case, the bolt is replaced, a new bolt is fastened, and the inspection is performed by the
On the other hand, when the worker removes the bolt, confirms the bolt fastening point, and does not cause a fastening failure such as seizing, information on the situation (fastening state is normal) is input to the
In step S13, it is determined whether or not the determination of whether or not the fastening state by the operator is good is input. If it is determined that the fastening state is good or not by the operator, the process proceeds to step S14. On the other hand, if it is determined that the determination of whether the fastening state by the operator is good or not is not input, step S13 is repeated. That is, the standby is performed until whether the input is good or not is determined.
Next, in step S14, a determination is made as to whether the fastening state by the operator is good or not, and this determination is transmitted to the
On the other hand, in the
Next, in step S10, the tightening waveform is accumulated in the waveform database 22 (see fig. 1). At this time, identification information of the nut tightener 1, tightening date and time, determination result, and the like are stored in the
Method for calculating a reference waveform
Next, a method of calculating a reference waveform stored in the
The reference waveform represents a temporal change of an ideal torque when the nut tightener 1 is used to tighten a bolt, and is calculated using, for example, clustering. Specifically, the reference waveform is calculated based on a plurality of past tightening waveforms. The conventional tightening waveform is obtained by tightening a bolt and a workpiece of the same specification in the same tightening pattern using the same nut tightening machine 1, for example. The number of the plurality of tightening waveforms is equal to or greater than a predetermined number, but for simplification of the description, the reference waveform is calculated using 100 past waveforms (past tightening waveforms) 1 to 100 in the following example.
First, the similarity F between the
Ftotal=k1×F1+k2×F2···(2)
Further, in the formula (2), k1And k2Is a coefficient determined by an experiment or the like.
Similarly, the similarity F is calculated for all combinations of the past waveforms 1 to 100total. An example of the results is shown in table 1 below.
[ TABLE 1 ]
And, based on the similarity FtotalThe past waveforms 1 to 100 are classified into a plurality of groups. For example, as shown in the following Table 2, the past waveforms 1 to 100 are classified into three groups 1 to 3.
[ TABLE 2 ]
In the specific examples shown in tables 1 and 2, the similarity F of the
The group 1 having the largest number of past waveforms among the three groups 1 to 3 is referred to as a normal group, and the average (average at each sampling time) of the
The calculation of the reference waveform may be performed in the
Capacity management of waveform databases
Next, capacity management of the
The
Effects-
In the present embodiment, as described above, by determining whether the fastening state of the bolt is good based on the similarity of the shape of the fastening waveform with respect to the shape of the reference waveform, it is possible to suppress the missing of the defect as compared with the case where the determination of whether the fastening state of the bolt is good is performed based on the final torque at the time of completion of the fastening. Further, by providing the
In the present embodiment, the past waveforms 1 to 100 are classified into groups 1 to 3 based on the similarity, the group 1 having the largest number of past waveforms among the groups 1 to 3 is set as a normal group, the average of the past waveforms of the group 1 as the normal group is set as a reference waveform, and the past waveform of the normal group has high probability of being a normal past waveform, and therefore, an appropriate reference waveform can be calculated.
In the present embodiment, the similarity between the shape of the reference waveform and the shape of the tightening waveform can be quantified by calculating five feature quantities related to the similarity between the shape of the reference waveform and the shape of the tightening waveform, and the determination of the acceptability can be made. That is, the similarity between the shape of the reference waveform and the shape of the tightening waveform can be numerically expressed as five feature quantities, and therefore, the acceptability determination can be performed using these five feature quantities.
In the present embodiment, two composite feature quantities F are calculated from five feature quantities1And F2The threshold for determining the quality can be graphically examined and set.
In the present embodiment, by deleting the tightening waveform determined to be normal from among the tightening waveforms stored in the
In the present embodiment, by allowing the
Other embodiments
The embodiments disclosed herein are illustrative in all respects and are not to be construed as limiting. Therefore, the technical scope of the present invention is defined not by the embodiments described above but by the claims. The technical scope of the present invention includes all modifications within the meaning and range equivalent to the claims.
For example, in the above embodiment, the reference waveform may be calculated for each of the plurality of nut tighteners 1 arranged in the production line, or the reference waveform may be shared among a part of the plurality of nut tighteners 1 arranged in the production line. For example, when the nut tighteners 1 are of the same type and tighten bolts and workpieces of the same specification in the same tightening pattern, the reference waveforms can be shared.
In the above embodiment, whether or not the fastening state of the bolt is good may be determined based on the similarity of the shape of the fastening waveform indicating the temporal change in the ideal rotation angle when the bolt is fastened to the reference waveform indicating the temporal change in the rotation angle when the worker fastens the bolt using the nut tightener 1. That is, whether or not the fastening state of the bolt is good may be determined based on the similarity of the shape of the fastening waveform with respect to the rotation angle of the bolt with respect to the shape of the reference waveform. With such a configuration, it is possible to further suppress defective omission.
In the above embodiment, an example of all the bolt fastening points on the inspection line is shown, but the present invention is not limited thereto, and a part of the bolt fastening points on the inspection line may be used.
In the above embodiment, the fastening mode in which fastening and stopping of the fastening are repeated three times is shown, but the present invention is not limited to this, and other fastening modes such as a fastening mode in which fastening is completed once may be used.
In the above embodiment, an example is shown in which the determination result is notified to the operator when a warning is required, but the present invention is not limited to this, and the determination result may be notified when it is determined that the fastening state is normal.
In the above-described embodiment, the example in which the
In the above embodiment, the
In the above embodiment, the method of removing the spike pulse may be any method, and for example, a sampling point at which a comparison value between the output of the median filter and the actual data exceeds a predetermined value may be selectively removed.
In the above embodiment, the fastening waveform may be separated by any method, for example, a sharp drop of the fastening waveform may be detected from a differential value of the fastening waveform, and the fastening waveform may be separated at the time of the sharp drop.
In the above-described embodiment, the standard deviation, the cross correlation coefficient, the quadratic regression coefficient, the frequency analysis, and the second order differential coefficient of the difference are shown as five feature amounts, but the present invention is not limited thereto, and the number of feature amounts may be any number, and feature amounts other than these may be used.
In addition, in the above-described embodiment, it is shown that two composite feature quantities F are calculated from five feature quantities1And F2The example of (3) is not limited to this, and one composite feature may be calculated from five features. In addition, the similarity may be directly determined from five feature amounts without calculating the composite feature amount.
In the above embodiment, the results of determination in which a plurality of stages are output with respect to the degree of similarity are shown, but the present invention is not limited thereto, and either normal or defective may be output as the results of determination.
In the above-described embodiment, the example of warning when it is determined that the performance is poor or when it is determined that the performance is high is shown, but the present invention is not limited thereto, and warning may be performed when it is determined that the performance is low. That is, the condition for performing warning may be changed as appropriate.
In the above embodiment, the correction coefficient m may be set when the determination result of the
In the above embodiment, when the tightening waveform determined to be normal is stored in the
In the above embodiment, the example of calculating the reference waveform using clustering is described, but the present invention is not limited thereto, and an ideal past waveform may be used as the reference waveform.
In the above embodiment, the example in which the tightening waveform determined to be normal is deleted from the
In the above embodiment, the communication between the nut tightener 1 and the
In the above-described embodiment, the example in which the present invention is applied to the inspection system for inspecting the fastening state of the bolt is shown, but the present invention is not limited to this, and the present invention may be applied to an inspection system for inspecting the fastening state of a screw other than the bolt.
The application requests priority over Japanese application No. 2018-126652 on 7/3/2018. By reference thereto, the entire content thereof is incorporated into the present application.
Industrial applicability
The present invention can be used in an inspection system for inspecting a fastening state of a screw.
- 上一篇:一种医用注射器针头装配设备
- 下一篇:一种飞机转接段密封圈更换工装及方法