JP2001198498A - Apparatus and method for detecting abnormality of coating gun - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an apparatus capable of automatically and easily detecting the abnormality of a coating gun. SOLUTION: The apparatus for detecting the abnormality of a coating gun has a vibration sensor 12 for detecting the vibration of a coating gun and an arithmetic unit 22 analyzing the frequency of the detected vibration and using a correlation evaluation function to calculate the correlation of the analyzed frequency and preliminarily stored frequency at a time of normality and judging the presence of abnormality in the coating gun when the calculation result is below a predetermined threshold value.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an apparatus and a method for detecting an abnormality of a paint gun for spraying paint.

[0002]

2. Description of the Related Art A coating gun sprays paint on an object to be coated by spraying. In such a coating gun, abnormalities such as clogging may occur during coating, which may cause uneven coating.

Conventionally, in order to detect such an abnormality,
For example, in Japanese Patent Application Laid-Open No. 8-29211, vibration of a paint gun during a painting operation is detected, the vibration is analyzed, and a frequency distribution pattern at that time is compared with a frequency distribution pattern in a normal state. It is disclosed to detect a clogging abnormality.

[0004]

As compared with the prior art, when comparing the frequency distribution patterns during normal operation and during the painting operation, when an operator looks at the patterns and compares them, an abnormality is determined from the difference between the two patterns. Can be.

However, when it is considered that this is automatically determined by a personal computer or the like, both patterns are compared as figures. The comparative study of figures requires a certain amount of time even with recent advances in computer technology, and thus has a problem in that the judgment is delayed. There is also a problem that it is difficult to compare the figures in the first place and to determine how much difference between the two patterns is determined to be abnormal. This is because, in order to automate the process of human judgment by looking at a figure on a personal computer, it is necessary to create a program to execute such a judgment process, and creating such a program is difficult unlike numerical comparison That's why.

SUMMARY OF THE INVENTION It is an object of the present invention to provide an apparatus and a method for detecting an abnormality of a coating gun which can automatically and easily detect an abnormality of the coating gun.

[0007]

The above object is achieved by the following means.

(1) Vibration detecting means for detecting the vibration of the coating gun, frequency analyzing means for analyzing the frequency of the vibration detected by the vibration detecting means, normal frequency analyzed by the frequency analyzing means, Calculating means for calculating a correlation with the frequency during coating analyzed by the frequency analyzing means using a correlation evaluation function, and calculating the correlation evaluation function, the coating result based on a predetermined threshold based on a calculation result of the correlation evaluation function. An abnormality detection device for a coating gun, comprising: a determination unit configured to determine abnormality of the gun.

(2) The frequency is in the range of 5 to 22 kHz.

(3) The frequency range of 5 to 22 kHz is extracted from the vibration detected by the vibration detecting means before being sent to the frequency analyzing means.

(4) The normal frequency is a result of analysis by the frequency analyzing means from vibration detected by the vibration detecting means in a normal state for 0.75 seconds or more and less than 12.5 seconds. I do.

(5) detecting the vibration of the coating gun in a normal state, performing a frequency analysis of the detected vibration in the normal state, and detecting the vibration of the coating gun during coating. Performing the frequency analysis of the vibration during the painting, the analyzed normal frequency, and calculating the correlation between the frequency during the painting using a correlation evaluation function, from the calculation result of the correlation evaluation function Determining an abnormality of the coating gun based on a predetermined threshold value.

(6) The frequency is in the range of 5 to 22 kHz.

(7) The frequency range of 5 to 22 kHz is extracted from the detected vibration before frequency analysis.

(8) The step of performing the frequency analysis of the vibration in the normal state is characterized in that the analysis is performed from the vibration detected in the normal state for 0.75 seconds or more and less than 12.5 seconds.

[0016]

The present invention has the following effects for each claim.

According to the first aspect of the present invention, the vibration means detects the vibration of the coating gun during the coating operation, the frequency analysis means decomposes the detected vibration for each frequency component, and the operation means The correlation between the normal frequency and the frequency during painting is calculated using a correlation evaluation function, and the judging means judges abnormality based on a predetermined threshold from the calculation result by the Seki evaluation function. Therefore, the abnormality of the coating gun can be detected by the numerical calculation. Therefore,
By using an arithmetic device such as a personal computer, it is possible to automatically and easily detect an abnormality of the coating gun.

According to the second aspect of the present invention, since the frequency is in the range of 5 to 22 kHz, high-frequency components caused by noise due to the power supply frequency and mechanical vibration are cut off.
An abnormality of the coating gun can be detected with higher accuracy.

According to the third aspect of the present invention, the above-mentioned 5-
Since the frequency range of 22 kHz is extracted before the frequency analysis, it is possible to reduce the load on the frequency analysis and the arithmetic processing, and to quickly determine the presence or absence of an abnormality.

According to the fourth aspect of the present invention, when an abnormality occurs during the coating by performing the frequency analysis in the normal state from the vibration detected for 0.75 seconds or more and less than 12.5 seconds, The correlation between the normal frequency and the abnormal frequency is clarified, making it easier to determine the abnormality.

According to the present invention, the vibration of the coating gun during the coating operation is detected, and the detected vibration is decomposed for each frequency component, and the correlation between the normal frequency and the frequency during coating is detected. Is calculated using the correlation evaluation function, and an abnormality is determined based on a predetermined threshold value from the calculation result, so that an abnormality of the coating gun can be detected by numerical calculation. Therefore, it is possible to automatically and easily detect an abnormality of the coating gun by using an arithmetic device such as a personal computer.

According to the sixth aspect of the present invention, since the frequency is in the range of 5 to 22 kHz, high-frequency components caused by noise due to the power supply frequency and mechanical vibration are cut off.
An abnormality of the coating gun can be detected with higher accuracy.

According to the present invention as set forth in claim 7, the aforementioned 5 to 5
Since the frequency range of 22 kHz is extracted before the frequency analysis, it is possible to reduce the load on the frequency analysis and the arithmetic processing, and to quickly determine the presence or absence of an abnormality.

According to the eighth aspect of the present invention, when an abnormality occurs during coating by performing frequency analysis in a normal state from vibration detected for 0.75 seconds to less than 12.5 seconds, The correlation between the normal frequency and the abnormal frequency is clarified, making it easier to determine the abnormality.

[0025]

BRIEF DESCRIPTION OF THE DRAWINGS FIG.
An embodiment of the present invention will be described.

Embodiment 1 FIG. 1 is a drawing showing a schematic configuration of a coating apparatus and an abnormality detection apparatus according to Embodiment 1 of the present invention.

The coating apparatus comprises a coating gun 11 attached to the end of the arm of the robot 10, a pump 13 for supplying coating to the coating gun 11, and a coating tank 14. The coating gun 11 is provided with a spray nozzle 12 for injecting a coating material at its tip.

The abnormality detecting device comprises a vibration sensor 21 attached to the spray nozzle 12, a computing device 22, and an output device 23 for outputting a result. Here, the arithmetic unit 22 is a so-called personal computer or the like, and executes a program created in accordance with a processing procedure described later to perform frequency analysis by fast Fourier transform (FFT) to determine whether there is an abnormality. The output device 23 is a display or a printer connected to a personal computer.

The vibration sensor 21 is an acceleration sensor,
The vibration when the paint is sprayed from the spray nozzle 12 is detected.

The painting operation is performed by the robot 10 operating the painting gun 11 according to a predetermined program so that a predetermined range is painted on a surface to be painted of a work (not shown). At this time, the pump 13 sends the paint in the paint tank 14 to the painting gun 11 at a predetermined pressure in conjunction with the operation of the robot 10. Thereby, the paint is sprayed from the spray nozzle 12 provided at the tip of the paint gun 11 to perform the painting.

Next, a procedure for detecting an abnormality of the coating gun during the coating operation will be described.

FIG. 2 is a flowchart showing the procedure for detecting an abnormality.

First, prior to the actual abnormality detection, vibrations in a normal coating state are detected (S1).

In this case, as shown in FIG. 3, in the normal vibration detection, as shown in FIG. 3, when one cut-out time for detecting vibration and performing frequency analysis is set to n, in the present embodiment, the first cut-out time n0 is determined. Sections are used.

The detected normal vibration is subjected to FFT (Fast Fourier Transform) analysis as frequency analysis (S2).
Subsequently, a specific frequency band is cut out, and a normal-time spectrum X0 is calculated (S3).

Here, the specific frequency band is, for example, 5 to 5.
A range of 22 kHz is preferred. This means that by cutting the frequency below 5 kHz, it is possible to remove the power frequency and other low-frequency vibration components (for example, vibration when a person or a machine moves) from the outside, while cutting the frequency exceeding 22 kHz. Thus, high frequency components such as machine oscillation can be removed.

FIG. 12 shows a raw waveform of the FFT signal when the spray nozzle 12 is clogged. As shown in the figure, the spectrum change region clearly appears within the frequency range of 5 to 22 kHz, and it can be understood that the range is a range that can be accurately detected without being affected by noise from externalization.

The calculation of the normal spectrum in step S3 is performed as follows.

Here, the frequency spectrum X (m, L) of the time-series time signal x (n) is expressed by the following (1).
Defined as shown in the equation.

[0040]

(Equation 1)

Where N is the block size, M is the overlap size, and ω (k) = when −K ≦ k ≦ K.
(1−cos (2πk / K)) / 2, where k is otherwise, ω (k) = 0, K = N / 2, m is an integer that increases in the time direction, and L increases in the frequency direction. (Hereinafter the same in each formula). Note that both time and frequency are discrete data.

From the equation (1), the spectrum X at normal time is
0 is defined as in the following equation (2).

[0043]

(Equation 2)

Here, in the equation, “the number of N included in n”
Is a number indicating how many analysis numbers are included in one cutout time n with the block size N for FFT analysis as one unit. Specifically, for example, if the block size N is 512 (the number of blocks),
This is a number indicating how many times the FFT analysis with 512 units as one unit has been performed during n hours (the same applies to each formula below).

After the calculation of the normal-time spectrum, an abnormality is detected in the actual painting operation.

First, vibration during the painting operation is detected (S
4). The detected vibration is subjected to FFT analysis at regular time intervals (S5), and a specific frequency band is cut out, and the above-mentioned (1)
The operating spectrum Xi is calculated by the following equation (3) using the equation (S6). Here, the specific frequency band is preferably 5 to 22 kHz as described above.

[0047]

(Equation 3)

After calculating the operating spectrum Xi, the correlation between the normal state and the operating state is calculated from the normal state spectrum X0 and the operating spectrum Xi by using a correlation evaluation function shown in the following equation (4) (S7).

[0049]

(Equation 4)

In equation (4), fU is the lower limit frequency (here, 5 Hz), and fL is the upper limit frequency (here, 2 Hz).
2 Hz). The “number between (fU−fL)” means that when FFT analysis is performed with a lower limit of 5 kHz and an upper limit of 22 kHz, actually obtained data becomes discrete data (Δf = (sampling frequency / block size)). It means the number of data that exists between the upper limit and the adjustment.

Subsequently, the calculation result ρ (i) of the correlation evaluation function
It is determined whether or not (correlation value) is less than a predetermined threshold value (S8). If the correlation value is less than the threshold value, an abnormal signal is output to the output device 23 (S9). , And the process ends. On the other hand, if the correlation value is not less than the threshold,
Returning to step S4, the abnormality detection is continued until the painting operation is completed.

Here, the threshold value varies depending on the block size of FFT analysis, how to take various constants, and the like. As a result of various experiments, for example, the block size 512,
Assuming that the observation time is 0.75 seconds and the overlap is 0, the threshold value is set to 0.85 if only full blockage is detected.
It is preferred that Further, in order to detect a half-clogging abnormality, it is preferable to set the threshold value of the correlation value to 0.9.

FIG. 4 is a drawing showing correlation values when a full clogging abnormality (FIG. 4A) and a half clogging abnormality (FIG. 4B) occur. In each drawing, the vertical axis represents the correlation value (calculation result of the correlation evaluation function), and the horizontal axis represents time (second).

As can be seen from the figure, the correlation value decreases when an abnormality occurs in both the full blockage and the half blockage. In the case of a full blockage abnormality (FIG. 4A), the correlation value gradually decreases, and eventually becomes zero. Semi-clogging error (Fig. 4
In B), the correlation value once decreased has increased again,
By setting the threshold value to 0.90, it is possible to detect a half-clogged state.

Embodiment 2 FIG. 5 is a drawing showing a schematic configuration of a coating apparatus and an abnormality detection apparatus according to Embodiment 2 of the present invention. Note that members having the same functions as those of the first embodiment are given the same reference numerals, and descriptions thereof will be omitted.

The structure of the coating apparatus is the same as that of the first embodiment, and the robot 10, the coating gun 11, the pump 13,
And a paint tank 14.
A spray nozzle 12 is provided at the tip. The coating operation by this coating apparatus is the same as that of the first embodiment, and the description is omitted.

The abnormality detecting device includes a vibration sensor 21 attached to the spray nozzle 12, and a low-pass filter (LP) for directly cutting out high-frequency components in an analog manner in order to extract a specific frequency component from a signal from the vibration sensor.
F) 31, a high-pass filter (HPF) 32 for cutting low-frequency components, an arithmetic unit 22, and an output unit 23.

Next, a procedure for detecting an abnormality of the coating gun during the coating operation will be described.

FIG. 6 is a flowchart showing the procedure for detecting an abnormality.

First, normal vibration is detected (S1).
1). Then, unnecessary frequency components are cut from the signal from the vibration sensor by the low-pass filter 31 and the high-pass filter 32 to extract only specific frequency components (S12). Here, as described above, only signals in the range of 5 to 22 kHz are passed.

Thereafter, the FFT is performed in the same manner as in the first embodiment.
The analysis is performed (S13), and the normal spectrum X0 is calculated (S14).

After the normal-time spectrum is calculated, the abnormality in the actual painting operation is detected.

First, vibration during the painting operation is detected (S1).
5) As described above, unnecessary frequency components are cut from the signal from the vibration sensor using the low-pass filter 31 and the high-pass filter 32 (S16).

Thereafter, as in the first embodiment, an FFT analysis is performed (S17), and an operating spectrum Xi is calculated (S18). Subsequently, the correlation is calculated using the correlation evaluation function shown in the above-described equation (4) (S19). continue,
It is determined from the result of the correlation calculation whether the correlation value is less than a predetermined threshold value (S20). If the correlation value is less than the threshold value, an abnormal signal is output to the output device 23 (S2).
1), end the process. On the other hand, if the correlation value is not less than the threshold value, the process returns to step S15, and the abnormality detection is continued until the painting operation ends.

Thus, in the second embodiment, similarly to the first embodiment, by setting the threshold value, it is possible to detect a full clogging abnormality or a half clogging abnormality of the coating gun.

In the second embodiment, unnecessary frequency components are cut off from the signal from the vibration sensor using the low-pass filter 31 and the high-pass filter 32 in an analog manner. Since only the frequency components necessary for the processing need to be processed, the load on the arithmetic unit is reduced.

Embodiment 3 In Embodiments 1 and 2 described above, the spectrum in the normal state is calculated using only the first cut-out time n1, but in Embodiment 3, instead of this, a plurality of times are calculated. The normal-time spectrum was calculated using the cut-out time of FIG. When this is conceptually illustrated, as shown in FIG. 7, a normal-time spectrum is calculated from a plurality of cut-out times n01 to n0i as a reference section t0, and then vibrations in the observation sections n1 to ni are detected to detect abnormalities. Is to make a judgment.

Therefore, the device configuration in the third embodiment may be the same as that of the first or second embodiment, and the entire operation procedure may be the same as that of the first or second embodiment. Therefore, in the description of the third embodiment, the description of the device configuration and the operation procedure will be omitted, and only the calculation of the normal-time spectrum unique to the third embodiment and the abnormality determination using the same will be described.

First, when calculating the normal time spectrum, the spectrum at one cut-out time n is calculated by the following equation (5).

[0070]

(Equation 5)

Then, one calculated cut-out time n
At a plurality of cut-out times (n01 to
n0i) is averaged according to the following equation (6), and a normal spectrum X0 (m) is calculated.

[0072]

(Equation 6)

Subsequently, for the normal spectrum, the correlation evaluation value ρ0 (i) of each spectrum X0i in the reference section is calculated by the following equation (7).

[0074]

(Equation 7)

The average of the correlation evaluation value ρ0 (i) is obtained by the following equation (8), and this is set as a variation function ρ0 in the reference section t0.

[0076]

(Equation 8)

Here, in the formula, the “reference spectrum number” is the number of spectrums when the paint is normally sprayed from the spray nozzle when the paint is applied.

Then, the correlation evaluation function ρ (i) between the normal state and the actual painting operation is calculated from the vibration detected for each cutting time n in the observation section by the following equation (9).

[0079]

(Equation 9)

By evaluating the value of the correlation evaluation function based on the threshold value, it is possible to determine the presence or absence of an abnormality in the observation section.

FIGS. 8 to 11 show the results of evaluation using the method of the third embodiment for various times in the reference section t0. In each drawing, the vertical axis represents the correlation value (calculation result of the correlation evaluation function), and the horizontal axis represents time (second). here,
The block size N is set to 512, one cutout time n is set to 0.25 seconds, and the overlap is set to 0.

First, as shown in FIG. 8A, when the reference section t0 for calculating the normal-state spectrum is set to one cut-out time of 0.25 seconds, a large fluctuation is observed in b when the half-clogging abnormality occurs. Therefore, it is possible to detect the occurrence of an abnormality. However, there is a relatively large variation portion a in the correlation value before that, and depending on the setting of the threshold value, this a portion may be determined to be abnormal.

On the other hand, as shown in FIG. 8B, by setting the reference section t0 to 0.5 seconds composed of two cut-out times, the variation of the portion a is reduced, and as shown in FIG. By setting the cutout time to 0.75 seconds, the variation in the a portion is further reduced. Further, as shown in FIGS. 9D to 11J, it can be seen that the difference between the normal part and the abnormal part becomes clearer by further increasing the time of the reference section t0. From these evaluation results, it can be understood that it is easier to determine an abnormality by increasing the normal time spectrum calculation time. As a result, it is preferable that the time of the reference section t0 be 0.75 seconds or more (three times as the cut-out time).
With the setting of 75, the abnormality can be clearly determined. Further, by setting the time of the reference section t0 to 1.25 seconds or more (the cutout time is five times, FIG. 9E), the difference between the minimum value a of the normal part and the value b at the time of occurrence of the abnormality is 0.2 or more. Therefore, it becomes easier to determine the abnormality.

On the other hand, the reference section t0 corresponds to FIG.
As shown in the figure, 12.5 seconds (the cut-out time is 50
If the correlation value exceeds 1), the correlation value is 1 even up to the abnormality occurrence part b.
(The difference between the correlation values between the minimum values a and b of the normal part is smaller than 0.2), which may make the determination difficult. Therefore, the upper limit of the reference section t0 for calculating the normal spectrum is 1
It is preferably less than 2.5 seconds (the cutout time is 50 times). Further, if the normal time spectrum calculation time is set to be too long, an abnormality is expected to occur during that time. From this viewpoint, it is more preferable to set the time to 10 seconds or less (the cutout time is 50 times) or less.

To summarize these results, the time of the reference section t0 for calculating the normal spectrum is preferably 0.75 seconds or more and less than 1.25 seconds, more preferably 1.25 seconds or more and 10 seconds. It can be said that:

The calculation of the normal-time spectrum may be performed before starting the painting operation, or may be performed immediately before starting the painting operation, and then proceed to the painting operation.

[Brief description of the drawings]

FIG. 1 is a schematic diagram for explaining a configuration of an abnormality detection device according to a first embodiment to which the present invention has been applied.

FIG. 2 is a flowchart illustrating an abnormality detection procedure according to the first embodiment.

FIG. 3 is a drawing for explaining a cut-out time of frequency analysis according to the first embodiment.

FIG. 4 is a drawing showing correlation values when a full clogging abnormality (FIG. 4A) and a semi-clogging abnormality (FIG. 4B) occur.

FIG. 5 is a schematic diagram for explaining a configuration of an abnormality detection device according to a second embodiment to which the present invention has been applied.

FIG. 6 is a flowchart illustrating an abnormality detection procedure according to the second embodiment.

FIG. 7 is a drawing for explaining cut-out times of frequency analysis according to the third embodiment.

FIG. 8 is a drawing showing evaluation results according to a third embodiment.

FIG. 9 is a drawing showing evaluation results according to a third embodiment.

FIG. 10 is a drawing showing evaluation results according to a third embodiment.

FIG. 11 is a drawing showing evaluation results according to a third embodiment.

FIG. 12 is a diagram showing a raw waveform of an FFT signal when a spray nozzle is clogged.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 11 Painting gun 12 Spray nozzle 21 Vibration sensor 22 Operation device 23 Output device 31 Low pass filter (LPF) 32 High pass filter (HPF)

 ──────────────────────────────────────────────────続 き Continued on the front page F term (reference) 2G024 AD50 BA27 CA13 2G064 AA12 AB22 BA02 CC06 CC43 CC47 CC52 DD09 4D075 AA01 AA80 AA81 4F035 AA03 BB31 4F042 DH02 DH09

Claims (8)

[Claims] A vibration detecting means for detecting a vibration of the coating gun; a frequency analyzing means for performing a frequency analysis of the vibration detected by the vibration detecting means; a normal frequency analyzed by the frequency analyzing means; Calculating means for calculating a correlation with the frequency during coating analyzed by the frequency analyzing means using a correlation evaluation function; and, based on a calculation result of the correlation evaluation function, the coating gun based on a predetermined threshold value. A determination means for determining an abnormality of the coating gun. 2. The apparatus according to claim 1, wherein the frequency is in a range of 5 to 22 kHz. 3. The abnormality detection of a coating gun according to claim 2, wherein the frequency range of 5 to 22 kHz is extracted from the vibration detected by the vibration detecting means before being sent to the frequency analyzing means. apparatus. 4. The normal frequency is a result of analysis by the frequency analyzing means from vibration detected by the vibration detecting means in a normal state for 0.75 seconds or more and less than 12.5 seconds. An apparatus for detecting an abnormality of a coating gun according to claim 1. 5. A step of detecting a vibration of the coating gun in a normal state, a step of performing frequency analysis of the detected vibration in a normal state, and a step of detecting the vibration of the coating gun during coating. Performing a frequency analysis of the vibration during painting, and the analyzed normal frequency, and calculating the correlation between the frequency during painting using a correlation evaluation function, and from the calculation result of the correlation evaluation function, Judging abnormality of the paint gun based on a predetermined threshold,
A method for detecting an abnormality of a coating gun, comprising: 6. The method according to claim 5, wherein the frequency is in a range of 5 to 22 kHz. 7. The method according to claim 6, wherein the frequency range of 5 to 22 kHz is extracted from the detected vibration before frequency analysis. 8. The method according to claim 5, wherein the step of performing the frequency analysis of the vibration in the normal state includes analyzing the vibration detected in the normal state for a period of 0.75 seconds or more and less than 12.5 seconds.
8. The method for detecting an abnormality of a coating gun according to any one of items 7 to 7.