JPH06262345A - Method and device for automatic brazing - Google Patents

Abstract

PURPOSE:To perform a brazing work with high reliability and a high quality with an automatic operation by measuring a temperature with a vision sensor, feeding the measured temperature back to a heating means, and controlling the temperature of the metal to be joined to prevent the metal to be joined from being overheated or underheated in a brazing work. CONSTITUTION:Metals 1a and 1b to be joined are heated by a heating means 2, light emitted from the metals 1a and 1b to be joined is detected by an industrial television camera 4, luminance values are summed up and calculated by an image processing means 5, and the heating state of the metals to be joined, is judged. From the result of the judgment, a controlling device 3 controls the heating means 2 to adjust the quantity of heating.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a brazing method for a metal member and an apparatus therefor, and more particularly to an automatic brazing method and an apparatus for controlling heating by judging a temperature of a brazing portion of a metal to be joined by image processing.

[0002]

2. Description of the Related Art A conventional automatic brazing apparatus is disclosed in Japanese Patent Laid-Open No.
As described in JP-A-70376, a long wavelength light transmission filter is added to a vision sensor to observe the infrared intensity from the joint, and the image signal is binarized to heat the metal to be joined depending on the area of the high brightness portion. There was something to judge the condition.

[0003]

The infrared rays emitted from the above-mentioned members to be joined are observed by a vision sensor, and the temperature of the joining portion of the members to be joined is determined by the area of the high-intensity part after binarizing the image signal. Since the judgment method is premised on processing the binarized image information, it only detects that the temperature of the metal to be joined has risen as a whole. No detailed determination has been made, and no attempt has been made to control the temperature of the metal to be joined by positively returning the image signal obtained by the vision sensor to the heating means. Also, if any obstacles intrude or exist in the observed image and they overlap with the temperature observation part of the joined member in the vision sensor's field of view, accurate temperature measurement becomes impossible and it is important for brazing quality. There was a problem that had a great influence. Furthermore, since the evaluation of the workability of the brazing part work was done by evaluating the work after the work was completed, the workability was evaluated only after the completion of one work, which is extremely efficient. It was bad and there was a problem that the evaluation could not be made during brazing.

The object of the present invention is to improve the reliability in automatic brazing by observing the light emitted by the temperature rise of the metal to be joined and processing the multi-valued image data of the metal to be joined. By determining the temperature rise of the brazing part and referring to this to adjust the heating amount to control the metal to be welded to a temperature suitable for brazing, it is possible to prevent overheating and underheating of the metal to be joined. An object of the present invention is to provide an automatic brazing method and apparatus for performing highly reliable automatic brazing operation.

Further, by removing information other than the metal to be joined, noise, etc. from the above observed image data,
The aim is to increase the accuracy of temperature measurement and to perform more reliable automatic brazing operation. Further, when the intensity of the light emitted from the metal to be bonded is near the threshold value set for removing information other than the metal to be bonded or disturbance such as noise, the judgment is made by a small amount of data. It is to prevent stable operation and perform stable automatic brazing operation. Further, when the disturbance is fluctuating, the threshold is automatically set to an optimum threshold according to the disturbance, and an automatic brazing operation that is not easily influenced by the disturbance is performed. Further, the portion where the image data below the threshold value is removed is complemented by using the image data other than the removed portion, and the temperature rise is judged by the complemented image data to further determine the temperature. It is intended to improve accuracy and perform highly reliable automatic brazing operation.

Furthermore, the temperature of the work being brazed is grasped and evaluated in real time to make an efficient judgment of the quality of brazing and effective guidance, and to quickly set the equipment during automatic operation. Even if there is a problem with brazing, it will be possible to avoid it,
The purpose is to perform high-quality brazing.

[0007]

In order to achieve the above object, an automatic brazing method and apparatus of the present invention include a heating means for heating a brazing portion of a metal to be joined, and a brazing metal brazing to be joined. Image sensing means for detecting the light emitted from the image forming portion, image processing means for calculating a representative value for determining the temperature of the brazing portion of the metal to be joined from the brightness value of the image signal input from the image sensing means, and the image And a controller for controlling the heating amount of the heating means so that the representative value output from the processing means becomes a target value corresponding to a predetermined temperature (claims 1 and 2).

The image processing means sets a threshold value for removing low luminance values corresponding to information other than the metal to be joined and noise from the luminance value of the image data input from the image pickup means. , A means for calculating the representative value by removing the luminance value whose luminance value is less than or equal to the threshold value (claim 3).

Further, the image processing means delays the calculation of the representative value when the number of brightness values exceeding the threshold value is less than a predetermined number and executes the processing in that case. When the number of luminance values exceeding the above is sufficient, means for outputting a "true" signal is provided, and a gate based on this judgment is provided at the final stage of the representative value calculating section of the image processing means. When the signal is "false", it has means for executing processing such as temporarily setting the representative value to a predetermined lower limit value (claim 4).

In order to automatically set the threshold value to an optimum value, the image processing means reads a change in brightness along the horizontal axis represented by the image signal, and clearly identifies a high brightness portion. When there is a low-brightness part, it has a means to measure the brightness change from the low-brightness part to the high-brightness part, and use the brightness value of the point with the largest change rate as the threshold value. (Claims 5 and 6).

Further, the image processing means complements the removed brightness value equal to or less than the threshold value with a complement value calculated from the non-removed brightness value, and then calculates the representative value from all the brightness values. Means (Claim 7).

Further, the image processing means monitors the temperature of the brazing portion of the metal to be joined during brazing by measuring the temperature based on the representative value, and evaluates the brazing work to judge the brazing quality. (Claim 8 and Claim 9).

[0013]

According to the automatic brazing method and apparatus of the present invention, the metal to be joined is heated by the heating means, and the light emitted when the brazing part of the metal to be joined is heated is detected by the imaging means. A representative value for judging the temperature of the brazing part of the metal to be joined by referring to the brightness value of each pixel of the image signal after inputting the image signal obtained by the image pickup means by the image processing means and digitizing it calculate. A representative value when the metal to be joined has a predetermined temperature suitable for brazing is registered as a target value in the image processing means, and the target value is compared with the representative value to heat the comparison result. Feed back to the controller of the means. The controller adjusts the heating amount of the heating means by this feedback signal and controls the metal to be joined to a predetermined temperature suitable for brazing. With the above operation, it is possible to prevent the metal to be joined from being overheated and insufficiently heated, and improve the reliability of the automatic brazing operation (claims 1 and 2).

Further, when the image processing means calculates the representative value by referring to the luminance value of each pixel from the image signal obtained by the image pickup means, it is used to remove information other than the metal to be joined, noise and the like. It is possible to improve the accuracy of the temperature measurement of the metal to be joined by setting the threshold value and removing the low luminance value below this threshold value and calculating the representative value (claim 3).

In addition, it is possible to prevent the temperature rise from being judged by a small number of brightness values when the brightness value corresponding to the intensity of the light emitted from the brazing part of the metal to be joined is near the threshold value or due to noise. Therefore, when the number of brightness values exceeding the threshold value is sufficient, a number judgment means for outputting a "true" signal is provided, and the gate based on this judgment is provided at the end of the representative value calculation section of the image processing means. It is installed on the steps. When the number is sufficient, that is, when the output of the number determining means is "true", the representative value is output, and when the number is insufficient, that is, the output of the number determining means is "false". In that case, a value such as "0" or "-1" is output to inform the subsequent processing stage. Therefore, when it is received that the number is insufficient in the next processing step, measures such as setting the above-mentioned representative value to the lower limit value are taken, and a representative value falsely calculated with a small number of luminance values is calculated and referred to. It is possible to perform stable automatic brazing operation (claim 4).

Further, in order to automatically set the threshold value to the optimum value by the image processing means, the change in the brightness is read along the horizontal axis represented by the image signal, and clearly the high brightness part and the low brightness part are read. , The luminance change from the low-luminance part to the high-luminance part is measured, and the point with the largest rate of change, that is, the maximum absolute value of the differential value with respect to the horizontal axis is obtained. It is possible to deal with an image other than an indeterminate metal to be joined by setting the threshold value with the luminance value of a given point (claim 5).

Further, after the image processing means removes the luminance value equal to or lower than the threshold value, the removed luminance value is complemented by using a complementary value calculated from a luminance value other than the removed luminance value. In order to calculate a representative value for temperature rise determination from all the brightness values after the complement, there is a memory that stores whether or not the brightness value of each pixel is removed when performing the image processing. The existence of the removed brightness value and its pixel position are confirmed by this memory, the removed brightness value is complemented by a complement value calculated from the brightness values other than the removed brightness value, and the total brightness after the complement By calculating the representative value for temperature determination from the value, the missing information can be corrected and the temperature can be measured with higher accuracy (claim 6).

Furthermore, the temperature of the brazing part of the metal to be joined during brazing is monitored by measuring the temperature by the above-mentioned representative value by the image means, and the brazing work is evaluated to judge the brazing quality. Even if there is a problem in the brazing work, a process for avoiding it can be executed to perform high-quality brazing (claim 7).
Claim 8).

[0019]

Embodiments of the present invention will be described below with reference to FIGS. FIG. 1 is a schematic configuration diagram showing an embodiment of an automatic brazing method and apparatus according to the present invention. In FIG. 1, the heating amount by the heating means 2 for heating the metal to be joined 1a, 1b can be arbitrarily changed by the control device 3. Further, the energy source of the heating means 2 is not shown. The metal 1a, 1b to be heated is fixed at a predetermined position by a gripping device (not shown). A vision sensor 4, which is an image pickup means, is installed so as to capture the metal to be joined 1a, 1b in its field of view, and this vision sensor 4 is provided with a neutral density filter on the front surface of the lens of an industrial black and white television camera. In addition, a visible light cut filter may be attached when visible light such as illumination of the installation place greatly affects the sensing. The image signal captured by the vision sensor 4 is digitized by the image processing means 5, and the brightness value of each pixel is stored. The brightness value is arithmetically processed to determine the heating status. Based on this determination, the host controller 6 is configured to issue a command to the control device 3 to adjust the heating amount by the heating means 2 and to advance the step of automatic operation.

The operation of the automatic brazing device having the above configuration is as follows. However, in this example, it is assumed that the metal to be joined 1a, 1b to be brazed is a copper pipe, which has been processed to a predetermined size in advance, a brazing metal has been mounted, and the metals to be joined have already been inserted. Further, in the image processing means 5, an area corresponding to the metal to be joined is set in advance in the image, and only the inside of this area is the target for calculating the representative value for temperature determination. FIG. 2 is an explanatory diagram showing a setting example of a region set on an image obtained by the vision sensor 4 which is the image pickup means of FIG. In FIG. 2, metal 1 to be joined in the image
An area 5a is set in the brazing parts of a and 1b. Further, in this embodiment, the average value of the brightness values of each pixel is adopted as the representative value. In addition, it is assumed that the image processing means 5 is set in advance with a target value of a representative value for temperature rise determination.

First, the metals 1a and 1b to be joined shown in FIG. 1 are conveyed and supported to a predetermined position by a gripping device (not shown). Next, the heating means 2 is set to the heating position. This heating is performed by high frequency induction, the heating means 2 has a coil for heating, and when a heating start command is issued from the host controller 6, the controller 3 energizes the metal to be welded 1a,
Heat 1b. The heating state of the brazing part of the metal to be joined is observed by the vision sensor 4 which is an image pickup means, and the light emitted from the metal to be joined is detected. The image signal obtained by the vision sensor 4 is input to the image processing means 5,
After being digitized, the brightness value of each pixel is stored in the frame memory 5A. Next, the area extracting means 5B performs the total calculation of the brightness values in the set area 5a. This tabulation calculation is achieved by simply adding the luminance values of the respective pixels stored in the frame memory 5A of the image processing means 5. However, when a great amount of processing time is required to collect all the pixels in the region 5a, it is possible to thin out the pixels for the purpose of time reduction. Then, the representative value calculating means 5C divides the total value by the number of brightness values (the number of pixels) to which the total value is added, whereby the average value of the brightness values in the region as a representative value for temperature determination can be calculated. The target value comparing means 5D compares the calculated representative value with the target value, and the next heating determining means 5E determines and outputs the heating status of the metal to be joined based on the comparison result. Based on this determination, the host controller 6 issues a command, and the controller 3 increases the heating amount when the heating power of the heating means 2 is insufficiently heated by this command, and conversely when the heating force of the heating means 2 is excessively heated. By reducing the amount, the metal to be joined 1a, 1b is controlled to the optimum temperature for brazing.

In this embodiment, since the frame memory 5A for image processing has a structure of 8 bits per pixel, the luminance value takes a decimal value from 0 to 255. The setting of the camera of the vision sensor 4 is performed by the metal to be joined 1a, 1b.
When the light emission state of 800 ° C., which is a temperature suitable for brazing, is captured, the aperture of the lens, the filter, etc. are adjusted so that the brightness value becomes 150, for example. And
By setting the target value in the image processing means 5 to 150 and performing a comparison calculation with the average value which is the representative value,
The metal to be joined can be controlled at 800 ° C. In this control system evaluation method, P
Although ID control or the like can be considered, P control is used in this embodiment to improve the calculation speed. After the evaluation of this control system, it is desired that the heating amount of the heating means 2 be increased or decreased linearly with the evaluation value, but a method of adjusting the heating amount in two or three stages is also conceivable. Further, in the case of a simplified control system, it is judged that the temperature rise of the metal to be joined is completed based on the fact that the average value which is a representative value has reached 150 without performing feedback control to the heating means 2. It is also possible to configure the entire sequence such that the subsequent heating is terminated or the heating amount is lowered until the wax penetrates thereafter to maintain a constant temperature. Since the brazing temperature has a temperature range that can be tolerated to some extent, the target value is set by dividing it into the maximum value and the minimum value, and the heating amount is set when the representative value is lower than the minimum value. It is also possible to use a method of increasing the amount of heat and decreasing the amount of heat when the temperature rises above the maximum value. The metal to be joined is not limited to the copper pipe. The representative value is not limited to the simple average value.
By controlling the temperature of the metal to be joined during brazing as described above, highly reliable automatic brazing operation can be performed.

Further, by the image processing means 5, information such as noise and obstacles other than the metals 1a and 1b to be joined are present in the temperature determination area 5a in the image signal from the vision sensor 4 which is the image pickup means. If so, the process for removing the information is executed as follows. Generally, the temperature of the obstacles other than the metal to be bonded is lower than that of the metal to be bonded 1a and 1b during heating, so a threshold value for removal is set. Perform aggregate calculation. The average value as the representative value can be calculated by dividing the summed value of the brightness values by the number of addition (pixel number). FIG. 3 is a diagram for explaining an example in the case where information that seems to be other than the metals 1a and 1b to be joined in FIG. 1 is mixed in the image signal. In FIG. 3, the area 5 set on the image
The information 5b other than the metal to be joined is shown in a along with the luminance distribution of the brazed portions of the metal to be joined 1a and 1b. In addition, FIG. 4 is an explanatory diagram showing an example in which the luminance value in the area 5a of the image including the information 5b other than the metal to be joined in FIG. 3 is extracted. In addition, FIG. 5 is a luminance value in the area when the luminance value equal to or less than the threshold value set in FIG. 4 is removed to remove the luminance value of the information 5b which is considered to be other than the metal to be joined. It is an explanatory view showing an example of. As shown in the above explanatory diagram, the accuracy of temperature measurement can be further improved by removing the brightness value of the information 5b other than the metal to be joined.

FIG. 6 is a schematic diagram showing another embodiment of heating the metal to be joined using the gas burner of the automatic brazing method and apparatus according to the present invention. In FIG. 6, the gas burner 2a constituting the heating means 2 is gripped by the industrial robot 2b. The mixed gas supplied to the burner 2a is generated by the gas mixer 2c from the propane gas and the oxygen gas enclosed in the gas cylinders 2h and 2i, respectively. Further, the pressure / flow rate adjustments and shutoffs thereof are performed by the respective flow rate control valves 2f, 2g and shutoff valves 2d, 2e.
Done by. The metals 1a and 1b to be welded, which are heated by the burner 2a, are fixed at predetermined positions by a gripping device (not shown). A vision sensor 4 which is an image pickup means is installed so as to catch the brazing part of the metal to be joined 1a, 1b in its field of view, and the image signal captured by the vision sensor 4 is digitized by the image processing means 5 and stored and processed. To be done. The processing result is input to the host controller 6, and the host controller issues a command in response to the input.
The controller 3 controls the industrial robot 2b in response to the command to control the temperature of the metal to be welded 1a, 1b by heating the burner 2a to perform highly reliable automatic brazing.

The operation of the automatic brazing device having the above-mentioned configuration is as follows. However, the metal to be welded 1
It is assumed that a and 1b are copper pipes, which are preliminarily processed into a predetermined size, brazed, and metals to be joined are already inserted. Further, it is assumed that appropriate amounts of propane gas and oxygen gas are respectively supplied from the cylinders 2h and 2i, mixed by the gas mixer 2c, and then ejected from the burner 2a and already ignited. In addition, the image processing means 5 includes regions 5a corresponding to the metals 1a and 1b to be bonded, which are shown in FIG.
Further, it is assumed that a threshold value or the like serving as a reference for processing determination of the image signal including the metal 5b other than the metal to be joined shown in FIGS. 3 to 5 is set.

First, the metals 1a and 1b to be joined shown in FIG. 6 are conveyed and supported to a predetermined position by a gripping device (not shown). Next, the industrial robot 2b constituting the heating means 2 is operated by the bar.
2a is set to the heating position. Then, the heated state is observed by the vision sensor 4 to detect the light emitted from the brazed portions of the metal to be joined 1a, 1b. The image signal obtained by the vision sensor 4 is input to the image processing means 5, digitized and then stored in the frame memory 5A, and the brightness of each pixel of the area 5a set in FIG. 2 by the area extracting means 5B. Aggregate calculation of values is performed. This total calculation and calculation of the average value as the representative value in the representative value calculating means 5C, the target value comparing means 5D and the heating degree judging means 5
The processing at E proceeds in the same way as the processing of the embodiment of FIG. 1 above. By the output of the image processing means 5, the control device 3 controls the heating amount of the gas burner 2a by the industrial robot 2b constituting the heating means 2 via the host controller 6. When the gas burner 2a is used, this heating amount can be controlled by increasing or decreasing the gas flow rate, moving the gas burner away, or swinging the gas burner. Further, as the information 5a other than the metal to be welded of the image signals of FIGS. 3 to 5, the burner tip, the nozzle for supplying the brazing material, the noise, etc. are conceivable, but they can be removed by setting the above threshold value. As described above, the temperature of the metal to be joined can be controlled even in the automatic brazing device using the gas burner,
Reliable automatic brazing operation can be performed.

FIG. 7 is a flow chart of processing in the image processing means 5 of FIGS. 1 and 6. In FIG. 7, in process S1, the image signal obtained by the vision sensor 4 of FIGS. 1 and 6 is taken into the image processing means 5, digitized, and then the luminance value of each pixel is stored in the frame memory 5A. In process S2, each variable at each pixel position in the memory is cleared, and process S
In step 3, the luminance value of one pixel in the area 5a is extracted. In the next process S4, it is determined whether the brightness value is equal to or higher than the threshold value. If the brightness value is equal to or lower than the threshold value, the process returns to process S3. ) Is added. In the next process S6, it is determined whether or not all pixels have been extracted, and if not completed, the process returns to process S3. If completed, in the next process S7 the added value of the brightness values is divided by the added value of the number to average the brightness values. A value is calculated, and the average value of the brightness values is used as a representative value for judging the temperature in the region 5a. In the next process S8, the average value which is the representative value is compared with the target value, and the comparison value is output in process S9.

FIG. 8 is a process in the case where the process for delaying the output of the completion signal is added when the number (the number of pixels) of the image processing means of FIGS. FIG. In FIG. 8, after performing the same processing S1 to processing S8 as in FIG. 7, here, in the next processing 10, when the number (pixel number) exceeding the threshold value in the processing S4 is small, the output of the completion signal is postponed. Therefore, it is determined whether or not the number is equal to or more than a set value, and a unit for outputting a “true” signal is provided when the number aggregated in the process 5 is equal to or more than a predetermined value, and the process in FIG. Similarly, the comparison value is output, but if the signal is "false", in step S11, a process of setting "0" as an output signal is performed, and a signal of "0" is received in the subsequent processing stage. In such a case, countermeasures such as setting the representative value to a predetermined lower limit value are taken to prevent the representative value from being calculated and referred to in a small number, so that stable automatic brazing operation is performed. 7 and 8 show the case where the average value of the brightness values is calculated as the representative value for temperature determination.

FIG. 9 is an explanatory view showing a typical setting example of the horizontal axis 5c for automatically setting the threshold value for removing the low luminance part in the image processing means 5 of FIGS. 1 and 6. It is a figure. Further, FIG. 10 shows changes in the brightness value 5d on the representative horizontal axis of FIG. 9 and its spatial differential value 5e.
It is explanatory drawing which shows the change of. 9 and 10,
The automatic setting of the threshold value for removing the low luminance value portion in the image processing means 5 of FIGS. 1 and 6 is performed as follows. First, in the image signal including the information from the specific bonding metals 1a and 1b in the region 5a in FIG. 9 and the information 5b other than the metals to be bonded, the luminance of each pixel position on the horizontal axis 5c represented in FIG. 10 is represented. A change in the spatial differential value 5e of each pixel position spatially differentiated with respect to the horizontal axis 5c of the value 5d is obtained. Next, the point 5 at the pixel position showing the maximum value of the spatial differential value 5e.
f is detected, and the luminance value 5g of the pixel at that point is automatically set as the threshold value for removing the low luminance portion.

FIG. 11 shows the image processing means 5 shown in FIGS. 1 and 6, after removing the luminance value equal to or lower than the threshold value, the removed low luminance portion is used by using the luminance values other than the removed portion. It is explanatory drawing which shows a mode that it complements with the calculated complement value. In this example, processing is executed for each horizontal axis scanned in the area 5a of the image signal. In the specific example, among the 13 brightness values extracted from the image signal on one horizontal axis, 4 of the information 5b other than the removed metal to be joined is selected.
A total of 6 brightness values are extracted from both sides of each brightness value portion, and the average value thereof is used as a complement value to obtain all the brightness values after the complement.
Such processing work is executed in units of one horizontal axis to perform complementary work, and a complementary image of the area 5a is obtained. Note that, in addition to this, a method of linearly interpolating between the brightness values on both sides with respect to the horizontal axis can be considered for the calculation of the complementary value.

FIG. 12 is a schematic view showing still another embodiment in the case of brazing work in the assembly process of the gas turbine blade of the automatic brazing method and apparatus according to the present invention. In this brazing work, it takes time for the brazing filler metal to penetrate, so it is necessary to control the temperature for a longer period of time. Therefore,
The temperature control according to the present invention is very effective. 12, the configuration and operation of the device are almost the same as those of the embodiments of FIGS. 1 and 6 described above, but the brazing part of the metal to be joined is the image range captured by the camera of the vision sensor 4 which is the imaging means. Since it is comparatively large, it is desirable to measure the temperature of the metal to be joined using the swinging device of the TV camera of the vision sensor 4 or the TV camera of the plurality of vision sensors 4.

According to the embodiments shown in FIGS. 1, 6 and 12, since the brazing work of the metal to be joined is performed as described above, the fluctuation of the ambient temperature and the initial temperature of the heatable member and Even if the temperature rise state changes due to the dimensional change of the metal to be joined,
Even when the information other than the metal to be joined is mixed in the vision sensor which is the image pickup means, the influence can be removed, and highly reliable work can be performed.

FIG. 13 is a measurement diagram illustrating the relationship between the luminance value of the image signal output from the image pickup means and the work temperature in the automatic brazing method and apparatus thereof shown in FIGS. In FIG. 13, the measurement points of the luminance value of the black-and-white TV image output from the image pickup means and the work temperature are on the straight line 5h, indicating that the two are in a proportional relationship. Therefore, it is possible to accurately measure the temperature from the image signal of the brazing part of the metal to be joined 1a, 1b detected by the vision sensor 4 which is the image pickup means in the present embodiment. In the above embodiment, as a representative value for judging the temperature of the brightness value of the brazing part of the metal to be joined, a simple average value of the brightness values of each pixel of the digitized multi-valued image data in the area 5a is used. However, it is possible to use other various average values, and the average values are not limited to these.

FIG. 14 is a measurement diagram illustrating the relationship between the brightness of the image signal output from the image pickup means and the heating time of the work in the automatic brazing method and apparatus shown in FIGS. In FIG. 14, the measurement points of the brightness value of the radiated light from the metal brazed portion of the black-and-white TV image output from the imaging means and the heating time of the workpiece by the heating means 2 are on one curve 5i, and both are unique. It shows that the heating state is not normal if the two have the relationship of the curve 5j shown by the broken line. Therefore, the image processing means 5 of the automatic brazing apparatus according to the embodiment measures the temperature by referring to the brightness value of the captured image using the black and white TV camera of the vision sensor 4 which is the above-mentioned image capturing means, thereby performing the brazing process. The temperature of the work is captured and evaluated in real time to judge the quality of the efficient brazing work and to provide effective guidance, as well as the setting of each part of the automatic operation, such as the position adjustment of the metals 1a and 1b to be welded and the like. The moving position adjustment of the heating means 2 and the burner 2a thereof and the position adjustment of the TV camera of the vision sensor 4 which is the image pickup means can be performed promptly. Further, by monitoring the work temperature during heating during the automatic operation, even if a temperature rise problem occurs, a process for avoiding the problem can be executed, and thereby high-quality brazing can be performed. In the above example, the case where the brazing work of the metal to be joined is performed by using the high frequency induction coil or the gas burner which is the heating means,
Although described, the present invention is not limited to these, and can be similarly applied to, for example, a processed portion by heating of laser processing or electric welding processing.

In the above embodiment, a contact type or non-contact type temperature sensor is provided, and the control device checks the temperature corresponding to the representative value by the image processing means 5 and the temperature detected by the temperature sensor. In addition, the amount of heating of the heating means 2 can be controlled, or the aperture of the camera of the vision sensor 4 which is the image pickup means can be adjusted to calibrate the output of the vision sensor 4 and adjust the reference.

[0036]

As described above, according to the present invention, the brazing temperature of the metal to be joined is evaluated by processing the brightness value captured by the image pickup means to determine the temperature rise of the metal to be joined. However, by feeding back to the heating means, it is possible to prevent insufficient heating or excessive heating of the metal to be joined and perform highly reliable automatic operation. In addition, by deleting information other than the metal to be joined from the image signal captured by the imaging means and determining the temperature rise, the accuracy of temperature measurement can be improved, and more reliable brazing can be performed. It can be carried out. Further, when the number of data exceeding the removal threshold value for canceling the low-brightness portion is small, a means for delaying the output of the representative value indicating the temperature is provided, so that malfunction due to noise or a small amount of data is provided. It is possible to prevent the unstable judgment due to the operation and to perform stable brazing. Further, by automatically setting the threshold value for removing the low-luminance portion, it is possible to deal with an image other than the indefinite metal to be joined, and it is possible to perform an automatic brazing operation that is less susceptible to disturbance.
Further, by determining the representative value for temperature rise determination from the corrected data obtained by complementing the removed portion of the image signal with the complementary value calculated from the brightness value other than the removed portion, It is possible to measure temperature with high accuracy and improve reliability of brazing. Furthermore, the temperature of the work being brazed is measured in real time by performing temperature measurement by referring to the brightness value of the image using the TV camera of the image pickup means, and thereby, it is possible to efficiently judge whether or not the brazing is good or bad. In addition to being able to provide effective guidance, it is possible to quickly set each part of automatic operation, and even if there is a problem in brazing by monitoring the work temperature during heating during automatic operation. There is an effect that high-quality brazing can be performed by executing processing to avoid it.

[Brief description of drawings]

FIG. 1 is a schematic configuration diagram showing an embodiment of an automatic brazing device according to the present invention.

FIG. 2 is an explanatory diagram showing an example of a region set on an image by the vision sensor of FIG.

FIG. 3 is an explanatory diagram showing an example of an image including information other than the metal to be joined in FIG. 1.

FIG. 4 is an explanatory diagram showing an example of information in a region including information other than the metal to be joined in FIG.

FIG. 5 is an explanatory diagram showing an example of information in an area when a value equal to or less than a set threshold value in FIG. 4 is removed.

FIG. 6 is a schematic configuration diagram showing another embodiment in which heating is performed by the gas burner of the automatic brazing device according to the present invention.

FIG. 7 is a flowchart of processing in the image processing means shown in FIGS. 1 and 6;

8 is a flow chart of processing for delaying output of a completion signal when the number of pixels exceeding a threshold for removing a low-luminance portion is small in the image processing means of FIGS. 1 and 6;

9 is an explanatory diagram showing a setting example of a representative horizontal axis for automatically setting a threshold value for removing a low-luminance portion in the image processing means of FIGS. 1 and 6. FIG.

FIG. 10 is an explanatory diagram showing a change in brightness value and a change in spatial differential value thereof on the representative horizontal axis for automatically setting the threshold value for removing the low brightness portion in FIG.

FIG. 11 is an explanatory diagram showing a manner in which the removed low-luminance portion in the image processing means of FIGS. 1 and 6 is complemented by the luminance of the periphery thereof.

FIG. 12 is a schematic structural view showing still another embodiment in the case of brazing a gas turbine blade of the automatic brazing device according to the present invention.

FIG. 13 is a measurement diagram showing the relationship between the brightness of image data and the work temperature in the automatic brazing apparatus shown in FIGS. 1, 6 and 12.

FIG. 14 is a measurement diagram showing the relationship between the brightness of the image signal and the work heating time in the automatic brazing device shown in FIGS. 1, 6 and 12.

[Explanation of symbols]

 1a, 1b Metal to be joined, 2 Heating means, 2a Burner 2b Industrial robot 2c Gas mixer 2d Propane gas shutoff valve 2e Oxygen shutoff valve 2f Propane gas flow control valve 2g Oxygen flow control valve 2h Propane gas cylinder 2i Oxygen cylinder 3 Control device 4 Imaging means 5 Image processing means 6 Upper controller

 ─────────────────────────────────────────────────── ─── Continuation of the front page (72) Inventor Nobuyuki Nagayasu No. 709 Tomita, Ohira-cho, Shimotsuga-gun, Tochigi Prefecture 2 Hitachi Tochigi Electronics Co., Ltd.

Claims (14)

[Claims] 1. The brightness of each pixel after heating the brazing part of the metal to be joined, detecting the light emitted by the temperature rise of the brazing part of the metal to be joined and digitizing the detected image signal. Storing a value, calculating a representative value for judging the temperature of the brazing part of the metal to be joined from the stored luminance value of each pixel, and referring to the calculated representative value, the representative value The automatic brazing method is characterized in that the heating amount for heating the brazing part of the metal to be joined is controlled so that is a target value corresponding to a predetermined temperature. 2. A heating means for heating a brazing part of a metal to be joined, an image pickup means for detecting light emitted by a temperature rise of the brazing part of the metal to be joined, and the detected image signal being inputted. Image processing means for storing each pixel brightness value in a frame memory, and calculating a representative value for judging the temperature of the brazing part of the metal to be joined from the stored brightness value of each pixel. And a controller for controlling the heating amount of the heating means so that the representative value becomes a target value corresponding to a predetermined temperature with reference to the calculated representative value. Attachment device. 3. The image processing means sets a threshold value for removing a low brightness value corresponding to information other than the metal to be joined, noise, etc. from the stored brightness value of each pixel, 3. The automatic brazing apparatus according to claim 2, wherein the representative value is calculated by removing the brightness value whose brightness value is less than or equal to the threshold value. 4. The image processing means monitors the number of counted luminance values when removing the luminance values equal to or less than the threshold value and totaling the luminance values, and when the number is less than a predetermined number. ,
The automatic brazing device according to claim 3, wherein the process of temporarily setting the representative value to a predetermined lower limit value is executed after delaying the calculation of the representative value. 5. The automatic brazing apparatus according to claim 3 or 4, wherein the image processing means automatically and variably sets the threshold value during brazing or after some brazing trial. 6. The image processing means according to claim 5, wherein the change in the brightness value is read along a horizontal axis represented by the image signal, and the brightness value at the maximum point of the change rate is used as the threshold value. Automatic brazing device. 7. The image processing means complements a removed brightness value equal to or less than the threshold value by using a complement value calculated from a brightness value other than the removed brightness value, and then from the whole brightness value. The automatic brazing device according to claim 3, wherein the representative value is calculated. 8. The brightness of each pixel after heating the brazing part of the metal to be joined, detecting the light emitted by the temperature rise of the brazing part of the metal to be joined and digitizing the detected image signal. A value is stored, a representative value for judging the temperature of the brazing part of the metal to be joined is calculated from the stored brightness value of each pixel, and the quality of the brazing part is evaluated by the representative value. With reference to the calculated representative value, the heating amount for heating the brazing part of the metal to be joined is controlled so that the representative value becomes a target value corresponding to a predetermined temperature, and the quality is referred to by referring to the quality evaluation. An automatic brazing method characterized by making improvements. 9. A heating means for heating a metal to be joined, an image pickup means for detecting light emitted by a temperature rise of a brazing part of the metal to be joined, and the detected image are inputted and digitized. After that, the brightness value of each pixel is stored in the frame memory, a representative value for judging the temperature of the brazing part of the metal to be joined is calculated from the stored brightness value of each pixel, and brazing is performed using the representative value. Image processing means for performing quality evaluation of the parts, and controlling the heating amount of the heating means so that the representative value becomes a target value corresponding to a predetermined temperature by referring to the calculated representative value and performing the quality evaluation. An automatic brazing device comprising a control device for performing quality improvement measures such as setting adjustment of the heating means with reference to the above. 10. A temperature sensor for detecting the temperature of a brazing part of the metal to be joined is provided, and the control device compares the temperature corresponding to the representative value by the image processing means with the temperature detected by the temperature sensor. In addition, the automatic brazing device according to any one of claims 2 to 7 or 9, wherein the heating amount of the heating means is controlled. 11. A temperature sensor for detecting the temperature of a brazing part of the metal to be joined is provided, and the control device adjusts the output of the imaging means according to the detection output of the temperature sensor. The automatic brazing apparatus according to claim 1, claim 9, or claim 10. 12. The automatic brazing apparatus according to claim 2, wherein the representative value is an average value of luminance values of corresponding pixels. 13. The automatic brazing apparatus according to claim 2, wherein the relationship between the brightness value and the corresponding temperature is linearly proportional. . 14. The relationship between the brightness value and the representative value and the heating time is a unique increasing relationship, and if the relationship deviates from this relationship, it is evaluated that brazing is defective.
14. The automatic brazing device according to claim 13.