Defect determination device

Abstract

To solve the problem that it is difficult to prepare teacher data on a measurement result including a defect when automatically determining the presence or absence of a defect pertaining to a bead of welding on the basis of a measurement result of a laser ultrasonic method.SOLUTION: A defect determination device 1 includes: a reception section 11 that receives a measurement result of a laser ultrasonic method pertaining to a bead of welding; a prediction section 13 that predicts, on the basis of the received measurement result, a measurement result of a position adjacent to a position corresponding to the measurement result; a comparison section 14 that compares a prediction result with a measurement result of a position corresponding to the prediction result; and a determination section 15 that determines that a defect is present in the bead according to a comparison result. Thus, the defect determination device can predict a measurement result and determine the presence or absence of the defect without preparing teacher data on a measurement result, including a defect.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] The present invention relates to a defect determination device that determines the presence or absence of defects in a weld bead using the measurement results of the laser ultrasonic method.

Background Art

[0002] Conventionally, an apparatus for inspecting a weld using the laser ultrasonic method (Laser Ultrasonic Technique) is known (see, for example, Patent Document 1). By using the laser ultrasonic method, the state of the welded portion can be measured non-contact, so that, for example, it is possible to measure even the bead immediately after welding.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] There is a desire to automatically find welding defects by applying the measurement results by the laser ultrasonic method to a learning device such as a neural network. However, in order to train such a learning device, it is necessary to prepare the measurement results of the defect-free portion and the measurement results of the defective portion as teacher data. However, there is a problem that it is difficult to prepare the measurement results of the defective portion of the weld in an amount required for learning.

[0005] The present invention has been made to solve the above problems, and an object of the present invention is to provide a defect determination device that can automatically determine the presence or absence of defects in a weld bead based on the measurement results of the laser ultrasonic method without preparing teacher data regarding the defective portion.

Means for Solving the Problems

[0006] To achieve the above object, a defect determination device according to an aspect of the present invention includes a reception unit that receives a measurement result of the laser ultrasonic method regarding a weld bead, a prediction unit that predicts a measurement result of a position adjacent to the position corresponding to the measurement result based on the measurement result received by the reception unit, a comparison unit that compares the prediction result by the prediction unit with the measurement result of the position corresponding to the prediction result, and a determination unit that determines whether there is a defect in the bead according to the comparison result by the comparison unit.

Effects of the Invention

[0007] According to the defect determination device according to an aspect of the present invention, it is possible to automatically determine the presence or absence of a defect according to the comparison result between the prediction result based on the measurement result of the laser ultrasonic method regarding the weld bead and the measurement result corresponding to the prediction result. In this way, it becomes possible to determine the presence or absence of a defect without preparing teacher data regarding the location of the welding defect.

Brief Description of the Drawings

[0008]

Figure 1

Figure 2

Figure 3A

Figure 3B

Figure 4

Figure 5A

Figure 5B

Embodiments for Carrying Out the Invention

[0009] Hereinafter, the defect determination apparatus according to the present invention will be described using embodiments. In the following embodiments, components and steps denoted by the same reference numerals are the same or corresponding, and repeated descriptions may be omitted. The defect determination apparatus according to the present embodiment determines the presence or absence of a welding defect according to the comparison result between the prediction result predicted using the measurement result of the laser ultrasonic method and the measurement result corresponding to the prediction result.

[0010] FIG. 1 is a schematic diagram showing the configuration of the robot control system 100 according to the present embodiment, and FIG. 2 is a block diagram showing the configuration of the defect determination apparatus 1 according to the present embodiment. As shown in FIG. 1, the robot control system 100 includes a defect determination apparatus 1, a manipulator 2, a robot control apparatus 3, and a welding power source 4. In FIG. 1, the defect determination apparatus 1 is shown as an independent apparatus, but the defect determination apparatus 1 may be incorporated in another apparatus such as the robot control apparatus 3, for example.

[0011] The defect determination apparatus 1 determines the presence or absence of a defect in the welded portion using the measurement result of the laser ultrasonic method. As shown in FIG. 2, the defect determination apparatus 1 includes a reception unit 11, a storage unit 12, a prediction unit 13, a comparison unit 14, a determination unit 15, and an output unit 16. The defect determination apparatus 1 may be connected to, for example, the laser ultrasonic measurement apparatus 2a of the manipulator 2 or the robot control apparatus 3. Each configuration of the defect determination apparatus 1 will be described later.

[0012] The manipulator 2 has a plurality of arms connected by joints driven by a motor. A welding torch is attached to the tip of the manipulator 2. When a welding wire is used for welding, the manipulator 2 may be equipped with a wire feeding device for feeding the welding wire. Further, a laser ultrasonic measuring device 2a is attached to the manipulator 2. The mounting position of the laser ultrasonic measuring device 2a on the manipulator 2 is not particularly limited, but it is preferably mounted at a position where the welding portion can be sensed from a desired direction by the laser ultrasonic measuring device 2a. Therefore, the laser ultrasonic measuring device 2a may be mounted, for example, on the tip side of the manipulator 2. The tip side of the manipulator 2 may be, for example, a position on the base end side of the welding torch as shown in FIG. 1. The manipulator 2 is not particularly limited, but may be, for example, a vertically articulated robot.

[0013] The laser ultrasonic measuring device 2a measures the welding bead by the laser ultrasonic method. The laser ultrasonic measuring device 2a may include, for example, an ultrasonic generating unit 2b and an ultrasonic detecting unit 2c. In the laser ultrasonic measuring device 2a, ultrasonic waves may be generated in the measurement target by condensing the pulsed laser light emitted from the ultrasonic generating unit 2b onto the measurement target. Then, for example, the ultrasonic detecting unit 2c, which is a laser interferometer, irradiates the measurement target with the laser light for ultrasonic detection, and captures the Doppler effect generated by the surface vibration that occurs when the generated ultrasonic waves reach the irradiation position of the laser light for detection with the laser interferometer to detect the ultrasonic waves. In addition, ultrasonic waves may be detected by a method other than using a laser interferometer, for example, the knife edge method.

[0014] In this embodiment, as shown in FIG. 4, the case where laser light for generating ultrasonic waves is irradiated at a plurality of locations and the ultrasonic waves generated accordingly are detected by irradiating the laser light at one location will be mainly described. FIG. 4 is a diagram showing the result of fillet welding of the lap joint, that is, the result of welding the joint portion between the upper plate work 5a and the lower plate work 5b. In FIG. 4, the x-axis is provided in the longitudinal direction of the bead 6. And at each position where the value of x is B1, B2, B3..., measurement using the laser ultrasonic method is performed. Hereinafter, each position where the value of x is B1, B2, etc. may also be referred to as measurement positions B1, B2, etc. It is assumed that the welding is performed in the direction of the x-axis, that is, the rightward direction in FIG. 4. Also, the measurement positions along the longitudinal direction of the bead 6 may be provided, for example, at equal intervals. In this embodiment, for example, for one measurement position B1, laser light for generating ultrasonic waves is sequentially irradiated to a plurality of irradiation positions T1-1 to T1-8, and the ultrasonic waves generated by the laser light are detected by irradiating the laser light for ultrasonic wave detection to one irradiation position R1. The measurement by the laser ultrasonic measurement device 2a may be performed, for example, in response to a measurement instruction from the defect determination device 1, or may not be. In the latter case, the laser ultrasonic measurement device 2a may repeatedly obtain a plurality of measurement results, for example, by irradiating laser light for generating ultrasonic waves to a plurality of irradiation positions at predetermined intervals (for example, for each of the measurement positions B1, B2) along the longitudinal direction of the bead 6. As shown in FIG. 4, the laser light may be irradiated, for example, on the bead 6, or may be irradiated on the works 5a and 5b. Also, in this embodiment, the case where the bead 6 to be measured by the laser ultrasonic method is a bead corresponding to one welding pass will be mainly described, but the bead to be measured may be, for example, a set of beads corresponding to a plurality of welding passes in multi-layer welding.

[0015] Note that the irradiation position of the laser beam in the laser ultrasonic method is not limited to the above description. For example, the laser beam for generating ultrasonic waves may be irradiated at one location, and the ultrasonic waves generated accordingly may be detected by irradiating the laser beam at a plurality of locations. In this way, the irradiation position of the laser beam for generating ultrasonic waves may be, for example, one location or a plurality of locations, and the irradiation position of the laser beam for detecting ultrasonic waves may be, for example, one location or a plurality of locations. As a result, as long as a plurality of measurement results in a direction perpendicular to the longitudinal direction of the bead 6 can be obtained for each measurement position, the measurement method using the laser ultrasonic method is not limited. Note that the laser ultrasonic measurement device 2a is already known, and a detailed description thereof is omitted.

[0016] The robot control device 3 controls the manipulator 2, and may control the manipulator 2 and the welding power source 4 so that welding is performed along the welding lines of the workpieces 5a and 5b, for example. Note that the configuration of the robot control device 3 is already known, and a detailed description thereof is omitted.

[0017] The welding power source 4 supplies the high voltage used in welding to the welding torch and the workpieces 5a and 5b. Further, when a welding wire is used for welding, the welding power source 4 may perform control related to the feeding of the welding wire. Note that the configuration of the welding power source 4 is already known, and a detailed description thereof is omitted.

[0018] The reception unit 11 of the defect determination device 1 receives the measurement results of the laser ultrasonic method regarding the weld bead. The reception unit 11 may receive, as one measurement result, the ultrasonic waves measured by the laser light for ultrasonic wave detection irradiated at another location when, for example, the laser light for generating ultrasonic waves is irradiated at a certain location. In the present embodiment, this case will be mainly described. The reception unit 11 may receive, in real time, the measurement results regarding the bead formed by welding from the laser ultrasonic measurement device 2a during welding or after welding is completed, or may receive a plurality of measurement results previously measured by the laser ultrasonic measurement device 2a collectively. The collective reception of the measurement results may be, for example, the reading of a plurality of measurement results from a recording medium or the reception of a plurality of measurement results.

[0019] The reception unit 11 may receive a plurality of measurement results in the short-hand direction of the bead for each measurement position in the longitudinal direction of the bead. For example, regarding the measurement position B1 shown in FIG. 4, the reception unit 11 may receive eight measurement results when the laser light for generating ultrasonic waves is irradiated at the irradiation positions T1-1 to T1-8 respectively. The number of measurement results received for one measurement position is not limited. The reception of a plurality of measurement results corresponding to such measurement positions may be repeated sequentially for each measurement position. The reception unit 11 may store the received measurement results in the storage unit 12. When storing, the reception unit 11 may store the measurement results in the storage unit 12 in association with, for example, information indicating the measurement position and the irradiation position of the laser light.

[0020] The reception unit 11 may receive measurement results from, for example, the ultrasonic detection unit 2c, may receive measurement results transmitted via a wired or wireless communication line, or may receive measurement results read from a predetermined recording medium (e.g., an optical disk, a magnetic disk, a semiconductor memory, etc.). The reception unit 11 may receive information other than measurement results. For example, the reception unit 11 may receive the measurement position of the laser ultrasonic measurement device 2a from the robot control device 3. Also, the measurement results may be received in response to the defect determination device 1 outputting a measurement instruction to the laser ultrasonic measurement device 2a. The measurement instruction may be output, for example, when the measurement position received from the robot control device 3 becomes a predetermined value (e.g., B1 or B2, etc.). In response to the measurement instruction, the laser ultrasonic measurement device 2a may irradiate a plurality of irradiation points at the measurement position with laser light, acquire a plurality of measurement results corresponding to the plurality of irradiation points, and transmit them to the defect determination device 1. And the plurality of measurement results may be received by the reception unit 11. Note that the reception unit 11 may or may not include a device (e.g., a modem or a network card, etc.) for reception. Also, the reception unit 11 may be realized by hardware or may be realized by software such as a driver for driving a predetermined device.

[0021] As described above, the measurement results may be stored in the storage unit 12. Also, the prediction results by the prediction unit 13 may be stored in the storage unit 12. Also, other information may be stored in the storage unit 12. Note that the storage unit 12 is preferably realized by a non-volatile recording medium, but may be realized by a volatile recording medium. The recording medium may be, for example, a semiconductor memory or a magnetic disk, etc.

[0022] Based on the measurement results received by the reception unit 11, the prediction unit 13 predicts the measurement results of the positions adjacent to the position corresponding to the measurement results. This prediction is a prediction in the case where there are no defects. For example, using the measurement results of the positions without defects, the measurement results without defects of the positions adjacent to that position will be predicted. Therefore, the prediction results obtained by predicting the measurement results of the positions adjacent to a position based on the measurement results of the positions with defects will be different from the prediction results in the case where there are no defects. The prediction unit 13 may, for example, predict the plurality of measurement results of the measurement positions adjacent to a certain measurement position from the plurality of measurement results at that measurement position. More specifically, as shown in FIG. 5A, from the eight measurement results measured using the irradiation positions T1-1 to T1-8 and the irradiation position R1, the eight measurement results measured using the irradiation positions T2-1 to T2-8 and the irradiation position R2 may be predicted. In this case, eight prediction results will be obtained using the eight measurement results. In the present embodiment, the case where such a prediction is made will be mainly described, and other predictions will be described later. The prediction results predicted by the prediction unit 13 may be stored, for example, in the storage unit 12.

[0023] The prediction unit 13 may perform prediction using a model such as a neural network, GBDT (Gradient Boosting Decision Tree), or multiple regression model, or may perform prediction by other methods. When prediction is performed using GBDT or a multiple regression model, in addition to the measurement results, statistical data of the measurement results (for example, average, variance, etc.) may be input to the model. Also, when predicting using GBDT or a multiple regression model, the feature information obtained from the measurement results may also be used for prediction. The feature information may have feature quantities such as HOG (Histogram of Oriented Gradients) and SIFT (Scale-Invariant Feature Transform) obtained for B-scope and C-scope images obtained from a plurality of measurement results for each measurement position, for example.

[0024] The model used for prediction by the prediction unit 13 may be, for example, one learned using at least teacher data without defects. That is, the model may be learned using teacher data without defects, or may be learned using teacher data regardless of the presence or absence of defects. Usually, since the possibility of defects existing in the beads is low, even in the latter case, most of the teacher data will be data without defects, and it is considered that learning similar to the former case can be performed. When prediction is performed as described above, for example, learning may be performed using a plurality of learning data that are pairs of learning input data that are a plurality of measurement results at the measurement position BN and learning output data that are a plurality of measurement results at the measurement position B(N + 1). Note that N is an integer of 1 or more. In this case, for example, when eight measurement results at a certain measurement position are input, a model that outputs eight measurement results at the measurement position adjacent to that measurement position will be learned.

[0025] The neural network is not particularly limited, and may be, for example, an RNN (Recurrent Neural Network). Also, the RNN is not particularly limited, and for example, an LSTM (Long Short-Term Memory) network may be used. Also, the GBDT is not particularly limited, and for example, LightGBM or XGBoost may be used.

[0026] The comparison unit 14 compares the prediction result by the prediction unit 13 with the measurement result at the position corresponding to the prediction result. This comparison may be, for example, a comparison between one prediction result and one measurement result, or a comparison between a plurality of prediction results and a plurality of measurement results. In the latter case, usually, the number of prediction results is the same as the number of measurement results. In the present embodiment, the case where a comparison between a plurality of prediction results and a plurality of measurement results is performed will be mainly described. The comparison unit 14 may calculate information indicating the difference between the prediction result and the measurement result. The information indicating the difference between the two (that is, the evaluation function described later) may be, for example, MSE (Mean Squared Error), RMSE (Root Mean Squared Error), PSNR (Peak Signal-to-Noise Ratio), SSIM (Structural Similarity Index Measure), or the like. Note that MSE and RMSE become smaller values as the comparison targets are more similar, and PSNR and SSIM become larger values as the comparison targets are more similar. Thus, the comparison unit 14 may obtain a comparison result that is the value of the function by inputting, for example, one or more measurement results and one or more prediction results into an evaluation function for comparing the two. The evaluation function may be a function other than the above. For example, the evaluation function for comparing a plurality of prediction results and a plurality of measurement results may be a function including the variance or standard deviation regarding the difference between the prediction result and the measurement result corresponding to the prediction result.

[0027] The comparison unit 14 may compare, for example, a plurality of measurement results and a plurality of prediction results corresponding to the plurality of measurement results for each measurement position. When prediction is performed as shown in FIG. 5A, the comparison unit 14 may compare, for example, eight prediction results and eight measurement results for each measurement position. That is, the comparison unit 14 may calculate, for example, MSE, PSNR, etc. for eight pairs of a prediction result and a measurement result at the position corresponding to the prediction result. More specifically, when calculating MSE, PSNR, etc., a difference between a measurement result obtained by measurement using the irradiation position T2-M and a prediction result obtained by predicting the measurement result using the irradiation position T2-M may be calculated. In the cases shown in FIGS. 4 and 5A, M is an integer from 1 to 8.

[0028] When the evaluation function is MSE, the comparison unit 14 may calculate the value of MSE for a certain measurement position using, for example, the following formula. Here, dp i,j is the prediction result, and dg i,j is the measurement result. Also, i is an index indicating the irradiation position of the laser beam, and j is an index indicating the reception time. Also, i max is the number of irradiation positions of the laser beam, and j max is the maximum value of the reception time. For example, in the case shown in FIG. 4, i max = 8. Also, for each irradiation position i, j max pieces of time-direction data will be measured.

Equation

[0029] The determination unit 15 determines whether there is a defect in the bead according to the comparison result by the comparison unit 14. Note that when there is no defect in both the position where the measurement result used to obtain the prediction result was obtained and the position where the measurement result compared with the prediction result was obtained, the prediction result and the measurement result are approximate information. On the other hand, when there is a defect in at least one of the two positions, the prediction result and the measurement result are deviated information. Therefore, when it is shown by the comparison result that the measurement result and the prediction result are approximate, the determination unit 15 may determine that there is no defect, and when it is shown that the measurement result and the prediction result are deviated, the determination unit 15 may determine that there is a defect. Therefore, the determination unit 15 may determine whether there is a defect, for example, by comparing the value of the evaluation function obtained by the comparison unit 14 with a predetermined threshold value. More specifically, the determination unit 15 may determine that there is no defect when the value of the evaluation function, which is MSE or RMSE, is less than the threshold value, and may determine that there is a defect when the value of the evaluation function exceeds the threshold value. Also, the determination unit 15 may determine that there is no defect when the value of the evaluation function, which is PSNR or SSIM, exceeds the threshold value, and may determine that there is a defect when the value of the evaluation function is less than the threshold value. Note that for any evaluation function, when the value of the evaluation function is equal to the threshold value, the determination unit 15 may, for example, determine that there is a defect or may determine that there is no defect. Also, when the comparison by the comparison unit 14 is performed for each measurement position, the determination unit 15 may determine whether there is a defect in the bead for each measurement position. Note that as shown by the cavity V in FIG. 4, when the defect is local, only a part of i will result in a large value for the squared part of the above formula, and if i max is a large value, the difference in the evaluation function according to the presence or absence of the defect may not be so large. Therefore, in the evaluation function, a predetermined number of the larger ones among the difference parts between the prediction result and the measurement result may be used for calculating the evaluation function. For example, when the evaluation function is the above formula, MSE may be calculated using only a predetermined number (for example, 10) of the larger values from the squared part.

[0030] Further, when the determination unit 15 determines that there is a defect in the bead, it may specify the defect position according to the measurement position corresponding to the comparison result used for the determination. The defect position specified in this way may be, for example, at least one of the position corresponding to the measurement result of the comparison target used for the determination and the position corresponding to the measurement result used for obtaining the prediction result. For example, based on a plurality of measurement results at the measurement position B5, a plurality of prediction results at the measurement position B6 are obtained. When it is determined that there is no defect according to the comparison result between the plurality of prediction results and the plurality of measurement results at the measurement position B6, it is considered that there is no defect at both the measurement position B5 and the measurement position B6. On the other hand, in the above situation, when it is determined that there is a defect according to the comparison result between the plurality of prediction results and the plurality of measurement results at the measurement position B6, it is considered that the deviation between the measurement result and the prediction result has become large due to the presence of a defect at at least one of the measurement positions B5 and B6. Therefore, in this case, the determination unit 15 may specify the measurement positions B5 and B6 as the defect positions. Before that determination, assuming that it has been determined that there is no defect according to the comparison result between the prediction result of the measurement position B5 predicted based on the measurement result of the measurement position B4 and the measurement result of the measurement position B5, it is considered that there is no defect at the measurement position B5. Therefore, in this case, the determination unit 15 may specify the measurement position B6 as the defect position. Thus, when it is determined that there is a defect according to the comparison result using the prediction result of the second measurement position predicted using the measurement result of the first measurement position where no defect was determined, it may be specified that there is a defect at the second measurement position. Note that the first measurement position and the second measurement position are assumed to be adjacent to each other.

[0031] Further, when the determination unit 15 determines that there is a defect in the bead, it may obtain the size of the defect according to the difference between the predicted result corresponding to the comparison result used for the determination and the measurement result. For example, the determination unit 15 may obtain a larger defect size as the deviation between the predicted result and the measurement result is larger. More specifically, when the determination unit 15 determines that there is a defect in the bead, the larger the absolute value of the difference between the value of the evaluation function and the threshold value, the larger the defect size obtained, and the smaller the absolute value of the difference between the two, the smaller the defect size obtained. The size of the defect may be obtained using information such as a table that associates the absolute value of the difference between the value of the evaluation function and the threshold value with the size of the defect, or may be obtained by calculating the value of an increasing function with the absolute value of the difference between the value of the evaluation function and the threshold value as an argument.

[0032] The output unit 16 may output the determination result. When outputting the determination result indicating the presence of a defect, the output unit 16 may output the size of the defect and the defect position together with the determination result. When real-time determination is repeatedly performed, for example, the output unit 16 may repeatedly output the determination result. This output may be, for example, a display on a display device (such as a liquid crystal display or an organic EL display), a transmission via a communication line to a predetermined device, a print by a printer, an audio output by a speaker, a storage in a recording medium, or a transfer to another component. Note that the output unit 16 may or may not include a device for output (such as a display device or a communication device). Also, the output unit 16 may be realized by hardware or by software such as a driver for driving those devices.

[0033] Next, the operation of the defect determination device 1 will be described using the flowchart of FIG. 3A. (Step S101) The reception unit 11 determines whether it has received the measurement result from the laser ultrasonic measurement device 2a. If it has received the result, it proceeds to step S102; otherwise, it repeats the process of step S101 until it receives the measurement result. Note that the received measurement result may be stored in the storage unit 12.

[0034] (Step S102) The prediction unit 13 determines whether the exploration of one column has been completed. If the exploration of one column has been completed, it proceeds to step S103; otherwise, it returns to step S101. Note that the prediction unit 13 may determine that the exploration of one column has been completed when the measurement results for all irradiation positions have been received for a certain measurement position. For example, when measurement is performed as shown in FIG. 4, the image conversion unit 12 may determine that the exploration of one column has been completed when eight measurement results have been received for a certain measurement position (for example, measurement position B1, etc.).

[0035] (Step S103) The prediction unit 13 obtains a prediction result by predicting the measurement results of the measurement positions adjacent to the measurement position corresponding to the received measurement results for the exploration of one column using the received measurement results. Note that the prediction result may be stored in the storage unit 12.

[0036] (Step S104) The comparison unit 14 determines whether to compare the measurement result and the prediction result. If it is to perform the comparison, it proceeds to step S105; otherwise, it returns to step S101. Note that the comparison unit 14 may determine to perform the comparison when a prediction result already exists for the latest measurement position corresponding to the measurement results for the exploration of one column, for example.

[0037] (Step S105) The comparison unit 14 compares the measurement result corresponding to a certain measurement position with the prediction result. For example, the prediction result and the measurement result of the latest measurement position may be compared.

[0038] (Step S106) The determination unit 15 determines whether there is a defect in the bead using the comparison result of Step S105. More specifically, when it is determined that the deviation between the measurement result and the prediction result is large based on the comparison result, it is determined that there is a defect in the bead, and if not, it may be determined that there is no defect. Also, when there is a defect, for example, the size and position of the defect may be specified.

[0039] (Step S107) The output unit 16 outputs the determination result by the determination unit 15. Note that the size and position of the defect may also be output together with the determination result.

[0040] (Step S108) The prediction unit 13 determines whether to end the prediction process. When the prediction process is to be ended, the series of processes for determining welding defects ends, and if not, it returns to Step S101. Note that when the prediction is completed up to the end of the bead, it may be determined to end the prediction process.

[0041] Note that in the flowchart of FIG. 3A, when it is determined that there is a defect, a series of processes such as accepting the measurement result and prediction may be ended. Also, when the laser ultrasonic measurement device 2a performs measurement in response to a measurement instruction, the flowchart of FIG. 3A may include a process of outputting the measurement instruction to the laser ultrasonic measurement device 2a. This measurement instruction may be output, for example, by an instruction unit (not shown) of the defect determination device 1. In this case, the reception unit 11 may accept the measurement result in response to the output of the measurement instruction. Also, in the flowchart of FIG. 3A, the output of the determination result may be performed only once when ending the series of processes. Also, the order of the processes in the flowchart of FIG. 3A is an example, and the order of each step may be changed if the same result can be obtained.

[0042] Next, the operation of the defect determination device 1 according to the present embodiment will be described using a specific example. In this specific example, a case will be described in which, while welding is performed by the manipulator 2, the defect determination device 1 sequentially determines the presence or absence of a defect using the measurement results obtained by the laser ultrasonic measurement device 2a. Further, in this specific example, it is assumed that the evaluation function used for comparison is the MSE.

[0043] First, the reception unit 11 sequentially receives the measurement results for each of the irradiation positions T1-1 to T1-8 corresponding to the measurement position B1 (steps S101, S102). Then, when the number of received measurement results reaches 8, which corresponds to one column of exploration, the prediction unit 13 uses the 8 measurement results of the measurement position B1 stored in the storage unit 12 to obtain 8 prediction results of the measurement position B2 and stores them in the storage unit 12 (step S103). At this point, since there is no prediction result for the measurement position B1, it is determined that no comparison is performed (step S104).

[0044] Next, the reception of each measurement result of the measurement position B2 and the acquisition of the prediction result of the measurement position B3 using those measurement results are performed (steps S101 to S103). At this point, since the prediction result of the measurement position B2 is stored in the storage unit 12, the comparison unit 14 determines to compare the prediction result and the measurement result of the measurement position B2, and calculates the value of the MSE (steps S104, S105). In this case, it is assumed that the value of the MSE is smaller than a predetermined threshold. Then, the determination unit 15 determines that there is no defect in the bead, and the determination result is output (steps S106, S107).

[0045] Thereafter, the measurement results at each measurement position B3 are received, and prediction results using these measurement results are obtained (Steps S101 to S103). At this point, since the prediction result of measurement position B3 is stored in the storage unit 12, the comparison unit 14 determines to compare the prediction result and the measurement result of measurement position B3, and calculates the value of MSE (Steps S104, S105). In this case, as shown in FIG. 4, since there is a cavity V at measurement position B3, the deviation between the prediction result corresponding to measurement position B3 and the measurement result is large, and it is assumed that the value of MSE exceeds the threshold. Then, the determination unit 15 determines that there is a defect in the bead (Step S106). And the output unit 16 outputs the determination result (Step S107). For example, welding may be interrupted according to the output. Note that, in this specific example, the case where the defect is a cavity has been described, but the defect may be other defects such as bead undercut other than the cavity.

[0046] As described above, according to the defect determination device 1 according to the present embodiment, even if a sufficient amount of teacher data cannot be prepared for the welding defect, by predicting the measurement result of the position adjacent to a certain position from the measurement result of that position, it becomes possible to determine the presence or absence of the welding defect. Further, when the determination unit 15 determines that a defect exists and obtains the size of the defect according to the comparison result, it also becomes possible to know the size of the defect. Furthermore, by specifying the defect position, which is the position of the bead corresponding to the comparison result used in the defect determination, it also becomes possible to know the position of the defect. Also, by performing processes such as prediction and comparison according to the measurement results received in real time during welding, it becomes possible to know the presence or absence of a defect during welding. For example, when a defect exists, by interrupting the welding, it is also possible to prevent unnecessary work from being performed thereafter.

[0047] In addition, in this embodiment, although the case where a plurality of prediction results are obtained by predicting a plurality of measurement results at a measurement position adjacent to a certain measurement position using a plurality of measurement results at one measurement position has been mainly described, it may not be the case. For example, a plurality of prediction results may be obtained by predicting a plurality of measurement results at a measurement position adjacent to two or more measurement positions using a plurality of measurement results at the two or more measurement positions. In this case, the two or more measurement positions may be, for example, consecutive measurement positions. Also, in this case, in order to create a model used for prediction, learning may be performed with a plurality of measurement results at two or more measurement positions as inputs and a plurality of measurement results at a measurement position adjacent to the two or more measurement positions as outputs. In the case shown in FIG. 4, for example, eight prediction results obtained by predicting eight measurement results at measurement position B3 using 16 measurement results at two measurement positions B1 and B2 may be obtained. Also, the two or more measurement positions corresponding to the measurement results used for prediction may not be consecutive, for example. In this case, for example, eight prediction results obtained by predicting eight measurement results at measurement position B2 using 16 measurement results at two measurement positions B1 and B3 may be obtained. As is clear from these explanations, the measurement position adjacent to two or more measurement positions may be, for example, a measurement position (e.g., B3) adjacent to two or more consecutive measurement positions (e.g., B1, B2), or may be a measurement position (e.g., B2) adjacent to each of two or more non-consecutive measurement positions (e.g., B1, B3). The same shall apply when prediction is performed as shown in FIG. 5B. Thus, the prediction unit 13 may predict a plurality of measurement results at a measurement position adjacent to the one or more measurement positions based on a plurality of measurement results at the one or more measurement positions. Also, the prediction unit 13 may perform the prediction for each measurement position.

[0048] In addition, in the present embodiment, a case where prediction is performed according to the direction in which welding is performed, that is, a case where the measurement result at the measurement position B2 is predicted using the measurement result at the measurement position B1 has been described, but this is not necessary. For example, the measurement result at the measurement position B1 may be predicted using the measurement result at the measurement position B2. Thus, the direction in which prediction is performed does not matter. When prediction is performed in the direction opposite to welding, a model corresponding thereto may be used. For example, a model generated by performing learning using a plurality of learning data that is a set of learning input data that is a plurality of measurement results at the measurement position B(N + 1) and learning output data that is a plurality of measurement results at the measurement position BN may be used for prediction in the direction opposite to welding.

[0049] In addition, in this embodiment, although the case where a prediction result is obtained by predicting the measurement result at a measurement position different from the measurement position using the measurement result at a certain measurement position has been described, it is not necessary to be so. For example, at a certain measurement position, a prediction result obtained by predicting the measurement result at an irradiation position adjacent to the irradiation position from the measurement result corresponding to the irradiation position may be obtained. In this case, for example, as shown in FIG. 5B, a prediction result obtained by predicting the measurement result using the irradiation position T1-2 and the irradiation position R2 from one measurement result measured using the irradiation position T1-1 and the irradiation position R2 may be obtained. Similarly, the same prediction may be made based on each measurement result using the irradiation positions T1-2 to T1-7. By doing so, for each of the irradiation positions T1-2 to T1-8, a pair of a measurement result and a prediction result can be obtained. Therefore, similar to the case of FIG. 5A, a comparison result can be obtained, and the presence or absence of a defect can be determined using the comparison result. In this way, the prediction unit 13 may perform, for each of a plurality of irradiation positions at the measurement position, predicting the measurement result at an irradiation position adjacent to the irradiation position corresponding to one or more measurement results from one or more measurement results at the measurement position. In this case, for example, a determination process may be performed as shown in the flowchart of FIG. 3B. In the flowchart of FIG. 3B, it is assumed that the processes other than steps S201 to S204 are the same as those in the flowchart of FIG. 3A. In the case shown in FIG. 4, in step S201, it may be determined to perform a prediction except when the measurement result using the irradiation position TN-8 is received. Note that N is an integer of 1 or more. In step S202, the measurement result using the irradiation position TN-(M+1) may be predicted from the measurement result using the irradiation position TN-M. Note that M is an arbitrary integer from 1 to 8. In step S203, it is determined whether the search for one column is completed. If the search for one column is completed, the process proceeds to step S204. Otherwise, the process returns to step S101. In step S204, a comparison is made between the seven measurement results using the irradiation positions TN-2 to TN-8 and the seven prediction results respectively corresponding to the seven measurement results.

[0050] In this case, for example, when obtaining a prediction result obtained by predicting the measurement result measured using the irradiation position TN-2 from the measurement result measured using the irradiation position TN-1, and the irradiation position TN-3 from the measurement result measured using the irradiation position TN-2 The model used when obtaining the prediction result obtained by predicting the measurement result to be measured may be the same or different, for example. Although not particularly limited, for example, in the case of a neural network model, it may be the former, and in the case of a model such as GBDT or a multiple regression model, it may be the latter. In the latter case, for example, in FIG. 5B, 7 models may be used to perform predictions for each irradiation position.

[0051] In addition, even when the prediction is performed as shown in FIG. 5B, a prediction result obtained by predicting the measurement result using the irradiation position adjacent to the irradiation position from the measurement result using one irradiation position may be obtained, or two or more irradiation positions may be obtained. A prediction result obtained by predicting the measurement result using the irradiation position adjacent to the two or more irradiation positions from the measurement results using the irradiation positions may be obtained. Also in this case, the two or more irradiation positions may or may not be consecutive irradiation positions. Also, the direction in which the prediction is performed may be the reverse of that in FIG. 5B. For example, a prediction result obtained by predicting the measurement result measured using the irradiation position T1-7 from the measurement result measured using the irradiation position T1-8 may be obtained.

[0052] Also, in this embodiment, although the case where a plurality of measurement results at one measurement position are compared with a plurality of prediction results has been described, it is not necessary. The objects to be compared may cross a plurality of measurement positions, or one measurement result and one prediction result may be compared. In the former case, for example, a plurality of measurement results measured using irradiation positions T1-2 to T20-2 and a plurality of prediction results corresponding to the plurality of measurement results may be compared, and a determination regarding the presence or absence of a defect may be made. Also, when one measurement result and one prediction result are compared, for example, one prediction result may be obtained from one measurement result, or one prediction result may be obtained from a plurality of measurement results, or a plurality of prediction results may be obtained from a plurality of measurement results. When one prediction result is obtained from one measurement result, for example, prediction may be performed as shown in FIG. 5B, or one prediction result obtained by predicting one measurement result at a measurement position adjacent to the measurement position from one measurement result at a certain measurement position may be obtained. In the latter case, for example, one prediction result obtained by predicting the measurement result measured using irradiation position T2-1 from the measurement result measured using irradiation position T1-1 may be obtained.

[0053] As described above, there are various arbitrary aspects regarding the method of predicting a prediction result from a measurement result, the method of comparison, etc. In prediction, it is preferable that one or more measurement results adjacent to the position corresponding to the one or more measurement results are predicted based on the one or more measurement results. That position may be, for example, a measurement position, an irradiation position of laser light, or another position. Also, a position adjacent to a certain position may be, for example, a measurement position or an irradiation position adjacent to that position. Also, in comparison, it is preferable that one or more measurement results and one or more prediction results corresponding to the one or more measurement results are compared. The prediction result corresponding to a certain measurement result may be the prediction result of the position corresponding to the measurement result. For example, the prediction result corresponding to the measurement result measured using irradiation position T1-2 may be the prediction result obtained by predicting the measurement result measured using irradiation position T1-2 from the measurement result measured using irradiation position T1-1.

[0054] Also, when welding and measurement are performed in parallel, in FIG. 4, for example, while the laser beam is sequentially irradiated to a plurality of irradiation positions at a certain measurement position, the irradiation position of the laser beam may shift little by little in the x-axis direction. In this case, for example, a plurality of irradiation positions T1-1 to T1-8 of the laser beam are arranged in a direction not perpendicular to the x-axis, and the irradiation position R1 can also be a plurality of points arranged in the x-axis direction. Thus, a plurality of measurement results in a direction not perpendicular to the longitudinal direction of the bead 6 may be obtained for each measurement position, and prediction, comparison, and determination processes may be performed. In this case, the measurement position may be, for example, a measurement position having a predetermined width (for example, a measurement position where the value of the x-axis is from B1 S to B1 E and so on).

[0055] Also, in this embodiment, the case where welding is mainly described when performed by the manipulator 2 having a welding torch has been mainly described, but the manipulator 2 is for moving the laser ultrasonic measurement device 2a and may not perform welding. In this case, the manipulator 2 may not be provided with a welding torch, and the robot control system 100 may not be provided with the welding power source 4.

[0056] In addition, in the above-described embodiment, each component may be configured by dedicated hardware, or for components that can be realized by software, they may also be realized by executing a program. For example, each component can be realized by a program execution unit such as a CPU reading and executing a software program recorded on a recording medium such as a hard disk or a semiconductor memory. At the time of its execution, the program execution unit may execute the program while accessing a storage unit or a recording medium. Also, the program may be executed by being downloaded from a server or the like, or may be executed by reading a program recorded on a predetermined recording medium. Further, the computer that executes the program may be singular or plural. That is, centralized processing may be performed, or distributed processing may be performed.

[0057] Furthermore, the present invention is not limited to the above-described embodiments, and various modifications are possible, and it goes without saying that those are also included within the scope of the present invention.

Explanation of Reference Numerals

[0058] 1 Defect determination device, 11 Reception unit, 12 Image conversion unit, 12 Storage unit, 13 Prediction unit, 14 Comparison unit, 15 Determination unit, 16 Output unit

Claims

1. A receiving unit that receives measurement results of the laser ultrasonic method regarding the bead of welding; A prediction unit that predicts measurement results of positions adjacent to the position corresponding to the measurement results based on the measurement results received by the receiving unit; A comparison unit that compares the prediction results by the prediction unit with the measurement results of the positions corresponding to the prediction results; A defect determination device including a determination unit that determines whether there is a defect in the bead according to the comparison result by the comparison unit.

2. The receiving unit receives a plurality of measurement results in the short direction of the bead for each measurement position in the longitudinal direction of the bead, The prediction unit predicts, for each measurement position, a plurality of measurement results of measurement positions adjacent to the one or more measurement positions based on the plurality of measurement results of the one or more measurement positions, The comparison unit compares, for each measurement position, a plurality of measurement results with a plurality of prediction results corresponding to the plurality of measurement results, The determination unit determines, for each measurement position, whether there is a defect in the bead. The defect determination device according to Claim 1.

3. The receiving unit receives a plurality of measurement results in the short direction of the bead for each measurement position in the longitudinal direction of the bead, The prediction unit predicts, for each of the plurality of irradiation positions at a certain measurement position, measurement results of irradiation positions adjacent to the irradiation position corresponding to the one or more measurement results from the one or more measurement results, The comparison unit compares, for each measurement position, a plurality of measurement results with a plurality of prediction results, The determination unit determines, for each measurement position, whether there is a defect in the bead. The defect determination device according to Claim 1.

4. When the determination unit determines that there is a defect in the bead, the defect determination device according to Claim 2 or Claim 3 specifies a defect position according to the measurement position corresponding to the comparison result used for the determination.

5. The defect determination device according to any one of claims 1 to 4, wherein when the determination unit determines that a defect exists in the bead, the determination unit obtains the size of the defect according to the difference between the predicted result corresponding to the comparison result used for the determination and the measurement result.

Images