Anomaly detection device

Abstract

To provide an abnormality detection device which can reduce the labor of an inspection work.SOLUTION: An abnormality detection device comprises: an acquisition unit 11 which acquires welding measurement data measured for each object when performing arc welding and acquires appearance data obtained for each object by an appearance inspection of the welded object; an abnormality detection unit 12 which detects an abnormality in the object on the basis of the acquired welding measurement data and the appearance data; and an information generation unit 13 which generates abnormality information of the object on the basis of the detected abnormality.SELECTED DRAWING: Figure 2

Description

[Technical Field]

[0001] The present invention relates to an abnormality detection device. [Background technology]

[0002] Patent Document 1 below discloses an inspection device for inspecting for welding defects in arc welding. This inspection device stores welding measurement data, such as welding current measured during arc welding, and appearance data, such as bead images captured during arc welding, in association with each other. With this inspection device, when an inspector displays one of the data and checks the welding condition of an object, the inspector can specify a portion of the data and display the other data corresponding to the specified portion alongside the one of the data. As a result, if a suspected welding defect is found in one of the data, the other data corresponding to the specified portion can be displayed alongside the other data, allowing the inspector to check the welding condition while comparing the two data. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Japanese Patent Publication No. 2020-121345 Summary of the Invention [Problem to be solved by the invention]

[0004] However, with the inspection device of Patent Document 1, an inspector must visually check the data of each object and compare both sets of data for each suspected welding point to determine whether the welding is good or bad. This means that inspecting the welding condition requires a great deal of effort.

[0005] SUMMARY OF THE INVENTION It is therefore an object of the present invention to provide an abnormality detection device that can reduce the labor required for inspection work. [Means for solving the problem]

[0006] An abnormality detection device according to one embodiment of the present invention includes a first acquisition unit that acquires welding measurement data measured for each object during arc welding, a second acquisition unit that acquires appearance data obtained for each object by visual inspection of the welded object, an abnormality detection unit that detects an abnormality in the object based on the acquired welding measurement data and appearance data, and an information generation unit that generates abnormality information for the object based on the detected abnormality.

[0007] According to this aspect, it is possible to detect abnormalities in each object based on welding measurement data measured during arc welding and appearance data obtained by visual inspection of the welded object, and to generate abnormality information related to the detected abnormalities, thereby enabling workers to perform repair work based on the generated abnormality information.

[0008] In the above aspect, the first acquisition unit may further acquire first position information corresponding to the welding measurement data, the second acquisition unit may further acquire second position information corresponding to the appearance data, and the information generation unit may identify the abnormality location where the abnormality was detected based on the first position information and the second position information corresponding to the detected abnormality, and generate the abnormality information by including information on the identified abnormality location.

[0009] According to this aspect, the worker can make repairs to the abnormal part included in the abnormality information, thereby improving the accuracy of the repair work.

[0010] In the above aspect, the abnormality information may include marking information for marking the target object, and the information generating unit may identify an occurrence pattern of the abnormality based on the abnormality detected based on the welding measurement data and the abnormality detected based on the appearance data, and generate marking information corresponding to the identified occurrence pattern of the abnormality.

[0011] According to this aspect, workers can recognize the pattern in which an abnormality occurs based on the marking information contained in the abnormality information, and can carry out repairs by performing work that corresponds to the recognized pattern, thereby improving work efficiency.

[0012] In the above aspect, the information generating unit may generate the anomaly information including information on a repair process for the object corresponding to the detected anomaly.

[0013] According to this aspect, the worker can understand the details of the repair work based on the information about the repair process and can carry out the repair based on the understood details, thereby making it possible to improve the efficiency of the repair work.

[0014] In the above aspect, the abnormality detection unit may detect an abnormality in the object by comparing the acquired welding measurement data with a threshold value for detecting an abnormality, and may detect an abnormality in the object by comparing the acquired appearance data with appearance data in a normal state.

[0015] According to this aspect, since an abnormality can be detected using a threshold value for detecting an abnormality or appearance data in a normal state, it is possible to improve the accuracy of detecting an abnormality. [Effects of the Invention]

[0016] According to the present invention, it is possible to provide an abnormality detection device that can reduce the labor required for inspection work. [Brief explanation of the drawings]

[0017] [Figure 1] 1 is a schematic diagram illustrating a schematic configuration of an abnormality detection system including an abnormality detection device according to an embodiment. [Figure 2] 2 is a block diagram illustrating the configuration of the abnormality detection device shown in FIG. 1. FIG. [Figure 3] 2 is a flowchart illustrating an example of the operation of the abnormality detection device shown in FIG. DETAILED DESCRIPTION OF THE INVENTION

[0018] A preferred embodiment of the present invention will be described with reference to the accompanying drawings. In each drawing, components with the same reference numerals have the same or similar configurations. Furthermore, since the drawings are schematic, the dimensions and proportions of each component may differ from those of the actual components.

[0019] [Anomaly Detection System 100] 1 is a schematic diagram illustrating the overall configuration of an anomaly detection system including an anomaly detection device according to an embodiment. As shown in the figure, the anomaly detection system 100 includes, for example, an anomaly detection device 1, a welding power source 21, a control device 22, and a welding robot 23 used in a welding process 2, a control device 31, a visual inspection robot 32, and a marking robot 33 used in a visual inspection process 3, and a display device 41 used in a repair process 4.

[0020] [Welding process 2] The welding process 2 is a process in which a welding robot 23 performs arc welding on a workpiece (object) W. In this welding process 2, for example, a welding power source 21, a control device 22, and a welding robot 23 are used.

[0021] Welding power source 21 controls the welding current, welding voltage, wire feed speed, etc. in accordance with the welding robot conditions set in control device 22, and generates an arc between the tip of the welding wire and the workpiece.

[0022] The welding conditions for the welding robot include, for example, the welding conditions, welding start position, welding end position, welding distance, and welding torch posture. The welding conditions include, for example, the welding current, welding voltage, welding speed, wire feed speed, and workpiece thickness.

[0023] The control device 22 is configured by a control unit including, for example, a processor, a memory, and a communication interface. The processor executes a predetermined program stored in the memory to control the welding power source 21 and the welding robot 23.

[0024] Welding robot 23 is an industrial robot that performs arc welding in accordance with welding robot working conditions set in control device 22. Specifically, welding robot 23 has a multi-joint arm provided on a base member that is fixed to the floor or the like of a factory, a welding torch connected to the tip of the multi-joint arm, a wire feeder fixed to the multi-joint arm, and a workbench on which workpiece W is placed.

[0025] In welding process 2, for example, when an operator instructs execution of a work program, control device 22 sequentially executes the commands of the work program corresponding to the instruction. For example, if the command instructed to be executed is a movement command, control device 22 controls the servo motor of the articulated arm of welding robot 23. If the command instructed to be executed is a welding start command, control device 22 transmits commands for the welding current and welding voltage to welding power source 21.

[0026] Welding power source 21 receives welding current and welding voltage commands from control device 22 and outputs the current and voltage corresponding to the commands to welding robot 23. Welding power source 21 transmits welding measurement data acquired during arc welding and coordinate data corresponding to the welding measurement data to abnormality detection device 1. The welding measurement data is welding-related data measured for each workpiece W during arc welding, and includes, for example, a workpiece identification number, welding current, welding voltage, welding speed, and wire feed speed. The workpiece identification number is identification information for identifying the workpiece W. For example, the workpiece W moving through the production line may be recorded by a counter and a number corresponding to the counter may be assigned as the workpiece identification number, or a barcode or two-dimensional code may be attached to the workpiece W and a number read from the code may be assigned.

[0027] [Appearance inspection process 3] The appearance inspection process 3 includes a process in which an appearance inspection robot 32 inspects the appearance of the workpiece W welded in the welding process 2, and a process in which a marking robot 33 marks the workpiece W after the appearance inspection. In this appearance inspection process 3, for example, a control device 31, an appearance inspection robot 32, and a marking robot 33 are used.

[0028] The control device 31 is configured by a control unit including, for example, a processor, a memory, and a communication interface. The processor executes a predetermined program stored in the memory, thereby controlling the appearance inspection robot 32 and the marking robot 33.

[0029] The visual inspection robot 32 is an industrial robot that performs visual inspection in accordance with the working conditions for the visual inspection robot set in the control device 31. Specifically, the visual inspection robot 32 has a multi-joint arm provided on a base member that is fixed to the floor or the like of a factory, a laser sensor or the like attached to the tip of the multi-joint arm, and a workbench on which the workpiece W is placed.

[0030] The marking robot 33 is an industrial robot that performs marking in accordance with the marking robot application conditions set in the control device 31. Specifically, the marking robot 33 has an articulated arm mounted on a base member that is fixed to the factory floor or the like, a marking tool attached to the tip of the articulated arm, and a workbench on which the workpiece W is placed. The marking tool includes, for example, a pen, chalk, or spray.

[0031] In the appearance inspection process 3, for example, when an operator instructs the execution of a work program, the control device 31 sequentially executes the instructions of the work program corresponding to the instruction. For example, if the instruction instructed to be executed is a movement instruction, the control device 31 controls the servo motors of the articulated arms of the appearance inspection robot 32 and the marking robot 33. If the instruction instructed to be executed is an inspection start instruction for the appearance inspection robot 32, the control device 31 transmits a command to start appearance inspection to the appearance inspection robot 32. If the instruction instructed to be executed is a marking start instruction for the marking robot 33, the control device 31 transmits a marking start command to the marking robot 33.

[0032] Upon receiving a command from the control device 22 to start an appearance inspection, the appearance inspection robot 32 uses a laser sensor to acquire appearance data of the weld bead on the workpiece W. The appearance data is generated for each workpiece W and includes a workpiece identification number. The appearance inspection robot 32 transmits the acquired appearance data and coordinate data corresponding to the appearance data to the abnormality detection device 1. Upon receiving a command from the control device 22 to start marking, the marking robot 33 starts marking the workpiece W using a marking tool.

[0033] Here, it is preferable to adjust the welding robot 23 and the visual inspection robot 32 so that their positions in their respective coordinate systems correspond to each other. In this case, for example, calibration may be performed so that the positional relationship between the end of the articulated arm of each of the welding robot 23 and the visual inspection robot 32 and the workpiece coincides. After aligning the two coordinate systems, for example, the same teaching program as that of the welding robot 23 may be imparted to the visual inspection robot 32. In this case, it becomes possible to identify an abnormality discovered by the visual inspection robot 32 as, for example, "a location that has moved XX distance from the start position of the weld bead."

[0034] Furthermore, information on the abnormality location may be transmitted between the welding robot 23 and the visual inspection robot 32 using the coordinate system of the workpiece W. In this case, for example, it is possible to transmit information such as "where the abnormality occurred as viewed from the reference point of the workpiece W" between the welding robot 23 and the visual inspection robot 32.

[0035] [Retouching process 4] The repair process 4 is a process in which a worker repairs the workpiece W based on the marks made in the marking process of the visual inspection process 3. In this repair process 4, for example, a display device 41 is used.

[0036] The display device 41 is a display device that displays text, images, etc. The display device 41 displays abnormality detection information. The abnormality detection information is generated for each workpiece identification number. The abnormality detection information includes, for example, information indicating an abnormality location where an abnormality has been detected for each welding measurement data and appearance data. The abnormality location can be represented by coordinate data corresponding to the abnormality location. For example, the position of the abnormality location can be represented by one piece of coordinate data, or the range of the abnormality location can be represented by a set of coordinate data corresponding to the start point of the abnormality and coordinate data corresponding to the end point of the abnormality.

[0037] In the repair process 4, the worker checks the abnormality detection information displayed on the display device 41 and repairs the marked portion of the workpiece.

[0038] [Anomaly detection device 1] The configuration of the abnormality detection device 1 will be described with reference to Fig. 2. The abnormality detection device 1 has, as functional components, a control unit 10, a storage unit 14, and a communication unit 15, for example.

[0039] The storage unit 14 stores various programs and various information required for executing processes in the abnormality detection device 1. The various information includes, for example, workpiece inspection information. This workpiece inspection information includes, for example, welding measurement data, appearance data, and abnormality detection information.

[0040] The communication unit 15 controls, for example, communication between the control device 22, the welding power source 21, the control device 31, and the display device 41. The communication unit 15 includes, for example, a function as a display unit that causes the display device 41 to display workpiece inspection information corresponding to the workpiece W in which an abnormality has been detected.

[0041] [Control unit 10] The control unit 10 realizes, for example, each function of an acquisition unit 11, an abnormality detection unit 12, and an information generation unit 13 by causing the processor to execute a predetermined program stored in the memory. Each function of the control unit 10 will be described below.

[0042] [Acquisition part 11] Acquiring unit 11 acquires welding measurement data measured for each workpiece W during arc welding and coordinate data corresponding to the welding measurement data as first coordinate data. Specifically, upon receiving a data acquisition start command from control unit 10, acquiring unit 11 stores the acquired welding measurement data and first coordinate data in a data buffer of memory unit 14. Upon receiving a data acquisition end command from control unit 10, acquiring unit 11 stops storing the data in the data buffer.

[0043] The acquisition unit 11 acquires appearance data obtained for each workpiece W by visual inspection of the welded workpiece W, and coordinate data corresponding to the appearance data, as second coordinate data. Specifically, when the acquisition unit 11 receives a data acquisition start command from the control unit 10, it stores the acquired appearance data and second coordinate data in a data buffer of the storage unit 14. When the acquisition unit 11 receives a data acquisition end command from the control unit 10, it stops storing the data in the data buffer.

[0044] The first coordinate data or the second coordinate data may be acquired when an abnormality is detected by the abnormality detection unit 12. In this case, the coordinate data corresponding to the welding measurement data in which the abnormality is detected is acquired as the first coordinate data, and the coordinate data corresponding to the appearance data in which the abnormality is detected is acquired as the second coordinate data.

[0045] [Anomaly detection unit 12] The abnormality detection unit 12 detects an abnormality in the workpiece W based on the welding measurement data and appearance data acquired by the acquisition unit 11.

[0046] Specifically, when the abnormality detection unit 12 receives an instruction to start the abnormality detection process from the control unit 10, it detects an abnormality in the workpiece W by comparing the welding measurement data stored in the data buffer with a threshold value for abnormality detection, and also detects an abnormality in the workpiece W by comparing the appearance data stored in the data buffer with the appearance data under normal conditions.

[0047] [Information generation unit 13] The information generating unit 13 generates abnormality information for the workpiece W based on the abnormality detected by the abnormality detecting unit 12.

[0048] The abnormality information includes, for example, the above-mentioned abnormality detection information, marking information, and rework process information. The marking information is information for marking the workpiece W in which an abnormality has been detected. The rework process information is information regarding rework to be performed on the workpiece W in which an abnormality has been detected. The rework process information may include, for example, a message indicating that rework is required for a welding defect.

[0049] The information generating unit 13 identifies an abnormality location where the abnormality was detected based on the first coordinate data and second coordinate data corresponding to the detected abnormality, and generates abnormality detection information based on information relating to the identified abnormality location.

[0050] Information generation unit 13 identifies an abnormality occurrence pattern for each workpiece W based on an abnormality detected based on the welding measurement data and an abnormality detected based on the appearance data, and generates marking information based on the identified abnormality occurrence pattern. The abnormality occurrence pattern may include, for example, (1) a first pattern in which an abnormality is detected in both the welding measurement data and the appearance data, (2) a second pattern in which an abnormality is detected only in the welding measurement data, and (3) a third pattern in which an abnormality is detected only in the appearance data. The marking information generated for each abnormality occurrence pattern will be described below.

[0051] (1) First pattern (anomalies are detected in both welding measurement data and appearance data) It is assumed that there is a high possibility that poor welding has occurred at the abnormal location corresponding to the first pattern. In this case, the information generating unit 13 generates marking information for marking from the start point of the abnormality to the end point of the abnormality using marks for the first pattern. The start point and end point of the abnormality can be identified based on coordinates corresponding to the abnormal location in the welding measurement data or the appearance data. The marks can be arbitrarily set using different colors, patterns, etc. for each pattern so that each pattern can be identified.

[0052] Here, when an abnormality is detected in both the welding measurement data and the appearance data, this includes not only the case where the abnormal locations in both data coincide, but also the case where the abnormal locations partially overlap. Considering the case where the abnormal locations partially overlap, it is preferable to mark the abnormal locations with marks so that they can be distinguished from each other. For example, different marks can be set for the abnormal locations in the welding measurement data and the abnormal locations in the appearance data, and the marks can be placed side by side.

[0053] (2) Second pattern (pattern in which anomalies are detected only from welding measurement data) At the abnormal location corresponding to the second pattern, the weld bead does not appear to have an abnormality in appearance, but it is assumed that there is a high possibility that a defect such as poor penetration has occurred inside the weld bead. In this case, information generator 13 generates marking information for marking from the start point of the abnormality to the end point of the abnormality using markers for the second pattern. The start point and end point of the abnormality can be identified based on the coordinates corresponding to the abnormal location in the welding measurement data.

[0054] (3) Third pattern (abnormalities detected only by appearance data) It is assumed that the abnormality location corresponding to the third pattern is likely to be a welding defect that does not show any change in the welding measurement data, such as a deviation in the target position or a defect (e.g., a pit or pinhole) that occurs after the arc has passed. In this case, the information generating unit 13 generates marking information for marking from the start point of the abnormality to the end point of the abnormality using marks for the third pattern. The start point and end point of the abnormality can be identified based on the coordinates corresponding to the abnormality location in the appearance data.

[0055] The information generating unit 13 generates repair process information to be displayed on the display device 41. For example, in the case of the first and third patterns, the information generating unit 13 generates the repair process information by including a message to alert the worker in the repair process that repair of the defective weld is necessary. In addition, in the case of the second pattern, the information generating unit 13 generates the repair process information by including a message to alert the worker in the repair process to check the condition of the back side of the workpiece W in particular. In this case, if the worker determines that there is poor penetration, for example, repair work will be performed.

[0056] [Operation of abnormality detection device 1] Next, the operation of the abnormality detection device 1 according to the embodiment will be described with reference to FIG.

[0057] First, the acquisition unit 11 of the abnormality detection device 1 acquires welding measurement data measured for each workpiece W while arc welding is being performed (step S101).

[0058] Next, the acquisition unit 11 of the abnormality detection device 1 acquires appearance data obtained for each workpiece W by visual inspection of the welded workpiece W (step S102).

[0059] Next, the abnormality detection unit 12 of the abnormality detection device 1 determines whether or not there is an abnormality based on the welding measurement data acquired in the above step S101 and the appearance data acquired in the above step S102 (step S103). If this determination is NO (step S103; NO), the abnormality detection device 1 ends this operation.

[0060] If it is determined in the above step S103 that an abnormality exists (step S103; YES), the information generation unit 13 of the abnormality detection device 1 identifies the abnormality occurrence pattern for each workpiece W based on the abnormality detected based on the welding measurement data and the abnormality detected based on the appearance data (step S104).

[0061] Subsequently, the information generating unit 13 of the abnormality detection device 1 generates marking information based on the abnormality occurrence pattern identified in step S104 (step S105).

[0062] Subsequently, the information generating unit 13 of the abnormality detection device 1 generates repair process information to be displayed on the display device 41 (step S106). Then, the abnormality detection device 1 ends this operation.

[0063] As described above, the abnormality detection device 1 according to the embodiment can acquire welding measurement data and appearance data, detect abnormalities in each workpiece W based on the acquired welding measurement data and appearance data, and generate abnormality information related to the detected abnormalities. This enables workers to perform repair work based on the abnormality information generated by the abnormality detection device 1.

[0064] For example, workers can now make repairs to the abnormal parts included in the anomaly information, thereby improving the accuracy of repair work.

[0065] Furthermore, workers can recognize the pattern of abnormality occurrence based on the marking information included in the abnormality information, and can carry out the work corresponding to the recognized pattern to make corrections. For example, workers only need to check the condition of the workpiece W and inspect it for defects in the case of the second pattern described above. This minimizes the inspection burden on workers and improves work efficiency.

[0066] In this way, the abnormality detection device 1 according to the embodiment can reduce the labor required for inspection work.

[0067] Furthermore, in the anomaly detection system 100 including the anomaly detection device 1 according to the embodiment, the welding robot 23, the visual inspection robot 32, and the marking robot 33 are provided separately and independently, making it possible to flexibly apply the system to situations in which the welding process 2 and the visual inspection process 3 are carried out on separate production lines.

[0068] [Variations] The above-described embodiments are intended to facilitate understanding of the present invention and are not intended to limit the present invention. The elements of the embodiments, as well as their arrangement, materials, conditions, shapes, sizes, etc., are not limited to those exemplified and can be modified as appropriate.

[0069] For example, in the above-described embodiment, the appearance inspection process 3 is performed using an appearance inspection robot 32 and a marking robot 33, but it is also possible to have a single robot perform both the process of inspecting the appearance of the workpiece and the process of marking the workpiece.

[0070] In the above-described embodiment, the process of inspecting the appearance of the workpiece and the process of marking the workpiece are carried out together in the appearance inspection process 3, but each process may be separated into separate processes and carried out independently. In this case, it is preferable to provide a control device for each process.

[0071] Furthermore, when transporting the workpiece W between each process or each robot in the above-described embodiment, a transport robot (handling robot) may be used to transport the workpiece W. For example, a first transport robot that transports the workpiece W from the working area of ​​the welding robot 23 to the working area of ​​the appearance inspection robot 32, a second transport robot that transports the workpiece W from the working area of ​​the appearance inspection robot 32 to the working area of ​​the marking robot 33, and a third transport robot that transports the workpiece W from the working area of ​​the marking robot 33 to the working area of ​​the reworking process 4 may be further provided.

[0072] Furthermore, in the above-described embodiment, the first coordinate data and the second coordinate data are used when identifying the abnormality location, but the method is not limited to using coordinate data to identify the abnormality location. It is sufficient to identify the abnormality location based on position information corresponding to any position on the workpiece W. For example, the elapsed time from the welding start position (position information) until the abnormality occurs may be measured, and the abnormality location may be identified by multiplying the measured time by the welding speed. In this case, the acquisition unit 11 acquires the welding measurement data and the first position information, as well as the appearance data and the second position information, and the information generation unit 13 identifies the abnormality location based on the first position information and the second position information corresponding to the abnormality detected by the abnormality detection unit 12.

[0073] Furthermore, in the above-described embodiment, a workpiece identification number may be assigned to the workpiece W using, for example, a barcode or a two-dimensional code, and a step of reading the workpiece identification number may be further provided before carrying out the welding process 2. In this case, for example, the read workpiece identification number may be transmitted to the abnormality detection device 1, which may generate workpiece inspection information based on the workpiece identification number, and the welding measurement data, appearance data, and abnormality detection information received in the subsequent process may be stored in association with the workpiece identification number.

[0074] Furthermore, in the above-described embodiment, when the abnormality detection unit 12 of the abnormality detection device 1 detects an abnormality in the workpiece W based on the welding measurement data and appearance data, the data is compared with a threshold value for detecting an abnormality and with normal appearance data, but the values ​​and data to be compared are not limited to these. Any values ​​and data that can detect an abnormality in the workpiece W may be used, and in addition to the above, for example, normal values ​​or absolute values ​​may be used for comparison. [Explanation of symbols]

[0075] 1...abnormality detection device, 2...welding process, 3...visual inspection process, 4...repair process, 10...control unit, 11...acquisition unit, 12...abnormality detection unit, 13...information generation unit, 14...storage unit, 15...communication unit, 21...welding power source, 22...control unit, 23...welding robot, 31...control unit, 32...visual inspection robot, 33...marking robot, 41...display unit, 100...abnormality detection system

Claims

1. a first acquisition unit that acquires welding measurement data measured for each object while performing arc welding; a second acquisition unit that acquires appearance data obtained for each of the welded objects by visual inspection of the welded objects; an abnormality detection unit that detects an abnormality in the object based on the acquired welding measurement data and the acquired appearance data; an information generating unit that generates abnormality information of the object based on the detected abnormality; Equipped with the first acquisition unit further acquires first position information corresponding to the welding measurement data; the second acquisition unit further acquires second location information corresponding to the appearance data; the information generation unit identifies an abnormality location where the abnormality is detected based on the first position information and the second position information corresponding to the detected abnormality, and generates the abnormality information by including information on the identified abnormality location. Anomaly detection device.

2. a first acquisition unit that acquires welding measurement data measured for each object while performing arc welding; a second acquisition unit that acquires appearance data obtained for each of the welded objects by visual inspection of the welded objects; an abnormality detection unit that detects an abnormality in the object based on the acquired welding measurement data and the acquired appearance data; an information generating unit that generates abnormality information of the object based on the detected abnormality; Equipped with the anomaly information includes marking information for marking the object, the information generating unit identifies an occurrence pattern of an abnormality based on the abnormality detected based on the welding measurement data and the abnormality detected based on the appearance data, and generates the marking information corresponding to the identified occurrence pattern of the abnormality. Anomaly detection device.

3. the information generating unit generates the anomaly information by including information on a repair process of the object corresponding to the detected anomaly.

3. The abnormality detection device according to claim 1 or 2.

4. the abnormality detection unit detects an abnormality in the object by comparing the acquired welding measurement data with an abnormality detection threshold, and detects an abnormality in the object by comparing the acquired appearance data with appearance data in a normal state. The abnormality detection device according to claim 1 .

Images