A monitoring and alarm system for welding production line

By designing the welding production line monitoring and alarm system on the welding production line, the welding arc and employee status are monitored in real time, the problem of arc abnormalities not being discovered in time is solved, and the welding effect and safety are improved.

CN118417764BActive Publication Date: 2024-11-29苏州诺克智能装备股份有限公司
View PDF 21 Cites 0 Cited by

Patent Information

Application Number
CN202410468240.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-04-18
Publication Date
2024-11-29
Estimated Expiration
2044-04-18

Smart Images

  • Figure CN118417764B_ABST
    Figure CN118417764B_ABST
Patent Text Reader

Abstract

The present invention discloses a welding production line monitoring and alarm system, which belongs to the field of production line monitoring systems, and includes a fixture and tooling data acquisition module, a welding arc data acquisition module, an employee image data acquisition module, a data receiving module, a data analysis module, a data import module, an early warning classification module, and an alarm module; the fixture and tooling data acquisition module is used to collect image data of the fixture and tooling when clamping a workpiece through a camera, and send the collected data to the data receiving module; the welding arc data acquisition module is used to collect image data of the arc generated during welding through a camera, and send the collected data to the data receiving module. The present invention can monitor the arc generated during welding, and timely discover and warn when the welding arc is abnormal, which not only avoids the decline of welding effect, but also avoids causing safety accidents.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention relates to the field of production line monitoring systems, in particular to a welding production line monitoring alarm system. Background Art

[0002] Welding production line refers to a production line that uses a series of welding equipment, fixtures, transmission devices and control systems to weld various parts according to the preset process procedures to complete product processing. This production line is widely used in many industries such as automobiles, machinery, electronics, and construction. Especially in the manufacturing of automobile body-in-white, welding automation production line is an indispensable part.

[0003] In order to ensure the stable operation of the welding production line, people have added a monitoring and alarm system to the welding production line. The traditional monitoring and alarm system simply monitors the welding current, welding time, and electrode pressure of the production line, but cannot monitor the arc generated during welding. If the welding arc is abnormal and the warning cannot be discovered in time, it will not only easily lead to a decline in welding effect, but also easily cause safety accidents.

[0004] Therefore, those skilled in the art provide a welding production line monitoring alarm system to solve the problems raised in the above background technology. Summary of the invention

[0005] The purpose of the present invention is to provide a welding production line monitoring and alarm system, which can monitor the arc generated during welding and promptly detect and warn when the welding arc is abnormal, thereby not only avoiding the decline of welding effect, but also avoiding the occurrence of safety accidents, so as to solve the problems raised in the above background technology.

[0006] To achieve the above object, the present invention provides the following technical solutions:

[0007] A welding production line monitoring and alarm system, comprising a fixture and tooling data acquisition module, a welding arc data acquisition module, a data receiving module, a data analysis module, a data import module, a warning classification module and an alarm module;

[0008] The fixture data acquisition module is used to collect image data of the fixture when clamping the workpiece through a camera, and send the collected data to the data receiving module; the welding arc data acquisition module is used to collect image data of the arc generated during welding through a camera, and send the collected data to the data receiving module; the data receiving module is used to receive all collected data and package it and send it to the data analysis module; the data import module is used to import preset data to the data analysis module; the data analysis module is used to receive all data for analysis, and output the analysis results to the early warning classification module; the early warning classification module is used to perform early warning classification according to the received analysis results, and send the early warning classification results to the alarm module; the alarm module is used to issue an alarm of a corresponding level according to the early warning classification results.

[0009] As a further solution of the present invention: the specific process of the data analysis module analyzing the image data of the worker's action behavior is as follows:

[0010] (1): Extract the posture image of the employee when walking from the collected image data and analyze the posture image. The specific analysis process is as follows:

[0011] (101): In the posture image, the employee's left ear is marked as P1, the employee's left shoulder is marked as P2, the employee's right ear is marked as P3, and the employee's right shoulder is marked as P4;

[0012] (102): Calculate the distance between P1 and P2, marked as p1, and then calculate the distance between P3 and P4, marked as p2;

[0013] (103): Match p1 with the preset value p3, and then match p2 with the preset value p4;

[0014] (104): If p1 matches p3 successfully, and p2 matches p4 successfully, it means that the employee's posture is normal, otherwise, go to the next step;

[0015] (105): Calculate the difference between p1 and p3 and mark it as Δp1, then calculate the difference between p2 and p4 and mark it as Δp2;

[0016] (106): If (Δp1+Δp2) is greater than the preset value p5, it means that the employee's posture is abnormal; if (Δp1+Δp2) is less than or equal to the preset value p5, it means that the employee's posture is normal;

[0017] (2): Extract the footsteps image of the employee when walking from the collected image data and analyze the footsteps image. The specific analysis process is as follows:

[0018] (201): In the footstep image, the starting foot-lifting position is marked as W1, the highest point of the foot-lifting position is marked as W2, and the landing position is marked as W3;

[0019] (202): Connect W1, W2, and W3 in sequence to obtain triangle w1;

[0020] (203): Compare the similarity between w1 and the preset triangle w2;

[0021] (204): If the similarity between w1 and w2 is less than 90%, it means that the employee's footsteps are abnormal; if the similarity between w1 and w2 is greater than or equal to 90%, it means that the employee's footsteps are normal;

[0022] (3): Extract the finger images of the employee with empty hands from the collected image data and analyze the finger images. The specific analysis process is as follows:

[0023] (301): In the finger image, the five fingertips of the employee's left hand are marked as U1, U2, U3, U4, and U5, and the five fingertips of the employee's right hand are marked as U6, U7, U8, U9, and U10;

[0024] (302): Connect U1, U2, U3, U4, and U5 in sequence to obtain the graph u1, and then connect U6, U7, U8, U9, and U10 in sequence to obtain the graph u2;

[0025] (303): Compare the similarity between u1 and the preset graph u3, and then compare the similarity between u2 and the preset graph u4;

[0026] (304): If the similarity between u1 and u3, and between u2 and u4 all reaches 90%, it means that the employee's finger is normal, otherwise it means that the employee's finger is abnormal;

[0027] (4): If the employee has any two of abnormal posture, abnormal footsteps, and abnormal fingers, the output analysis result is that the employee status is abnormal.

[0028] As a further solution of the present invention: the specific process of the data analysis module analyzing the arc image data is as follows:

[0029] S1: construct a quantity analysis sub-model, input the arc quantity identified from the arc image into the model for analysis, and output the first analysis result;

[0030] S2: construct a shape analysis sub-model, input the arc shape identified from the arc image into the model for analysis, and output a second analysis result;

[0031] S3: construct a color analysis sub-model, input the arc color identified from the arc image into the model for analysis, and output a third analysis result;

[0032] S4: constructing a splash distance analysis sub-model, inputting the arc splash distance identified from the arc image into the model for analysis, and outputting a fourth analysis result;

[0033] S5: Integrate the first analysis result, the second analysis result, the third analysis result, and the fourth analysis result to output a total analysis result.

[0034] As a further solution of the present invention: in S1, the specific analysis process of the quantitative analysis sub-model is:

[0035] S101: Mark the number of arcs appearing per second during the welding process as Ai, i=1...n, where n is a positive integer;

[0036] S102: Mark the number of arc flashes per second in the preset data as ai, where i=1...n;

[0037] S103: Match Ai and ai one by one, and then calculate the absolute value of the difference between the matched Ai and ai, marked as αi, i=1···n;

[0038] S104: Compare αi with a first preset value in the preset data. If αi is greater than the first preset value, select the αi and send it to the first sequence. Count the number of all αi in the first sequence and mark them as β.

[0039] S105: If β>0, the first analysis result is outputted as the number of arc flashes is abnormal, and the degree of abnormality is (β*10%).

[0040] As a further solution of the present invention: in S2, the specific analysis process of the shape analysis sub-model is:

[0041] S201: Mark all arc shapes appearing in the welding process as Bi, i=1···n;

[0042] S202: Marking the preset arc shape in the preset data as bi, i=1...n;

[0043] S203: perform similarity comparison between any Bi and each bi, obtain the highest similarity and mark it as θi, i=1···n;

[0044] S204: Compare θi with a second preset value in the preset data. If θi is less than the second preset value, filter out θi and send it to the second sequence. Count the number of all θi in the second sequence and mark them as δ.

[0045] S205: If δ>0, the second analysis result is outputted as the arc shape is abnormal, and the degree of abnormality is (δ*10%).

[0046] As a further solution of the present invention: in S3, the specific analysis process of the color analysis sub-model is:

[0047] S301: Mark all arc colors appearing in the welding process as Ci, where i=1···n;

[0048] S302: Mark the preset arc color in the preset data as ci, i=1...n;

[0049] S303: Compare the similarity between Ci and bi in hue, purity and brightness. If the similarity between Ci and any bi in hue, purity and brightness is less than the third preset value in the preset data, the Ci is selected and sent to the third sequence. The number of all Ci in the third sequence is counted and marked as &;

[0050] S304: If &>0, the third analysis result is output as the arc light color is abnormal, and the degree of abnormality is (&*10%).

[0051] As a further solution of the present invention: in S4, the specific analysis process of the splash distance analysis sub-model is:

[0052] S401: Mark the maximum arc splash distance occurring per second during the welding process as Di, i=1···n;

[0053] S402: Mark the maximum arc splash distance per second preset in the preset data as di, i=1...n;

[0054] S403: Match Di and di one by one, and then calculate the difference value γi between the matched Di and di, γi = (Di-di);

[0055] S404: Compare γi with a fourth preset value in the preset data. If γi is greater than the fourth preset value, select the γi and send it to the fourth sequence. Count the number of all γi in the fourth sequence and mark them as ξ.

[0056] S405: If ξ>0, the fourth analysis result is outputted as the arc splash distance is abnormal, and the degree of abnormality is (ξ*10%).

[0057] As a further solution of the present invention: in S5, the overall analysis result is specifically:

[0058] If there is no abnormality in the first analysis result, the second analysis result, the third analysis result and the fourth analysis result, the overall analysis result is output as normal welding arc;

[0059] If any one of the first analysis result, the second analysis result, the third analysis result, and the fourth analysis result is abnormal, the total analysis result is output as welding arc abnormality, and the total abnormality degree is calculated, and the total abnormality degree is the sum of the abnormality degrees.

[0060] As a further solution of the present invention: the specific process of the fixture data acquisition module analyzing the image data when the fixture clamps the workpiece is as follows:

[0061] The image of the fixture clamping the workpiece acquired in real time is marked as X;

[0062] The image after the workpiece is clamped by the fixture preset in the preset data is marked as Y;

[0063] Compare the similarity between X and Y, and mark the similarity between the two as Z;

[0064] If Z is greater than or equal to the fifth preset value in the preset data, the output analysis result is that the clamping degree of the fixture is normal; if Z is less than the fifth preset value, the output analysis result is that the clamping degree of the fixture is abnormal, and the degree of abnormality is (1-Z).

[0065] As a further solution of the present invention: the specific process of the early warning classification module for performing early warning classification is:

[0066] If the analysis results show any of the following: abnormal welding arc, abnormal clamping degree of fixtures, or abnormal employee status, the output warning level will be level one;

[0067] If the analysis results show any two of the following: abnormal welding arc, abnormal clamping degree of fixtures and tools, or abnormal employee status, the output warning level will be level 2;

[0068] If the analysis results show all of the following: abnormal welding arc, abnormal clamping degree of fixtures and abnormal employee status, the output warning level will be level three.

[0069] Compared with the prior art, the present invention has the following beneficial effects:

[0070] 1. This application monitors the arc light generated during the welding process, analyzes whether there is any abnormality in the arc light from multiple aspects such as the number, shape, color and spatter distance of the arc light, and promptly discovers and calculates the degree of abnormality when the arc light is abnormal. This not only makes it convenient for staff to understand the state of the arc light, but also allows them to quickly know the severity of the abnormality when an abnormality occurs.

[0071] 2. This application performs similarity analysis on the image after the fixture clamps the workpiece to quickly determine whether the shape of the workpiece after the fixture clamps it is different from the preset shape. If the two are very different, it means that the fixture has an abnormality and its clamping degree does not meet the preset requirements.

[0072] 3. This application collects and analyzes monitoring images of employees on the welding production line, and conducts multi-angle analysis of the employees' posture, footsteps, and fingers to determine whether there is any abnormality in the current state of the employees, so as to timely detect and warn when the employees are overly tired to avoid accidents. It is especially suitable for monitoring and alarming of night shift workers. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 This is a structural block diagram of a welding production line monitoring and alarm system. DETAILED DESCRIPTION

[0074] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0075] As mentioned in the background technology of this application, the inventors have discovered through research that the existing monitoring and alarm system simply monitors the welding current, welding time, and electrode pressure of the production line, but cannot monitor the arc generated during welding. If an abnormality in the welding arc cannot be detected and warned in time, it will not only easily lead to a decline in welding results, but also easily cause safety accidents, which has certain defects.

[0076] In order to solve the above-mentioned defects, the present application discloses a welding production line monitoring and alarm system, which can monitor the arc generated during welding and promptly detect and warn when abnormalities occur in the welding arc, thereby not only avoiding a decrease in welding effect, but also avoiding safety accidents.

[0077] The following will describe in detail how the solution of the present application solves the above technical problems in conjunction with the accompanying drawings.

[0078] See also Figure 1In an embodiment of the present invention, a monitoring and alarm system for a welding production line includes a fixture data acquisition module, a welding arc data acquisition module, a data receiving module, a data analysis module, a data import module, an early warning classification module, and an alarm module; the fixture data acquisition module is used to collect image data of the fixture when the workpiece is clamped by a camera, and send the collected data to the data receiving module; the welding arc data acquisition module is used to collect image data of the arc generated during welding by a camera, and send the collected data to the data receiving module; the data receiving module is used to receive all collected data and package it and send it to the data analysis module; the data import module is used to import preset data to the data analysis module; the data analysis module is used to receive all data for analysis, and output the analysis results to the early warning classification module; the early warning classification module is used to perform early warning classification according to the received analysis results, and send the early warning classification results to the alarm module; the alarm module is used to issue an alarm of a corresponding level according to the early warning classification results. The present application can monitor the arc generated during welding, and timely detect and warn when the welding arc is abnormal, which not only avoids the decline of welding effect, but also avoids causing safety accidents.

[0079] In this embodiment, the specific process of the data analysis module analyzing the image data of the staff's action behavior is as follows: (1): extracting the posture image of the employee when walking from the collected image data, and analyzing the posture image. The specific analysis process is as follows: (101): marking the employee's left ear as P1, the employee's left shoulder peak as P2, the employee's right ear as P3, and the employee's right shoulder peak as P4 in the posture image; (102): calculating the distance between P1 and P2, marking it as p1, and then calculating the distance between P3 and P4, marking it as p2; (103): matching p1 with the preset value p3, and then matching p2 with the preset value p4; (104): matching the distance between P1 and P2, and then matching the distance between P3 and P4 ... ): If p1 matches p3 successfully, and p2 matches p4 successfully, it means that the employee's posture is normal, otherwise, go to the next step; (105): Calculate the difference between p1 and p3 and mark it as Δp1, then calculate the difference between p2 and p4 and mark it as Δp2; (106): If (Δp1+Δp2) is greater than the preset value p5, it means that the employee's posture is abnormal, if (Δp1+Δp2) is less than or equal to the preset value p5, it means that the employee's posture is normal; (2): Extract the footstep image of the employee when walking from the collected image data, and analyze the footstep image. The specific analysis process is: (201): Mark the starting foot lift position as W1 in the footstep image, Mark the highest point of the raised foot as W2, and the landing point as W3; (202): Connect W1, W2, and W3 in sequence to obtain triangle w1; (203): Compare w1 with the preset triangle w2 for similarity; (204): If the similarity between w1 and w2 is less than 90%, it means that the employee's footsteps are abnormal; if the similarity between w1 and w2 is greater than or equal to 90%, it means that the employee's footsteps are normal; (3): Extract the finger image of the employee when he / she is empty-handed from the collected image data, and analyze the finger image. The specific analysis process is as follows: (301): Mark the five fingertips of the employee's left hand as U1, U2, U3, U4, and U5 in the finger image, and then compare the employee's footsteps with the original fingertips. The five fingertips of the employee's right hand are marked as U6, U7, U8, U9, and U10 respectively; (302): Connect U1, U2, U3, U4, and U5 in sequence to obtain the graph u1, and then connect U6, U7, U8, U9, and U10 in sequence to obtain the graph u2; (303): Compare the similarity between u1 and the preset graph u3, and then compare the similarity between u2 and the preset graph u4; (304): If the similarity between u1 and u3, and between u2 and u4 all reaches 90%, it means that the employee's fingers are normal, otherwise it means that the employee's fingers are abnormal; (4): If the employee has any two of abnormal posture, abnormal footsteps, and abnormal fingers, the output analysis result is that the employee status is abnormal.This application collects and analyzes the monitoring images of employees on the welding production line, analyzes the employees' postures, footsteps, and fingers from multiple angles, and determines whether there is any abnormality in the current state of the employees, so as to timely detect and warn when the employees are overly tired, and avoid accidents. It is particularly suitable for monitoring and alarming night shift workers. It should be noted that when a person walks tiredly, the angle between the back and the head will change. This is mainly due to physical fatigue and weakened muscle strength. When the human body feels tired, the muscles of the back and shoulders may become loose, causing the back posture to become less straight. At the same time, the head may also sag due to fatigue of the neck muscles, causing the angle between the back and the head to change. This change is usually manifested as the back becoming more curved and the head drooping, which will change the distance from the acromion to the ear. This application monitors this value to determine whether the employee is in a state of fatigue. Gait changes are also a common manifestation of physical fatigue. Commonly, the steps become shorter and the height of the foot lift decreases. Therefore, this application constructs a triangular figure for each step for detection based on these points to determine whether the employee's footsteps are abnormal due to fatigue. In addition, when a person is tired, his hands will relax. When the hands are in a relaxed state, the fingers may naturally droop and no longer maintain the original bending or stretching posture. This state of weakness and drooping may be caused by muscle fatigue, which makes the fingers unable to maintain the previous posture. Based on this, this application collects the fingertip positions and connects them into a graph for detection to analyze whether the employee's hands are relaxed due to fatigue. In addition, the preset value p3, preset value p4, preset value p5, preset triangle w2, preset graph u3, and preset graph u4 are all preset data imported by the data import module.

[0080] In this embodiment: the specific process of the data analysis module analyzing the arc image data is as follows: S1: constructing a quantity analysis sub-model, inputting the arc quantity identified from the arc image into the model for analysis, and outputting the first analysis result; S2: constructing a shape analysis sub-model, inputting the arc shape identified from the arc image into the model for analysis, and outputting the second analysis result; S3: constructing a color analysis sub-model, inputting the arc color identified from the arc image into the model for analysis, and outputting the third analysis result; S4: constructing a spatter distance analysis sub-model, inputting the arc spatter distance identified from the arc image into the model for analysis, and outputting the fourth analysis result; S5: synthesizing the first analysis result, the second analysis result, the third analysis result, and the fourth analysis result, and outputting the total analysis result. By monitoring the arc generated during the welding process, the arc is analyzed from the aspects of the number, shape, color, and spatter distance of the arc to determine whether it is abnormal. When the arc is abnormal, it is timely discovered and the degree of abnormality is calculated, which not only facilitates the staff to understand the state of the arc, but also enables them to quickly know whether the degree of abnormality is serious when an abnormality occurs. It should be noted that during the welding process, an abnormal arc may indicate a variety of problems. Specifically, it may represent the following:

[0081] (1) The mixed gas is not completely eliminated: Before welding begins, ensure that the mixed gas in the welding area is completely eliminated to avoid sudden release of gas during welding, resulting in abnormal arc.

[0082] (2) Acetylene pressure is too low or gas flow is insufficient: If the pressure of acetylene or other gases is insufficient, or the gas flow does not meet the standard, it may cause abnormal arc during welding.

[0083] (3) Welding tip problems: The expansion, deformation, or blockage of the welding tip aperture may cause abnormal welding arc. These problems may be caused by long-term use or improper maintenance.

[0084] (4) Other equipment problems: For example, poor performance of the temper fuse or poor performance of the pressure regulator may also cause abnormal welding arc.

[0085] When the welding arc is abnormal, the welder should stop the welding operation immediately and check the welding equipment and related components to find and solve the problem. At the same time, in order to avoid possible safety risks, the welder should always wear appropriate protective equipment during the welding process and ensure the safety of the working area.

[0086] In this embodiment: In S1, the specific analysis process of the quantity analysis sub-model is as follows: S101: mark the number of arcs appearing per second during welding as Ai, i=1···n, where n is a positive integer; S102: mark the number of arcs appearing per second preset in the preset data as ai, i=1···n; S103: match Ai and ai one by one, and then calculate the absolute value of the difference between the matched Ai and ai, marked as αi, i=1···n; S104: compare αi with the first preset value in the preset data. If αi is greater than the first preset value, the αi is filtered out and sent to the first sequence, and the number of all αi in the first sequence is counted and marked as β; S105: If β>0, the first analysis result is output as abnormal arc quantity, and the degree of abnormality is (β*10%). This setting can quickly know whether there is an abnormality in the number of arcs. A sudden increase or decrease in the number of welding arcs means that parameters such as welding current, voltage or welding speed are improperly set and need to be adjusted.

[0087] In this embodiment: In S2, the specific analysis process of the shape analysis submodel is: S201: mark all arc shapes appearing in the welding process as Bi, i=1···n; S202: mark the preset arc shapes in the preset data as bi, i=1···n; S203: compare the similarity of any Bi with each bi, obtain the highest similarity and mark it as θi, i=1···n; S204: compare θi with the second preset value in the preset data. If θi is less than the second preset value, filter out the θi and send it to the second sequence, count the number of all θi in the second sequence and mark it as δ; S205: If δ>0, output the second analysis result as abnormal arc shape, and the degree of abnormality is (δ*10%). This setting can quickly know whether there is an abnormality in the shape of the arc. Under normal circumstances, the welding arc should present a stable and continuous shape. If the arc shape becomes irregular, intermittent or scattered, this may mean that there is a problem in the welding process.

[0088] In this embodiment: In S3, the specific analysis process of the color analysis sub-model is: S301: Mark all arc colors that appear in the welding process as Ci, i=1···n; S302: Mark the preset arc colors in the preset data as ci, i=1···n; S303: Compare the similarity of Ci and bi in hue, purity, and brightness. If the similarity between Ci and any bi in hue, purity, and brightness is less than the third preset value in the preset data, then filter out the Ci and send it to the third sequence, count the number of all Ci in the third sequence and mark it as &; S304: If &>0, then output the third analysis result as abnormal arc color, and the degree of abnormality is (&*10%). This setting can quickly know whether there is an abnormality in the color of the arc. The welding arc is usually bright white or bluish white. If the arc color changes to yellow, red or other abnormal colors, this may indicate a problem with the welding temperature, gas composition or material state.

[0089] In this embodiment: In S4, the specific analysis process of the spatter distance analysis submodel is as follows: S401: Mark the maximum arc spatter distance that occurs per second during the welding process as Di, i=1···n; S402: Mark the maximum arc spatter distance per second preset in the preset data as di, i=1···n; S403: Match Di and di one by one, and then calculate the difference between the matched Di and di, γi, γi=(Di-di); S404: Compare γi with the fourth preset value in the preset data. If γi is greater than the fourth preset value, the γi is screened out and sent to the fourth sequence, and the number of all γi in the fourth sequence is counted and marked as ξ; S405: If ξ>0, the fourth analysis result is output as an abnormal arc spatter distance, and the degree of abnormality is (ξ*10%). This setting can quickly know whether there is an abnormality in the arc spatter distance. Under normal circumstances, welding arc spatter should be controlled within a certain range. If the arc spatter is too large, it may lead to a decline in welding quality and even cause safety hazards to welders.

[0090] In this embodiment: In S5, the overall analysis result is specifically: if there is no abnormality in the first analysis result, the second analysis result, the third analysis result, and the fourth analysis result, then the overall analysis result is output as normal welding arc; if any one of the first analysis result, the second analysis result, the third analysis result, and the fourth analysis result is abnormal, then the overall analysis result is output as abnormal welding arc, and the overall abnormality degree is calculated, and the overall abnormality degree is the sum of the abnormality degrees. This setting can comprehensively analyze the analysis results of the number, shape, color, and spatter distance of the arc to determine whether the arc is abnormal.

[0091] In this embodiment: the specific process of the fixture data acquisition module analyzing the image data of the fixture clamping the workpiece is as follows: mark the image of the fixture clamping the workpiece collected in real time as X; mark the image of the fixture clamping the workpiece in the preset data as Y; compare X and Y for similarity, and mark the similarity between the two as Z; if Z is greater than or equal to the fifth preset value in the preset data, then the output analysis result is that the clamping degree of the fixture is normal; if Z is less than the fifth preset value, then the output analysis result is that the clamping degree of the fixture is abnormal, and the degree of abnormality is (1-Z). By performing similarity analysis on the image after the fixture clamps the workpiece, it is quickly known whether the shape of the fixture after clamping the workpiece is different from the preset shape. If the difference between the two is large, it means that the fixture is abnormal and its clamping degree does not meet the preset requirements.

[0092] In this embodiment: the specific process of the early warning classification module for early warning classification is: if the analysis result shows any one of abnormal welding arc, abnormal clamping degree of fixtures and abnormal employee status, the output early warning level is level one; if the analysis result shows any two of abnormal welding arc, abnormal clamping degree of fixtures and abnormal employee status, the output early warning level is level two; if the analysis result shows all of abnormal welding arc, abnormal clamping degree of fixtures and abnormal employee status, the output early warning level is level three. This setting can facilitate the staff to quickly know the importance of the early warning.

[0093] In this embodiment, the alarm module includes an audible and visual alarm. When the warning level is level one, the audible and visual alarm flashes yellow light and broadcasts the corresponding degree of abnormality; when the warning level is level two, the audible and visual alarm flashes red light and broadcasts the corresponding degree of abnormality; when the warning level is level three, the audible and visual alarm flashes red and yellow light alternately and broadcasts the corresponding degree of abnormality.

[0094] The present invention monitors the arc light generated during the welding process, analyzes whether the arc light is abnormal from multiple aspects such as the number, shape, color and splash distance of the arc light, and promptly discovers and calculates the degree of abnormality when the arc light is abnormal, which not only facilitates the staff to understand the state of the arc light, but also enables the staff to quickly know whether the degree of abnormality is serious when an abnormality occurs. By performing a similarity analysis on the image after the workpiece is clamped by the fixture, it is quickly known whether the shape of the workpiece after the fixture is clamped is different from the preset shape. If the difference between the two is large, it means that the fixture is abnormal and its clamping degree does not meet the preset requirements. The present application collects and analyzes the monitoring images of employees on the welding production line, performs multi-angle analysis on the employees' postures, steps and fingers, and determines whether the current employee's state is abnormal, so as to promptly discover and warn when the employee is overly tired, and avoid accidents. It is particularly suitable for monitoring and alarming night shift workers.

[0095] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

[0096] What is described above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A welding production line monitoring and alarm system, characterized in that: It includes fixture and tooling data acquisition module, welding arc data acquisition module, employee image data acquisition module, data receiving module, data analysis module, data import module, early warning classification module and alarm module; The fixture data acquisition module is used to collect image data of the fixture when clamping the workpiece through a camera, and send the collected data to the data receiving module; The welding arc light data acquisition module is used to collect image data of arc light generated during welding through a camera, and send the collected data to the data receiving module; The employee image data acquisition module is used to collect image data of the action behavior of the staff on the welding production line through a camera, and send the collected data to the data receiving module; The data receiving module is used to receive all collected data and package it to send to the data analysis module; the data import module is used to import preset data to the data analysis module; the data analysis module is used to receive all data for analysis and output the analysis results to the early warning classification module; the early warning classification module is used to perform early warning classification according to the received analysis results and send the early warning classification results to the alarm module; the alarm module is used to issue an alarm of a corresponding level according to the early warning classification results; The specific process of the data analysis module analyzing the image data of the staff's action behavior is as follows: (1): Extract the posture image of the employee when walking from the collected image data and analyze the posture image. The specific analysis process is as follows: (101): In the posture image, the employee's left ear is marked as P1, the employee's left shoulder is marked as P2, the employee's right ear is marked as P3, and the employee's right shoulder is marked as P4; (102): Calculate the distance between P1 and P2, marked as p1, and then calculate the distance between P3 and P4, marked as p2; (103): Match p1 with the preset value p3, and then match p2 with the preset value p4; (104): If p1 matches p3 successfully, and p2 matches p4 successfully, it means that the employee's posture is normal, otherwise, go to the next step; (105): Calculate the difference between p1 and p3 and mark it as Δp1, then calculate the difference between p2 and p4 and mark it as Δp2; (106): If (Δp1+Δp2) is greater than the preset value p5, it means that the employee's posture is abnormal; if (Δp1+Δp2) is less than or equal to the preset value p5, it means that the employee's posture is normal; (2): Extract the footsteps image of the employee when walking from the collected image data and analyze the footsteps image. The specific analysis process is as follows: (201): In the footstep image, the starting foot-lifting position is marked as W1, the highest point of the foot-lifting position is marked as W2, and the landing position is marked as W3; (202): Connect W1, W2, and W3 in sequence to obtain triangle w1; (203): Compare the similarity between w1 and the preset triangle w2; (204): If the similarity between w1 and w2 is less than 90%, it means that the employee's footsteps are abnormal; if the similarity between w1 and w2 is greater than or equal to 90%, it means that the employee's footsteps are normal; (3): Extract the finger images of the employee with empty hands from the collected image data and analyze the finger images. The specific analysis process is as follows: (301): In the finger image, the five fingertips of the employee's left hand are marked as U1, U2, U3, U4, and U5, and the five fingertips of the employee's right hand are marked as U6, U7, U8, U9, and U10; (302): Connect U1, U2, U3, U4, and U5 in sequence to obtain the graph u1, and then connect U6, U7, U8, U9, and U10 in sequence to obtain the graph u2; (303): Compare the similarity between u1 and the preset graph u3, and then compare the similarity between u2 and the preset graph u4; (304): If the similarity between u1 and u3, and between u2 and u4 all reaches 90%, it means that the employee's finger is normal, otherwise it means that the employee's finger is abnormal; (4): If the employee has any two of abnormal posture, abnormal footsteps, or abnormal fingers, the output analysis result is that the employee status is abnormal; The specific process of the data analysis module analyzing the arc image data is as follows: S1: construct a quantity analysis sub-model, input the arc quantity identified from the arc image into the model for analysis, and output the first analysis result; S2: construct a shape analysis sub-model, input the arc shape identified from the arc image into the model for analysis, and output a second analysis result; S3: construct a color analysis sub-model, input the arc color identified from the arc image into the model for analysis, and output a third analysis result; S4: constructing a splash distance analysis sub-model, inputting the arc splash distance identified from the arc image into the model for analysis, and outputting a fourth analysis result; S5: Integrate the first analysis result, the second analysis result, the third analysis result, and the fourth analysis result to output a total analysis result; In S1, the specific analysis process of the quantitative analysis sub-model is as follows: S101: Mark the number of arcs appearing per second during the welding process as Ai, i=1...n, where n is a positive integer; S102: Mark the number of arc flashes per second in the preset data as ai, where i=1...n; S103: Match Ai and ai one by one, and then calculate the absolute value of the difference between the matched Ai and ai, marked as αi, i=1···n; S104: Compare αi with a first preset value in the preset data. If αi is greater than the first preset value, select the αi and send it to the first sequence. Count the number of all αi in the first sequence and mark them as β. S105: If β>0, the first analysis result is outputted as the number of arc flashes is abnormal, and the degree of abnormality is (β*10%); In S2, the specific analysis process of the shape analysis sub-model is as follows: S201: Mark all arc shapes appearing in the welding process as Bi, i=1···n; S202: Marking the preset arc shape in the preset data as bi, i=1...n; S203: perform similarity comparison between any Bi and each bi, obtain the highest similarity and mark it as θi, i=1···n; S204: Compare θi with a second preset value in the preset data. If θi is less than the second preset value, filter out θi and send it to the second sequence. Count the number of all θi in the second sequence and mark them as δ. S205: If δ>0, the second analysis result is outputted as the arc shape is abnormal, and the degree of abnormality is (δ*10%); In S3, the specific analysis process of the color analysis sub-model is as follows: S301: Mark all arc colors appearing in the welding process as Ci, where i=1···n; S302: Mark the preset arc color in the preset data as ci, i=1...n; S303: Compare the similarity between Ci and bi in hue, purity and brightness. If the similarity between Ci and any bi in hue, purity and brightness is less than the third preset value in the preset data, the Ci is selected and sent to the third sequence. The number of all Ci in the third sequence is counted and marked as &; S304: If &>0, the third analysis result is outputted as the arc light color is abnormal, and the abnormality degree is (&*10%); In S4, the specific analysis process of the splash distance analysis sub-model is as follows: S401: Mark the maximum arc splash distance occurring per second during the welding process as Di, i=1···n; S402: Mark the maximum arc splash distance per second preset in the preset data as di, i=1...n; S403: Match Di and di one by one, and then calculate the difference value γi between the matched Di and di, γi = (Di-di); S404: Compare γi with a fourth preset value in the preset data. If γi is greater than the fourth preset value, select the γi and send it to the fourth sequence. Count the number of all γi in the fourth sequence and mark them as ξ. S405: If ξ>0, the fourth analysis result is outputted as the arc splash distance is abnormal, and the degree of abnormality is (ξ*10%).

2. A welding production line monitoring and alarm system according to claim 1, characterized in that: In S5, the overall analysis result is specifically: If there is no abnormality in the first analysis result, the second analysis result, the third analysis result and the fourth analysis result, the overall analysis result is output as normal welding arc; If any one of the first analysis result, the second analysis result, the third analysis result, and the fourth analysis result is abnormal, the total analysis result is output as welding arc abnormality, and the total abnormality degree is calculated, and the total abnormality degree is the sum of the abnormality degrees.

3. A welding production line monitoring and alarm system according to claim 2, characterized in that: The specific process of the fixture data acquisition module analyzing the image data when the fixture clamps the workpiece is as follows: The image of the fixture clamping the workpiece acquired in real time is marked as X; The image after the workpiece is clamped by the fixture preset in the preset data is marked as Y; Compare the similarity between X and Y, and mark the similarity between the two as Z; If Z is greater than or equal to the fifth preset value in the preset data, the output analysis result is that the clamping degree of the fixture is normal; if Z is less than the fifth preset value, the output analysis result is that the clamping degree of the fixture is abnormal, and the degree of abnormality is (1-Z).

4. A welding production line monitoring and alarm system according to claim 3, characterized in that: The specific process of the early warning classification module for early warning classification is as follows: If the analysis results show any of the following: abnormal welding arc, abnormal clamping degree of fixtures, or abnormal employee status, the output warning level will be level one; If the analysis results show any two of the following: abnormal welding arc, abnormal clamping degree of fixtures and tools, or abnormal employee status, the output warning level will be level 2; If the analysis results show all of the following: abnormal welding arc, abnormal clamping degree of fixtures and abnormal employee status, the output warning level will be level three.

Citation Information

Patent Citations

  • Welding process for production of a steel pipe

    CA1125383A

  • Abnormal falling behavior detection method based on space-time motion characteristics

    CN110321780A

  • Online monitoring device and monitoring method for arc welding process parameters

    CN110802305A

  • Welding technology performance evaluation method and system

    CN111037056A

  • Intelligent working condition monitoring equipment

    CN113411542A