Production quality detection system of motor

Abstract

The invention relates to the technical field of motor detection, in particular to a production quality detection system of a motor. The device comprises a machine table; the acquisition module is used for shooting the surface of the motor shell to obtain a test video; the processing module is used for analyzing the test video and judging whether the motor shell breaks down or not; the control module is used for controlling the execution module to execute the corresponding instruction according to the processing result of the processing module; the acquisition module comprises a slide rail fixedly installed on the top of the machine table, a support slidably installed on the slide rail, and an industrial camera installed on the support. According to the invention, the fault of the motor housing can be comprehensively detected.

Description

Technical Field

[0001] The present invention relates to the technical field of motor detection, and in particular to a production quality detection system for a motor. Background Art

[0002] A series of quality inspections are required during the production process and before the motor leaves the factory. For example, appearance inspection: check whether the rotor has defects such as cracks, deformation, rust, etc., ensure that the shell surface is flat and smooth, the coating is uniform, and there are no obvious scratches and peeling; verify whether the information on the nameplate is accurate and complete, including motor model, rated power, rated voltage, rated current, speed, manufacturing date, etc.

[0003] The invention patent with application number 202311422636.0 involves the field of image data processing technology, and proposes a motor housing quality detection method based on machine vision, including: obtaining a grayscale image of the motor front end shell and a brightness image of the motor front end shell; obtaining the gloss anomaly of each motor front end shell area according to the gloss characteristics between the motor front end shell areas on the front end shell brightness image; obtaining the structural difference value of each motor front end shell area according to the connected domain analysis results of the motor front end shell area in the front end shell brightness image; determining the defect significance coefficient of each motor front end shell area based on the gloss anomaly and structural difference value of each motor front end shell area; obtaining the quality detection result of the motor front end shell based on the significance detection result of the motor front end shell and the defect significance coefficient of the motor front end shell area. The above invention solves the problem of false detection of the robust principal component analysis algorithm when facing tiny defects on the motor front end shell, and improves the detection accuracy of defects in complex areas in the motor front end shell.

[0004] However, this patent can only detect the front end housing of the motor and cannot perform a comprehensive test.

[0005] Therefore, it is necessary to provide a new technical solution to overcome the above-mentioned defects. Summary of the invention

[0006] The object of the present invention is to provide a motor production quality inspection system which can effectively solve the above technical problems.

[0007] In order to achieve the purpose of the present invention, the following technical scheme is adopted:

[0008] A production quality inspection system for a motor comprises: a machine platform; an acquisition module for photographing the surface of a motor housing to obtain a test video; a processing module for analyzing the test video to determine whether a fault occurs in the motor housing; a control module for controlling an execution module to execute corresponding instructions according to the processing result of the processing module; the acquisition module comprises: a slide rail fixedly mounted on the top of the machine platform, a bracket slidably mounted on the slide rail, and an industrial camera mounted on the bracket; the execution module comprises: a power assembly; the power assembly comprises: a third mounting seat; a telescopic cylinder fixedly mounted on the third mounting seat; a connecting rod fixedly connected to the piston rod of the telescopic cylinder, and the connecting rod is fixedly connected to the bracket; a rack is fixedly connected to the bottom of the bracket; the rack is meshingly connected to a first gear; the first gear is coaxially connected to a material-prying and air-blowing member, and the material-prying and air-blowing member is connected to an air nozzle.

[0009] Furthermore, the execution module also includes: a clamping assembly; the clamping assembly includes: a first mounting seat, a first clamping plate rotatably mounted on the first mounting seat, a first motor that drives the first clamping plate to rotate; a second mounting seat slidably mounted on the top of the machine; a second clamping plate rotatably mounted on the second mounting seat; a second motor that drives the second mounting seat to slide; and a third motor that drives the second clamping plate to rotate.

[0010] Furthermore, the processing module determines whether the motor housing has a fault in the following steps:

[0011] S1. Determine a coordinate set of a key inspection area on the surface of the motor housing; the key inspection area includes: a location prone to scratches, pits, deformation or other defects;

[0012] Acquire an image of each key detection area according to the coordinate set;

[0013] S2, saving the images of all key detection areas obtained by the acquisition module into a first sequence;

[0014] Sort the images in the image set according to a preset rule to obtain a second sequence;

[0015] The preset rules include: Rule 1, sorting according to the importance of key areas; Rule 2, sorting according to the probability of defects and the severity of defects;

[0016] S3, identifying each image in the second sequence according to the preset rule to determine whether the image meets the quality standard;

[0017] S4. Outputting the recognition result, wherein the recognition result includes the type, location, and severity assessment of the defect; the defect is classified into mild, moderate, severe, etc. according to the severity of the defect;

[0018] S5. Output the recognition result, and control the execution module to execute the corresponding process based on the severity.

[0019] Furthermore, the step S3 also includes the following steps:

[0020] S3.1. When the rule 1 is adopted, when an image that does not meet the quality standard is identified, the unqualified position is marked until all the inspection items are completed. After all the inspections are completed, S4 is executed;

[0021] S3.2. When the second rule is adopted, when an image that does not meet the quality standard is identified, the unqualified position is marked and it is determined whether it is a serious defect;

[0022] S3.2.1. If it is not a serious defect, continue to inspect the high-probability area and randomly inspect the low-probability area. After the inspection is completed, execute S4;

[0023] S3.2.2: If it is a serious defect, go directly to S4;

[0024] S3.3. If each image in the second sequence meets the quality standard, execute S5.

[0025] Furthermore, the step S2 further includes the following steps:

[0026] S2.1. Initialize the weights of the rule 1 and the rule 2, respectively setting the regional importance weight Wi and the defect probability weight Wp. In the initial state, set Wi>Wp;

[0027] S2.2. After each inspection, the inspection results are saved in the database to form a historical defect data set; the data set contains the following information: coordinates of the inspection area; importance of the inspection area; defect type; frequency of defect occurrence; severity of the defect, etc.;

[0028] S2.3. During the detection process, the weights are dynamically adjusted according to the detection results.

[0029] Further, step S2.3 includes the following steps: let the defect probability value be P, ranging from 0 to 1; let the regional importance be S, 1 for slight, 2 for medium, and 3 for severe; let the low probability threshold be T; let the priority coefficient be D, ranging from 0 to 1; let the comprehensive ranking value be W, then

[0030] When S=3, 100% detection is performed unconditionally;

[0031] When S=2, the defects in the general area need to be paid special attention to, where the lowest weight of the general area is A, and the highest weight of the general area is B. A is between 0.1 and 0.3; B is between 0.5 and 0.9;

[0032] When S=1, random inspections are conducted on secondary areas or low-probability areas.

[0033] Furthermore, it also includes:

[0034] S2.4. Analyze the data set, establish a prediction model, calculate the defect probability for each area, and dynamically adjust the detection priority to predict the probability of defects in each area;

[0035] S2.5. Regularly update the training model and the defect probability of each area;

[0036] S2.6. Assign confidence levels to the results of each test based on the accuracy of past tests. Areas with low confidence levels can be retested in subsequent tests. After the test is completed, if it is found that the test results are significantly different from the actual defect situation, reduce the confidence level of the test model for that area.

[0037] Further, the control module controls the execution of the module to execute the corresponding instruction including:

[0038] S5.1. When the product is qualified, control the execution module to execute mode 1;

[0039] S5.2. When the product has a slight defect, the execution module is controlled to execute mode 2;

[0040] S5.3. When the product has serious defects, control the execution module to execute mode three;

[0041] The first mode is to loosen the motor housing and transport it to the first channel; the second mode is to loosen the motor housing and transport it to the second channel; the third mode is to loosen the motor housing and transport it to the third channel.

[0042] Compared with the prior art, the present invention has the following beneficial effects: the present invention can perform a relatively comprehensive detection of the faults of the motor housing. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The accompanying drawings are used to provide further understanding of the present invention and constitute a part of the specification. They are used to explain the present invention together with the embodiments of the present invention and do not constitute a limitation of the present invention.

[0044] Figure 1 It is a schematic diagram of the axial side of the present invention;

[0045] Figure 2It is a structural schematic diagram of the present invention;

[0046] Figure 3 It is a structural schematic diagram of the acquisition module of the present invention;

[0047] Figure 4 It is a schematic structural diagram of the clamping assembly of the present invention;

[0048] Figure 5 It is a schematic diagram of the structure of the power assembly of the present invention;

[0049] Figure 6 It is a schematic diagram of the internal structure of the material blowing component of the present invention.

[0050] In the figure: 1. machine; 2. slide rail; 3. bracket; 4. industrial camera; 5. light source; 6. clamping assembly; 7. power assembly; 8. channel one; 9. channel two; 10. channel three; 11. first mounting seat; 12. first clamping plate; 13. first motor; 14. second mounting seat; 15. second clamping plate; 16. second motor; 17. third motor; 18. turntable; 20. second gear; 21. third gear; 22. fourth motor; 23. third mounting seat; 24. telescopic cylinder; 25. connecting rod; 26. rack; 27. first gear; 28. material blowing member; 29. ​​hose; 30. air storage chamber; 31. guide plate. DETAILED DESCRIPTION

[0051] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are partial embodiments of the present invention, rather than all embodiments.

[0052] In the description of the present invention, it should be understood that the terms "center", "lateral", "longitudinal", "front", "back", "left", "right", "up", "down", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the scope of protection of the present invention. When a component is referred to as being "fixed to" another component, it may be directly on the other component or there may also be a centered component. When a component is considered to be "connected" to another component, it may be directly connected to the other component or there may be a centered component at the same time. When a component is considered to be "set on" another component, it may be directly set on the other component or there may be a centered component at the same time. The terms "vertical", "horizontal", "left", "right" and similar expressions used herein are for illustrative purposes only.

[0053] like Figures 1 to 6 As shown, a motor production quality inspection system of the present invention comprises: a machine 1, a simulation module, a collection module, a processing module, a control module, and an execution module arranged on the machine 1;

[0054] Simulation module: used to simulate different temperature environments, such as different ambient temperatures in summer and winter, which may affect the test results; the simulation module includes: a temperature controller for adjusting the temperature and a temperature sensor for detecting the temperature.

[0055] Acquisition module: used to shoot the surface of the motor casing under different temperature environments to obtain test videos under different temperature environments to fully reflect the conditions of the motor casing surface under different temperature environments.

[0056] The acquisition module includes: a slide rail 2 fixedly mounted on the top of the machine platform 1, a bracket 3 slidably mounted on the slide rail 2, an industrial camera 4 mounted on the bracket 3, and a light source 5 mounted on the bracket 3;

[0057] Processing module: used to analyze the test video under each simulation environment to determine whether the motor housing has a fault under each simulation environment.

[0058] Control module: used to control the execution module to execute corresponding instructions according to the processing results of the processing module.

[0059] The steps for the processing module to determine whether the motor housing is faulty in each simulated environment are as follows:

[0060] S1. Determine the coordinate set of the key inspection areas on the surface of the motor housing; the key inspection areas include: areas that are prone to scratches, pits, deformation or other defects, such as the edges, corners and flat areas of the housing; and the nameplate recognition area, which is used to identify and confirm whether the logo on the nameplate is clear and complete.

[0061] The image of each key inspection area is obtained according to the coordinate set. The image is used as the basis for subsequent analysis to ensure accurate inspection of key parts of the motor housing surface.

[0062] S2, saving the images of all key detection areas obtained by the acquisition module into the first sequence;

[0063] Sort the images in the image set according to a preset rule to obtain a second sequence;

[0064] The preset rules include: Rule 1, sort by the importance of key areas, and give priority to the analysis of important areas; Rule 2, sort by the probability of defects and the severity of defects; among them, give priority to analyzing areas with high probability of failure or important areas, and conduct random inspections on low probability or general areas or secondary areas; sort the probabilities in random inspections to save working time; all major faults that can be detected are inspected, and all problems with high probability are inspected.

[0065] S3, identifying each image in the second sequence according to preset rules to determine whether the image meets the quality standards; the quality standards include: whether there are scratches, pits, deformation, uniform color, etc.;

[0066] S3.1. When rule 1 is used, when an image that does not meet the quality standard is identified, the unqualified position is marked until all inspection items are completed. After all inspections are completed, S4 is executed;

[0067] S3.2. When Rule 2 is used, when an image that does not meet the quality standard is identified, the unqualified position is marked and it is determined whether it is a serious defect;

[0068] S3.2.1. If it is not a serious defect, continue to inspect the high-probability area and randomly inspect the low-probability area. After the inspection is completed, execute S4;

[0069] S3.2.2: If it is a serious defect, go directly to S4;

[0070] S3.3. If each image in the second sequence meets the quality standard, execute S5.

[0071] S4. Output recognition results, which include defect type (such as scratches, pits, deformation, etc.), location (specific coordinate area), and severity assessment; according to the severity of the defect, it is divided into minor, moderate, severe and other levels for subsequent corresponding processing.

[0072] S5, output the recognition result, and control the execution module to execute the corresponding process based on the severity;

[0073] S5.1. When the product is qualified, the control execution module executes mode 1;

[0074] S5.2. When there are slight defects in the product, the control execution module executes mode 2;

[0075] S5.3. When the product has serious defects, the control execution module executes mode three.

[0076] Preferably, in step S2, the steps of preferentially selecting areas with high probability of failure or important areas for analysis and randomly sampling areas with low probability or secondary areas are as follows:

[0077] S2.1. Initialize the weights of the two sorting rules, set them as the regional importance weight Wi and the defect probability weight Wp respectively. In the initial state, set Wi>Wp, that is, the sorting of important areas according to rule one is initially preferred. For example, in the initial state, set Wi=1; when the detection starts, the system does not have enough historical data to accurately predict the defect probability of certain areas. At this time, rule one is used for comprehensive detection.

[0078] S2.2. After each inspection, the inspection results are saved in the database to form a historical defect data set; the data set contains the following information: the coordinates of the inspection area; the importance of the inspection area; the defect type (such as scratches, pits, etc.); the frequency of defect occurrence; the severity of the defect, etc.

[0079] S2.3, dynamically adjust the weight according to the test results during the test process;

[0080] S2.3.1, let the defect probability value be P, ranging from 0 to 1; the regional importance be S, 1 for slight, 2 for medium, and 3 for severe; the low probability threshold be T; the priority coefficient be D, ranging from 0 to 1; the comprehensive ranking value be W, then:

[0081]

[0082] When S=3, 100% inspection is performed unconditionally, because defects in important areas have a great impact on the product and must be discovered;

[0083] When S=2, the defects in the general area need to be paid special attention to. The lowest weight of the general area is set to A, and the highest weight of the general area is set to B. A is between 0.1 and 0.3; B is between 0.5 and 0.9;

[0084] When S=1, weights are set according to the probability for minor areas or areas with low probability of failure, and random inspections are performed;

[0085] The defect probability value P is obtained based on the number of times the specified area is detected and the number of times the defect occurs. For example, if the number of times the specified area is detected is 100 and the number of times the defect occurs in the area is 1, then P is 1÷100=0.01.

[0086] In the initial detection stage, the default weight Wi is relatively large. During each detection, the data accumulated as the defect detection is recorded. If the frequency of defects in certain areas is high, the weight value of the area in the sorting is increased; Wi and Wp are dynamically adjusted by the number of detection rounds, the severity of the defects and the frequency of occurrence, so that the system can automatically adjust the priority according to the actual working conditions; as the number of detections increases, the system accumulates a large amount of historical data, and at this time it can rely more on the defect probability weight; the defect probability weight helps the system dynamically adapt to the actual detection situation, giving priority to the detection of important areas with high risks, and avoiding wasting resources on secondary areas with low risks; if only based on the importance of the area, the system may ignore some high-risk but less important areas, resulting in delays in key issues; if only based on the defect probability, the system may rely too much on historical data and ignore some structurally critical but usually problem-free areas (such as nameplates and edges).

[0087] S2.4. Analyze the data set, establish a prediction model, calculate the defect probability for each area, dynamically adjust its detection priority to predict the probability of defects in each area, and dynamically adjust the sorting during actual detection. Predict the probability of defects in each area through statistical models and machine learning models, and adjust the sorting logic in Rule 2 according to the defect probability output by the model.

[0088] S2.5. As the test data accumulates, the training model is updated regularly, and the defect probability of each area is updated, so that the sorting rules are updated in real time; regularly means after a certain number of tests or after a certain period of time, for example: updating the sorting rules every hundred times, or updating the sorting rules every other day.

[0089] The display module includes: a display and a user control panel. The user can view the detection status in real time on the display and select the detection mode in real time on the control panel. The defect-frequently occurring areas, the confidence level of the detection results and the sorting priority are displayed to the user through the display to form a visual feedback. The user can also manually view the detection history and adjust the detection strategy.

[0090] S2.6. Based on the accuracy of past tests, the system can assign confidence levels to the results of each test. Areas with low confidence levels can be retested in subsequent tests. After the test is completed, if it is found that the test results are significantly different from the actual defect situation, the confidence level of the test model for that area will be reduced.

[0091] The failure probabilities of different parts of the motor casing are different. Prioritizing the detection of areas with high failure probabilities or important areas can quickly locate defects, optimize the detection process, concentrate resources to improve detection quality, reduce time and steps, and focus limited visual inspection equipment and inspection personnel's energy on areas with high failure probabilities or important areas to ensure detection quality while avoiding waste; at the same time, random sampling of low-probability areas to discover potential faults, thereby effectively balancing costs and benefits, ensuring comprehensive quality monitoring, and reducing the risk of quality accidents. Random sampling can comprehensively monitor quality, discover potential and uncommon fault conditions, and avoid quality accidents caused by ignoring problems; comprehensive detection of low-probability areas is costly, and random sampling can strike a balance between costs and benefits, and can serve as a supplementary means of quality risk assessment to facilitate timely adjustment of detection strategies.

[0092] Preferably, in step S5, mode one is to loosen the motor housing and transport the motor housing to the first channel; mode two is to loosen the motor housing and transport the motor housing to the second channel; mode three is to loosen the motor housing and transport the motor housing to the third channel.

[0093] The execution module includes: a clamping assembly 6, used for clamping the motor housing or driving the motor housing to rotate; a power assembly 7, used for pushing the motor housing toward channel one 8 in execution mode one, or pushing the motor housing toward channel two 9 in execution mode two, or pushing the motor housing toward channel three 10 in execution mode three.

[0094] The clamping assembly 6 includes: a first mounting seat 11, a first clamping plate 12 rotatably mounted on the first mounting seat 11, a first motor 13 fixedly connected to the first clamping plate 12, and the first motor 13 fixedly mounted on the first mounting seat 11; a second mounting seat 14 slidably mounted on the top of the machine table 1; a second clamping plate 15 rotatably mounted on the second mounting seat 14; a second motor 16 is threadedly connected to the second mounting seat 14; a third motor 17 is fixedly mounted on the second mounting seat 14, and the third motor 17 is connected to the second clamping plate 15 through a belt drive; the second motor 16 drives the second mounting seat 14 to move closer to or away from the first mounting seat 11, and drives the second clamping plate 15 to rotate through the third motor 17, and drives the first clamping plate 12 and the second clamping plate 15 to rotate synchronously through the first motor 13.

[0095] A turntable 18 is provided between the first mounting seat 11 and the second mounting seat 14; a second gear 20 is coaxially connected to the bottom of the turntable 18; the second gear 20 is connected to the third gear 21 through a synchronous belt transmission; the third gear 21 is connected to the fourth motor 22; the turntable 18 is driven to rotate by the fourth motor 22; and the motor housing is then driven to rotate.

[0096] The power assembly 7 includes: a third mounting seat 23; a telescopic cylinder 24 fixedly mounted on the top of the third mounting seat 23; a connecting rod 25 fixedly connected to the piston rod of the telescopic cylinder 24, and the connecting rod 25 is fixedly connected to the bracket 3; a rack 26 is fixedly connected to the bottom of the bracket 3; the rack 26 is meshingly connected to the first gear 27; the first gear 27 is coaxially connected to a material-prying and air-blowing member 28, and the material-prying and air-blowing member 28 is located on the feeding side of the first channel; the material-prying and air-blowing member 28 is connected to an air nozzle through a hose 29, and the air nozzle is used to blow air to the lens of the industrial camera 4; the material-prying and air-blowing member 28 is composed of an air storage chamber 30 and a guide plate 31.

[0097] When the telescopic cylinder 24 drives the bracket 3 to slide along the slide rail 2, the bracket 3 drives the rack 26 to slide synchronously, and the rack 26 drives the first gear 27 to rotate, thereby driving the material-pickling and air-blowing member 28 to swing. At the same time, the material-pickling and air-blowing member 28 contacts the second channel or the third channel and is squeezed (a block may also be provided to squeeze the material-pickling and air-blowing member 28), so that the gas is blown from the air nozzle toward the industrial camera 4 to clean the floating dust at the shooting end of the industrial camera 4.

[0098] The standard parts used in the present invention can all be purchased from the market, and special-shaped parts can be customized according to the description in the specification and the drawings. The specific connection methods of each part adopt conventional means such as mature bolts, rivets, welding, etc. in the prior art. Machinery, parts and equipment all adopt conventional models in the prior art, and the circuit connection adopts the conventional connection method in the prior art, which will not be described in detail here. The content not described in detail in this specification belongs to the prior art known to professional and technical personnel in this field.

[0099] It should be understood that those skilled in the art can make improvements or changes based on the above description, and all these improvements and changes should fall within the scope of protection of the appended claims of the present invention.

Claims

1. A motor production quality inspection system, characterized in that: include: Machine; Acquisition module: used to shoot the surface of the motor housing to obtain the test video; Processing module: used to analyze the test video and determine whether the motor housing is faulty; Control module: used to control the execution module to execute corresponding instructions according to the processing result of the processing module; The acquisition module includes: a slide rail fixedly mounted on the top of the machine, a bracket slidably mounted on the slide rail, and an industrial camera mounted on the bracket; The execution module includes: a power assembly; the power assembly includes: a third mounting seat; a telescopic cylinder fixedly mounted on the third mounting seat; a connecting rod fixedly connected to the piston rod of the telescopic cylinder, and the connecting rod is fixedly connected to the bracket; a rack is fixedly connected to the bottom of the bracket; the rack is meshingly connected to a first gear; the first gear is coaxially connected to a material-prying and air-blowing piece, and the material-prying and air-blowing piece is connected to an air nozzle.

2. A motor production quality inspection system as claimed in claim 1, characterized in that: The execution module also includes: a clamping assembly; the clamping assembly includes: a first mounting seat, a first clamping plate rotatably mounted on the first mounting seat, a first motor that drives the first clamping plate to rotate; a second mounting seat slidably mounted on the top of the machine; a second clamping plate rotatably mounted on the second mounting seat; a second motor that drives the second mounting seat to slide; and a third motor that drives the second clamping plate to rotate.

3. A motor production quality inspection system as claimed in claim 1, characterized in that: The steps of the processing module to determine whether the motor housing is faulty are as follows: S1. Determine a coordinate set of a key detection area on the surface of the motor housing; the key detection area includes: a location prone to scratches, pits, deformation or other defects; Acquire an image of each key detection area according to the coordinate set; S2, saving the images of all key detection areas obtained by the acquisition module into a first sequence; Sort the images in the image set according to a preset rule to obtain a second sequence; The preset rules include: Rule 1, sorting according to the importance of key areas; Rule 2, sorting according to the probability of defects and the severity of defects; S3, identifying each image in the second sequence according to the preset rule to determine whether the image meets the quality standard; S4. Outputting the recognition result, wherein the recognition result includes the type, location, and severity assessment of the defect; the defect is classified into mild, moderate, severe, etc. according to the severity of the defect; S5. Output the recognition result, and control the execution module to execute the corresponding process based on the severity.

4. A motor production quality inspection system as claimed in claim 3, characterized in that: The step S3 also includes the following steps: S3.

1. When the rule 1 is adopted, when an image that does not meet the quality standard is identified, the unqualified position is marked until all the inspection items are completed. After all the inspections are completed, S4 is executed; S3.

2. When the second rule is adopted, when an image that does not meet the quality standard is identified, the unqualified position is marked and it is determined whether it is a serious defect; S3.2.

1. If it is not a serious defect, continue to inspect the high-probability area and randomly inspect the low-probability area. After the inspection is completed, execute S4; S3.2.2: If it is a serious defect, go directly to S4; S3.

3. If each image in the second sequence meets the quality standard, execute S5.

5. A motor production quality inspection system as claimed in claim 3, characterized in that: The step S2 further comprises the following steps: S2.

1. Initialize the weights of the rule 1 and the rule 2, respectively setting the regional importance weight Wi and the defect probability weight Wp. In the initial state, set Wi>Wp; S2.

2. After each inspection, the inspection results are saved in the database to form a historical defect data set; the data set contains the following information: coordinates of the inspection area; importance of the inspection area; defect type; frequency of defect occurrence; severity of the defect, etc.; S2.

3. During the detection process, the weights are dynamically adjusted according to the detection results.

6. A motor production quality inspection system as claimed in claim 5, characterized in that: The step S2.3 includes the following steps: let the defect probability value be P, ranging from 0 to 1; let the regional importance be S, 1 for slight, 2 for medium, and 3 for severe; let the low probability threshold be T; let the priority coefficient be D, ranging from 0 to 1; The comprehensive ranking value is W, then When S=3, 100% detection is performed unconditionally; When S=2, the defects in the general area need to be paid special attention to, where the lowest weight of the general area is A, and the highest weight of the general area is B. A is between 0.1 and 0.3; B is between 0.5 and 0.9; When S=1, random inspections are conducted on secondary areas or low-probability areas.

7. A motor production quality inspection system as claimed in claim 6, characterized in that: Also includes: S2.

4. Analyze the data set, establish a prediction model, calculate the defect probability for each area, and dynamically adjust the detection priority to predict the probability of defects in each area; S2.

5. Regularly update the training model and the defect probability of each area; S2.

6. Assign confidence levels to the results of each test based on the accuracy of past tests; areas with lower confidence levels can be retested in subsequent tests; After the inspection is completed, if it is found that the inspection results are significantly different from the actual defect situation, the confidence of the inspection model in this area is reduced.

8. A motor production quality inspection system as claimed in claim 3, characterized in that: The control module controls the execution of the module to execute the corresponding instruction including: S5.

1. When the product is qualified, control the execution module to execute mode 1; S5.

2. When the product has a slight defect, the execution module is controlled to execute mode 2; S5.

3. When the product has serious defects, control the execution module to execute mode three; The first mode is to loosen the motor housing and transport it to the first channel; the second mode is to loosen the motor housing and transport it to the second channel; the third mode is to loosen the motor housing and transport it to the third channel.

Images