An automatic detection system for processed parts based on data analysis

Abstract

The present invention relates to the field of processing and detection, and discloses an automatic detection system for processed parts based on data analysis, including a detection module for obtaining detection data; a detection data processing and analysis module for determining whether the outer surface quality of the processed part is qualified or unqualified; a secondary verification module for performing secondary verification on the processed part with unqualified outer surface quality. If the outer surface quality of the processed part is determined to be qualified after secondary determination, no alarm is given. Otherwise, it is determined that the outer surface quality of the processed part is unqualified; an alarm module for sending an alarm signal to the processed part determined to have unqualified outer surface quality by the secondary verification module. The present invention first makes a preliminary judgment on the detection data through the detection data processing and analysis module, and then performs a review through the secondary verification module, so as to prevent detection errors caused by impurities on the surface of the processed part. Finally, an alarm is issued through the alarm module, and the processed part is placed in the unqualified area.

Description

Technical Field

[0001] The present invention relates to the field of processing and detection, and particularly relates to an automatic detection system for processed parts based on data analysis. Background Art

[0002] Existing mechanical processed parts need to be detected during and after processing. The detection is divided into internal detection and appearance detection. The purpose of internal detection is to detect whether there are hidden damages or cracks inside the processed parts, while appearance detection basically includes aspects such as cracks, surface quality, and dimensions. Generally speaking, both internal detection and appearance detection determine the ex-factory quality of the processed parts.

[0003] For round-shaped rod-shaped, shaft-shaped, or disk-shaped processed parts, whether the shape of the outer surface is regular is an important factor affecting the quality of the processed parts. However, in the actual detection process, the general method for detecting the outer cylindrical surface is to measure it by placing it in a go-no-go gauge. However, the measurement result of the go-no-go gauge will be affected by impurities attached to the surface of the processed part, resulting in detection errors and inaccurate overall detection results. Summary of the Invention

[0004] The purpose of the present invention is to provide an automatic detection system for processed parts based on data analysis to solve the above technical problems.

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] An automatic detection system for processed parts based on data analysis, comprising:

[0007] A detection module for detecting the outer surface of the processed part and obtaining detection data;

[0008] A detection data processing and analysis module for processing the detection data to obtain the outer surface quality coefficient of the processed part, and comparing the outer surface quality coefficient with a preset outer surface quality range. If the outer surface quality coefficient is within the outer surface quality range, it is determined that the outer surface quality of the processed part is qualified; otherwise, it is determined that the outer surface quality of the processed part is unqualified;

[0009] A secondary verification module for performing secondary verification on the processed part with unqualified outer surface quality. The process of the secondary verification is as follows:

[0010] Clean the outer surface of the current processed part through a pre-installed cleaning module, and then perform secondary detection through the detection module to obtain secondary detection data;

[0011] After processing the secondary detection data through the detection data processing and analysis module, obtain the secondary outer surface quality coefficient;

[0012] If the secondary outer surface quality coefficient is within the outer surface quality range, the outer surface quality of the workpiece is judged to be qualified for the second time; otherwise, the outer surface quality of the workpiece is judged to be unqualified for the second time.

[0013] An alarm module that sends an alarm signal for workpieces whose outer surface quality is judged to be unqualified by the secondary verification module.

[0014] As a further technical solution, the detection module includes:

[0015] Laser emitters, multiple laser emitters are provided and evenly distributed on the outer periphery of one end of the current workpiece;

[0016] A rotating disk, the laser emitters are installed on the rotating disk, a display board is provided at the other end of the workpiece, and the display board is used to display the laser trajectory obtained by the rotation of the laser emitters along the outer periphery of the workpiece;

[0017] A driving member, the driving member is in transmission connection with the rotating disk and is used to drive the rotating disk and the laser emitters to rotate;

[0018] A camera unit for collecting the image formed by the laser trajectory and using it as detection data.

[0019] As a further technical solution, the process of processing the detection data is as follows:

[0020] The image is processed into a standard format to obtain an actual picture, and the actual picture is compared with a preset standard picture;

[0021] Obtain the ratio K of the area of the overlapping part of the actual picture and the standard picture to the total area and the arc length L of the missing part in the actual picture;

[0022] Input K and L into the quality detection model to obtain the outer surface quality coefficient F of the current workpiece.

[0023] As a further technical solution, the expression of the quality detection model is:

[0024]

[0025] Among them, x(L) is a judgment function regarding the arc length L of the missing part in the actual picture. When L > 0, x(L) = 1 + L * δ b , when L = 0, x(L) = 1, δ a 、δ b are preset proportionality coefficients, K0 is a standard value, and F is the outer surface quality coefficient.

[0026] As a further technical solution, the system further includes a marking module for marking the processed parts judged to be unqualified. The markings include red markings and yellow markings. Among them, the objects of the red markings are the processed parts judged to be unqualified in external surface quality by the secondary verification module, and the objects of the yellow markings are the processed parts judged to be unqualified in external surface quality by the detection data processing and analysis module.

[0027] As a further technical solution, the system further includes a quality prediction module for the processed parts, which performs quality prediction and analysis on the processed parts based on the historical data of the marking module, and compares the quality prediction value Q calculated according to the quality prediction and analysis with the quality prediction threshold Q0;

[0028] If Q≥Q0, judge the quality trend difference of the processed parts in the next unit time. If Q<Q0, judge the quality trend of the processed parts in the next unit time is good.

[0029] As a further technical solution, the process of performing quality prediction and analysis on the processed parts based on the historical data of the marking module is as follows:

[0030] Obtain the number M of red markings within m consecutive unit times traced back from the current moment by the marking module i , the number M of yellow markings j ;

[0031] Obtain the number M of the total detected processed parts within each unit time z ;

[0032] Draw the curve M of the number of red markings changing with time i (t), the curve M of the number of yellow markings changing with time j (t) and the curve M of the number of the total detected processed parts changing with time z (t);

[0033] Through the formula:

[0034]

[0035] Calculate the quality prediction value Q;

[0036] Among them, A r is the change amount of the red markings, B r is the change amount of the yellow markings, C r is the change amount of the total detected processed parts; θ, μ, σ are weight coefficients, and r is the r-th unit time.

[0037] As a further technical solution, the expression of the A r is:

[0038] The said B r The expression of is:

[0039] The said C r The expression of is:

[0040] Wherein, t0 and t1 are respectively the starting time and the ending time of the unit time.

[0041] Advantages of the present invention:

[0042] In the present invention, the detection data processing and analysis module first makes a preliminary judgment on the detection data, and then the secondary verification module conducts a review, so as to prevent detection errors caused by impurities on the surface of the workpiece. According to the judgment result of the detection data processing and analysis module, the cleaning module is selectively started, aiming to avoid detection errors caused by possible impurity attachment on the outer surface. On this basis, detection and analysis are carried out again. If it is still judged that the quality is unqualified, it means that the workpiece is indeed unqualified. Finally, an alarm is issued through the alarm module, and the workpiece is placed in the unqualified area. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] The present invention will be further described below with reference to the drawings.

[0044] Figure 1 is the system logic schematic diagram of the present invention;

[0045] Figure 2 is the working schematic diagram of the detection module. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0046] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0047] Please refer to Figure 1 - Figure 2 as shown, the present invention is an automatic detection system for workpieces based on data analysis, including:

[0048] A detection module for detecting the outer surface of the workpiece and obtaining detection data;

[0049] A detection data processing and analysis module for processing the detection data to obtain the outer surface quality coefficient of the workpiece, and comparing the outer surface quality coefficient with a preset outer surface quality range. If the outer surface quality coefficient is within the outer surface quality range, it is determined that the outer surface quality of the workpiece is qualified; otherwise, it is determined that the outer surface quality of the workpiece is unqualified.

[0050] The secondary verification module performs secondary verification on the workpieces whose outer surface quality is judged to be unqualified. The process of the secondary verification is as follows:

[0051] The outer surface of the current workpiece is cleaned by the pre-installed cleaning module, and then a secondary inspection is performed by the inspection module to obtain secondary inspection data;

[0052] After processing the secondary test data through the test data processing and analysis module, the secondary outer surface quality coefficient is obtained;

[0053] If the secondary outer surface quality coefficient is within the outer surface quality range, the outer surface quality of the workpiece is secondarily judged to be qualified, otherwise the outer surface quality of the workpiece is secondarily judged to be unqualified;

[0054] The alarm module sends an alarm signal to the processed parts whose outer surface quality is judged to be unqualified by the secondary verification module.

[0055] In this embodiment, the detection data is first preliminarily judged by the detection data processing and analysis module, and then reviewed by the secondary verification module, so as to prevent detection errors caused by impurities on the surface of the workpiece, and selectively start the cleaning module according to the judgment result of the detection data processing and analysis module. The cleaning module can use a high-pressure air pump to blow out airflow for cleaning or use other similar cleaning effects, which are not limited here; there should be no impurities on the surface of the workpiece after cleaning by the cleaning module. Compared with the method of cleaning before detection, the cleaning module is a newly added module on the original processing production line to prevent detection errors. In actual use, if each time Cleaning all parts may slow down the detection efficiency, so only the workpieces that are initially judged to be unqualified are cleaned, with the aim of avoiding detection errors caused by impurities that may be attached to the outer surface. On this basis, inspection and analysis are performed again. If the quality is still judged to be unqualified, it means that the workpiece is indeed unqualified. Finally, an alarm is issued through the alarm module, and the workpiece is placed in the unqualified area. Through the above technical scheme, dual detection of the detection data processing and analysis module and the secondary verification module is utilized, so that on the one hand, detection errors caused by non-processing problems on the outer surface of the workpiece can be avoided as much as possible, and on the other hand, the alarm accuracy can be improved through double inspection to prevent misjudgment.

[0056] The detection module comprises:

[0057] A laser emitter, wherein a plurality of laser emitters are provided and are evenly spaced and distributed on the periphery of one end of the workpiece being processed;

[0058] A rotating disk, on which the laser emitter is mounted. At the other end of the workpiece, there is a display board for displaying the laser trajectory obtained by the rotation of the laser emitter along the outer circumference of the workpiece.

[0059] A driving member, which is in transmission connection with the rotating disk and is used to drive the rotation of the rotating disk and the laser emitter.

[0060] An imaging unit for collecting the image formed by the laser trajectory and serving as detection data.

[0061] In this embodiment, the specific working mode of the detection module is provided. The outer circular surface of the workpiece is detected by driving the rotating disk to drive multiple laser emitters on the rotating disk. Preferably, there are 1 - 2 laser emitters. The laser beam emitted by the laser emitter is close to the outer circular surface of the workpiece and rotates one full circle or half a circle starting from the starting point. According to the moving trajectory of the laser beam, it can be judged whether the outer circular surface of the workpiece is regular. Because when there are local protrusions on the outer surface of the workpiece, the laser beam will be blocked, resulting in a break in the moving trajectory of the laser beam at that place, thus achieving the purpose of detection. The moving trajectory of the laser beam is compared with the standard trajectory. The moving trajectory of the laser beam tends to be circular. The higher the coincidence degree with the standard circle, the more standard it is.

[0062] The process of processing the detection data is as follows:

[0063] The image is processed into a standard format to obtain the actual picture, and the actual picture is compared with the preset standard picture.

[0064] Obtain the ratio K of the area of the overlapping part of the actual picture and the standard picture to the total area and the arc length L of the missing part in the actual picture.

[0065] Input K and L into the quality detection model to obtain the outer surface quality coefficient F of the current workpiece.

[0066] The expression of the quality detection model is:

[0067]

[0068] Among them, x(L) is a judgment function regarding the arc length L of the missing part in the actual picture. When L > 0, x(L) = 1 + L * δ b , when L = 0, x(L) = 1, δ a 、δ b are preset proportionality coefficients, K0 is the standard value, and F is the outer surface quality coefficient.

[0069] In this embodiment, a method for processing detection data is provided. By detecting the acquired picture and the standard picture, the difference between the two pictures can be compared, and the ratio K of the area of the overlapping part to the total area and the arc length L of the missing part in the actual picture can be extracted from the comparison process. The above two features can directly reflect the quality of the outer circular surface of the workpiece. Then, through the quality detection model the outer surface quality coefficient of the workpiece can be obtained. If it is within the interval, it means the quality of the workpiece is qualified; otherwise, it means the quality of the workpiece is unqualified. l i is the arc length occupied by each break in the trajectory image, and i is the i-th one. Obviously, the larger the total arc length L, the worse the outer surface quality; the larger the |K - K0|, the better the outer surface quality.

[0070] The system further includes a marking module for marking the workpieces judged to be unqualified. The markings include red markings and yellow markings. Among them, the objects of the red markings are the workpieces whose outer surface quality is judged to be unqualified by the secondary verification module, and the objects of the yellow markings are the workpieces whose outer surface quality is judged to be unqualified by the detection data processing and analysis module.

[0071] The system further includes a workpiece quality prediction module for performing quality prediction and analysis of the workpiece based on the historical data of the marking module, and comparing the quality prediction value Q calculated according to the quality prediction and analysis with the quality prediction threshold Q0;

[0072] If Q ≥ Q0, it is judged that the quality trend of the workpiece in the next unit time is poor; if Q < Q0, it is judged that the quality trend of the workpiece in the next unit time is good.

[0073] The process of performing quality prediction and analysis of the workpiece based on the historical data of the marking module is as follows:

[0074] Obtain the number M of red markings of the marking module continuously traced back m unit times backward from the current moment i and the number M of yellow markings j ;

[0075] Obtain the number M of the total detected workpieces in each unit time z ;

[0076] Draw the curve M i (t) of the number of red markings changing with time, the curve M j (t) of the number of yellow markings changing with time, and the curve M z (t) of the number of the total detected workpieces changing with time;

[0077] Through the formula:

[0078]

[0079] Calculate the quality prediction value Q;

[0080] Wherein, A r is the change amount of the red mark, B r is the change amount of the yellow mark, C r is the change amount of the total detected workpieces; θ, μ, σ are weight coefficients, and r is the r-th unit time.

[0081] The said A r has the following expression:

[0082] The said B r has the following expression:

[0083] The said C r has the following expression:

[0084] Wherein, t0 and t1 are respectively the starting time and the ending time of the unit time.

[0085] In this embodiment, the marking module marks the workpieces that are initially judged unqualified and secondarily judged unqualified in the processing, so as to evaluate the cleaning degree of the workpieces before subsequent detection, and at the same time, it is also convenient for the regional processing of the workpieces and easy to distinguish; at the same time, according to the change situation of the number of the two marks in the historical data, the overall change amount is comprehensively obtained, so as to predict the situation of the workpieces according to the size of the change amount. Through the formula calculate the quality prediction value Q. Obviously, the larger the quality prediction value Q is, the worse the quality trend of the workpieces in the next unit time is. Therefore, it is necessary to check back the problems in the processing equipment to prevent the quality problems of batch workpieces and avoid the expansion of the loss scale.

[0086] It should be noted that: the calculation formulas and each parameter participating in the operation in the present invention have been pre-dimensionless processed, and the process of dimensionless processing is well known in the industry and will not be described here.

[0087] The above has described a detailed description of an embodiment of the present invention, but the content described is only the preferred embodiment of the present invention and cannot be considered as limiting the implementation scope of the present invention. All equivalent changes and improvements made according to the scope of the present invention application should still fall within the scope covered by the patent of the present invention.

Claims

1. An automatic detection system for workpieces based on data analysis, characterized in that, Including: A detection module, which is used to detect the outer surface of the workpiece and obtain detection data; A detection data processing and analysis module, which is used to process the detection data, obtain the outer surface quality coefficient of the workpiece, and compare the outer surface quality coefficient with a preset outer surface quality range. If the outer surface quality coefficient is within the outer surface quality range, it is determined that the outer surface quality of the workpiece is qualified; otherwise, it is determined that the outer surface quality of the workpiece is unqualified; The process of processing the detection data is as follows: The image is processed into a standard format to obtain an actual image, and the actual image is compared with a preset standard image; Obtain the ratio of the area of the overlapping part of the actual image and the standard image to the total area and the arc length of the missing part in the actual image ; Input and into the quality inspection model to obtain the outer surface quality coefficient of the current workpiece being processed ; The expression of the quality detection model is: ; Among them, is the arc length of the missing part in the actual picture is the judgment function. When > 0, , when = 0, , , is a preset proportionality coefficient, is the standard value, is the outer surface quality coefficient, is the arc length occupied by each fracture in the trajectory image, is the th; The secondary verification module performs secondary verification on the processed parts with unqualified outer surface quality. The process of the secondary verification is as follows: The outer surface of the current workpiece is cleaned by a pre-installed cleaning module, and then secondary detection is performed by the detection module to obtain secondary detection data; After the secondary detection data is processed by the detection data processing and analysis module, a secondary outer surface quality coefficient is obtained; If the secondary outer surface quality coefficient is within the outer surface quality range, it is secondarily determined that the outer surface quality of the workpiece is qualified; otherwise, it is secondarily determined that the outer surface quality of the workpiece is unqualified; An alarm module, which sends an alarm signal to the workpiece whose outer surface quality is judged to be unqualified by the secondary verification module.

2. The automatic detection system for workpieces based on data analysis according to claim 1, characterized in that The detection module includes: Laser emitters, multiple laser emitters are provided and evenly distributed on the outer periphery of one end of the current workpiece; A rotating disk, the laser emitters are installed on the rotating disk, and a display board is provided at the other end of the workpiece. The display board is used to display the laser trajectory obtained by the laser emitters rotating along the outer periphery of the workpiece; A driving member, the driving member is in transmission connection with the rotating disk and is used to drive the rotating disk and the laser emitters to rotate; An imaging unit, which is used to collect the image formed by the laser trajectory and use it as detection data.

3. The automatic detection system for workpieces based on data analysis according to claim 1, characterized in that, The system further includes a marking module, which is used to mark the workpieces judged to be unqualified. The markings include red markings and yellow markings. Among them, the objects of the red markings are the workpieces whose outer surface quality is judged to be unqualified by the secondary verification module, and the objects of the yellow markings are the workpieces whose outer surface quality is judged to be unqualified by the detection data processing and analysis module.

4. The automatic detection system for workpieces based on data analysis according to claim 3, wherein The system further includes a workpiece quality prediction module, which performs quality prediction analysis on the workpiece based on the historical data of the marking module, and calculates the quality prediction value obtained from the quality prediction analysis and the quality prediction threshold for comparison; If , judge the quality trend difference of the workpiece to be processed in the next unit time. If , judge that the quality trend of the workpiece to be processed in the next unit time is good.

5. The automatic detection system for workpieces based on data analysis according to claim 4, wherein The process of quality prediction and analysis of workpieces based on the historical data of the marking module is as follows: Obtain the number of red marks within consecutive unit times traced back from the current moment by the marking module , and the number of yellow marks ; Obtain the number of total detected and processed parts per unit time ; Plot the curve of the number of red markers over time per unit time , the curve of the number of yellow markers over time and the curve of the number of total detected workpieces over time ; By the formula: calculate to obtain the mass prediction value ; wherein, is the change amount of the red mark, is the change amount of the yellow mark, is the change amount of the total detected workpieces; and and are the weight coefficients, is the th unit time.

6. The automatic detection system for workpieces based on data analysis according to claim 5, characterized in that The said The expression of: ; The expression is: ; The said has the expression: ; Among them, and are the starting time and the ending time per unit time, respectively.

Images