Circuit board welding quality detection system

Abstract

The invention discloses a circuit board welding quality detection system, and relates to the technical field of circuit board detection. The circuit board image is divided into areas according to function modules, the welding spot circularity is analyzed for each sub-area, and the welding spots with abnormal shapes are found through comparison with a reference sample; the welding quality is accurately evaluated from the shape and the position, omissions of general evaluation are avoided, the detection precision is improved, and the problems that in the prior art, only general evaluation is conducted on the welding quality of the circuit board, targeted evaluation cannot be conducted on all functional sub-areas of the circuit board, and the detection precision is poor are solved. And the intelligent degree of quality detection is relatively low.

Description

Technical Field

[0001] The present application relates to the technical field of circuit board detection, and in particular to a circuit board welding quality detection system. Background Art

[0002] During the production process of circuit boards, the welding quality directly affects the performance and reliability of circuit boards. With the continuous improvement of the integration level of circuit boards, higher requirements are put forward for the accuracy and efficiency of welding quality detection.

[0003] However, the existing circuit board welding quality detection systems still have the following deficiencies in the actual application process:

[0004] The detection systems in the existing technology only conduct a broad and general evaluation of the circuit board welding quality, and fail to conduct targeted analysis in each functional sub-region of the circuit board. Since the circuit structures and welding process requirements of different functional modules on the circuit board, such as power supply modules, signal processing modules, storage modules, etc., are significantly different, the degree of intelligence of quality detection is low;

[0005] In addition, in terms of electrical performance evaluation, the detection systems in the existing technology ignore the influence of transmission delay on the performance of circuit boards. In high-speed circuits, if the transmission delay does not meet the requirements, it will cause signal asynchronization, leading to problems such as data transmission errors and unstable system operation. On the other hand, the detection methods for component parameters such as resistors, capacitors, and inductors are not scientific and rigorous enough. They often simply measure a single value without considering measurement errors and random interference factors.

[0006] Therefore, a circuit board welding quality detection system is introduced. Summary of the Invention

[0007] In order to solve the problems proposed in the background art, the present application provides a circuit board welding quality detection system.

[0008] A circuit board welding quality detection system provided by the present application adopts the following technical solutions: A circuit board welding quality detection system includes:

[0009] Quality detection module: including a sample screening unit, a welding analysis unit, and an electrical detection unit;

[0010] The sample screening unit is used to count the total number of circuit boards produced in the current batch, and determine the quantity extraction ratios of welding analysis samples and electrical detection samples according to the total number range, so as to extract the corresponding proportional quantities of welding analysis samples and electrical detection samples;

[0011] The welding analysis unit is used to collect images of the extracted welding analysis samples and preprocess each group of collected sample images; after the preprocessing is completed, comprehensive analysis is performed on each group of sample images to obtain the welding evaluation index hju of the circuit boards in the current production batch.

[0012] The electrical detection unit is used to collect electrical parameters of the extracted electrical detection samples; the electrical parameters include resistance value, capacitance value, inductance value, and transmission delay; after the collection is completed, comprehensive evaluation is performed on the electrical parameters of each electrical detection sample to obtain the electrical performance index hjr of the circuit boards in the current production batch.

[0013] Optionally, the extraction ratio of the number of welding analysis samples and electrical detection samples is determined according to the total number range, specifically:

[0014] For each group of quantities corresponding to the preset number of circuit boards, a range of each group of quantities is set, and each range of each group of quantities corresponds to a set of extraction ratios; the set of extraction ratios includes the extraction ratios corresponding to the welding analysis samples and the electrical detection samples respectively.

[0015] Match the total number of circuit boards produced in the current batch with the preset ranges of each group of quantities to obtain the set of extraction ratios for the circuit boards produced in the current batch; multiply the total number of circuit boards produced in the current batch by the extraction ratio of the welding analysis samples in the set of extraction ratios to determine the extraction quantity of the welding analysis samples.

[0016] Multiply the total number of circuit boards produced in the current batch by the extraction ratio of the electrical detection samples in the set of extraction ratios to determine the extraction quantity of the electrical detection samples.

[0017] Optionally, after the preprocessing is completed, comprehensive analysis is performed on each group of sample images, specifically:

[0018] Pre-input the circuit board image samples with qualified welding quality as the reference quality samples for the circuit boards in the current production batch.

[0019] According to the pre-set divided areas, divide each group of sample images into respective functional sub-areas.

[0020] For each functional sub-area of each group of sample images, extract the solder joint edge contours of each functional sub-area in each group of sample images and obtain the roundness of each solder joint in each functional sub-area of each group of sample images.

[0021] From each functional sub-area of the reference quality samples, respectively extract the roundness with the largest difference from the integer one among the roundness of each solder joint as the passing roundness of each functional sub-area of the sample images.

[0022] Compare the circularity of each solder joint in each functional sub-region of each group of sample images with the corresponding passing circularity. If the circularity of a solder joint in a certain functional sub-region of a certain group of sample images is lower than the corresponding passing circularity, it is determined as a solder joint with abnormal shape.

[0023] Count the number of solder joints with abnormal shape in each functional sub-region of each group of sample images, denoted as the number of shape abnormalities; set a weight coefficient corresponding to the number of shape abnormalities for each different functional sub-region; multiply the number of shape abnormalities in each functional sub-region of each group of sample images by the corresponding weight coefficient, and then sum to obtain the welding shape abnormality index of each group of sample images.

[0024] Optionally, obtain the welding evaluation index hju of the current production batch of circuit boards, specifically:

[0025] For the reference quality samples, obtain the geometric center of each solder joint contour as the center coordinates of each solder joint.

[0026] At the same time, obtain the center coordinates of each solder joint in each group of sample images, match the center coordinates of each solder joint in each group of sample images with the corresponding solder joint center coordinates in the reference quality samples, and after the matching is completed, obtain the solder joint pairs of each solder joint in each group of sample images, that is, (sample image solder joint center coordinates, reference solder joint center coordinates).

[0027] Calculate the differences in the horizontal and vertical coordinates for each group of solder joint pairs respectively, and set the allowable position deviation thresholds corresponding to the horizontal and vertical coordinates.

[0028] Compare the differences in the horizontal and vertical coordinates of each group of solder joint pairs with the corresponding set allowable position deviation thresholds respectively. If there is a group of differences greater than the corresponding set allowable position deviation threshold, mark the solder joint pairs with differences greater than the allowable position deviation threshold as solder joints with abnormal position.

[0029] Count the number of solder joints with abnormal position in each functional sub-region of each group of sample images, denoted as the number of position abnormalities; set a weight coefficient corresponding to the number of position abnormalities for each different functional sub-region, multiply the number of position abnormalities in each functional sub-region of each group of sample images by the corresponding weight coefficient, and then sum to obtain the welding position abnormality index of each group of sample images.

[0030] Extract the welding shape abnormality index and welding position abnormality index corresponding to each group of sample images, and mark them as E1 and E2 respectively; according to the formula Perform weighted calculation on the welding shape abnormality index E1 and welding position abnormality index E2 of each group of sample images to obtain the welding effect index E3 corresponding to each group of sample images.

[0031] Take the mean value of the welding effect index E3 of each group of sample images to obtain the welding evaluation index hju of the current production batch of circuit boards.

[0032] Optionally, comprehensively evaluate the electrical parameters of each electrical test sample, specifically as follows:

[0033] The number of measurements G corresponding to the preset resistance value, capacitance value, and inductance value;

[0034] According to the number of measurements G, obtain G groups of resistance values, capacitance values, and inductance values of each electrical test sample; and calculate the mean values of the G groups of resistance values, capacitance values, and inductance values of each electrical test sample respectively to obtain the resistance comprehensive value, capacitance comprehensive value, and inductance comprehensive value of each electrical test sample;

[0035] Preset the nominal values corresponding to the resistance comprehensive value, capacitance comprehensive value, and inductance comprehensive value respectively, and calculate the resistance deviation rate D R and capacitance deviation rate D C and inductance deviation rate D L ; that is, calculate through the formula where R 综合 and R 标称 represent the resistance comprehensive value and the corresponding preset nominal value respectively, C 综合 and C 标称 represent the capacitance comprehensive value and the corresponding preset nominal value respectively, L 综合 and L 标称 represent the inductance comprehensive value and the corresponding preset nominal value respectively;

[0036] Set the weight coefficients corresponding to the resistance deviation rate D R , capacitance deviation rate D C and inductance deviation rate D L respectively, multiply the resistance deviation rate D R , capacitance deviation rate D C and inductance deviation rate D L of each electrical test sample by the corresponding set weight coefficients respectively, and then sum to obtain the component performance index Dt of each electrical test sample.

[0037] Optionally, obtain the electrical performance index hjr of the current production batch of circuit boards, specifically as follows:

[0038] Input a set excitation signal to each electrical test sample, mark the time point of inputting the excitation signal as the starting point, mark the time point of outputting the excitation signal as the ending point, and use the time difference between the starting point and the ending point as the transmission delay;

[0039] Set the number of times of inputting the excitation signal, and obtain the transmission delay corresponding to each input of the excitation signal for each electrical test sample, calculate the mean value of each group of transmission delays of each electrical test sample to obtain the transmission delay comprehensive value De of each electrical test sample;

[0040] According to the formula Perform weighted calculation on the component performance index Dt and the comprehensive transmission delay value De of each electrical test sample to obtain the electrical comprehensive index Dy of each electrical test sample; where Da and Db are respectively the preset performance allowable index and delay allowable value; β1 and β2 are respectively the influence weight factors of the component performance index Dt and the comprehensive transmission delay value De.

[0041] Take the mean value of the electrical comprehensive index Dy of each electrical test sample to obtain the electrical performance index hjr of the circuit board of the current production batch.

[0042] Optionally, it further includes:

[0043] Quality assessment module: Receive the welding assessment index hju and the electrical performance index hjr of the circuit board of the current production batch, so as to output the welding quality grade and the electrical quality grade of the circuit board of the current production batch, and send the output welding quality grade and electrical quality grade to the management personnel.

[0044] Optionally, output the welding quality grade and the electrical quality grade of the circuit board of the current production batch, specifically:

[0045] Three groups of index value ranges corresponding to the preset welding assessment index hju and the electrical performance index hjr are set. Each group of index value ranges corresponding to the welding assessment index hju corresponds to a welding quality grade, and each group of index value ranges corresponding to the electrical performance index hjr corresponds to an electrical quality grade; where the welding quality grades include poor welding grade, passing welding grade, and excellent welding grade, and the electrical quality grades include poor electrical grade, passing electrical grade, and excellent electrical grade;

[0046] Match the welding assessment index hju and the electrical performance index hjr of the circuit board of the current production batch with the corresponding index value ranges respectively, so as to determine the welding quality grade and the electrical quality grade of the circuit board of the current production batch.

[0047] In summary, the present application includes at least one of the following beneficial technical effects:

[0048] The present invention divides the circuit board image into regions according to functional modules, analyzes the roundness of solder joints for each sub-region, compares with the reference sample to find solder joints with abnormal shapes; matches the center coordinates of solder joints to calculate the deviation to determine solder joints with abnormal positions, accurately evaluates the welding quality from the shape and position, avoids the omission of general evaluation, improves the detection accuracy, and solves the problem that in the prior art, most of them only generally evaluate the welding quality of the circuit board and cannot conduct targeted evaluation for each functional sub-region of the circuit board, resulting in a low degree of intelligence in quality detection.

[0049] The present invention measures the resistance, capacitance, inductance values and transmission delay, takes the average value of multiple measurements to calculate the comprehensive value, evaluates the component performance by comparing the deviation rate with the nominal value, and combines the weighted calculation of the transmission delay comprehensive value to comprehensively and accurately evaluate the electrical performance of the circuit board, making up for the one-sidedness existing in the prior art;

[0050] The present invention determines the sampling ratio according to the range of the total number of circuit boards, extracts the samples for welding and electrical testing, and the scientific sampling scheme enables the samples to represent the quality of the whole batch. While ensuring the accuracy of the testing, it greatly reduces the testing volume, improves the overall production efficiency, and avoids the waste of time and resources for full inspection;

[0051] The present invention calculates the welding effect index by calculating the welding shape difference index and the welding position difference index, takes the average value to obtain the welding evaluation index, quantifies the welding quality, and intuitively reflects the welding quality level of the circuit boards in the production batch. At the same time, it calculates the component performance index and the transmission delay comprehensive value to obtain the electrical comprehensive index, and takes the average value to obtain the electrical performance index, quantifying the electrical performance and providing accurate data support for the electrical quality evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0052] In the following description of the exemplary embodiments in conjunction with the drawings, more details, features and advantages of the present application are disclosed. In the drawings:

[0053] Figure 1 is the principle block diagram of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0054] The following will describe several embodiments of the present application in more detail with reference to the drawings so that those skilled in the art can implement the present application. The present application can be embodied in many different forms and purposes and should not be limited to the embodiments described herein. These embodiments are provided to make the present application comprehensive and complete, and fully convey the scope of the present application to those skilled in the art. The embodiments do not limit the present application.

[0055] Unless otherwise defined, all terms (including technical terms and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which the present application belongs. It will be further understood that terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the relevant art and / or the context of this specification, and will not be interpreted in an idealized or overly formal sense unless clearly defined herein.

[0056] Please refer to Figure 1 as shown, a circuit board welding quality detection system includes a quality detection module and a quality evaluation module;

[0057] The quality detection module includes a sample screening unit, a welding analysis unit and an electrical detection unit;

[0058] The sample screening unit is used to count the total number of circuit boards produced in the current batch, and determine the extraction ratio of the number of welding analysis samples and electrical detection samples according to the total number range, so as to extract the corresponding ratio of welding analysis samples and electrical detection samples;

[0059] Specifically:

[0060] The intervals where the quantities of each group corresponding to the preset number of circuit boards are located are set, and each interval where the quantity of a group is located corresponds to a set of extraction ratios; the set of extraction ratios includes the extraction ratios corresponding to the welding analysis samples and the electrical detection samples respectively;

[0061] Match the total number of circuit boards produced in the current batch with the preset intervals where the quantities of each group are located, so as to obtain the set of extraction ratios for the circuit boards produced in the current batch; multiply the total number of circuit boards produced in the current batch by the extraction ratio of the welding analysis samples in the set of extraction ratios to determine the extraction quantity of the welding analysis samples;

[0062] Multiply the total number of circuit boards produced in the current batch by the extraction ratio of the electrical detection samples in the set of extraction ratios to determine the extraction quantity of the electrical detection samples;

[0063] It should be noted that if welding analysis and electrical detection are carried out on each circuit board, it will consume a large amount of time and detection resources, resulting in a significant reduction in production efficiency and a significant increase in detection costs. By determining the extraction ratio according to the total number range of circuit boards for sampling detection, it is possible to greatly reduce the workload of detection, improve the overall production efficiency, and reduce the detection cost while ensuring a certain detection accuracy;

[0064] The preset quantity intervals and the corresponding sets of extraction ratios are scientifically analyzed and verified by practice. It is possible to select a suitable sampling ratio according to different production scales. The welding analysis samples and electrical detection samples extracted in this way can represent the quality status of the entire batch of circuit boards to a certain extent, so as to infer the quality of the entire batch of products through the detection results of the samples and provide a reliable basis for quality control.

[0065] The welding analysis unit is used to collect images of the extracted welding analysis samples; for the extracted welding analysis samples, use a high-resolution industrial camera to collect images from multiple angles, and reasonably adjust the position, angle and focal length of the camera according to the size and solder joint distribution of the circuit board to ensure clear and complete solder joint images are obtained. During the collection process, various lighting methods are adopted, such as ring light sources, coaxial light sources, etc., to eliminate shadows and reflections and ensure the image quality; and preprocess the collected group of sample images; the preprocessing includes grayscale conversion, denoising, and contrast adjustment, etc.; after the preprocessing is completed, comprehensively analyze the group of sample images to obtain the welding evaluation index hju of the circuit boards in the current production batch;

[0066] It should be noted that the collected original images may have problems such as noise and uneven illumination. First, grayscale processing is performed to convert the color images into grayscale images, which is convenient for subsequent calculations and analyses. Then, filtering algorithms such as Gaussian filtering are used to remove the noise in the images and smooth the images. At the same time, methods such as histogram equalization are used to adjust the brightness and contrast of the images, enhance the features of the solder joints in the images, and make the subsequent feature extraction and analysis more accurate.

[0067] Specifically:

[0068] Pre-input the circuit board image samples with qualified welding quality as the reference quality samples for the circuit boards of the current production batch;

[0069] According to the pre-set divided regions, each group of sample images is divided into each functional sub-region; the regional division is carried out according to different functional modules on the circuit board. For example, the power supply module, signal processing module, storage module, etc. are each divided into a region; because the circuit structures and welding process requirements of different functional modules may be different, such division is convenient for analyzing the welding conditions of different functional regions

[0070] For each functional sub-region of each group of sample images, use the edge detection algorithm to extract the solder joint edge contours of each functional sub-region in each group of sample images, and obtain the roundness of each solder joint in each functional sub-region of each group of sample images;

[0071] From each functional sub-region of the reference quality samples, respectively extract the roundness with the largest difference from the integer one among the roundness of each solder joint as the passing roundness of each functional sub-region of the sample image; that is, the lowest roundness of the solder joints in each functional sub-region;

[0072] Compare the roundness of each solder joint in each functional sub-region of each group of sample images with the corresponding passing roundness. If the roundness of a certain solder joint in a certain functional sub-region of a certain group of sample images is lower than the corresponding passing roundness, it is determined as a solder joint with abnormal shape;

[0073] Count the number of solder joints with abnormal shape in each functional sub-region of each group of sample images, denoted as the number of shape abnormalities; set a weight coefficient corresponding to each functional sub-region for the number of shape abnormalities; multiply the number of shape abnormalities in each functional sub-region of each group of sample images by the corresponding weight coefficient, and then sum to obtain the welding shape abnormality index of each group of sample images;

[0074] It should be noted that by dividing the circuit board image into different functional sub-regions and analyzing the roundness of the solder joints for each sub-region, the previous general evaluation method of the entire circuit board is changed, and the refined evaluation of the welding quality from the dimension of different functional regions is realized.

[0075] For the reference quality samples, obtain the geometric center of each solder joint profile as the center coordinates of each solder joint.

[0076] At the same time, obtain the center coordinates of each solder joint in each group of sample images, match the center coordinates of each solder joint in each group of sample images with the corresponding solder joint center coordinates in the reference quality samples, and after the matching is completed, obtain the solder joint pairs of each solder joint in each group of sample images, that is, (sample image solder joint center coordinates, reference solder joint center coordinates).

[0077] Calculate the differences in the horizontal and vertical coordinates for each group of solder joint pairs respectively, that is According to the soldering process standard of the circuit board, set the allowable position deviation thresholds corresponding to the horizontal and vertical coordinates respectively.

[0078] Compare the differences in the horizontal and vertical coordinates of each group of solder joint pairs with the corresponding set allowable position deviation thresholds respectively. If there is a group of differences greater than the corresponding set allowable position deviation threshold, mark the solder joint pairs with differences greater than the allowable position deviation threshold as solder joints with abnormal positions.

[0079] Count the number of solder joints with abnormal positions in each functional sub-region of each group of sample images, denoted as the number of abnormal positions; set a weight coefficient for the number of abnormal positions corresponding to each different functional sub-region, multiply the number of abnormal positions in each functional sub-region of each group of sample images by the corresponding weight coefficient, and then sum to obtain the soldering position anomaly index of each group of sample images.

[0080] It should be noted that by matching and calculating the differences in the center coordinates of each solder joint in the sample image with the corresponding solder joint center coordinates in the reference quality sample, the solder joints with positions deviating from the standard can be accurately found; this precise matching and calculation method avoids the omissions and misjudgments that may occur in manual inspection, greatly improving the accuracy and efficiency of the inspection; dividing the circuit board into different functional sub-regions for counting the solder joints with abnormal positions makes the positioning of soldering problems more refined.

[0081] Extract the soldering shape anomaly index and soldering position anomaly index corresponding to each group of sample images, and mark them as E1 and E2 respectively; according to the formula Perform weighted calculation on the soldering shape anomaly index E1 and soldering position anomaly index E2 of each group of sample images to obtain the soldering effect index E3 corresponding to each group of sample images.

[0082] Take the mean value of the soldering effect index E3 of each group of sample images to obtain the soldering evaluation index hju of the circuit boards in the current production batch.

[0083] It should be noted that taking the mean value of the soldering effect index of each group of sample images to obtain the soldering evaluation index of the circuit boards in the current production batch can reflect the soldering quality level of the circuit boards in this production batch as a whole.

[0084] The electrical detection unit is used to collect electrical parameters of the extracted electrical detection samples; the electrical parameters include resistance value, capacitance value, inductance value, and transmission delay; after the collection is completed, the electrical parameters of each electrical detection sample are comprehensively evaluated to obtain the electrical performance index hjr of the circuit board of the current production batch.

[0085] Specifically:

[0086] The number of measurements G corresponding to the preset resistance value, capacitance value, and inductance value, where G > 3, is specifically set by the technical personnel.

[0087] According to the number of measurements G, G groups of resistance values, capacitance values, and inductance values of each electrical detection sample are obtained; and the average values of the G groups of resistance values, capacitance values, and inductance values of each electrical detection sample are calculated respectively to obtain the resistance comprehensive value, capacitance comprehensive value, and inductance comprehensive value of each electrical detection sample; for resistance, capacitance, and inductance, the test clips of the LCR tester are respectively connected to both ends of the corresponding components, and through the preset test frequency and test level, the actually measured resistance value, capacitance value, and inductance value are read and recorded.

[0088] The nominal values corresponding to the preset resistance comprehensive value, capacitance comprehensive value, and inductance comprehensive value are preset, and the resistance deviation rate D R , capacitance deviation rate D C , and inductance deviation rate D L are calculated; that is, calculated through the formula , where R 综合 and R 标称 respectively represent the resistance comprehensive value and the corresponding preset nominal value, C 综合 and C 标称 respectively represent the capacitance comprehensive value and the corresponding preset nominal value, L 综合 and L 标称 respectively represent the inductance comprehensive value and the corresponding preset nominal value.

[0089] The weight coefficients corresponding to the set resistance deviation rate D R , capacitance deviation rate D C , and inductance deviation rate D L are set, and the resistance deviation rate D R , capacitance deviation rate D C , and inductance deviation rate D L of each electrical detection sample are respectively multiplied by the corresponding set weight coefficients, and then summed to obtain the component performance index Dt of each electrical detection sample.

[0090] It should be noted that by presetting the number of measurements G (G > 3) and obtaining G groups of resistance values, capacitance values, and inductance values, and then calculating the mean values to obtain the comprehensive resistance value, comprehensive capacitance value, and comprehensive inductance value, the influence of measurement errors and random interference can be effectively reduced;

[0091] By setting the weight coefficients corresponding to the resistance deviation rate, capacitance deviation rate, and inductance deviation rate respectively, and calculating the component quality index through calculation, the quantitative evaluation of the component quality is realized;

[0092] Input a set excitation signal to each electrical test sample; the excitation signal is a periodic pulse signal or a clock signal; mark the time point when the excitation signal is input as the starting point, and mark the time point when the excitation signal is output as the ending point, and take the time difference between the starting point and the ending point as the transmission delay;

[0093] Set the number of times the excitation signal is input, and obtain the transmission delay corresponding to each input of the excitation signal for each electrical test sample. Calculate the mean value of each group of transmission delays of each electrical test sample to obtain the comprehensive transmission delay value De of each electrical test sample;

[0094] It should be noted that incorporating the comprehensive transmission delay value into the evaluation system takes into account the signal transmission ability of the circuit board.

[0095] According to the formula Perform weighted calculation on the component performance index Dt and the comprehensive transmission delay value De of each electrical test sample to obtain the electrical comprehensive index Dy of each electrical test sample; where Da and Db are the preset performance allowable index and delay allowable value respectively; β1 and β2 are the influence weight factors of the component performance index Dt and the comprehensive transmission delay value De respectively;

[0096] Take the mean value of the electrical comprehensive index Dy of each electrical test sample to obtain the electrical performance index hjr of the current production batch of circuit boards;

[0097] It should be noted that by calculating the electrical comprehensive index, two important aspects of component performance and transmission delay are integrated, realizing the comprehensive evaluation of the electrical performance of the circuit board.

[0098] The quality evaluation module is used to receive the welding evaluation index hju and the electrical performance index hjr of the current production batch of circuit boards, so as to output the welding quality grade and electrical quality grade of the current production batch of circuit boards, and send the output welding quality grade and electrical quality grade to the management personnel. The management personnel selectively trigger the repeat signaling after receiving it. If the repeat signaling is triggered, the triggered repeat signaling is sent to the quality detection module;

[0099] It should be noted that if the management believes that the welding quality level is "poor welding level" or the electrical quality level is "poor electrical level" and has doubts about the results of quality inspection, a repeat signal can be triggered. After triggering, the repeat signal is sent to the quality inspection module. After receiving the repeat signal, the quality inspection module can re-extract welding analysis samples and electrical test samples from this batch of circuit boards for testing to further confirm the quality status of the current batch of circuit boards.

[0100] Specifically:

[0101] Three sets of index value ranges corresponding to the preset welding evaluation index hju and electrical performance index hjr are set. Each set of index value ranges corresponding to the welding evaluation index hju corresponds to a welding quality level, and each set of index value ranges corresponding to the electrical performance index hjr corresponds to an electrical quality level; among them, the welding quality levels include poor welding level, passing welding level, and excellent welding level, and the electrical quality levels include poor electrical level, passing electrical level, and excellent electrical level;

[0102] The welding evaluation index hju and electrical performance index hjr of the current production batch of circuit boards are respectively matched with the corresponding index value ranges to determine the welding quality level and electrical quality level of the current production batch of circuit boards;

[0103] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the present invention to the specific implementation manners. Obviously, many modifications and changes can be made according to the content of this specification. These embodiments are selected and specifically described in this specification to better explain the principles and practical applications of the present invention, so that those skilled in the art in the relevant technical field can well understand and utilize the present invention. The present invention is only limited by the claims and their full scope and equivalents.

Claims

1. A circuit board welding quality detection system, characterized in that: include: Quality inspection module: including sample screening unit, welding analysis unit and electrical inspection unit; The sample screening unit is used to count the total number of circuit boards produced in the current batch, and determine the extraction ratio of welding analysis samples and electrical inspection samples according to the total number range, so as to extract the welding analysis samples and electrical inspection samples of corresponding proportion; The welding analysis unit is used to collect images of the extracted welding analysis samples and pre-process each group of sample images collected; After the preprocessing is completed, each group of sample images is comprehensively analyzed to obtain the welding evaluation index hju of the current production batch of circuit boards; The electrical detection unit is used to collect electrical parameters of the extracted electrical detection samples; The electrical parameters include resistance, capacitance, inductance and transmission delay. After the collection is completed, the electrical parameters of each electrical test sample are comprehensively evaluated to obtain the electrical performance index hjr of the current production batch of circuit boards.

2. A circuit board welding quality detection system according to claim 1, characterized in that: Determine the sampling ratio of welding analysis samples and electrical inspection samples according to the total number range, specifically: The preset circuit board quantity corresponds to the interval of each group quantity, and each group quantity interval is set to correspond to an extraction ratio set; The extraction ratio set includes the extraction ratios corresponding to the welding analysis samples and the electrical inspection samples respectively; Match the total number of circuit boards produced in the current batch with the preset intervals of the numbers of each group, so as to obtain the extraction ratio set of the circuit boards produced in the current batch; multiply the total number of circuit boards produced in the current batch with the extraction ratio of the welding analysis samples in the extraction ratio set to determine the extraction quantity of the welding analysis samples; The total number of circuit boards produced in the current batch is multiplied by the extraction ratio of the electrical test samples in the extraction ratio set to determine the number of electrical test samples to be extracted.

3. A circuit board welding quality detection system according to claim 2, characterized in that: After the preprocessing is completed, the sample images of each group are comprehensively analyzed, specifically: Input the circuit board image samples with qualified welding quality in advance as the reference quality samples of the current production batch of circuit boards; According to the pre-set division areas, each group of sample images is divided into various functional sub-areas; For each functional sub-region of each group of sample images, extract the edge contour of the solder joint in each functional sub-region in each group of sample images, and obtain the circularity of each solder joint in each functional sub-region in each group of sample images; From each functional sub-region of the reference quality sample, the circularity with the largest difference between the distance integer and the circularity of each solder joint is extracted respectively as the qualified circularity of each functional sub-region of the sample image; The circularity of each solder joint in each functional sub-region in each group of sample images is compared with the corresponding qualified circularity. If the circularity of a solder joint in a functional sub-region in a group of sample images is lower than the corresponding qualified circularity, it is determined to be a solder joint with abnormal shape. The number of abnormally shaped welds in each functional sub-region in each group of sample images is counted and recorded as the number of shape anomalies; a weight coefficient corresponding to a number of shape anomalies is set for each functional sub-region; the number of shape anomalies in each functional sub-region in each group of sample images is multiplied by the corresponding weight coefficient, and then the sum is calculated to obtain the welding shape anomaly index of each group of sample images.

4. A circuit board welding quality detection system according to claim 3, characterized in that: Get the soldering evaluation index hju of the current production batch of circuit boards, specifically: For the reference quality sample, the geometric center of each solder joint contour is obtained as the center coordinate of each solder joint; At the same time, the center coordinates of each solder joint in each group of sample images are obtained, and the center coordinates of each solder joint in each group of sample images are matched with the center coordinates of the corresponding solder joints in the reference quality sample. After the matching is completed, the solder joint pairs of each solder joint in each group of sample images are obtained, that is, (center coordinates of solder joints in sample images, center coordinates of reference solder joints); Calculate the difference of each group of welding points on the horizontal and vertical coordinates, and set the allowable position deviation thresholds corresponding to the horizontal and vertical coordinates respectively; The difference between the horizontal coordinate and the vertical coordinate of each group of welding point pairs is compared with the corresponding set allowable position deviation threshold. If there is a group of differences that is greater than the corresponding set allowable position deviation threshold, the welding point pair greater than the allowable position deviation threshold is marked as an abnormal position welding point; The number of abnormal welds in each functional sub-region in each group of sample images is counted and recorded as the number of abnormalities; a weight coefficient corresponding to a number of abnormalities is set for each functional sub-region, the number of abnormalities in each functional sub-region in each group of sample images is multiplied by the corresponding weight coefficient, and then the sum is calculated to obtain the welding abnormality index of each group of sample images; The welding shape index and welding position index corresponding to each group of sample images are extracted and marked as E1 and E2 respectively; according to the formula The welding shape index E1 and welding position index E2 of each group of sample images are weightedly calculated to obtain the welding effect index E3 corresponding to each group of sample images; The average value of the welding effect index E3 of each group of sample images is taken to obtain the welding evaluation index hju of the current production batch of circuit boards.

5. A circuit board welding quality detection system according to claim 4, characterized in that: Comprehensively evaluate the electrical parameters of each electrical test sample, specifically: The number of measurements G corresponding to the preset resistance value, capacitance value and inductance value; According to the measurement number G, obtain the resistance value, capacitance value and inductance value of group G of each electrical test sample; The resistance value, capacitance value and inductance value of group G of each electrical test sample are calculated respectively to obtain the comprehensive resistance value, comprehensive capacitance value and comprehensive inductance value of each electrical test sample; Preset the nominal values ​​corresponding to the comprehensive resistance value, comprehensive capacitance value, and comprehensive inductance value, and calculate the resistance deviation rate D R , capacitance deviation rate D C And the inductance deviation rate D L ; That is, through the formula Calculate and get; where R 综合 and R 标称 Respectively represent the comprehensive resistance value and the corresponding preset nominal value, C 综合 and C 标称 Respectively represent the comprehensive capacitance value and the corresponding preset nominal value, L 综合 and L 标称 Respectively represent the comprehensive value of inductance and the corresponding preset nominal value; Set the resistance deviation rate D R , capacitance deviation rate D C And the inductance deviation rate D L The corresponding weight coefficients are used to calculate the resistance deviation rate D of each electrical test sample. R , capacitance deviation rate D C And the inductance deviation rate D L They are multiplied by the corresponding set weight coefficients respectively, and then the sum is obtained to obtain the component performance index Dt of each electrical test sample.

6. A circuit board welding quality detection system according to claim 5, characterized in that: Get the electrical performance index hjr of the current production batch of circuit boards, specifically: Input a set excitation signal to each electrical test sample, mark the time point of inputting the excitation signal as the starting point, mark the time point of outputting the excitation signal as the ending point, and use the time difference between the starting point and the ending point as the transmission delay; The number of input times of the excitation signal is set, and the transmission delay of each electrical detection sample corresponding to each input excitation signal is obtained, and the mean value of each group of transmission delays of each electrical detection sample is calculated to obtain the comprehensive value De of the transmission delay of each electrical detection sample; According to the formula The component performance index Dt and the transmission delay comprehensive value De of each electrical test sample are weightedly calculated to obtain the electrical comprehensive index Dy of each electrical test sample; wherein Da and Db are respectively the preset performance allowable index and delay allowable value; β1 and β2 are respectively the influencing weight factors of the component performance index Dt and the transmission delay comprehensive value De; Take the average value of the comprehensive electrical index Dy of each electrical test sample to obtain the electrical performance index hjr of the current production batch of circuit boards.

7. A circuit board welding quality detection system according to claim 6, characterized in that: Also includes: Quality assessment module: receiving the welding assessment index hju and electrical performance index hjr of the current production batch of circuit boards, thereby outputting the welding quality grade and electrical quality grade of the current production batch of circuit boards, and sending the output welding quality grade and electrical quality grade to the management personnel.

8. A circuit board welding quality detection system according to claim 7, characterized in that: Output the soldering quality level and electrical quality level of the current production batch of circuit boards, specifically: Three groups of index value ranges corresponding to the welding evaluation index hju and the electrical performance index hjr are preset, and each group of index value ranges corresponding to the welding evaluation index hju corresponds to a welding quality grade, and each group of index value ranges corresponding to the electrical performance index hjr corresponds to an electrical quality grade; The welding quality grades include poor welding grade, qualified welding grade and excellent welding grade, and the electrical quality grades include poor electrical grade, qualified electrical grade and excellent electrical grade; The soldering evaluation index hju and the electrical performance index hjr of the current production batch of circuit boards are matched with the corresponding index value ranges respectively, so as to determine the soldering quality grade and the electrical quality grade of the current production batch of circuit boards.

Images