Automatic detection method and device for capacitor pins

Abstract

The invention provides an automatic detection method and device for capacitor pins, and relates to the technical field of capacitor detection. Sorting the detection items according to the fault repair feasibility to obtain a to-be-detected item sequence; scheduling the detection device to execute detection data collection to obtain a sample detection data set; performing pin defect identification according to the sample detection data set to obtain sample defect information; aggregating the sample defect information to obtain sample detection defect distribution; performing quality evaluation on the sample capacitor according to the sample defect information to obtain batch quality distribution; comprehensively analyzing sample detection defect distribution and batch quality distribution, and outputting a quality detection result. According to the method, the technical problems of low detection efficiency and insufficient batch quality processing accuracy caused by fixed detection process and no consideration of fault repair feasibility in the prior art are solved, the quality detection efficiency is improved, and the batch capacitor quality control accuracy is enhanced.

Description

Technical Field

[0001] The present application relates to the technical field of capacitor detection, and in particular to an automated detection method and device for capacitor pins. Background Art

[0002] As a common component in electronic devices, the quality of capacitor pins has an important impact on product performance and service life. The current capacitor pin detection methods mainly include manual detection, fixed process automatic detection and statistical sampling detection. Manual detection relies on operator observation and tool assistance. Although it is highly flexible, it is inefficient and subjective, and it is difficult to adapt to large-scale detection needs. Fixed process automatic detection uses automated equipment to detect each characteristic of the capacitor pin one by one. Although it improves the detection speed, the detection process is fixed and the feasibility of defect repair is not considered, which easily leads to resource waste. Statistical sampling detection infers batch quality distribution through the detection of a small number of samples. It is suitable for large-scale detection, but its defect analysis accuracy is limited, and it is impossible to conduct in-depth analysis of defect distribution and it is difficult to fully reflect the batch quality. These existing methods cannot fully meet the needs of modern manufacturing for intelligent and efficient detection. Summary of the invention

[0003] The present application provides an automated detection method and device for capacitor pins, which solves the technical problems of low detection efficiency and insufficient batch quality processing accuracy due to the fixed detection process and failure to consider the feasibility of fault repair in the prior art, thereby achieving the technical effect of improving the efficiency of capacitor pin quality detection and enhancing the accuracy of batch capacitor quality control.

[0004] In view of the above problems, on the one hand, the present application provides an automated detection method for capacitor pins, the method comprising: collecting detection samples for batch capacitors to obtain K sample capacitors, wherein the K sample capacitors and the batch capacitors meet a preset sampling ratio; sorting the detection items according to the feasibility of fault repair to obtain a sequence of items to be detected; using the sequence of items to be detected as a constraint, scheduling the detection device to perform detection data set collection on K groups of capacitor pins of the K sample capacitors to obtain K sample detection data sets; performing pin defect identification based on the K sample detection data sets to obtain K sample defect information, wherein the sample defect information includes a sample defect label sequence and a sample defect level sequence; aggregating the K sample defect information to obtain a sample detection defect distribution; performing quality evaluation on the K sample capacitors based on the K sample defect information to obtain a batch quality distribution; comprehensively analyzing the sample detection defect distribution and the batch quality distribution, and outputting the quality detection results of the batch capacitors.

[0005] On the other hand, the present application also provides an automatic detection device for capacitor pins, the device comprising: A sample collection module is used to collect test samples of batch capacitors to obtain K sample capacitors, wherein the K sample capacitors and the batch capacitors meet a preset sampling ratio; a test item sorting module is used to sort the test items according to the feasibility of fault repair to obtain a sequence of items to be tested; a test data set collection module is used to schedule the test device to perform test data collection on K groups of capacitor pins of the K sample capacitors with the sequence of items to be tested as a constraint to obtain K sample test data sets; a pin defect identification module is used to perform pin defect identification based on the K sample test data sets to obtain K sample defect information, wherein the sample defect information includes a sample defect label sequence and a sample defect level sequence; a defect information aggregation module is used to aggregate the K sample defect information to obtain a sample test defect distribution; a quality evaluation module is used to perform quality evaluation on the K sample capacitors based on the K sample defect information to obtain a batch quality distribution; a test result output module is used to comprehensively analyze the sample test defect distribution and the batch quality distribution, and output the quality test result of the batch capacitors.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: By presetting the sampling ratio, the batch capacitors are tested and sampled to obtain K sample capacitors, ensuring that the collected samples are representative and can reflect the quality status of the entire batch capacitors, providing basic data for subsequent testing. The test items are sorted according to the feasibility of fault repair, and the test process is optimized to give priority to defects that are easy to repair and have a greater impact, ensuring the importance and practicality of the test items. According to the optimized test item sequence, the test device is scheduled to collect data from the pins of the sample capacitors to ensure the accuracy and completeness of the test data and provide data support for subsequent defect identification. By analyzing the test data, the defects of the pins are identified, and the defects are labeled and graded to achieve refined management of the defects and provide detailed information for quality evaluation. The defect information of multiple samples is aggregated to form a sample test defect distribution, providing macro data for batch quality evaluation, and facilitating the discovery of common problems in batch production. Based on the sample defect information, the sample capacitors are evaluated for quality, forming a batch quality distribution, providing comprehensive quality information, and facilitating the timely discovery and resolution of quality problems. The sample test defect distribution and batch quality distribution are comprehensively analyzed to output the final quality test results, providing intuitive and comprehensive quality evaluation, avoiding the one-sidedness of single indicator judgment.

[0007] In summary, this application achieves accurate identification and graded screening of capacitor pin defects through dynamic sorting of detection items based on the feasibility of fault repair; accurately evaluates the quality distribution of batch capacitors through sample collection and defect information aggregation; and optimizes the processing method of batch capacitors by combining with an intelligent decision-making output mechanism, significantly improving the efficiency of capacitor pin detection and the accuracy of quality assessment, and ensuring the rationality of batch capacitor pin processing.

[0008] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0009] Figure 1 A schematic flow chart of an automated detection method for capacitor pins provided in an embodiment of the present application.

[0010] Figure 2 A schematic diagram of a flow chart of obtaining a sequence of items to be detected in the automated detection method for capacitor pins provided in an embodiment of the present application.

[0011] Figure 3 A schematic diagram of a process for obtaining K sample defect information in the automated detection method for capacitor pins provided in an embodiment of the present application.

[0012] Figure 4 A schematic diagram of the structure of an automated detection device for capacitor pins provided in an embodiment of the present application.

[0013] Explanation of the reference numerals: sample collection module 10 , detection item sorting module 20 , detection data collection module 30 , pin defect recognition module 40 , defect information aggregation module 50 , quality evaluation module 60 , detection result output module 70 . DETAILED DESCRIPTION

[0014] The embodiments of the present application provide an automated detection method and device for capacitor pins, thereby solving the technical problems of low detection efficiency and insufficient batch quality processing accuracy due to the fixed detection process and failure to consider the feasibility of fault repair in the prior art. This achieves the technical effect of improving the efficiency of capacitor pin quality detection and enhancing the accuracy of batch capacitor quality control.

[0015] Embodiment 1, as Figure 1 As shown, the embodiment of the present application provides an automated detection method for capacitor pins, the method comprising: Step S1: collecting test samples of batch capacitors to obtain K sample capacitors, wherein the K sample capacitors and the batch capacitors satisfy a preset sampling ratio.

[0016] Specifically, batch capacitors refer to a large number of capacitors that need to be tested, which are the overall object of the inspection task. According to the pre-set sampling ratio, a certain number of capacitors are randomly selected from the batch capacitors as sample capacitors to obtain K sample capacitors. Here K is a positive integer, which refers to the total number of sample capacitors selected. For example, if the preset sampling ratio is 10%, and there are 1000 batch capacitors, then 100 capacitors need to be randomly selected as sample capacitors (ie K=100), and each capacitor is numbered using a simple numbering system. Then, a random number generator is used to randomly determine 100 numbers between 1 and 1000, and the capacitors corresponding to these numbers are extracted as sample capacitors.

[0017] The above sampling can reduce the detection workload, improve detection efficiency, and avoid the waste of resources caused by testing all capacitors while ensuring that the samples are representative.

[0018] Step S2: Sort the inspection items according to the feasibility of fault repair to obtain a sequence of items to be inspected.

[0019] Specifically, the feasibility of fault repair refers to the difficulty and cost of repairing a capacitor pin after it fails. Classify the possible faults of the capacitor pins, and define the feasibility of repair for each defect based on historical data or engineering experience. Sort the inspection items corresponding to each type of fault according to the repair feasibility, and determine the sequence of items to be inspected. The detection device adjusts the action flow according to the sequence, such as giving priority to detecting fractures and analyzing the pin surface shape or cracks through an optical microscope. Optimizing the inspection process by sorting the inspection items can improve the inspection efficiency.

[0020] Step S3: using the sequence of items to be detected as a constraint, scheduling the detection device to perform detection data set collection on K groups of capacitor pins of the K sample capacitors to obtain K sample detection data sets.

[0021] Specifically, according to the sequence of items to be tested obtained in step S2, the corresponding testing devices are used in turn to test the K groups of capacitor pins of the K sample capacitors, and the corresponding test data are collected to generate K sample test data sets. Each sample capacitor corresponds to a group of capacitor pins, that is, a sample test data set. The sample test data set contains all the test data records generated by each sample capacitor, such as feature data such as pin shape, size, and surface state.

[0022] By scheduling the detection devices for data collection in a specific order, the validity and accuracy of the data are ensured, providing a reliable data source for subsequent defect identification and helping to accurately determine the status of the capacitor pins.

[0023] Step S4: performing pin defect identification according to the K sample detection data sets to obtain K sample defect information, wherein the sample defect information includes a sample defect label sequence and a sample defect level sequence.

[0024] Specifically, data analysis software or algorithms are used to analyze K sample inspection data sets to determine whether there are defects in the pins and the type and degree of the defects. For example, for the data on the appearance of the pins, if the bending angle of the pins exceeds a certain threshold, it is marked as a "bending" defect label, and the defect level is determined as mild, moderate or severe based on the degree of bending. K sample defect information is generated based on the analysis and judgment results of the K sample inspection data sets, and each sample defect information includes a sample defect label sequence and a sample defect level sequence. Among them, the sample defect label sequence is used to mark the defect type of each sample capacitor pin; the sample defect level sequence is used to mark the defect severity of each sample capacitor pin.

[0025] By accurately identifying the defects of each sample capacitor pin, detailed basic information is provided for subsequent overall analysis, which helps to accurately evaluate the quality of the sample capacitors.

[0026] Step S5: Aggregate the K sample defect information to obtain sample detection defect distribution.

[0027] Specifically, statistical software is used to count the data in the defect label sequence and defect level sequence of each sample, and the proportion of each defect type in all samples and the proportion distribution of different defect levels are calculated to obtain the sample detection defect distribution. The sample detection defect distribution describes the distribution of all sample capacitor pin defects in the overall batch of capacitors, including the proportion of different defect types and defect levels.

[0028] By aggregating the defect information of K samples and obtaining the sample detection defect distribution, we can grasp the defect situation of the sample capacitor pins as a whole, discover the concentration trend of defects, common defect types and severity distribution, and provide a reference for evaluating the quality of the entire batch.

[0029] Step S6: performing quality evaluation on the K sample capacitors according to the defect information of the K samples to obtain batch quality distribution.

[0030] Specifically, the quality evaluation is performed based on the pre-set quality standards and the defect information of each sample capacitor. For example, if the defect level of a sample capacitor is mild and does not affect the overall function, it may be judged as qualified; if the defect level is severe and difficult to repair, it is judged as scrapped; those in between are judged to be repaired. Then, based on the quality evaluation results of the sample capacitors, the quality distribution of the entire batch of capacitors is inferred by statistical methods to obtain the batch quality distribution, including the proportion of different quality states such as qualified, to be repaired, and scrapped. For example, probability theory and mathematical statistics methods can be used for inference.

[0031] By obtaining the batch quality distribution through quality evaluation, the quality of the samples can be reasonably extended to the entire batch, providing a basis for comprehensively evaluating the quality of the entire batch of capacitors and helping to make correct processing decisions.

[0032] Step S7: Comprehensively analyze the sample detection defect distribution and batch quality distribution, and output the quality detection results of the batch capacitors.

[0033] Specifically, the data in the sample inspection defect distribution and batch quality distribution are compared, weighed, and analyzed, and the final quality inspection results are determined based on the two distributions, such as whether they are qualified and what kind of treatment is required (such as full rework, partial repair, etc.). By comprehensively considering multiple aspects of information, the one-sidedness of single indicator judgment is avoided, and the quality inspection results of batch capacitors can be accurately output to ensure that the treatment decision for batch capacitors is scientific and reasonable.

[0034] Further, such as Figure 2 As shown, step S2 includes: Step S21: Using the capacitor model of the batch capacitors as a search instruction, performing network data call to obtain multiple repair feasibility evaluation indicators for multiple sample defect types, wherein the repair feasibility evaluation indicators include repair difficulty, repair cost, repair life loss, repair performance loss and repair time.

[0035] Step S22: Synchronize the multiple repair feasibility evaluation indicators to the repair feasibility evaluation function to obtain multiple defect repair feasibility coefficients.

[0036] Step S23: obtaining a plurality of sample detection devices of the plurality of sample defect types by matching.

[0037] Step S24: Aggregate the multiple sample detection devices based on device consistency to obtain M types of sample detection devices.

[0038] Step S25: According to the M types of sample detection devices and multiple defect repair feasibility coefficients, the multiple sample defect types are prioritized for parallel detection, and the sequence of items to be detected is output.

[0039] Specifically, the model of the batch capacitors to be tested is determined, and a search instruction is generated according to the capacitor model. The historical data is called through an online database (such as a production record database or an industry fault analysis platform) to extract the common defects of the capacitor model and the corresponding repair feasibility evaluation indicators for each defect. These repair feasibility evaluation indicators are key parameters for measuring the possibility of defect repair, including the following: repair difficulty, that is, the complexity of the repair operation; repair cost, that is, the material and labor costs required for repair; repair life loss, that is, the shortening of the capacitor life after repair; repair performance loss, that is, the decline in the performance of the capacitor after repair; repair time, that is, the time required to complete the repair.

[0040] The repair feasibility evaluation function is a mathematical model or algorithm used to comprehensively evaluate multiple indicators (such as difficulty, cost, and time consumption) and output a repair feasibility coefficient. Exemplarily, a multi-indicator comprehensive evaluation model (such as weighted average method or fuzzy comprehensive evaluation method) can be used to construct a repair feasibility evaluation function, and multiple repair feasibility evaluation indicators can be converted into a defect repair feasibility coefficient. This defect repair feasibility coefficient is a numerical result used to indicate the feasibility of repairing a certain sample defect type. The higher the value, the higher the repair feasibility. Combining multiple repair feasibility evaluation indicators into one coefficient simplifies the evaluation of defect repair feasibility, allowing the repair feasibility between different sample defect types to be directly compared, and provides a unified quantitative standard for subsequent sorting.

[0041] According to different sample defect types, the corresponding detection device is matched from the pre-established detection device database. This database stores the mapping relationship between various sample defect types and corresponding detection devices. For example, the database records the relationship of "pin breakage-X-ray detection device", "capacitance value deviation-capacitance tester", etc. Accurately matching the detection device corresponding to the sample defect type ensures the accuracy and effectiveness of subsequent detection, and provides hardware equipment-level support for parallel detection priority sorting.

[0042] The matched multiple sample detection devices are analyzed, and the detection devices are aggregated according to their similarities or correlations in terms of functions, principles, detection ranges, etc., to obtain M types of sample detection devices. Here, M is a positive integer, which represents the number of sample detection devices after aggregation. Exemplarily, a device consistency evaluation matrix can be established, in which rows and columns represent different sample detection devices, respectively, and the matrix elements represent the degree of consistency between the two devices (which can be a value between 0 and 1, 0 represents complete inconsistency, and 1 represents complete consistency). Then, sample detection devices with high consistency are aggregated together through a clustering algorithm (such as a K-means clustering algorithm) to obtain M types of sample detection devices. For example, if the optical microscope and the electron microscope have a high degree of consistency, the two can be aggregated into a type of sample detection device for detecting microscopic defects on the surface of the pin. By aggregating the sample detection devices, the types of detection devices are reduced, the arrangement of subsequent detection is simplified, and it is also conducive to the optimal configuration of detection resources and the improvement of detection efficiency.

[0043] Comprehensively consider the availability, efficiency and feasibility coefficients of multiple defect repair of M sample detection devices, prioritize multiple sample defect types for parallel detection, determine the order of detection of multiple sample defect types, and consider the situation that multiple tests can be performed simultaneously (parallel detection) when sorting, and output the determined order as the sequence of items to be detected. Determine a reasonable sequence of items to be detected, and in the case of parallel detection, give priority to those sample defect types with high repair feasibility and convenient detection, improve the overall detection efficiency, and reduce the waste of detection time and resources.

[0044] Further, step S25 includes: Step S251: Aggregate the multiple defect repair feasibility coefficients according to the M types of sample detection devices to obtain M groups of defect repair feasibility coefficients.

[0045] Step S252: Obtain M comprehensive repair feasibility coefficients by adding the M groups of defect repair feasibility coefficients within the group.

[0046] Step S253: Serialize the M types of sample detection devices according to the M comprehensive repair feasibility coefficients to obtain the scheduling priority of the detection device.

[0047] Step S254: performing parallel detection priority sorting on the multiple sample defect types according to the M groups of defect repair feasibility coefficients to obtain M parallel detection priorities.

[0048] Step S255: associate and store the detection device scheduling priority and the M parallel detection priorities to generate the sequence of items to be detected.

[0049] Specifically, first, a mapping relationship is established to associate each sample detection device with the defect repair feasibility coefficient of the corresponding sample defect type. For example, if there are three sample detection devices (M=3), they are a microscope for detecting the appearance of the pin, a multimeter for detecting electrical performance, and an X-ray machine for detecting the internal structure. For each detection device, find the defect repair feasibility coefficient of the sample defect type associated with it. Assuming that the sample defect types corresponding to the microscope are pin bending and pin surface scratches, the defect repair feasibility coefficients of these two sample defect types are grouped together, and so on, and finally M groups of defect repair feasibility coefficients are obtained. Through this aggregation method, the defect repair feasibility coefficient is linked to the sample detection device, which provides a grouping data basis for the subsequent comprehensive consideration of the detection device and the repair feasibility for priority sorting.

[0050] For the obtained M groups of defect repair feasibility coefficients, perform the intra-group summation operation respectively to obtain M comprehensive repair feasibility coefficients corresponding to the M sample detection devices. Among them, each sample detection device corresponds to a comprehensive repair feasibility coefficient, which represents the overall degree of repair feasibility of the sample defect type related to the sample detection device. For example, for a group of defect repair feasibility coefficients (a 1 , a 2 , a 3 ), then the comprehensive repair feasibility coefficient A=a 1 +a 2 +a 3 The representation of the repair feasibility associated with each sample detection device is simplified by calculating the comprehensive repair feasibility coefficient, which facilitates comparison and priority ranking among different sample detection devices.

[0051] The M types of sample detection devices are sorted according to the size of the M comprehensive repair feasibility coefficients. For example, if the larger the comprehensive repair feasibility coefficient is, the higher the priority for detection is, then the M types of sample detection devices are sorted from large to small (or from small to large, according to the specific priority definition) according to the comprehensive repair feasibility coefficients to obtain the detection device scheduling priority, which is used to arrange the scheduling order of the detection devices. A sorting algorithm (such as a bubble sort algorithm) can be used to implement this sorting process.

[0052] Based on the M groups of defect repair feasibility coefficients, the sample defect types in each group are prioritized. For example, for the sample defect types corresponding to a group of defect repair feasibility coefficients, their order in parallel detection is determined according to the size of the coefficient. A sorting method similar to step S253 can be used to sort the sample defect types according to specific priority rules (such as the larger the coefficient, the earlier the detection), and M parallel detection priorities are obtained. According to these parallel detection priorities, when performing parallel detection, it can be carried out in a reasonable order, and those sample defect types with higher repair feasibility can be detected first, thereby improving the overall detection efficiency.

[0053] The detection device scheduling priority and the M parallel detection priorities are stored in association to generate a sequence of items to be detected. A data structure (such as a linked list or a dictionary) can be used to implement the association storage. For example, in the dictionary, the detection device scheduling priority is used as the key and the corresponding M parallel detection priorities are used as the value. In this way, the two can be associated to form a complete sequence of items to be detected.

[0054] By integrating the sequential information of inspection device scheduling and parallel inspection of sample defect types, a sequence of items to be inspected is generated, which provides a clear operating guide for subsequent inspection work, ensures that the inspection process is carried out in a reasonable order, and improves the accuracy and efficiency of inspection.

[0055] Further, step S3 includes: Step S31: scheduling a first detection device entity of a first sample detection device based on the detection device scheduling priority in the sequence of items to be detected, wherein the first detection device entity is an image acquisition device.

[0056] Step S32: parsing the first parallel detection priority to obtain an image acquisition requirement sequence, wherein the first parallel detection priority includes H sample defect types.

[0057] Step S33: using the image acquisition requirement sequence as a constraint, running the first detection device entity to perform automatic acquisition of detection data on the first group of capacitor pins of the first sample capacitor to obtain a first acquisition detection data set, wherein the first acquisition detection data set includes a first detection image sequence corresponding to the image acquisition requirement sequence.

[0058] Step S34: Similarly, analyze the M-1 parallel detection priority, and run the M-1 detection device entity of the M-1 sample detection device to perform automatic detection data acquisition on the first group of capacitor pins according to the analysis result to obtain the M-1 acquisition detection data set, wherein the first acquisition detection data set and the M-1 acquisition detection data set constitute the first sample detection data set.

[0059] Step S35: Similarly, detection data collection is performed on the K groups of capacitor pins to obtain K sample detection data sets.

[0060] Specifically, the scheduling priority of the detection device is obtained from the sequence of items to be detected, and the first detection device entity of the first sample detection device to be scheduled first is determined according to this priority. In actual operation, the scheduling process can be implemented through an automated control system. For example, in a detection equipment management system, the rules for scheduling priorities of the detection devices are pre-set. When the detection starts, the system automatically selects and starts the first detection device entity according to this rule, that is, an image acquisition device, such as an industrial camera or other device for acquiring images of capacitor pins.

[0061] The first parallel detection priority is the parallel detection priority corresponding to the sample defect type of the first sample detection device, which includes the priority order of H types of sample defect types. Among them, H is a positive integer less than or equal to M. The first parallel detection priority is parsed, and information related to image acquisition is extracted from it, and this information is arranged in order of priority to form an image acquisition requirement sequence. This image acquisition requirement sequence is the order of image acquisition requirements determined according to the first parallel detection priority, and clarifies the image acquisition order for different sample defect types. For example, if the first parallel detection priority indicates that the appearance bending of the pin should be detected first, followed by the scratches on the surface of the pin, then in the image acquisition requirement sequence, the image for detecting the bending of the pin will be arranged first, followed by the image for detecting the scratches on the pin.

[0062] A sample capacitor for detection is randomly selected from K sample capacitors, which is recorded as the first sample capacitor. The first sample capacitor has a first group of capacitor pins. The image acquisition requirement sequence is used as a constraint condition and input into the first detection device entity (image acquisition device). The image acquisition device performs image acquisition on the first group of capacitor pins of the first sample capacitor according to this requirement sequence. For example, in the control program of an industrial camera, according to the instructions in the image acquisition requirement sequence, the camera's shooting angle, focal length and other parameters are adjusted, and images under different requirements are collected in sequence to form a first detection image sequence. These images together constitute the first acquisition detection data set. Among them, each image in the first detection image sequence is applied to a defect detection project.

[0063] Similarly, the remaining M-1 parallel detection priorities are parsed, and according to the parsing results, the corresponding M-1 detection device entities of the M-1 sample detection devices are run to automatically collect detection data for the first group of capacitor pins to obtain M-1 collection detection data sets. The first collection detection data set and the M-1 collection detection data sets are summarized to form the first sample detection data set. For example, after the first sample detection device is scheduled to collect data on the pins of the first sample capacitor, the remaining sample detection devices are scheduled one by one according to the detection device scheduling priority to collect detection data on the pins of the first sample capacitor.

[0064] According to the method similar to the above-mentioned method of collecting the first sample detection data set, the detection data set of each sample capacitor pin is collected in turn to obtain the corresponding K sample detection data sets. For example, after scheduling the first sample detection device, the image data of the remaining sample capacitor pins are collected; then the next sample detection device is scheduled according to the detection device scheduling priority to collect detection data for each sample capacitor pin, and this process is repeated until the detection data set collection of K groups of capacitor pins of K sample capacitors is completed, and finally K sample detection data sets are obtained.

[0065] Through the above steps, the inspection data of K sample capacitor pins are collected in an orderly and comprehensive manner, providing a complete data basis for subsequent pin defect identification, quality evaluation and other operations, ensuring the accuracy and effectiveness of the entire inspection process.

[0066] Further, such as Figure 3 As shown, step S4 includes: Step S41: constructing H pin defect recognition networks according to the H sample defect types, wherein the input data of the pin defect recognition network is the sample detection image, and the output data is the sample defect label and the sample defect level.

[0067] Step S42: using the first parallel detection priority as a constraint, setting the model scheduling priority of the H pin defect recognition networks.

[0068] Step S43: After synchronizing the first detection image sequence to the H pin defect recognition networks, starting the H pin defect recognition networks one by one to perform pin surface defect recognition based on the model scheduling priority, and outputting a first sample defect label sequence and a first sample defect level sequence.

[0069] Step S44: associate and store the first sample defect label sequence and the first sample defect level sequence to obtain first sample defect information.

[0070] Step S45: Similarly, pin defect identification is performed based on the K sample detection data sets to obtain the K sample defect information.

[0071] Specifically, a pin defect recognition network is constructed for each sample defect type. During the construction process, a deep learning framework such as TensorFlow or PyTorch can be used. First, the structure of the network is determined, such as a combination of multiple convolutional layers, pooling layers, and fully connected layers. Then a large number of labeled sample detection images (containing known sample defect labels and sample defect levels) are used to train the network. These labeled data can be obtained through manual annotation or from an existing capacitor pin defect database. By constructing a pin defect recognition network for different sample defect types, the defect type and level of the pin can be identified more accurately, improving the accuracy and specificity of defect recognition.

[0072] According to the order of the H sample defect types in the first parallel detection priority, set the model scheduling priority for the corresponding H pin defect recognition networks to reasonably arrange the startup order of each network. For example, if the detection order of a certain sample defect type is ranked first in the first parallel detection priority, then the model scheduling priority of the corresponding pin defect recognition network is higher. These priority relationships can be stored and managed through a priority list or configuration file. Determine the running order of the pin defect recognition network so that the network can be called in a reasonable order when performing pin surface defect recognition, improve recognition efficiency and match the previous detection order.

[0073] The first inspection image sequence is input into H pin defect recognition networks, and these networks are started one by one according to the model scheduling priority. During the network operation, the network analyzes and calculates according to the input image data, and outputs the corresponding sample defect labels and sample defect levels, that is, the first sample defect label sequence and the first sample defect level sequence. For example, in a network built using the TensorFlow framework, a session mechanism can be used to start the network and obtain the output results.

[0074] A data structure (such as a dictionary or a data table) is used to associate and store the first sample defect label sequence and the first sample defect level sequence. For example, a dictionary can be created with the sample defect label as the key and the corresponding sample defect level as the value to obtain the first sample defect information.

[0075] According to the method similar to obtaining the first sample defect information, pin defect recognition is performed on K sample test data sets respectively. For each sample test data set, a corresponding pin defect recognition network is constructed, the model scheduling priority is set, pin surface defect recognition is performed, defect labels and level sequences are associated and stored, and finally K sample defect information is obtained.

[0076] The above steps realize the automatic identification and storage of sample pin defects by building a defect recognition network, priority scheduling and batch testing. The construction of a dedicated network realizes the accurate identification of different defect types; the reasonable allocation of computing resources through priority scheduling improves the detection efficiency, and the unified storage of detection results provides complete data support for subsequent analysis.

[0077] Furthermore, the repair feasibility evaluation function is as follows: ; Among them, R is the defect repair feasibility coefficient, D is the repair difficulty, C is the repair cost, L is the repair life loss, P is the repair performance loss, and T is the repair time.

[0078] Specifically, the repair feasibility function is given by and Two items. This one, It is the product of the difficulty of repair and the cost of repair, reflecting the cost of repair from the perspective of difficulty and cost. The larger the value of this item, the more resources and technical support are required for repair, which may make the repair more challenging and costly.

[0079] In the term, the molecule It is the product of the repair life loss and the repair performance loss, taking into account the impact of repair on the life and performance of the capacitor; the denominator Makes repair time-consuming The effect on the results is nonlinear. As the value of this part increases, it will decrease rapidly. Then multiply these two parts and take the square root to get an intermediate result that takes multiple factors into consideration. The longer the repair takes, the greater the impact of the loss of life and performance after the repair on the feasibility of the repair. Longer repair time may result in poor performance and life of the repaired equipment.

[0080] The term takes into account the situation where the repair cost changes with the repair difficulty. The increase in repair difficulty usually leads to an increase in repair cost. When the repair difficulty is high, the impact on the repair feasibility is greater, reflecting the positive correlation between difficulty and cost.

[0081] The repair feasibility evaluation function quantifies multiple important factors in the repair process, including repair difficulty, cost, impact on life and performance, time consumption, and the changing relationship between repair cost and repair difficulty. The defect repair feasibility coefficient R calculated by this function can quantitatively evaluate the repair feasibility of different sample defect types, providing a scientific and reasonable basis for subsequent operations such as determining the priority of the inspection project.

[0082] Furthermore, step S5 further includes: Step S51: using the capacitor model of the batch capacitors as a search instruction, performing network data call to obtain historical quality inspection data, wherein the historical quality inspection data includes multiple historical defect detection sequences of multiple historical defective pins.

[0083] Step S52: obtaining a defect association relationship map by performing defect association identification on the multiple historical defect detection sequences.

[0084] Step S53: generating defect detection normalization rules based on the defect association relationship map.

[0085] Step S54: using the defect detection normalization rule to traverse the K sample defect information, remove redundant defect detection results, and obtain K updated defect information.

[0086] Step S55: Aggregate the K updated defect information according to the multiple sample defect types to obtain the sample detection defect distribution, wherein the sample detection defect distribution includes multiple groups of sample defect level percentages of the multiple sample defect types.

[0087] Specifically, a search instruction is generated according to the capacitor model, and the historical quality inspection data related to the capacitor model is called through a networked database (such as a production record database or an industry fault analysis platform). These historical quality inspection data include multiple historical defect detection sequences of multiple historical defect pins, that is, the test result sequences recorded for multiple historical defect pins, such as the type and grade of the detected defects. By obtaining historical quality inspection data, more references are provided for subsequent defect correlation analysis and quality assessment, and historical experience can be used to optimize the current inspection results.

[0088] Defect association identification is performed on defect types in multiple historical defect detection sequences, the relationships between these defect types are found, and then a defect association relationship map is constructed based on the analyzed relationships. The map is a data structure that graphically represents the association relationship between defect types, where nodes represent defect types and edges represent the association relationship between defect types. Exemplarily, data mining and graph algorithms can be used to implement defect association identification. First, the defect types in multiple historical defect detection sequences are extracted and counted. Then, information such as the frequency of simultaneous occurrence of different defect types is analyzed. For example, if defect type A and defect type B often appear in the same historical defect detection sequence at the same time, it can be considered that there is an association relationship between them. Use a graph building tool (such as the NetworkX library, in Python) to construct a defect association relationship map, with defect types as nodes and association relationships as edges. The defect association relationship map intuitively shows the relationship between different defect types, which helps to deeply understand the intrinsic connection of capacitor pin defects and provides a basis for formulating defect detection normalization rules.

[0089] Defect detection normalization rules are generated based on the relationship between nodes and edges in the defect association graph. The normalization rules are used to standardize and unify defect detection results. For example, if defect type A and defect type B are found to have a strong correlation in the graph and can be regarded as different manifestations of the same defect under certain conditions, then the defect detection normalization rules can stipulate that when A and B are detected at the same time, they are merged or processed uniformly in some way. Defect detection normalization rules can improve the accuracy and consistency of defect detection results and avoid differences in results caused by different detection angles or standards.

[0090] Traverse each of the K sample defect information, and for each defect type and level, make a judgment according to the defect detection normalization rule. If it is found that there is a need to merge or eliminate, perform the corresponding operation. For example, if according to the rules, two defect types of a sample are regarded as the same defect, then they are merged into one defect, and the relevant information such as the sample defect level is updated. For example, if bending and electrical parameter abnormalities are detected at the same time, the sample defect level corresponding to the bending is retained. By eliminating redundant defect detection results, more accurate and refined information about the defects of the K sample capacitor pins, that is, K updated defect information, can be obtained, which improves the accuracy of the sample defect information, reduces the interference caused by redundant information, and makes the subsequent defect distribution analysis more reliable.

[0091] For K updated defect information, count the number of occurrences of each sample defect type in different samples and the corresponding defect level. Then calculate the defect level percentage of each sample defect type in all samples to generate the sample detection defect distribution. For example, for sample defect type A, count that n samples have A defects among K samples, and classify and count them according to different defect levels to calculate the percentage of each level. Data statistics and analysis tools (such as the Pandas library in Python) can be used to implement these calculations. The final sample detection defect distribution can comprehensively reflect the distribution and severity of various sample defect types in the sample, providing an important basis for capacitor quality assessment, production process improvement, etc.

[0092] Further, step S7 includes: Step S71: configuring defect interference weights for the multiple sample defect types to obtain multiple sample defect interference weights.

[0093] Step S72: configuring defect level weights for the multiple sample defect types to obtain multiple groups of sample defect level weights.

[0094] Step S73: using the plurality of sample defect interference weights and the plurality of groups of sample defect level weights, weighted processing is performed on the plurality of groups of sample defect level percentages to obtain a defect comprehensive interference coefficient.

[0095] Step S74: extracting the sampling qualified percentage from the batch quality distribution, wherein the batch quality distribution also includes the sampling scrap percentage and the sampling repair percentage.

[0096] Step S75: preset the sampling acceptance threshold and defect comprehensive interference threshold.

[0097] Step S76: The quality inspection result is qualified if and only if the defect comprehensive interference coefficient and the sampling qualified percentage meet the sampling qualified threshold and the defect comprehensive interference threshold respectively.

[0098] Specifically, defect interference weights are configured for a variety of sample defect types to obtain a plurality of sample defect interference weights. These sample defect interference weights are used to measure the degree of interference of each sample defect type on the overall performance or quality of the capacitor. The interference weight of each sample defect type can be determined based on empirical data, industry standards or through experiments. For example, if a broken pin has a very large impact on the performance of the capacitor, it may be assigned a higher interference weight, such as 0.8; while a slight scratch on the surface of the pin has a relatively small impact on the performance, it may be assigned a lower interference weight, such as 0.2. The sample defect interference weight can more accurately reflect the degree of influence of different defect types on the quality of the capacitor, and provide important parameters for the subsequent comprehensive evaluation of the quality of the capacitor.

[0099] Similarly, defect level weights are configured for multiple sample defect types to obtain multiple groups of sample defect level weights to measure the impact of the same sample defect type of different levels on the overall performance or quality of the capacitor. For example, for the defect type of pin bending, a slight bend (level 1) may be assigned a lower level weight, such as 0.3; a moderate bend (level 2) may be assigned 0.6; and a severe bend (level 3) may be assigned 0.9. Sample defect level weights help refine the impact assessment of defects of different levels, making quality assessment more detailed and accurate.

[0100] Determine the actual proportion of each sample defect type in the sample based on the defect grade percentage of multiple groups of samples. Then, for each sample defect type, multiply its sample defect interference weight by the corresponding sample defect grade weight, and then multiply it by the sample defect grade percentage of the defect type. Finally, add the calculated results of all sample defect types to obtain the defect comprehensive interference coefficient. For example, there are two sample defect types A and B, and the sample defect grade percentage of A is p A , the interference weight is w A , the level weight is w A−level ; The percentage of defect level of sample B is p B , the interference weight is w B , the level weight is w B−level , then the defect comprehensive interference coefficient C=w A × A−level ×p A +w B × B−level ×p B The defect comprehensive interference coefficient can comprehensively evaluate the impact of defects on capacitor quality by integrating multiple factors, and provides a comprehensive quantitative indicator for judging whether the capacitor quality is qualified.

[0101] The batch quality distribution includes information such as sampling qualified percentage, sampling scrap percentage and sampling repair percentage. The sampling qualified percentage indicator is extracted from the batch quality distribution data.

[0102] The sampling pass threshold and defect comprehensive interference threshold are pre-set according to product quality requirements or industry standards. Among them, the sampling pass threshold is used to determine whether the sampling pass percentage is qualified. The defect comprehensive interference threshold is used to determine whether the defect comprehensive interference coefficient is within an acceptable range. For example, the sampling pass threshold can be set to 90%, indicating that the proportion of qualified products in the sampled products should be at least 90%; the defect comprehensive interference threshold can be set to 0.3, indicating that the defect comprehensive interference coefficient should not exceed 0.3. These thresholds can be stored in configuration files or system parameters, providing clear standards for judging whether the capacitor quality test results are qualified, making the quality assessment objective and operational.

[0103] Compare the defect comprehensive interference coefficient and the defect comprehensive interference threshold, and compare the sampling qualified percentage and the sampling qualified threshold. If the defect comprehensive interference coefficient is less than or equal to the defect comprehensive interference threshold, and the sampling qualified percentage is greater than or equal to the sampling qualified threshold, the quality inspection result is qualified; otherwise, it is unqualified. When the quality inspection result is unqualified, the batch of capacitors will be collectively scrapped.

[0104] The above steps use a dual threshold rule to quantitatively evaluate the quality inspection results of batch capacitors through a comprehensive analysis of defect interference weights, grade weights and batch quality distribution, and comprehensively consider the impact of defect types and grades to achieve a more scientific quality assessment, providing data basis and operational guidance for product quality control and further optimization.

[0105] In summary, the automated detection method for capacitor pins provided in the embodiments of the present application has the following technical effects: The embodiment of the present application realizes efficient and accurate detection of batch capacitor pins through an automated detection device and a refined defect identification and quality evaluation method. First, representative samples are collected by presetting the sampling ratio to ensure the representativeness of the test data. Secondly, the test items are sorted according to the feasibility of fault repair, the test process is optimized, and the test efficiency is improved. Then, the scheduling detection device collects data from the pins of the sample capacitor to ensure the accuracy and completeness of the test data. Then, by analyzing the test data, the defects of the pins are identified, and the defects are labeled and graded to achieve refined management of the defects. Finally, the sample detection defect distribution and batch quality distribution are comprehensively analyzed, and the final quality inspection results are output, providing comprehensive and intuitive quality information, which is convenient for timely discovery and resolution of quality problems. Overall, the embodiment of the present application not only improves the quality inspection efficiency of capacitor pins, but also improves the quality inspection accuracy, thereby enhancing the quality control accuracy of batch capacitors and meeting the needs of precise quality control.

[0106] Embodiment 2, as Figure 4 As shown, based on the same inventive concept as the above-mentioned embodiment 1, the embodiment of the present application provides an automatic detection device for capacitor pins, the device comprising: The sample collection module 10 is used to collect test samples from batch capacitors to obtain K sample capacitors, wherein the K sample capacitors and the batch capacitors meet a preset sampling ratio.

[0107] The detection item sorting module 20 is used to sort the detection items according to the feasibility of fault repair and obtain a sequence of items to be detected.

[0108] The detection data set acquisition module 30 is used to schedule the detection device to perform detection data acquisition on the K groups of capacitor pins of the K sample capacitors based on the sequence of items to be detected as a constraint, so as to obtain K sample detection data sets.

[0109] The pin defect recognition module 40 is used to perform pin defect recognition according to the K sample detection data sets to obtain K sample defect information, wherein the sample defect information includes a sample defect label sequence and a sample defect level sequence.

[0110] The defect information aggregation module 50 is used to aggregate the K sample defect information to obtain sample detection defect distribution.

[0111] The quality evaluation module 60 is used to perform quality evaluation on the K sample capacitors according to the K sample defect information to obtain batch quality distribution.

[0112] The test result output module 70 is used to comprehensively analyze the sample test defect distribution and batch quality distribution, and output the quality test results of the batch capacitors.

[0113] Furthermore, the detection item sorting module 20 of the embodiment of the present application is also used to perform the following steps: Using the capacitor model of the batch capacitors as a search instruction, network data call is performed to obtain multiple repair feasibility evaluation indicators for multiple sample defect types, wherein the repair feasibility evaluation indicators include repair difficulty, repair cost, repair life loss, repair performance loss and repair time consumption; the multiple repair feasibility evaluation indicators are synchronized to the repair feasibility evaluation function to obtain multiple defect repair feasibility coefficients; multiple sample detection devices of the multiple sample defect types are matched to obtain; the multiple sample detection devices are aggregated based on device consistency to obtain M types of sample detection devices; according to the M types of sample detection devices and the multiple defect repair feasibility coefficients, the multiple sample defect types are prioritized for parallel detection, and the sequence of items to be detected is output.

[0114] Furthermore, the detection item sorting module 20 of the embodiment of the present application is also used to perform the following steps: Aggregating the multiple defect repair feasibility coefficients according to the M types of sample detection devices to obtain M groups of defect repair feasibility coefficients; obtaining M comprehensive repair feasibility coefficients by intra-group summing up the M groups of defect repair feasibility coefficients; serializing the M types of sample detection devices according to the M comprehensive repair feasibility coefficients to obtain detection device scheduling priorities; performing parallel detection priority sorting on the multiple sample defect types according to the M groups of defect repair feasibility coefficients to obtain M parallel detection priorities; and associatively storing the detection device scheduling priorities and the M parallel detection priorities to generate the sequence of items to be detected.

[0115] Furthermore, the detection data collection module 30 of the embodiment of the present application is also used to perform the following steps: Guided by the scheduling priority of the detection device in the sequence of items to be detected, the first detection device entity of the first sample detection device is scheduled, wherein the first detection device entity is an image acquisition device; the first parallel detection priority is parsed to obtain an image acquisition requirement sequence, wherein the first parallel detection priority includes H types of sample defect types; with the image acquisition requirement sequence as a constraint, the first detection device entity is run to perform automatic acquisition of detection data on a first group of capacitor pins of a first sample capacitor to obtain a first acquisition detection data set, wherein the first acquisition detection data set includes a first detection image sequence corresponding to the image acquisition requirement sequence; by analogy, the M-1 parallel detection priority is parsed, and according to the parsing result, the M-1 detection device entity of the M-1 sample detection device is run to perform automatic acquisition of detection data on the first group of capacitor pins to obtain an M-1 acquisition detection data set, wherein the first acquisition detection data set and the M-1 acquisition detection data set constitute a first sample detection data set; by analogy, detection data set acquisition is performed on the K groups of capacitor pins to obtain K sample detection data sets.

[0116] Furthermore, the pin defect identification module 40 of the embodiment of the present application is also used to perform the following steps: H pin defect recognition networks are constructed according to the H types of sample defects, wherein the input data of the pin defect recognition network is the sample detection image, and the output data is the sample defect label and the sample defect level; the model scheduling priority of the H pin defect recognition networks is set with the first parallel detection priority as a constraint; after synchronizing the first detection image sequence to the H pin defect recognition networks, the H pin defect recognition networks are started one by one with the model scheduling priority as a constraint to perform pin apparent defect recognition, and a first sample defect label sequence and a first sample defect level sequence are output; the first sample defect label sequence and the first sample defect level sequence are associated and stored to obtain the first sample defect information; and by analogy, pin defect recognition is performed according to the K sample detection data sets to obtain the K sample defect information.

[0117] Furthermore, the repair feasibility evaluation function is as follows: ; Among them, R is the defect repair feasibility coefficient, D is the repair difficulty, C is the repair cost, L is the repair life loss, P is the repair performance loss, and T is the repair time.

[0118] Furthermore, the defect information aggregation module 50 in the embodiment of the present application is also used to perform the following steps: Using the capacitor model of the batch capacitors as a retrieval instruction, network data is called to obtain historical quality inspection data, wherein the historical quality inspection data includes multiple historical defect detection sequences of multiple historical defect pins; defect association identification is performed on the multiple historical defect detection sequences to obtain a defect association relationship map; defect detection normalization rules are generated based on the defect association relationship map; the K sample defect information is traversed using the defect detection normalization rules to eliminate redundant defect detection results to obtain K updated defect information; according to the multiple sample defect types, the K updated defect information is aggregated to obtain the sample detection defect distribution, wherein the sample detection defect distribution includes multiple groups of sample defect level percentages of the multiple sample defect types.

[0119] Furthermore, the detection result output module 70 of the embodiment of the present application is also used to perform the following steps: Defect interference weights are configured for the multiple sample defect types to obtain multiple sample defect interference weights; defect level weights are configured for the multiple sample defect types to obtain multiple groups of sample defect level weights; the multiple sample defect interference weights and multiple groups of sample defect level weights are used to weight the multiple groups of sample defect level percentages to obtain a comprehensive defect interference coefficient; the sampling qualified percentage is extracted from the batch quality distribution, wherein the batch quality distribution also includes the sampling scrap percentage and the sampling repair percentage; a sampling qualified threshold and a defect comprehensive interference threshold are preset; the quality inspection result is qualified if and only if the defect comprehensive interference coefficient and the sampling qualified percentage meet the sampling qualified threshold and the defect comprehensive interference threshold respectively.

[0120] Through the detailed description of the automated detection method for capacitor pins in the foregoing specification, those skilled in the art can clearly understand the automated detection device for capacitor pins in this embodiment. For the device disclosed in Embodiment 2, since it corresponds to the method disclosed in Embodiment 1 and has corresponding functional modules and beneficial effects, the relevant parts can be referred to the description of the method part.

[0121] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An automated detection method for capacitor pins, characterized in that: The method comprises: Collecting test samples of the batch capacitors to obtain K sample capacitors, wherein the K sample capacitors and the batch capacitors satisfy a preset sampling ratio; Sort the inspection items according to the feasibility of fault repair to obtain a sequence of items to be inspected; Taking the sequence of items to be detected as a constraint, scheduling the detection device to perform detection data set collection on K groups of capacitor pins of the K sample capacitors to obtain K sample detection data sets; Perform pin defect recognition according to the K sample detection data sets to obtain K sample defect information, wherein the sample defect information includes a sample defect label sequence and a sample defect level sequence; Aggregating the K sample defect information to obtain sample detection defect distribution; Performing quality evaluation on the K sample capacitors according to the K sample defect information to obtain batch quality distribution; The sample inspection defect distribution and batch quality distribution are comprehensively analyzed, and the quality inspection results of the batch capacitors are output.

2. The automated detection method for capacitor pins according to claim 1, characterized in that: The detection items are sorted according to the feasibility of fault repair to obtain a sequence of items to be detected, and the method includes: Using the capacitor model of the batch capacitor as a search instruction, network data is called to obtain multiple repair feasibility evaluation indicators for multiple sample defect types, wherein the repair feasibility evaluation indicators include repair difficulty, repair cost, repair life loss, repair performance loss and repair time consumption; Synchronizing the multiple repair feasibility assessment indicators to a repair feasibility evaluation function to obtain multiple defect repair feasibility coefficients; Matching and obtaining a plurality of sample detection devices of the plurality of sample defect types; Aggregating the multiple sample detection devices based on device consistency to obtain M types of sample detection devices; According to the M types of sample detection devices and a plurality of defect repair feasibility coefficients, the plurality of sample defect types are prioritized for parallel detection, and the sequence of items to be detected is output.

3. The automated detection method for capacitor pins according to claim 2, characterized in that: According to the M types of sample detection devices and a plurality of defect repair feasibility coefficients, the plurality of sample defect types are prioritized for parallel detection, and the sequence of items to be detected is outputted. The method comprises: Aggregating the plurality of defect repair feasibility coefficients according to the M types of sample detection devices to obtain M groups of defect repair feasibility coefficients; By adding the M groups of defect repair feasibility coefficients within the group, M comprehensive repair feasibility coefficients are obtained; Serialize the M types of sample detection devices according to the M comprehensive repair feasibility coefficients to obtain the scheduling priority of the detection device; According to the M groups of defect repair feasibility coefficients, the plurality of sample defect types are prioritized for parallel detection to obtain M parallel detection priorities; The detection device scheduling priority and the M parallel detection priorities are stored in association to generate the sequence of items to be detected.

4. The automated detection method for capacitor pins according to claim 3, characterized in that: Taking the sequence of items to be detected as a constraint, scheduling a detection device to perform detection data set collection on K groups of capacitor pins of the K sample capacitors to obtain K sample detection data sets, the method comprising: Based on the scheduling priority of the detection device in the sequence of items to be detected, scheduling a first detection device entity of a first sample detection device, wherein the first detection device entity is an image acquisition device; Parsing a first parallel detection priority to obtain an image acquisition requirement sequence, wherein the first parallel detection priority includes H sample defect types; Taking the image acquisition requirement sequence as a constraint, operating the first detection device entity to automatically acquire detection data on a first group of capacitor pins of a first sample capacitor to obtain a first acquired detection data set, wherein the first acquired detection data set includes a first detection image sequence corresponding to the image acquisition requirement sequence; By analogy, the M-1 parallel detection priority is parsed, and according to the parsing result, the M-1 detection device entity of the M-1 sample detection device is operated to automatically collect detection data on the first group of capacitor pins to obtain an M-1 collection detection data set, wherein the first collection detection data set and the M-1 collection detection data set constitute a first sample detection data set; By analogy, detection data collection is performed on the K groups of capacitor pins to obtain K sample detection data sets.

5. The automated detection method for capacitor pins according to claim 4, characterized in that: Performing pin defect recognition according to the K sample detection data sets to obtain K sample defect information, wherein the sample defect information includes a sample defect label sequence and a sample defect level sequence, the method comprising: Constructing H pin defect recognition networks according to the H sample defect types, wherein the input data of the pin defect recognition network is the sample detection image, and the output data is the sample defect label and the sample defect level; Taking the first parallel detection priority as a constraint, setting the model scheduling priority of the H pin defect recognition networks; After synchronizing the first detection image sequence to the H pin defect recognition networks, starting the H pin defect recognition networks one by one to perform pin apparent defect recognition with the model scheduling priority as a constraint, and outputting a first sample defect label sequence and a first sample defect level sequence; Associatively storing the first sample defect label sequence and the first sample defect level sequence to obtain first sample defect information; By analogy, pin defect identification is performed based on the K sample detection data sets to obtain the K sample defect information.

6. The automated detection method for capacitor pins according to claim 2, characterized in that: The repair feasibility evaluation function is as follows: ； Among them, R is the defect repair feasibility coefficient, D is the repair difficulty, C is the repair cost, L is the repair life loss, P is the repair performance loss, and T is the repair time.

7. The automated detection method for capacitor pins according to claim 5, characterized in that: The method further comprises: Using the capacitor model of the batch capacitor as a search instruction, performing online data call to obtain historical quality inspection data, wherein the historical quality inspection data includes multiple historical defect detection sequences of multiple historical defective pins; By performing defect correlation identification on the multiple historical defect detection sequences, a defect correlation relationship map is obtained; Generating defect detection normalization rules based on the defect association relationship map; The K sample defect information is traversed using the defect detection normalization rule to eliminate redundant defect detection results to obtain K updated defect information; The K updated defect information are aggregated according to the multiple sample defect types to obtain the sample detection defect distribution, wherein the sample detection defect distribution includes multiple groups of sample defect level percentages of the multiple sample defect types.

8. The automated detection method for capacitor pins according to claim 7, characterized in that: Comprehensively analyzing the sample detection defect distribution and batch quality distribution, and outputting the quality detection results of the batch capacitors, the method includes: Performing defect interference weight configuration on the multiple sample defect types to obtain multiple sample defect interference weights; Performing defect level weight configuration on the multiple sample defect types to obtain multiple groups of sample defect level weights; Using the plurality of sample defect interference weights and the plurality of groups of sample defect grade weights, weighted processing is performed on the plurality of groups of sample defect grade percentages to obtain a defect comprehensive interference coefficient; Extracting a sampling qualified percentage from the batch quality distribution, wherein the batch quality distribution also includes a sampling scrap percentage and a sampling repair percentage; Preset sampling acceptance threshold and defect comprehensive interference threshold; The quality inspection result is qualified if and only if the defect comprehensive interference coefficient and the sampling qualified percentage meet the sampling qualified threshold and the defect comprehensive interference threshold respectively.

9. An automated detection device for capacitor pins, characterized in that: The device is used to execute the automatic detection method of capacitor pins according to any one of claims 1 to 8, comprising: A sample collection module, used to collect test samples of batch capacitors to obtain K sample capacitors, wherein the K sample capacitors and the batch capacitors meet a preset sampling ratio; A test item sorting module is used to sort the test items according to the feasibility of fault repair and obtain a sequence of items to be tested; A detection data set acquisition module, used to schedule the detection device to perform detection data acquisition on K groups of capacitor pins of the K sample capacitors based on the sequence of items to be detected as a constraint, so as to obtain K sample detection data sets; A pin defect recognition module, used to perform pin defect recognition according to the K sample detection data sets to obtain K sample defect information, wherein the sample defect information includes a sample defect label sequence and a sample defect level sequence; A defect information aggregation module is used to aggregate the K sample defect information to obtain the sample detection defect distribution; A quality evaluation module, used to perform quality evaluation on the K sample capacitors according to the K sample defect information to obtain batch quality distribution; The test result output module is used to comprehensively analyze the sample test defect distribution and batch quality distribution, and output the quality test results of the batch capacitors.

Images