Circuit board testing and quality screening automation method and system

Abstract

The invention discloses a circuit board testing and quality screening automation method and system, and aims to solve the problems of low efficiency, insufficient precision and imperfect quality control of traditional circuit board testing. Precise positioning of the circuit board is achieved through the high-precision X / Y / Z three-dimensional positioning mechanical arm, parallel testing is conducted in combination with equipment such as a multi-channel oscilloscope, a signal source and a spectrum analyzer, and the testing efficiency is improved. Temperature and humidity control testing is carried out through the environment simulation bin, the thermal imager scans temperature distribution of the circuit board, a high-precision instrument is used for detecting signal integrity and analyzing electromagnetic compatibility, and the performance of the high-speed circuit board is ensured. A machine learning algorithm predicts potential faults, a test process is optimized, an automatic sorting system is adopted to separate qualified and unqualified circuit boards, and high consistency and high reliability of product quality are ensured. Through the automatic and intelligent testing process, the production efficiency, the quality control capability and the testing precision are greatly improved, and the labor cost and the production risk are reduced.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of circuit board testing, in particular to a circuit board testing and quality screening automation method and system. BACKGROUND

[0002] Traditional circuit board testing and quality screening usually rely on manual operation. Manual testing is not only inefficient, but also easily affected by factors such as operator skills, experience, fatigue, etc., leading to inconsistent test results and even missing some potential quality problems. Modern circuit board designs are becoming increasingly complex, containing multiple high-speed circuits, radio frequency circuits, precision components, etc. Traditional testing equipment and methods are difficult to meet the needs of fine and multi-dimensional testing. For example, in testing high-speed signals, electrical performance, thermal characteristics, etc., a single traditional method cannot ensure high precision and stability. Circuit boards may exhibit different performance under different working environments (such as temperature and humidity changes), especially in harsh temperature and humidity conditions, where the circuit board may exhibit unstable performance or fail. Traditional testing methods often only test under standard conditions, making it difficult to simulate environmental stress in actual applications. Traditional circuit board testing methods can only detect known faults, but it is difficult for testing systems to detect potential hidden faults in advance. These hidden faults may cause device performance degradation or even failure during later use, thereby affecting product quality and user experience. SUMMARY

[0003] A circuit board testing and quality screening automation method, comprising the following steps, S1. Circuit board testing system design: high-precision X / Y / Z three-dimensional positioning mechanical arm is used for precise positioning of the circuit board, oscilloscope, signal source, spectrum analyzer with multiple channels are used to ensure parallel testing of multiple signal channels, improve testing efficiency, and design environmental simulation testing, including temperature and humidity control unit, climate simulation chamber, environmental stress testing: temperature variation, humidity verification of circuit board performance in extreme environments, precise positioning, automatic adjustment of probe contact points to ensure accurate testing of key areas such as solder joints and pins in multiple tests; S2. Electrical performance testing and fault diagnosis: electrical performance testing and fault diagnosis, high voltage / current testing, using automated instruments to test the circuit board's resistance to high voltage and high current, verifying its anti-interference and stability, timing analysis and synchronous testing, using a timing analyzer to capture signals from the circuit board in synchronization, ensuring that the timing of multiple signal channels fully complies with design specifications, multi-point parallel testing, using a multi-channel oscilloscope or scanning instrument to measure and compare multiple points simultaneously, greatly reducing testing time, self-diagnosis function, the system uses intelligent algorithms to analyze test data and automatically detect common fault points in the circuit board: virtual soldering, short circuit, open circuit, noise; S3. Signal integrity and precision testing: For high-speed circuit boards (such as RF circuits, digital circuits, etc.), use high-precision signal analyzers (such as spectrum analyzers, logic analyzers, etc.) to detect the integrity and quality of signals; through the analysis of reflection loss (S parameter) and transmission loss, ensure the stability of signal transmission; use electromagnetic compatibility (EMC) test equipment to test the radiation and anti-interference ability of the circuit board; S4. Thermal analysis and environmental adaptability testing: Use an infrared thermal imager to scan the surface temperature distribution of the circuit board, detect whether there are overheating areas, and combine with heat source simulation test environment for continuous operation test, observe the performance change of the circuit board in harsh environment under high and low temperature humidity change cycle test; S5. Automated visual inspection: Use high-definition CCD cameras and image processing technology to detect the soldering quality, component position, and printing defects of the circuit board, automatically classify and evaluate the solder joints, identify false soldering, open soldering, and excessive soldering problems, automatically detect and classify defects in images, improve the accuracy and efficiency of defect identification, and identify and judge small defects (such as scratches, foreign matter, discoloration, etc.) on the surface of the circuit board; S6. Data collection, analysis and decision support: The data (such as voltage, current, signal waveform, etc.) of each circuit board testing process is uploaded to the cloud or local server in real time to ensure that the data will not be lost, the data from multiple systems such as electrical testing, visual inspection, thermal analysis, environmental stress testing, etc. are integrated to form a unified test report, through the data analysis platform, real-time monitoring of various indicators, once an abnormality (such as a parameter exceeding the allowed range) is found, the system will immediately issue an alarm and stop or take measures to adjust, based on historical test data, using machine learning algorithms to predict test results, predict potential failure areas and automatically adjust test processes, the system can record the test results of each circuit board and generate detailed quality reports, through statistical analysis, identify bottlenecks or potential problems in the production process; S7. Automated sorting and quality screening: Automatic handling of unqualified products, according to the automatic test results, the automatic equipment sorts the qualified products and unqualified products to different areas to ensure that the qualified products flow to the production line, through automatic machine vision and function retest, the unqualified products are re-screened to ensure that the unqualified products are accurately rejected, for the unqualified circuit board, the system can analyze the specific defect type and provide detailed repair suggestions or guidance for engineers to reference, and through real-time analysis of a large amount of production data, the system can identify regular problems in production and provide improvement suggestions to help optimize production processes, test standards, automatic equipment, and test methods are optimized and updated regularly to meet new technical requirements and production challenges; S8. System maintenance and upgrade: Regular automated equipment detection: Regular detection and calibration of mechanical arms, test equipment, vision systems to ensure stable operation of the system, regular software upgrades to the control system, addition of new test standards, defect detection algorithms or new fault diagnosis models, the system has good scalability, supports the access of new equipment and new technologies, and can quickly adapt to the testing needs of new models or new technologies of circuit boards.

[0004] Further, a circuit board testing and quality screening automation method, In step S1, the environmental simulation test design is performed, including a temperature and humidity control unit and a climate simulation chamber, as follows: The temperature and humidity control unit and the climate simulation chamber provide temperature variation and humidity control environment stress testing to test the reliability of the circuit board under extreme environmental conditions. The environmental simulation can be adjusted by a PLC programmable logic controller to simulate various environmental stress conditions in different test stages.

[0005] Further, a circuit board testing and quality screening automation method, In step S2, the timing analysis and synchronization test is performed, in which a timing analyzer is used to capture the signals of the circuit board synchronously, as follows: The timing analyzer is used to capture the multi-channel signals in the circuit board synchronously to ensure that the timing between different signal channels fully complies with the design specifications. At the same time, the timestamp and data acquisition technology are used to analyze the data of the multi-signal parallel test in real time.

[0006] Further, a circuit board testing and quality screening automation method, In step S6, based on historical test data, a machine learning algorithm is used to predict the test results, predict potential fault areas and automatically adjust the test process, as follows: S61. Data preparation and preprocessing: Collect historical data from the automated test system, including test parameters of the circuit board: voltage, current, timing, fault types: virtual welding, open circuit, short circuit, test environment conditions: temperature, humidity, integrate different sources of data: electrical performance test, visual inspection, thermal analysis into a unified database, ensure that the data has a unified format and identification, and perform data cleaning on the database, and standardize / normalize the data; S62. Feature selection and extraction: Select features from the original test data that are helpful for predicting potential faults: signal waveforms, timing, temperature that affect the performance of the circuit board, design some new features: timing difference, waveform anomaly based on the working principle of the circuit board and the test data to improve the prediction performance of the model; S63. Constructing machine learning model: divide historical test data into training set and test set, train gradient boosting machine algorithm using training set data, adjust hyperparameters, and evaluate model effectiveness according to performance indicators obtained during training process: accuracy, precision, recall; S64. Potential fault prediction and analysis: use trained machine learning model to predict new circuit board, identify potential fault area, test newly produced circuit board, and input test results: voltage, current, temperature into trained model, output prediction results including fault type, fault severity, and fault location, based on model output, combine test data characteristics, further analyze which areas have greater deviation from normal mode of historical data, may have potential faults, overheat heat map and other visualization techniques display potential fault area, help engineers accurately locate problem area; S65. Automatic adjustment of test process: according to predicted fault area, automatically narrow test range or increase test frequency of specific area, when model predicts that certain area has virtual welding problem, system can increase probe contact times in these areas, perform more detailed testing, add new test points to possible fault area, ensure key detection of these areas; S66. Continuous optimization and feedback: after each test is completed, record actual test results and compare with prediction results of machine learning model to evaluate prediction accuracy, based on newly collected feedback data, continuously update and optimize machine learning model, so that model can more accurately predict potential faults over time, improve system adaptability under different test conditions.

[0007] A circuit board testing and quality screening automation system for implementing a circuit board testing and quality screening automation method; the circuit board testing and quality screening automation system comprises: Circuit board positioning and operation module: responsible for positioning, clamping and mechanical operation of circuit board, ensuring accuracy and stability of circuit board during testing process; Multi-dimensional testing and quality screening module: performs electrical performance testing, timing analysis, synchronization testing and fault diagnosis, signal integrity and precision testing, thermal analysis and environmental adaptability testing, and automated visual inspection of circuit board; Data acquisition and analysis module: acquires data from various test modules, centrally stores all test data, and based on historical data and machine learning algorithms, performs potential fault prediction, generates test report, and adjusts test process according to prediction results; Automatic sorting and quality screening module: according to the test results, the qualified and unqualified circuit boards are automatically sorted, the qualified products enter the production line, and the unqualified products are detected again or rejected. For unqualified circuit boards, the system generates repair suggestions according to the predicted fault types of machine learning for engineers to refer.

[0008] The beneficial effects of the present application: through efficient mechanical arm operation, parallel multi-channel testing, automated defect identification and other technologies, the test efficiency of the circuit board is greatly improved. Compared with manual detection or traditional testing methods, the system can complete the precise testing of a large number of circuit boards in a short time, significantly shortening the production cycle. Through precise equipment (such as spectrum analyzer, infrared thermal imager, machine vision, etc.) and automated testing process, the system can perform high-precision electrical performance testing, signal integrity analysis and environmental adaptability testing. The automated visual inspection system can also accurately identify small defects such as welding defects and component position errors, ensuring the consistency and reliability of the test results. It can perform real-time fault detection and also has a fault prediction function based on historical data. Machine learning algorithms can analyze historical test data and predict potential fault areas, take preventive measures in advance, and reduce maintenance and repair costs. BRIEF DESCRIPTION OF DRAWINGS

[0009] Figure 1 Flowchart of a circuit board testing and quality screening automation method; DETAILED DESCRIPTION

[0010] A circuit board testing and quality screening automation method, comprising the following steps, S1. Circuit board testing system design: high-precision X / Y / Z three-dimensional positioning mechanical arm is used for precise positioning of the circuit board, oscilloscope, signal source, spectrum analyzer with multiple channels are used to ensure parallel testing of multiple signal channels, improve test efficiency, and design environmental simulation test, including temperature and humidity control unit, climate simulation warehouse, environmental stress test: temperature change, humidity verification circuit board performance in extreme environment, through precise positioning, automatic adjustment of probe contact point, ensure that each solder joint, pin and other key areas can be accurately tested in multiple tests; S2. Electrical performance testing and fault diagnosis: Electrical performance testing and fault diagnosis, high voltage / current testing, using automated instruments to test the circuit board's resistance to high voltage and high current, verifying its anti-interference and stability, timing analysis and synchronization testing, using a timing analyzer to capture signals from the circuit board in synchronization, ensuring that the timing of multiple signal channels fully complies with design specifications, multi-point parallel testing, using a multi-channel oscilloscope or scanning instrument to measure and compare multiple points simultaneously, greatly reducing test time, self-diagnosis function, the system uses intelligent algorithms to analyze test data and automatically detect common faults in the circuit board: virtual welding, short circuit, open circuit, noise; S3. Signal integrity and precision testing: For high-speed circuit boards (such as RF circuits, digital circuits, etc.), use high-precision signal analyzers (such as spectrum analyzers, logic analyzers, etc.) to detect signal integrity and quality; through the analysis of reflection loss (S parameter) and transmission loss, ensure the stability of signal transmission; use electromagnetic compatibility (EMC) test equipment to test the radiation and anti-interference ability of the circuit board; S4. Thermal analysis and environmental adaptability testing: Use an infrared thermal imager to scan the surface temperature distribution of the circuit board to detect whether there are overheating areas, and combine heat source simulation test environment for continuous operation test, observe the performance change of the circuit board in harsh environment under high and low temperature humidity change cycle test in the environmental simulation box; S5. Automated visual inspection: Use high-definition CCD cameras and image processing technology to detect the soldering quality, component position, and printing defects of the circuit board, automatically classify and evaluate the solder joints, identify virtual welding, open welding, and excessive welding problems, automatically detect and classify defects in the image, improve the accuracy and efficiency of defect identification, and identify and judge small defects (such as scratches, foreign matter, discoloration, etc.) on the surface of the circuit board; S6. Data collection, analysis and decision support: The data (such as voltage, current, signal waveform, etc.) of each circuit board testing process is uploaded to the cloud or local server in real time to ensure that the data will not be lost, the data from multiple systems such as electrical testing, visual inspection, thermal analysis, and environmental stress testing are integrated to form a unified test report, through the data analysis platform, real-time monitoring of various indicators, once an abnormality (such as a parameter exceeding the allowed range) is found, the system will immediately issue an alarm and stop or take measures to adjust, based on historical test data, using machine learning algorithms to predict test results, predicting potential failure areas and automatically adjusting the test process, the system can record the test results of each circuit board and generate a detailed quality report, through statistical analysis, identify bottlenecks or potential problems in the production process; S7. Automated sorting and quality screening: automated handling of defective products, according to the automatic test results, the automatic equipment sorts the qualified products and the unqualified products to different areas, ensures that the qualified products flow to the production line, through the automatic machine vision and function retest, the unqualified products are re-screened, and the unqualified products are accurately removed, for the unqualified circuit board, the system can analyze the specific defect type and provide detailed repair suggestions or guidance for reference by engineers, and through real-time analysis of a large amount of production data, the system can identify regular problems in production and provide improvement suggestions to help optimize production processes, and the test standards, automatic equipment and test methods are optimized and updated regularly to meet new technical requirements and production challenges. S8. System maintenance and upgrade: regular automated equipment detection: regular detection and calibration of mechanical arms, test equipment and vision systems to ensure stable system operation, regular software upgrades for control systems, addition of new test standards, defect detection algorithms or new fault diagnosis models, the system has good scalability, supports the access of new equipment and new technologies, and can quickly adapt to new model or new technology circuit board test requirements.

[0011] Further, a circuit board testing and quality screening automation method, In step S1, the environmental simulation test design is performed, including a temperature and humidity control unit and a climate simulation chamber, as follows: The temperature and humidity control unit and the climate simulation chamber provide temperature variation and humidity control environment stress testing to test the reliability of the circuit board under extreme environmental conditions. The environmental simulation can be adjusted by a PLC programmable logic controller to simulate various environmental stress conditions in different test stages.

[0012] Further, a circuit board testing and quality screening automation method, In step S2, the timing analysis and synchronization test is performed, in which a timing analyzer is used to capture the signals of the circuit board synchronously, as follows: The timing analyzer is used to capture the multi-channel signals in the circuit board synchronously to ensure that the timing between different signal channels fully complies with the design specifications, and at the same time, the timestamp and data acquisition technology are used to analyze the data of the multi-signal parallel test in real time.

[0013] Further, a circuit board testing and quality screening automation method, In step S6, based on historical test data, a machine learning algorithm is used to predict test results, predict potential fault areas and automatically adjust test procedures, as follows: S61. Data preparation and preprocessing: Collect historical data from automated test systems, including test parameters of circuit boards: voltage, current, timing, fault types: virtual welding, open circuit, short circuit, test environment conditions: temperature, humidity, integrate data from different sources: electrical performance test, visual inspection, thermal analysis into a unified database, ensure that the data has a unified format and identification, and clean the database, and standardize / normalize the data; S62. Feature selection and extraction: Select features from the original test data that are helpful for predicting potential faults: signal waveforms, timing, temperature that affect circuit board performance, design some new features: timing differences, waveform abnormalities based on the working principle of the circuit board and the test data to improve the prediction performance of the model; S63. Building machine learning models: Divide the historical test data into training set and test set, use the training set data to train the gradient boosting machine algorithm, adjust the hyperparameters, and evaluate the effectiveness of the model according to the performance indicators: accuracy, precision, recall obtained during the training process; S64. Potential fault prediction and analysis: Use the trained machine learning model to predict new circuit boards, identify potential fault areas, test new production circuit boards, and input test results: voltage, current, temperature into the trained model, output prediction results, including fault type, fault severity, and fault location, based on model output, combined with test data features, further analyze which areas have a large deviation from the normal mode of historical data, which may have potential faults, and visualize the potential fault area using heat maps and other visualization techniques to help engineers accurately locate the problem area; S65. Automatic adjustment of test process: According to the predicted fault area, automatically narrow the test range or increase the test frequency of specific areas, when the model predicts that some areas have virtual welding problems, the system can increase the number of probe contacts in these areas for more detailed testing, and add new test points to possible fault areas to ensure focused detection of these areas; S66. Continuous optimization and feedback: After each test is completed, record the actual test results and compare them with the prediction results of the machine learning model to evaluate the accuracy of the prediction, based on the newly collected feedback data, continuously update and optimize the machine learning model, so that the model can more accurately predict potential faults over time, and improve the system's adaptability under different test conditions.

[0014] A circuit board testing and quality screening automation system for implementing a circuit board testing and quality screening automation method; the circuit board testing and quality screening automation system comprises: Circuit board positioning and operation module: responsible for the positioning, clamping and mechanical operation of the circuit board, to ensure the accuracy and stability of the circuit board during testing; Multi-dimensional testing and quality screening module: performs electrical performance testing, timing analysis, synchronous testing and fault diagnosis, signal integrity and precision testing, thermal analysis and environmental adaptability testing, and automated visual inspection of the circuit board; Data acquisition and analysis module: collects data from various test modules, stores all test data centrally, and based on historical data and machine learning algorithms, performs potential fault prediction, generates test reports, and adjusts the test process according to the prediction results; Automatic sorting and quality screening module: automatically sorts qualified and unqualified circuit boards according to test results, qualified products enter the production line, unqualified products are detected again or rejected, for unqualified circuit boards, the system generates repair suggestions based on machine learning predicted fault types for engineers to reference.

Claims

1. A method for automated testing and quality screening of circuit boards, comprising: Comprising the following steps, S1. Circuit board test system design: Use high-precision X / Y / Z three-dimensional positioning mechanical arm for precise positioning of the circuit board, use oscilloscope, signal source, spectrum analyzer with multiple channels to ensure parallel testing of multiple signal channels, improve test efficiency, and design environmental simulation test, including temperature and humidity control unit, climate simulation warehouse, environmental stress test: temperature change, humidity verification circuit board performance in extreme environment, through precise positioning, automatic adjustment of probe contact point, accurate test of each solder joint, pin key area in multiple tests; S2. Electrical performance testing and fault diagnosis: Electrical performance testing and fault diagnosis, high voltage, current test, use automated instruments to test the resistance of the circuit board under high voltage and high current, verify its anti-interference and stability, timing analysis and synchronous test, through the timing analyzer to capture the signals of the circuit board, ensure that the timing of multiple signal channels completely meets the design specification, multi-point parallel test, through the multi-channel oscilloscope and scanner, implement simultaneous measurement and comparison of multiple points, shorten the test time, self-diagnosis function, the system preliminary analyzes the test data through intelligent algorithm, automatically detects common fault points in the circuit board: virtual welding, short circuit, open circuit, noise; S3. Signal integrity and precision testing: For high-speed circuit boards: RF circuits, digital circuits, use high-precision signal analyzers: spectrum analyzers to detect signal integrity and quality; Through reflection loss: S parameter and transmission loss analysis, analyze the stability of signal transmission; Use electromagnetic compatibility EMC test equipment to test the radiation and anti-interference ability of the circuit board; S4. Thermal analysis and environmental adaptability test: Use infrared thermal imager to scan the surface temperature distribution of the circuit board, detect whether there are overheating areas, and combine with heat source simulation test environment for continuous work test, observe the performance change of the circuit board in harsh environment in the environmental simulation box under high and low temperature and humidity change cycle test; S5. Automated visual inspection: Use high-definition CCD camera and image processing technology to detect the soldering quality, component position, and silk screen defects of the circuit board, automatically classify and evaluate the solder joints, identify virtual welding, open welding, and excessive welding problems, automatically detect and classify images to improve the accuracy and efficiency of defect recognition, identify and judge small defects on the surface of the circuit board: scratches, foreign matter, discoloration; S6. Data collection, analysis and decision support: The data of each circuit board testing process: voltage, current, signal waveform is uploaded to the local server in real time, the data from the electrical test, visual inspection, thermal analysis, environmental stress test multiple system is integrated, the unified test report is formed, through the data analysis platform, real-time monitoring of each index, when finding abnormal: a parameter exceeds the allowed range, the system immediately sends an alarm, and stops, takes measures to adjust, and based on historical test data, using machine learning algorithm to predict the test results, predict the potential failure area and automatically adjust the test process, record the test results of each circuit board, and generate detailed quality report, through statistical analysis, identify the bottlenecks in the production process, potential problems; S7. Automatic sorting and quality screening: automatic processing of unqualified products, according to the automatic test results, the automatic equipment sorts the qualified products and unqualified products to different areas, ensures that the qualified products flow to the production line, through automatic machine vision and function retest, the unqualified products are re-screened, for the unqualified circuit board, the system can analyze the specific defect type, and provide detailed repair suggestion, guidance, for the reference of engineers, the test standard, automatic equipment, test method is optimized and updated regularly, to meet the new technical requirements and production challenges; S8. System maintenance and upgrade: regular automatic equipment detection, regular detection and calibration of mechanical arm, test equipment, vision system, regular software upgrade of control system, addition of new test standards, defect detection algorithm, new fault diagnosis model, support for access of new equipment and new technology, can quickly adapt to new model and new technology circuit board test demand.

2. The circuit board testing and quality screening automation method of claim 1, wherein the environmental simulation test design in step S1 includes a temperature and humidity control unit and a climate simulation chamber, and specifically as follows: The temperature and humidity control unit and the climate simulation chamber provide temperature variation and humidity control environment stress testing to test the reliability of the circuit board under extreme environments. The environmental simulation can be adjusted by a PLC programmable logic controller to simulate various environmental stress conditions in different test stages.

3. The circuit board testing and quality screening automation method of claim 1, wherein the timing analysis and synchronous testing in step S2 is specifically as follows: The timing analyzer is used to synchronously capture the signals in the circuit board, to ensure that the timing between different signal channels completely meets the design specifications, and to use timestamps and data acquisition technology to analyze the data of the multi-signal parallel test in real time.

4. The circuit board testing and quality screening automation method of claim 1, wherein the machine learning algorithm is used to predict the test results based on historical test data, to predict the potential failure area and automatically adjust the test process, and the specific steps are as follows: ​ ​ ​ S61. Data preparation and preprocessing: Collect historical data from automated test systems, including test parameters of circuit boards: voltage, current, timing, fault types: virtual welding, open circuit, short circuit, test environment conditions: temperature, humidity, integrate data from different sources: electrical performance test, visual inspection, thermal analysis into a unified database, ensure that the data has a unified format and identification, and clean the database, and standardize and normalize the data; S62. Feature selection and extraction: Select features from the original test data that are helpful for predicting potential faults: signal waveforms, timing, temperature that affect circuit board performance, design new features: timing differences, waveform abnormalities based on the working principle of the circuit board and test data to improve model prediction performance; S63. Constructing a machine learning model: Divide the historical test data into training set and test set, use the training set data to train the gradient boosting machine algorithm, adjust the hyperparameters, and evaluate the effectiveness of the model according to the performance indicators: accuracy, precision, recall obtained during training; S64. Potential fault prediction and analysis: Use the trained machine learning model to predict new circuit boards, identify potential fault areas, test new production circuit boards, and input test results: voltage, current, temperature into the trained model, the model outputs prediction results, including fault type, fault severity and fault location, based on model output, combined with test data features, further analyze the normal mode of different regions of test data and historical data, there is a potential fault, the potential fault area is displayed through heat map visualization technology, helping engineers accurately locate the problem area; S65. Automatically adjust the test process: According to the predicted fault area, automatically narrow the test range, increase the test frequency of specific areas, when the model predicts that some areas have virtual welding problems, the system increases the number of probe contacts in these areas, conducts detailed testing, and adds new test points to possible fault areas to ensure focused detection of these areas; S66. Continuous optimization and feedback: After each test is completed, record the actual test results and compare them with the prediction results of the machine learning model to evaluate the accuracy of the prediction, based on the newly collected feedback data, continuously update and optimize the machine learning model, so that the model can more accurately predict potential faults over time, and improve the system's adaptability under different test conditions.

5. A circuit board testing and quality screening automation system, characterized by, The circuit board test and quality screening automation system is used to implement any one of the circuit board test and quality screening automation methods of claims 1-4; the circuit board test and quality screening automation system comprises: Circuit board positioning and operation module: responsible for the positioning, clamping and mechanical operation of the circuit board, ensuring the accuracy and stability of the circuit board during testing; Multi-dimensional test and quality screening module: performs electrical performance test, timing analysis, synchronous test and fault diagnosis, signal integrity and precision test, thermal analysis and environmental adaptability test, and automatic visual inspection of circuit boards. Data collection and analysis module: Collect data from various test modules, store all test data centrally, and based on historical data and machine learning algorithms, make potential fault prediction, generate test report, and adjust test process according to prediction results; Automatic sorting and quality screening module: According to the test results, the qualified and unqualified circuit boards are automatically sorted, the qualified products enter the production line, and the unqualified products are detected again and rejected. For unqualified circuit boards, the system generates repair suggestions based on machine learning predicted fault types for engineers to refer.

Images