Online fully automatic detection method and system for printed circuit board defects based on data analysis

Abstract

The present invention discloses a method and system for online fully automatic detection of printed circuit board defects based on data analysis, and relates to the technical field of printed circuit board performance testing. The invention realizes a comprehensive evaluation of circuit board performance through steps one to four. By configuring the test system and setting the test conditions, the accurate installation and debugging of each test module are ensured; by calculating the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz, the comprehensive evaluation index Pc is obtained by correlation calculation, and the performance evaluation is performed to effectively determine whether the circuit board is qualified. When the comprehensive evaluation index Pc is lower than the first performance threshold value X, a performance unqualified label is generated, and an in-depth hierarchical evaluation is performed through the second multi-level evaluation module to promote the identification of specific performance defects.

Description

Technical Field

[0001] The present invention relates to the technical field of printed circuit board performance testing, and in particular to an online fully automatic detection method and system for printed circuit board defects based on data analysis. Background Art

[0002] Printed circuit boards are core components of computer systems, undertaking key tasks such as computing, data processing, communication, and storage management. Typical circuit boards include motherboards, graphics cards, network cards, storage controllers, etc., and their performance directly determines the overall stability and operating efficiency of computer systems. However, in actual applications, circuit boards may be affected by a variety of environmental and workload factors, causing performance defects such as signal integrity issues, thermal failures, electromagnetic interference sensitivity, power consumption anomalies, etc. These problems not only affect the stable operation of the computer, but may also cause system failures or hardware damage.

[0003] Existing circuit board testing methods usually focus on a single performance indicator, such as focusing only on temperature, operating voltage or basic current load, but fail to comprehensively consider the impact of multiple environmental variables on the overall performance of the circuit board. For example, power supply instability testing may be limited to low-frequency voltage fluctuations, but fail to simulate complex situations such as high-speed voltage transients, ripple interference, and dynamic load changes; thermal stability testing may only focus on surface temperature distribution, while ignoring power consumption hot spots, solder joint fatigue caused by local overheating, and other issues. In addition, electromagnetic compatibility (EMC) testing often only evaluates external interference sources, but fails to effectively analyze problems such as signal coupling, reflection interference, and ground bounce noise inside the circuit board.

[0004] Since traditional testing methods have limitations in simulating real usage environments and cannot accurately capture performance defects that may occur in circuit boards under complex working conditions, there is an urgent need to propose an online fully automatic detection method and system for printed circuit board defects based on data analysis. Summary of the invention

[0005] In view of the deficiencies in the prior art, the present invention provides a method and system for online fully automatic detection of printed circuit board defects based on data analysis to solve the problems mentioned in the background technology.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a fully automatic online detection system for printed circuit board defects based on data analysis, comprising:

[0007] The power stability test module is used to use a pluggable power module, replace the power module under different frequency and voltage conditions to simulate an unstable power environment, collect and measure the working status of the circuit board under different voltage fluctuations, and construct a power stability index Dyw;

[0008] Dynamic power switching and fault tolerance test module, used to test the power switching capability and fault tolerance of the circuit board under dual power supply conditions, simulate power switching and fault environment through the fault injection system, and build the power switching fault tolerance index Qhrc;

[0009] Thermal stress and thermal fatigue test module, used to use environmental simulation box to simulate -20℃, -10℃, 20℃, 40℃ and 60℃ air temperature conditions, test the working state of the circuit board in high and low temperature cycle, collect temperature stress data, and construct thermal stress index Ryl;

[0010] Shock test module, used to simulate the mechanical vibration and shock conditions encountered by the circuit board during transportation or operation, test the working state of the circuit board under mechanical vibration and shock conditions, and construct the mechanical stress index Mvz;

[0011] The electromagnetic compatibility and anti-interference test module applies electromagnetic waves of different intensities and frequencies around the circuit board through an electromagnetic interference generator to test the circuit board's anti-interference ability, collect electromagnetic compatibility data, and construct the electromagnetic compatibility index Jrx;

[0012] Resource contention and multi-task processing test module, by simulating a multi-task parallel environment, tests the responsiveness of the circuit board under resource contention, collects multi-task processing data, and constructs a resource contention index Zyzq;

[0013] The abnormal state and self-repair capability test module tests the self-detection and repair capabilities of the circuit board by intentionally introducing abnormal states such as power failure, signal interference, and communication interruption, collects abnormal processing data, and constructs a self-repair index Xfz;

[0014] The first evaluation module is used to process the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz dimensionlessly, and then calculate and obtain the comprehensive evaluation index Pc; when the comprehensive evaluation index Pc is lower than the first performance threshold X, it indicates that the comprehensive evaluation result is unqualified, and a second multi-level evaluation command is generated;

[0015] The second multi-level evaluation module receives the second multi-level evaluation command, and performs graded evaluation on the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz under the comprehensive evaluation index Pc, obtains the corresponding test evaluation results, and generates corresponding debugging instructions.

[0016] Preferably, the power supply stability test module includes a high-frequency voltage fluctuation simulation unit, a power supply noise injection unit, a transient voltage drop simulation unit, and a transient voltage spike simulation unit;

[0017] The high-frequency voltage fluctuation simulation unit is used to use a fluctuation generator to simulate the voltage fluctuation of the power supply at high frequency, test the circuit board's adaptability to frequent voltage changes, monitor the circuit board's voltage response through a data recorder, and calculate the high-frequency voltage fluctuation response index V using the following formula fluct :

[0018]

[0019] Where Pg i represents the task processing speed of the circuit board at the ith measurement point in a high-frequency voltage environment, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V ferq Indicates the amplitude of high-frequency voltage fluctuation, V Z Indicates the reference voltage value under normal conditions; n indicates the total number of measurements;

[0020] The power supply noise injection unit is used to use a noise generator and a noise injector to simulate power supply noise signals of different frequencies and intensities and inject them into the power supply input terminal to monitor the noise response changes of the circuit board under noise interference. The noise interference response index R is calculated by the following formula noise :

[0021]

[0022] Where Pz i represents the task processing speed of the circuit board at the ith measurement point under noise conditions, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V noise Indicates the voltage amplitude including the noise signal;

[0023] The transient voltage drop simulation unit is used to use a signal generator to generate a sudden voltage drop signal and inject it into the power input terminal of the circuit board, and monitor the response of the circuit board under the voltage drop condition. The transient voltage drop response index R is calculated by the following formula voltage，drop :

[0024]

[0025] Where Pd i represents the task processing speed of the circuit board at the i-th measurement point under transient voltage drop conditions, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V drop Indicates the voltage value of transient voltage drop;

[0026] The transient voltage spike simulation unit is used to inject the transient voltage spike signal of the signal generator into the power input terminal of the circuit board, and monitor the response of the circuit board under the voltage spike condition. The transient voltage spike response index R is calculated by the following formula voltage，spike :

[0027]

[0028] Where Ps i represents the task processing speed of the circuit board at the ith measurement point under transient voltage spike conditions, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V drop Indicates the voltage value of a transient voltage spike.

[0029] Preferably, the power supply stability test module further includes a power supply comprehensive analysis unit, which is used to convert the high-frequency voltage fluctuation response index V fluct , Noise interference response index R noise , transient voltage drop response index R voltage，drop and transient voltage spike response index R voltage，spike The power stability index Dyw is obtained by associating the following formula:

[0030]

[0031] Where a1, a2, a3 and a4 are high frequency voltage fluctuation response index V fluct , Noise interference response index R noise , transient voltage drop response index R voltage，drop and transient voltage spike response index R voltage，spike The weight value of , C1 is represented as the first correction constant.

[0032] Preferably, the power switching fault tolerance index Qhrc is calculated by the following formula:

[0033]

[0034] In the formula, ΔI surge,i It represents the sudden change value of the circuit board power supply current during the jth circuit board power supply switching process, I norm Indicates the reference current value under normal working conditions, ΔV fluc,j Indicates the voltage fluctuation of the circuit board power supply during the jth circuit board power supply switching process, through ΔV fluc,j =|VV Z |Calculate and obtain, V Z Indicates the reference voltage value under normal conditions; Frt jFrt is the time required for the circuit board to resume normal operation after the jth power switch. max Indicates the maximum allowable recovery time of the preset circuit board; w1, w2 and w3 represent weight coefficients.

[0035] Preferably, the thermal stress index Ryl is calculated by the following formula:

[0036]

[0037] Where P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, P T1 Indicates the task processing speed of the circuit board at an ambient temperature of -20°C, P T2 It indicates the task processing speed of the circuit board at an ambient temperature of -10℃, P T3 It indicates the task processing speed of the circuit board at an ambient temperature of 20°C, P T4 It indicates the task processing speed of the circuit board at 40℃ ambient temperature, P T5 It indicates the task processing speed of the circuit board under 60℃ air temperature condition. The thermal stress index Ryl comprehensively summarizes the performance fluctuation of the task processing speed under different temperature conditions relative to the reference temperature.

[0038] Preferably, the mechanical stress index Mvz is calculated by the following formula:

[0039]

[0040] Where PcJ represents the task processing speed of the circuit board under the jth vibration or shock condition, P1 represents the benchmark task processing speed without vibration or shock, and k represents the total number of vibration or shock tests;

[0041] The electromagnetic compatibility index Jrx is calculated using the following formula:

[0042]

[0043] Where, PE j represents the circuit board task processing speed under the jth electromagnetic interference condition, P2 represents the benchmark task processing speed under the condition of no electromagnetic interference, and y represents the total number of electromagnetic interference tests; the resource contention index Zyzq is calculated and obtained by the following formula;

[0044]

[0045] Where, Pmt j represents the circuit board task processing speed in the jth multitasking environment, P3 represents the circuit board task processing speed in the benchmark environment without task contention, Rrc jrepresents the resource contention rate in the j-th multitasking environment, R ref represents the resource contention rate in the benchmark environment, which is set to 1, and q is the total number of tests in different multi-tasking environments;

[0046] The self-repair index Xfz is calculated by the following formula:

[0047]

[0048] Where Rfix j Rerr represents the repair success rate under the jth abnormal state. j Trep represents the error rate under the jth abnormal state. j represents the time required to repair the jth abnormal state, T max Indicates the maximum allowed repair time, and B indicates the total number of abnormal state tests.

[0049] Preferably, the first evaluation module includes a first association unit and a first evaluation unit;

[0050] The first correlation unit is used to process the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz dimensionlessly, and obtain the comprehensive evaluation index Pc through the following correlation calculation:

[0051]

[0052] Among them, b1, b2, b3, b4, b5, b6 and b7 represent the weight values ​​of the power supply stability index Dyw, the power supply switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repairing index Xfz, respectively, which are adjusted and set by the user;

[0053] The first evaluation unit is used to preset a first performance threshold value X, and compare the comprehensive evaluation index Pc with the first performance threshold value X, including a first evaluation result, including:

[0054] When the comprehensive evaluation index Pc ≥ the first performance threshold X, it means that the circuit board has passed the comprehensive performance test, and the circuit board is a qualified product and is allowed to circulate in the market;

[0055] When the comprehensive evaluation index Pc is less than the first performance threshold value X, it indicates that the comprehensive performance test fails, a performance failure label is marked, and a second multi-level evaluation command is generated.

[0056] Preferably, the second multi-level assessment module includes a hierarchical assessment unit and a strategy unit;

[0057] The hierarchical evaluation unit is used to perform hierarchical evaluation on the power stability index Dyw, power switching fault tolerance index Qhrc, thermal stress index Ryl, mechanical stress index Mvz, electromagnetic compatibility index Jrx, resource contention index Zyzq and self-repair index Xfz under the comprehensive evaluation index Pc after receiving the second multi-level evaluation command, and obtain corresponding test evaluation results, including:

[0058] When the power supply stability index Dyw ≥ the first stability threshold X1, it indicates that the performance of the circuit board under the conditions of high-frequency voltage fluctuation, noise interference, transient voltage drop and transient voltage spike is qualified, and a first qualified label is generated;

[0059] When the power supply stability index Dyw is less than the first stability threshold value X1, it indicates that the circuit board power supply stability is unqualified, and the strategy unit generates a first debugging instruction, including: optimizing the circuit board power supply voltage regulator module or optimizing the power supply design, with the target improvement range being to improve the power supply stability by 10%;

[0060] When the power switching fault tolerance index Qhrc ≥ the second stable threshold value X2, it indicates that the circuit board has qualified performance in the power switching, and the recovery time after the power switching is within the expected range, and a second qualified label is generated;

[0061] When the power switching fault tolerance index Qhrc is less than the second stable threshold value X2, it indicates that the circuit board fails to meet the performance requirements in the power switching, and the strategy unit generates a second debugging instruction, including: improving the power design, with the goal of reducing the recovery time after the power switching by 15%;

[0062] When the thermal stress index Ryl ≥ the third stable threshold value X3, it indicates that the performance of the circuit board is qualified under temperature changes, and a third qualified label is generated;

[0063] When the thermal stress index Ryl is less than the third stable threshold value X3, it indicates that the performance of the circuit board is unqualified under temperature changes, and the strategy unit generates a third debugging instruction, including: optimizing the heat dissipation design, with the goal of improving the heat dissipation efficiency by 20%;

[0064] When the mechanical stress index Mvz ≥ the fourth stable threshold value X4, it indicates that the circuit board has qualified performance under mechanical vibration and impact conditions, and a fourth qualified label is generated;

[0065] When the mechanical stress index Mvz is less than the fourth stable threshold value X4, it indicates that the performance of the circuit board under mechanical vibration and impact conditions is unqualified, and the strategy unit generates a fourth debugging instruction, including: optimizing the installation and fixing design, adding 10% of shockproof materials, with the goal of improving the impact resistance by 20%;

[0066] When the electromagnetic compatibility index Jrx ≥ the fifth stable threshold value X5, it indicates that the performance of the circuit board under the electromagnetic interference condition is qualified; the fifth qualified label is generated;

[0067] When the electromagnetic compatibility index Jrx is less than the fifth stable threshold value X5, it indicates that the performance of the circuit board under the electromagnetic interference condition is unqualified; the strategy unit generates a fifth debugging instruction, including: adding 10% electromagnetic shielding material, the goal is to reduce the impact of interference signals on performance by 20%;

[0068] When the resource contention index Zyzq ≥ the sixth stable threshold value X6, it indicates that the performance of the circuit board in the multi-tasking and resource contention environment is qualified, and a sixth qualified label is generated;

[0069] When the resource contention index Zyzq is less than the sixth stable threshold value X6, it indicates that the performance of the circuit board in the multi-tasking and resource contention environment is unqualified, and the policy unit generates a sixth debugging instruction, including: increasing 10% of spare processor memory and storage, using data parallel processing technology, dividing data into multiple parts and processing them simultaneously on different processing units;

[0070] When the self-repair index Xfz ≥ the seventh stable threshold value X7, it indicates that the self-repair capability of the circuit board in an abnormal state is qualified, and the seventh qualified label is generated;

[0071] When the self-repair index Xfz is less than the seventh stable threshold value X7, it indicates that the circuit board fails to meet the performance requirements under abnormal conditions, and the strategy unit generates the seventh debugging instruction, including: strengthening fault detection and repair capabilities, shortening repair time, with the goal of reducing repair time by 20%.

[0072] The method for online fully automatic detection of printed circuit board defects based on data analysis includes the following steps:

[0073] Step 1: Configure the test system, install and debug the power stability test module, dynamic power switching and fault tolerance test module, thermal stress and thermal fatigue test module, impact test module, electromagnetic compatibility and immunity test module, resource contention and multitasking test module, abnormal state and self-healing ability test module, and set the test conditions;

[0074] Step 2: Collect the circuit board test working status data under each simulation environment in step 1, analyze and calculate to obtain the power stability index Dyw, power switching fault tolerance index Qhrc, thermal stress index Ryl, mechanical stress index Mvz, electromagnetic compatibility index Jrx, resource contention index Zyzq and self-repair index Xfz;

[0075] Step 3: After the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz are dimensionlessly processed, a comprehensive evaluation index Pc is obtained through an associated calculation formula and evaluated to obtain a first evaluation result;

[0076] Step 4. When the comprehensive evaluation index Pc is lower than the first performance threshold X, it indicates that the comprehensive performance test fails, a performance failure label is marked, and a second multi-level evaluation command is generated. The power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz are graded and evaluated through the second multi-level evaluation module to obtain corresponding test evaluation results and generate corresponding debugging instructions.

[0077] The present invention provides a method and system for online automatic detection of printed circuit board defects based on data analysis.

[0078] Beneficial effects:

[0079] (1) Traditional circuit board testing methods usually only focus on a single performance dimension, such as temperature or voltage. This method cannot fully reflect the comprehensive performance of the circuit board in the actual environment. Through the multi-dimensional intelligent testing system of the present invention, the circuit board can be comprehensively tested from multiple aspects (including power supply stability, thermal stress, mechanical shock, electromagnetic compatibility, resource contention and self-repair ability). This comprehensive testing method can more accurately evaluate the performance and stability of the circuit board in the actual use environment and ensure the reliability of the system.

[0080] (2) The system of the present invention integrates multiple test modules and can perform comprehensive performance testing on a single platform. This integrated test platform improves the efficiency and accuracy of testing and avoids the complexity of multiple independent test devices and steps required in traditional testing methods. Through automated data collection and analysis, comprehensive test results can be obtained more quickly, shortening the test cycle and reducing test costs.

[0081] (3) Traditional testing methods are often not accurate enough or limited to basic situations when simulating adverse environmental conditions. For example, the power supply stability test may be limited to basic voltage fluctuations, while ignoring high-frequency voltage changes and transient voltage drops. The present invention uses a special power supply stability test module, including a high-frequency voltage fluctuation simulation unit, a transient voltage drop simulation unit, etc., to more accurately simulate various power supply instability situations and provide more comprehensive performance data. By using a data recorder and formulas to calculate various response indices, the present invention can quantify the processing capabilities of the circuit board under different power supply abnormal conditions. Specifically, the high-frequency voltage fluctuation response index V fluct , Noise interference response index R noise , transient voltage drop response index R voltage，drop and transient voltage spike response index R voltage，spike The task processing speed of the circuit board in these environments was evaluated respectively. This quantitative method makes the evaluation of circuit board performance more objective and scientific, and helps to find potential performance bottlenecks and improvement points.

[0082] (4) The online fully automatic detection method for printed circuit board defects based on data analysis realizes a comprehensive evaluation of the performance of the circuit board through steps one to four. First, by configuring the test system and setting the test conditions, the accurate installation and debugging of each test module are ensured, thereby providing a reliable data basis for subsequent analysis. Secondly, the method ensures the comprehensiveness of the test by calculating multiple key performance indicators such as the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz. After the comprehensive evaluation step processes each indicator dimensionlessly, the comprehensive evaluation index Pc is obtained through the associated calculation formula, and the performance evaluation is performed to effectively determine whether the circuit board meets the market standards. When the comprehensive evaluation index Pc is lower than the first performance threshold X, the system can generate a performance failure label and conduct an in-depth hierarchical evaluation through the second multi-level evaluation module. This mechanism can not only identify specific performance defects, but also generate targeted debugging instructions based on the evaluation results to optimize the circuit board design and improve its comprehensive performance and reliability. This method not only improves the accuracy of testing, but also can quickly locate problems and guide optimization, thereby significantly improving the overall quality and market competitiveness of circuit boards. BRIEF DESCRIPTION OF THE DRAWINGS

[0083] Figure 1 It is a schematic diagram of the system flow of the online fully automatic detection method for printed circuit board defects based on data analysis of the present invention;

[0084] Figure 2 The figure is a schematic diagram of the steps of the online fully automatic detection method for printed circuit board defects based on data analysis of the present invention.

[0085] In the figure: 10, power supply stability test module; 20, dynamic power supply switching and fault tolerance test module; 30, thermal stress and thermal fatigue test module; 40, impact test module; 50, electromagnetic compatibility and immunity test module; 60, resource contention and multi-tasking test module; 70, abnormal state and self-repair capability test module; 80, first evaluation module; 90, second multi-level evaluation module; 11, high-frequency voltage fluctuation simulation unit; 12, power supply noise injection unit; 13, transient voltage drop simulation unit; 14, transient voltage spike simulation unit; 15, power supply comprehensive analysis unit; 81, first associated unit; 82, first evaluation unit; 91, hierarchical evaluation unit; 92, strategy unit. DETAILED DESCRIPTION

[0086] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0087] Example 1

[0088] See also Figure 1 The present invention provides a fully automatic online detection system for printed circuit board defects based on data analysis, comprising:

[0089] The power stability test module 10 is used to use a pluggable power module to replace the power module under different frequency and voltage conditions to simulate an unstable power environment. The high-frequency voltage detector, noise analyzer, voltage recorder and performance analysis tool Profilers collect and measure the working state of the circuit board under different voltage fluctuations to construct a power stability index Dyw;

[0090] The dynamic power switching and fault tolerance test module 20 is used to test the power switching capability and fault tolerance of the circuit board under dual power supply conditions, simulate the power switching and fault environment through the fault injection system, collect the power switching capability and fault tolerance data of the circuit board under dual power supply conditions through the current detector, voltage detector and performance analysis tool Profilers, and construct the power switching fault tolerance index Qhrc;

[0091] The thermal stress and thermal fatigue test module 30 is used to use an environmental simulation box to simulate the air temperature conditions of -20°C, -10°C, 20°C, 40°C and 60°C, collect the working state of the test circuit board in the high and low temperature cycle through the temperature sensor and the performance analysis tool Profilers, collect the temperature stress data, and construct the thermal stress index Ryl;

[0092] The shock test module 40 is used to simulate the mechanical vibration and shock conditions encountered by the circuit board during transportation or operation, and to test the working state of the circuit board under the mechanical vibration and shock conditions through accelerometers, shock sensors and performance analysis tools Profilers, and to construct a mechanical stress index Mvz;

[0093] The electromagnetic compatibility and immunity test module 50 applies electromagnetic waves of different strengths and frequencies around the circuit board through an electromagnetic interference generator, tests the anti-interference ability of the circuit board through the performance analysis tool Profilers, collects electromagnetic compatibility data, and constructs the electromagnetic compatibility index Jrx;

[0094] The resource contention and multi-tasking test module 60 tests the responsiveness of the circuit board under resource contention by simulating a multi-tasking parallel environment and using resource monitoring tools, including CPU and memory monitoring tools, to collect multi-tasking data and construct a resource contention index Zyzq;

[0095] The abnormal state and self-repair capability test module 70 tests the self-detection and repair capability of the circuit board by intentionally introducing abnormal states such as power failure, signal interference and communication interruption, collects abnormal processing data through the error detection and repair log system, and constructs a self-repair index Xfz;

[0096] In actual applications, the circuit board may encounter various abnormal conditions, such as power failure, signal interference, etc. Traditional testing methods usually do not include testing of the self-repair ability of the circuit board. Through the abnormal state and self-repair ability test module 70, the present invention can evaluate the self-detection and repair ability of the circuit board under abnormal conditions and generate targeted repair instructions. This function can improve the self-recovery ability of the circuit board in actual use and enhance the stability and reliability of the system.

[0097] The first evaluation module 80 is used to process the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz dimensionlessly, and then calculate and obtain the comprehensive evaluation index Pc; when the comprehensive evaluation index Pc is lower than the first performance threshold X, it indicates that the comprehensive evaluation result is unqualified, and a second multi-level evaluation command is generated;

[0098] The second multi-level evaluation module 90 receives the second multi-level evaluation command, and performs graded evaluation on the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz under the comprehensive evaluation index Pc, obtains corresponding test evaluation results, and generates corresponding debugging instructions.

[0099] Traditional circuit board testing methods usually only focus on a single performance dimension, such as temperature or voltage. This method cannot fully reflect the comprehensive performance of the circuit board in the actual environment. Through the multi-dimensional intelligent testing system of the present invention, the circuit board can be comprehensively tested from multiple aspects including power supply stability, thermal stress, mechanical shock, electromagnetic compatibility, resource contention and self-repair ability. This comprehensive testing method can more accurately evaluate the performance and stability of the circuit board in the actual use environment and ensure the reliability of the system. The system of the present invention integrates a variety of test modules and can perform comprehensive performance tests on a single platform. This integrated testing platform improves the efficiency and accuracy of the test and avoids the complexity of multiple independent test equipment and steps required in traditional testing methods. Through automated data acquisition and analysis, comprehensive test results can be obtained more quickly, shortening the test cycle and reducing testing costs.

[0100] Using convolutional neural networks, an initial model of the convolutional neural network is constructed, and the initial model of the convolutional neural network is trained and tested with the original data set. The trained initial model of the convolutional neural network is used as a circuit board test evaluation model. At the same time, the intermediate layer output of the equipment operation status model is used as a feature vector to identify feature information. The circuit board test evaluation model is trained and tested through the acquired feature information, and the trained circuit board test evaluation model is used as data operation prediction.

[0101] Example 2

[0102] This example is an explanation of Example 1. Figure 1 , specifically, the power supply stability test module 10 includes a high-frequency voltage fluctuation simulation unit 11, a power supply noise injection unit 12, a transient voltage drop simulation unit 13, and a transient voltage spike simulation unit 14;

[0103] The high-frequency voltage fluctuation simulation unit 11 is used to use a fluctuation generator to simulate the voltage fluctuation of the power supply at high frequency, test the adaptability of the circuit board to frequent voltage changes, monitor the voltage response of the circuit board through a data recorder, and calculate the high-frequency voltage fluctuation response index V by the following formula fluct :

[0104]

[0105] Where Pg i represents the task processing speed of the circuit board at the ith measurement point in a high-frequency voltage environment, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V ferq Indicates the amplitude of high-frequency voltage fluctuation, V Z Indicates the reference voltage value under normal conditions; n indicates the total number of measurements;

[0106] The power supply noise injection unit 12 is used to use a noise generator and a noise injector to simulate power supply noise signals of different frequencies and intensities and inject them into the power supply input terminal to monitor the noise response change of the circuit board under noise interference, and obtain the noise interference response index R by calculating the following formula: noise :

[0107]

[0108] Where Pz i represents the task processing speed of the circuit board at the ith measurement point under noise conditions, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V noise Indicates the voltage amplitude including the noise signal;

[0109] The transient voltage drop simulation unit 13 is used to use a signal generator to generate a sudden voltage drop signal and inject it into the power input terminal of the circuit board, and monitor the response of the circuit board under the voltage drop condition, and obtain the transient voltage drop response index R by calculating the following formula: voltage，drop :

[0110]

[0111] Where Pd i represents the task processing speed of the circuit board at the i-th measurement point under transient voltage drop conditions, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V drop Indicates the voltage value of transient voltage drop;

[0112] The transient voltage spike simulation unit 14 is used to inject a transient voltage spike signal from a signal generator into the power input terminal of the circuit board, and monitor the response of the circuit board under the voltage spike condition, and obtain the transient voltage spike response index R by calculating the following formula: voltage，spike :

[0113]

[0114] Where Ps i represents the task processing speed of the circuit board at the ith measurement point under transient voltage spike conditions, P ZRepresents the processing speed of the circuit board's reference tasks under normal environmental conditions, V drop Represents the voltage value of the transient voltage spike.

[0115] When traditional testing methods simulate adverse environmental conditions, they are often not precise enough or limited to basic situations. For example, power supply stability testing may be limited to basic voltage fluctuations, ignoring high-frequency voltage changes and transient voltage dips. Through the dedicated power supply stability testing module 10 of the present invention, including the high-frequency voltage fluctuation simulation unit 11, the transient voltage dip simulation unit 13, etc., it can more precisely simulate various power supply instability situations and provide more comprehensive performance data. By using a data recorder and formula to calculate various response indices, the present invention can quantify the processing ability of the circuit board under different power supply abnormal conditions. Specifically, the high-frequency voltage fluctuation response index V fluct , the noise interference response index R noise , the transient voltage dip response index R voltage，drop and the transient voltage spike response index R voltage，spike respectively evaluate the task processing speed of the circuit board in these environments. This quantification method makes the evaluation of the circuit board's performance more objective and scientific, helping to discover potential performance bottlenecks and improvement points. For example, if the high-frequency voltage fluctuation response index is low, it may indicate that the circuit board has insufficient adaptability to power supply fluctuations and may require optimizing the power supply voltage regulation design; if the noise interference response index is low, then anti-interference design may be needed. Such a feedback mechanism can support the performance optimization of the circuit board and improve the overall quality and reliability of the product.

[0116] The power supply stability testing module 10 also includes a power supply comprehensive analysis unit 15. The power supply comprehensive analysis unit 15 is used to correlate the high-frequency voltage fluctuation response index V fluct , the noise interference response index R noise , the transient voltage dip response index R voltage，drop and the transient voltage spike response index R voltage，spike to obtain the power supply stability index Dyw through the following formula:

[0117]

[0118] In the formula, a1, a2, a3, and a4 are the weight values of the high-frequency voltage fluctuation response index V fluct , the noise interference response index R noise , the transient voltage dip response index R voltage，drop and the transient voltage spike response index R voltage，spike , which are adjusted and set by the user, 0 < a1 < 1, 0 < a2 < 1, 0 < a3 < 1, 0 < a4 < 1, and a1 + a2 + a3 + a4 = 1. C1 represents the first correction constant;

[0119] In this embodiment, the power comprehensive analysis unit 15 generates a comprehensive power stability index Dyw by comprehensively calculating the high-frequency voltage fluctuation response index V fluct , the noise interference response index R noise , the transient voltage drop response index R voltage，drop , and the transient voltage spike response index R voltage，spike . This comprehensive evaluation method can comprehensively consider the impact of various power anomalies on the performance of the circuit board, thereby providing a more comprehensive evaluation of power stability.

[0120] Embodiment 3

[0121] This embodiment is an explanatory description based on Embodiment 1. Please refer to Figure 1 . Specifically, the power supply switching fault tolerance index Qhrc is calculated by the following formula:

[0122]

[0123] In the formula, ΔI surge,i represents the mutation value of the circuit board power supply current during the jth power supply switching of the circuit board power supply, I norm represents the reference current value in the normal working state, ΔV fluc,j represents the voltage fluctuation amplitude of the circuit board power supply during the jth power supply switching of the circuit board power supply, and is obtained by calculating ΔV fluc,j = |V - V Z |, where V Z represents the reference voltage value in the normal environment; Frt j represents the time required for the circuit board to return to normal operation after the jth power supply switching, and Frt max represents the preset maximum allowable recovery time of the circuit board; w1, w2, and w3 represent weight coefficients, which are adjusted and set by the user, 0 < w1 < 1, 0 < w2 < 1, 0 < w3 < 1, and w1 + w2 + w3 = 1.

[0124] The dynamic power switching and fault tolerance test module 20 can accurately evaluate the power switching capability of the circuit board under dual power supply conditions. By simulating the power switching environment, the system can measure current mutations, voltage fluctuations and recovery time, thereby evaluating whether the circuit board can operate stably during the switching process, ensuring the reliability of the circuit board in practical applications. The power switching fault tolerance index Qhrc can comprehensively analyze the fault tolerance of the circuit board during the power switching process. By considering multiple factors such as current mutations, voltage fluctuations and recovery time, the system can evaluate in detail the fault tolerance of the circuit board in the face of power switching, thereby helping engineers understand the performance of the circuit board in the face of power failures. By calculating the time required for the circuit board to resume normal operation after the power switching, the Qhrc indicator can quantify the recovery ability of the circuit board. This indicator can ensure that the circuit board can quickly resume normal operation after a power failure or switching, reducing downtime caused by power problems.

[0125] The thermal stress index Ryl is calculated by the following formula:

[0126]

[0127] Where P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, P T1 Indicates the task processing speed of the circuit board at an ambient temperature of -20°C, P T2 It indicates the task processing speed of the circuit board at an ambient temperature of -10℃, P T3 It indicates the task processing speed of the circuit board at an ambient temperature of 20°C, P T4 It indicates the task processing speed of the circuit board at 40℃ ambient temperature, P T5 It indicates the task processing speed of the circuit board under the air temperature of 60℃. The thermal stress index Ryl comprehensively considers the performance fluctuation of the task processing speed under different temperature conditions relative to the reference temperature. If Ryl is closer to 1, the performance stability of the circuit board in the high and low temperature cycle is better. If Ryl is low, it means that the temperature change has a greater impact on the performance of the circuit board, indicating that its tolerance is weak.

[0128] Simulating high and low temperature cycles (-20℃, -10℃, 20℃, 40℃, 60℃) goes a step further than traditional single surface temperature distribution testing, taking into account the impact of temperature cycle stress on the circuit board. This means that your system can analyze the stress of the circuit board during temperature changes, rather than just measuring the surface temperature. By quantifying thermal stress, it indirectly involves power consumption hot spots or local overheating problems.

[0129] The thermal stress and thermal fatigue test module 30 can comprehensively evaluate the impact of temperature changes on circuit board performance by testing circuit boards under multiple temperature environments. By simulating extreme low and high temperature environments, the system can identify the performance stability of the circuit board under these conditions and ensure the reliability of the circuit board in actual use. The thermal stress index Ryl combines the circuit board performance data at different temperatures to provide a quantitative indicator to evaluate the performance of the circuit board in high and low temperature cycles. This data can help engineers understand the performance fluctuations of the circuit board under different temperature conditions, and then identify potential problems in the design. The data obtained through thermal stress testing can be used to optimize the design of the circuit board to ensure that it can adapt to various temperature conditions in actual applications. These data help to identify and improve the temperature management strategy of the circuit board in the design, thereby improving the reliability and stability of the overall system.

[0130] The mechanical stress index Mvz is calculated by the following formula:

[0131]

[0132] Where PcJ represents the task processing speed of the circuit board under the jth vibration or shock condition, P1 represents the benchmark task processing speed without vibration or shock, and k represents the total number of vibration or shock tests;

[0133] The shock test module 40 can effectively evaluate the performance of the circuit board under mechanical vibration and shock conditions encountered during transportation or operation. These tests can simulate the mechanical stress in the actual use environment to ensure that the circuit board can still work stably when facing physical shock and vibration. The mechanical stress index Mvz comprehensively considers the performance of the circuit board under different vibration and shock conditions, and provides a quantitative indicator to evaluate the stability of the circuit board under these conditions. Through this data, engineers can understand the mechanical stress conditions that the circuit board may face during actual transportation and operation. Through mechanical stress testing, engineers can identify the weak links of the circuit board under mechanical vibration and shock conditions, so as to optimize the circuit board design and increase protection measures. This helps to improve the durability and shock resistance of the circuit board.

[0134] The electromagnetic compatibility index Jrx is calculated using the following formula:

[0135]

[0136] Where, PE j represents the circuit board task processing speed under the jth electromagnetic interference condition, P2 represents the benchmark task processing speed under the condition of no electromagnetic interference, and y represents the total number of electromagnetic interference tests;

[0137] The electromagnetic compatibility and immunity test module 50 can accurately evaluate the performance of the circuit board under different electromagnetic interference conditions. By applying electromagnetic waves of different intensities and frequencies, the anti-interference ability of the circuit board in various electromagnetic interference environments is tested to ensure that it can work stably in practical applications. The electromagnetic compatibility index Jrx provides a quantitative indicator for evaluating the performance fluctuation of the circuit board under electromagnetic interference by comprehensively considering the performance of the circuit board under different electromagnetic interference conditions. The test results can confirm the stability of the circuit board in the electromagnetic interference environment. By simulating the electromagnetic interference conditions in the actual working environment, the reliability of the circuit board in this environment can be verified, reducing the risk of system failure caused by electromagnetic interference.

[0138] The resource contention index Zyzq is calculated by the following formula:

[0139]

[0140] Where, Pmt j represents the circuit board task processing speed in the jth multitasking environment, P3 represents the circuit board task processing speed in the benchmark environment without task contention, Rrc j represents the resource contention rate in the j-th multitasking environment, R ref represents the resource contention rate in the benchmark environment, which is set to 1, and q is the total number of tests in different multi-tasking environments;

[0141] The resource contention and multi-tasking test module 60 can effectively evaluate the performance of the circuit board when facing multi-tasking parallel processing. By simulating various multi-tasking environments, the responsiveness and task processing speed of the circuit board under resource contention can be tested to understand its performance in actual applications. The resource contention index Zyzq provides a quantitative indicator for evaluating the impact of resource contention on circuit board performance. By comparing the performance in a multi-tasking environment with the performance in a baseline environment, the test results can help engineers identify the performance bottlenecks of the circuit board under resource contention, thereby optimizing the allocation and management of system resources.

[0142] The self-repair index Xfz is calculated by the following formula:

[0143]

[0144] Where Rfix j Rerr represents the repair success rate under the jth abnormal state. j Trep represents the error rate under the jth abnormal state. j represents the time required to repair the jth abnormal state, T max Indicates the maximum allowed repair time, and B indicates the total number of abnormal state tests.

[0145] The abnormal state and self-repair capability test module 70 can accurately evaluate the self-repair capability of the circuit board in the face of abnormal states such as power failure, signal interference and communication interruption. By testing the reaction and repair capability of the circuit board in various abnormal states, its reliability in practical applications can be understood. The self-repair index Xfz provides a quantitative indicator for measuring the repair success rate of the circuit board in abnormal states. By calculating the repair success rate, error occurrence rate and repair time, the self-repair performance of the circuit board can be fully understood. By introducing abnormal states such as power failure, signal interference and communication interruption, the self-repair capability of the test circuit board can ensure that the system can continue to operate stably when encountering abnormal conditions. This helps to improve the reliability and robustness of the system. The test results can reveal the advantages and disadvantages of the repair mechanism of the circuit board in abnormal states. Engineers can optimize the self-repair mechanism based on the test data to improve the repair efficiency and success rate of the circuit board in abnormal states.

[0146] Specifically, the calculation of each performance index in Example 2 and Example 3 is based on the preprocessing and normalization of the original test data to ensure the accuracy and reliability of the test results. The normalization process is to standardize the data of different test indicators to a common unit range, so that effective comparison and comprehensive evaluation can be performed. The normalized formulas will be summarized into a dimensionless standard form, so that the results of different indicators can be calculated and compared under the same unit.

[0147] Example 4

[0148] This example is an explanation of Example 1. Figure 1 , specifically, the first evaluation module 80 includes a first associating unit 81 and a first evaluating unit 82;

[0149] The first associating unit 81 is used to process the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz dimensionlessly, and obtain the comprehensive evaluation index Pc through the following associating calculation:

[0150]

[0151] Among them, b1, b2, b3, b4, b5, b6 and b7 respectively represent the weight values ​​of the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz, which are adjusted and set by the user; and the sum of the weights is 1.

[0152] The first evaluation unit 82 is used to preset a first performance threshold value X, and compare the comprehensive evaluation index Pc with the first performance threshold value X, including a first evaluation result, including:

[0153] When the comprehensive evaluation index Pc ≥ the first performance threshold X, it means that the circuit board has passed the comprehensive performance test, and the circuit board is a qualified product and is allowed to circulate in the market;

[0154] When the comprehensive evaluation index Pc is less than the first performance threshold value X, it indicates that the comprehensive performance test fails, a performance failure label is marked, and a second multi-level evaluation command is generated.

[0155] In this embodiment, the first evaluation unit 82 automatically determines the test result of the circuit board by comparing the comprehensive evaluation index Pc with the preset first performance threshold X. This reduces the need for manual intervention and improves the efficiency and consistency of the evaluation process. By comparing the comprehensive evaluation index Pc with the first performance threshold X, the system clarifies whether the circuit board meets the standards for market circulation. When Pc ≥ X, the circuit board is marked as a qualified product and can be circulated in the market; when Pc < X, the circuit board is marked as unqualified and requires further multi-level evaluation. This clear standard helps to ensure the quality of products on the market. The automated comprehensive evaluation and decision-making process can significantly improve test efficiency and reduce the time for manual analysis and judgment. This helps to speed up product testing and market launch to meet market demand. This helps to maintain consistent evaluation standards between different test cycles and different products.

[0156] Example 5

[0157] This example is an explanation of Example 4. Figure 1 , specifically, the second multi-level evaluation module 90 includes a hierarchical evaluation unit 91 and a strategy unit 92;

[0158] The hierarchical evaluation unit 91 is used to perform hierarchical evaluation on the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz under the comprehensive evaluation index Pc after receiving the second multi-level evaluation command, and obtain corresponding test evaluation results, including:

[0159] When the power supply stability index Dyw ≥ the first stability threshold X1, it indicates that the performance of the circuit board under the conditions of high-frequency voltage fluctuation, noise interference, transient voltage drop and transient voltage spike is qualified, and a first qualified label is generated;

[0160] When the power supply stability index Dyw is less than the first stability threshold value X1, it indicates that the circuit board power supply stability is unqualified, and the strategy unit 92 generates a first debugging instruction, including: optimizing the circuit board power supply voltage regulator module or optimizing the power supply design, with the target improvement range being to improve the power supply stability by 10%;

[0161] When the power switching fault tolerance index Qhrc ≥ the second stable threshold value X2, it indicates that the circuit board has qualified performance in the power switching, and the recovery time after the power switching is within the expected range, and a second qualified label is generated;

[0162] When the power switching fault tolerance index Qhrc is less than the second stable threshold value X2, it indicates that the circuit board fails to meet the performance requirements in the power switching, and the strategy unit 92 generates a second debugging instruction, including: improving the power design, with the goal of reducing the recovery time after the power switching by 15%;

[0163] When the thermal stress index Ryl ≥ the third stable threshold value X3, it indicates that the performance of the circuit board is qualified under temperature changes, and a third qualified label is generated;

[0164] When the thermal stress index Ryl is less than the third stable threshold value X3, it indicates that the performance of the circuit board is unqualified under temperature changes, and the strategy unit 92 generates a third debugging instruction, including: optimizing the heat dissipation design, with the goal of improving the heat dissipation efficiency by 20%;

[0165] When the mechanical stress index Mvz ≥ the fourth stable threshold value X4, it indicates that the circuit board has qualified performance under mechanical vibration and impact conditions, and a fourth qualified label is generated;

[0166] When the mechanical stress index Mvz is less than the fourth stable threshold value X4, it indicates that the circuit board does not meet the performance requirements under mechanical vibration and impact conditions, and the strategy unit 92 generates a fourth debugging instruction, including: optimizing the installation and fixing design, adding 10% of shockproof materials, with the goal of improving the impact resistance by 20%;

[0167] When the electromagnetic compatibility index Jrx ≥ the fifth stable threshold value X5, it indicates that the performance of the circuit board under the electromagnetic interference condition is qualified; the fifth qualified label is generated;

[0168] When the electromagnetic compatibility index Jrx is less than the fifth stable threshold value X5, it indicates that the performance of the circuit board under the electromagnetic interference condition is unqualified; the strategy unit 92 generates a fifth debugging instruction, including: adding 10% electromagnetic shielding material, the goal is to reduce the impact of the interference signal on the performance by 20%;

[0169] When the resource contention index Zyzq ≥ the sixth stable threshold value X6, it indicates that the performance of the circuit board in the multi-tasking and resource contention environment is qualified, and a sixth qualified label is generated;

[0170] When the resource contention index Zyzq < the sixth stable threshold value X6, indicating that the performance of the circuit board in the multi-tasking and resource contention environment is unqualified, the policy unit 92 generates a sixth debugging instruction, including: increasing 10% of spare processor memory and storage, using data parallel processing technology, dividing the data into multiple parts and processing them simultaneously on different processing units;

[0171] When the self-repair index Xfz ≥ the seventh stable threshold value X7, it indicates that the self-repair capability of the circuit board in an abnormal state is qualified, and the seventh qualified label is generated;

[0172] When the self-repair index Xfz is less than the seventh stable threshold value X7, it indicates that the circuit board fails to meet the performance requirements under abnormal conditions. The strategy unit 92 generates a seventh debugging instruction, including: strengthening fault detection and repair capabilities, shortening repair time, with the goal of reducing repair time by 20%.

[0173] In this embodiment, the second multi-level evaluation module 90 provides a more detailed performance analysis by performing a graded evaluation on each index. This detailed analysis helps to accurately identify the performance of the circuit board under different test conditions and ensure that the performance of each aspect meets the expected standards. The comparison of the graded evaluation results of each test indicator with the stability threshold can generate a clear qualified label. This clear identification makes the product quality judgment clearer and helps to quickly determine whether the product meets the market circulation standards. When the test results do not meet the stability threshold, the debugging instructions generated by the strategy unit 92 have clear improvement goals. For example, improving power supply stability, improving heat dissipation efficiency, optimizing electromagnetic compatibility, etc. These specific goals make the improvement measures operational and can effectively improve the performance of the circuit board. By setting specific target ranges such as improving power supply stability by 10%, optimizing heat dissipation efficiency by 20%, etc., the performance of the circuit board can be improved in a targeted manner. Such precise improvement measures can significantly improve the reliability and market competitiveness of the product.

[0174] Data example:

[0175]

[0176] Original data, n = 5;

[0177] Pg i = [105, 110, 98, 102, 108];

[0178] P Z =100;

[0179] V ferq =240;

[0180] V Z =230;

[0181] Substitute into the formula to get the calculation result: V fluct =11.03;

[0182]

[0183] Original data, n = 5;

[0184] Pz i = [100, 97, 104, 96, 101];

[0185] P Z =100;

[0186] V nise =225;

[0187] V Z =230;

[0188] Substitute the formula to obtain the calculation result: R noise =4.85;

[0189]

[0190] Original data, n = 5;

[0191] Pd i =[92,94,90,89,93];

[0192] P Z =100;

[0193] V drop =210;

[0194] V Z =230;

[0195] Substitute the formula to obtain the calculation result: R voltage，drop =16.07;

[0196]

[0197] Original data, n = 5;

[0198] Ps i =[108,106,104,109,107];

[0199] P Z =100;

[0200] V spike =250;

[0201] V Z =230;

[0202] Substitute the formula to obtain the calculation result: Rvoltage，drop =14.19;

[0203]

[0204] Combined with V fluct =11.03; R noise =4.85; R voltage，drop =16.07; R voltage，drop =14.19;

[0205] Set a1 = 0.3; a2 = 0.2; a3 = 0.25; a4 = 0.25; C1 = 0.046;

[0206]

[0207] The calculation result of the power stability index Dyw is 11.89.

[0208] Power switching fault tolerance index Qhrc data example:

[0209] Assuming that three power switchings are performed, the following are the data of current, voltage changes and recovery time during each switching process:

[0210] 1st time:

[0211] The first current mutation value ΔI surge,i =5.2A;

[0212] Reference current I norm =5.0;

[0213] Voltage V1 = 230; ΔV fluc,j Then ΔV fluc,j =|VV Z |;

[0214] Reference voltage V Z =240V;

[0215] First recovery time Frt j =3.5s;

[0216] Maximum allowed recovery time Frt max =5s;

[0217] 2nd time:

[0218] Current mutation value ΔI surge,i =4.9A;

[0219] Voltage V2 = 235V;

[0220] Second recovery time Frt j =2.0s;

[0221] 3rd time:

[0222] Current mutation value ΔI surge,i =5.1A;

[0223] Voltage V3 = 238 V;

[0224] The third recovery time Frt j =1.5s;

[0225] Set the weight coefficients to w1 = 0.4; w2 = 0.3; w3 = 0.3; calculate the first power switching:

[0226]

[0227] Substituting into the formula, the calculation for the first switch is:

[0228]

[0229] Calculate the second power switch:

[0230]

[0231] Substituting into the formula, the calculation for the second switch is:

[0232]

[0233] The result of comprehensive calculation to obtain the power switching fault tolerance index Qhrc is 0.55242;

[0234]

[0235] Thermal stress index Ryl calculation example:

[0236] Original data: Assume that the test obtains the following data:

[0237] Benchmark task processing speed:

[0238] P Z = 100 units (e.g., number of tasks processed per second)

[0239] Task processing speed at different temperatures:

[0240] P T1 =95(-20℃);

[0241] P T2 =97(-10℃);

[0242] P T3 =100(20℃);

[0243] P T4=102(40℃);

[0244] P T5 =98 (60°C);

[0245] Substitute into the formula to obtain:

[0246]

[0247] The result of thermal stress index Ryl is 0.976;

[0248] The mechanical stress index Mvz is calculated by the following formula:

[0249]

[0250] Where PcJ represents the task processing speed of the circuit board under the jth vibration or shock condition, P1 represents the benchmark task processing speed without vibration or shock, and k represents the total number of vibration or shock tests;

[0251] Data example: Data example:

[0252] Assume that 4 vibration or shock tests are performed and the following task processing speed data is obtained:

[0253] Benchmark task processing speed P1 = 1000tasks / sec;

[0254] The first vibration PcJ1 = 920 tasks / sec;

[0255] The second impact PcJ2 = 880 tasks / sec;

[0256] The third vibration PcJ3 = 950 tasks / sec;

[0257] The fourth impact PcJ4 = 870 tasks / sec;

[0258] Substitute into the formula to calculate:

[0259] The final result of mechanical stress index Mvz was 0.905;

[0260] The electromagnetic compatibility index Jrx is calculated by the following formula:

[0261]

[0262] Where, PE j represents the circuit board task processing speed under the jth electromagnetic interference condition, P2 represents the benchmark task processing speed under the condition of no electromagnetic interference, and y represents the total number of electromagnetic interference tests;

[0263] Data example:

[0264] Assume that 3 electromagnetic interference tests are performed and the following task processing speed data is obtained:

[0265] Benchmark task processing speed P2 = 1000tasks / sec;

[0266] The first electromagnetic interference PE1 = 940tasks / sec;

[0267] The second electromagnetic interference PE2 = 910tasks / sec;

[0268] The third electromagnetic interference PE3 = 890 tasks / sec;

[0269] Substitute into the formula to calculate:

[0270] The final result of the electromagnetic compatibility index Jrx is 0.913;

[0271] The resource contention index Zyzq is calculated and obtained by the following formula:

[0272]

[0273] Where, Pmt j represents the circuit board task processing speed in the jth multitasking environment, P3 represents the circuit board task processing speed in the benchmark environment without task contention, Rrc j represents the resource contention rate in the j-th multitasking environment, R ref represents the resource contention rate in the benchmark environment, which is set to 1, and q is the total number of tests in different multi-tasking environments;

[0274] Data example:

[0275] Assume that three multi-tasking environment tests are performed, and the following task processing speed and resource contention rate data are obtained:

[0276] Benchmark task processing speed P3 = 1000tasks / sec;

[0277] The first multitasking process Pmt1 = 850 tasks / sec, resource contention rate Rrc1 = 0.4;

[0278] The second multitasking process Pmt2 = 820tasks / sec, resource contention rate Rrc2 = 0.6;

[0279] The third multitasking process Pmt2 = 780tasks / sec, resource contention rate Rrc3 = 0.5;

[0280] Substitute the formula to calculate the resource contention index Zyzq:

[0281]

[0282] The final result of resource contention index Zyzq is 0.4093;

[0283] The self-repair index Xfz is calculated by the following formula:

[0284]

[0285] Where Rfix j Rerr represents the repair success rate under the jth abnormal state. j Trep represents the error rate under the jth abnormal state. j represents the time required to repair the jth abnormal state, T max Indicates the maximum allowed repair time, and B indicates the total number of abnormal state tests.

[0286] Data example: Assume that three abnormal state tests are performed and the following repair success rate, error rate, and repair time data are obtained:

[0287] The first abnormal state Rfix1 = 0.9; Rerr1 = 0.05;

[0288] Repair time Trep1 = 2 hours;

[0289] Maximum allowed time Tmax = 5 hours;

[0290] The second abnormal state Rfix2 = 0.8; Rerr2 = 0.1;

[0291] Repair time Trep2 = 3 hours;

[0292] The third abnormal state Rfix3 = 0.7; Rerr3 = 0.15;

[0293] Repair time Trep3 = 4 hours;

[0294] Substitute the formula to calculate the self-repair index Xfz:

[0295]

[0296] The final result of the self-healing index Xfz is 4.978;

[0297] The results obtained by combining all the previous data examples are:

[0298] The calculation result of the power stability index Dyw is 11.89. The comprehensive calculation result of the power switching fault tolerance index Qhrc is 0.55242; the thermal stress index Ryl result is 0.976; the mechanical stress index Mvz final result is 0.905; the electromagnetic compatibility index Jrx final result is 0.913; the resource contention index Zyzq final result is 0.4093; the self-repair index Xfz final result is 4.978;

[0299] Set weight values ​​b1 = 0.2; b2 = 0.15; b3 = 0.1; b4 = 0.1; b5 = 0.15; b6 = 0.2; b7 = 0.1;

[0300] Substitute into the calculation to obtain the comprehensive evaluation index Pc:

[0301] Pc=0.2*11.89+0.15*0.55242+0.1*0.976+0.1*0.905+0.15*0.913

[0302] +0.2*0.4093+0.1*4.978=2.378+0.082863+0.0976+0.0905

[0303] +0.13695+0.08186+0.4978=3.365573;

[0304] Evaluation Example:

[0305] Set the first performance threshold X=3.5;

[0306] Substitute the comprehensive evaluation index Pc=3.365573 for evaluation. When Pc<X, that is, 3.365573<3.5, it means that the comprehensive performance test of the circuit board fails and is marked as a performance failure label;

[0307] And set the first performance threshold X=3.5;

[0308] Substitute the comprehensive evaluation index Pc=3.365573 for evaluation. When Pc<X, that is, 3.365573<3.5, it means that the comprehensive performance test of the circuit board fails, and it is marked as a performance failure label, and the second multi-level evaluation command is triggered;

[0309] The hierarchical evaluation unit is used to perform hierarchical evaluation on the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz under the comprehensive evaluation index Pc after receiving the second multi-level evaluation command, and obtain corresponding test evaluation results;

[0310] The first stable threshold X1 is set to 10;

[0311] The second stable threshold X2 is set to 0.8;

[0312] The third stable threshold X3 is set to 0.9;

[0313] The fourth stable threshold X4 is set to 0.85;

[0314] The fifth stability threshold X5 is set to 0.9;

[0315] The sixth stable threshold X6 is set to 0.6;

[0316] The seventh stability threshold X7 is set to 4.5;

[0317] The calculation result of the power supply stability index Dyw is 11.89>10 of the first stability threshold value X1, indicating that the power supply stability of the circuit board is qualified, and a first qualified label is generated.

[0318] The result of comprehensive calculation to obtain the power switching fault tolerance index Qhrc is 0.55242<0.8 of the second stable threshold X2; it indicates that the power switching fault tolerance of the circuit board is weak and unqualified; the second debugging instruction is generated by the strategy unit 92, including: optimizing the power switching mechanism, with the goal of reducing the recovery time after power switching by 15% and improving the fault tolerance rate.

[0319] The result of the thermal stress index Ryl is 0.976>0.9 of the third stable threshold value X3, indicating that the performance of the circuit board is qualified under the thermal stress condition, and a third qualified label is generated.

[0320] The final result of the mechanical stress index Mvz is 0.905>0.85 of the fourth stable threshold value X4, indicating that the circuit board performs well under the mechanical stress condition, and a fourth qualified label is generated.

[0321] The final result of the electromagnetic compatibility index Jrx is 0.913>0.9 of the fifth stable threshold value X5; it indicates that the circuit board has qualified performance in the electromagnetic interference environment, and the fifth qualified label is generated.

[0322] The final result of the resource contention index Zyzq is 0.4093<0.6 of the sixth stable threshold value X6, which means that the circuit board performs poorly under the resource contention condition and is unqualified.

[0323] The sixth debugging instruction is generated by the policy unit 92, including: increasing processor memory and improving scheduling efficiency, with the goal of increasing resource utilization by 10%.

[0324] The final result of the self-repair index Xfz is 4.978>4.5 of the seventh stable threshold value X7, indicating that the circuit board's self-repair capability is qualified, and the seventh qualified label is generated.

[0325] Example 6

[0326] Please refer to Figure 2 , a method for online fully automatic detection of printed circuit board defects based on data analysis, comprising the following steps:

[0327] Step 1: Configure the test system, install and debug the power stability test module 10, the dynamic power switching and fault tolerance test module 20, the thermal stress and thermal fatigue test module 30, the impact test module 40, the electromagnetic compatibility and immunity test module 50, the resource contention and multitasking test module 60, the abnormal state and self-repair capability test module 70, and set the test conditions;

[0328] Step 2: Collect the circuit board test working status data under each simulation environment in step 1, analyze and calculate to obtain the power stability index Dyw, power switching fault tolerance index Qhrc, thermal stress index Ryl, mechanical stress index Mvz, electromagnetic compatibility index Jrx, resource contention index Zyzq and self-repair index Xfz;

[0329] Step 3: After the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz are dimensionlessly processed, a comprehensive evaluation index Pc is obtained through an associated calculation formula and evaluated to obtain a first evaluation result;

[0330] Step 4: When the comprehensive evaluation index Pc is lower than the first performance threshold X, it indicates that the comprehensive performance test fails, a performance failure label is marked, and a second multi-level evaluation command is generated. The power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz are graded and evaluated through the second multi-level evaluation module 90 to obtain corresponding test evaluation results and generate corresponding debugging instructions.

[0331] In this embodiment, through steps one to four, a comprehensive evaluation of the performance of the circuit board is achieved. First, by configuring the test system and setting the test conditions, the accurate installation and debugging of each test module are ensured, thereby providing a reliable data basis for subsequent analysis. Secondly, the method ensures the comprehensiveness of the test by calculating multiple key performance indicators such as the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz. After the comprehensive evaluation step processes each indicator dimensionlessly, the comprehensive evaluation index Pc is obtained through the associated calculation formula, and the performance evaluation is performed to effectively determine whether the circuit board meets the market standards. When the comprehensive evaluation index Pc is lower than the first performance threshold X, the system can generate a performance failure label and perform an in-depth hierarchical evaluation through the second multi-level evaluation module 90. This mechanism can not only identify specific performance defects, but also generate targeted debugging instructions based on the evaluation results to optimize the circuit board design and improve its comprehensive performance and reliability. This method not only improves the accuracy of testing, but also can quickly locate problems and guide optimization, thereby significantly improving the overall quality and market competitiveness of circuit boards.

[0332] The threshold is set to facilitate comparison. The threshold depends on the amount of sample data and the number of bases set by technicians in this field for each set of sample data, as long as it does not affect the proportional relationship between the parameter and the quantized value.

[0333] The above formulas are obtained by collecting a large amount of data for software simulation and selecting a formula that is close to the actual value. The coefficients in the formula are set by technical personnel in this field according to actual conditions. The above is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited to this. Any technical personnel familiar with the technical field within the technical scope disclosed by the present invention, according to the technical solution and the inventive concept of the present invention, make equivalent replacement or change, which should be covered within the protection scope of the present invention.

Claims

1. The online fully automatic detection system for printed circuit board defects based on data analysis is characterized by: include: A power stability test module (10) is used to use a pluggable power module, replace the power module under different frequency and voltage conditions, so as to simulate an unstable power environment, collect and measure the working state of the circuit board under different voltage fluctuations, and construct a power stability index Dyw; A dynamic power switching and fault tolerance test module (20) is used to test the power switching capability and fault tolerance of the circuit board under dual power supply conditions, simulate power switching and fault environment through a fault injection system, and construct a power switching fault tolerance index Qhrc; A thermal stress and thermal fatigue test module (30) is used to use an environmental simulation box to simulate air temperatures of -20°C, -10°C, 20°C, 40°C and 60°C, test the working state of the circuit board in high and low temperature cycles, collect temperature stress data, and construct a thermal stress index Ryl; An impact test module (40) is used to simulate the mechanical vibration and impact conditions encountered by the circuit board during transportation or operation, test the working state of the circuit board under the mechanical vibration and impact conditions, and construct a mechanical stress index Mvz; The electromagnetic compatibility and anti-interference test module (50) applies electromagnetic waves of different strengths and frequencies around the circuit board through an electromagnetic interference generator to test the anti-interference ability of the circuit board, collect electromagnetic compatibility data, and construct an electromagnetic compatibility index Jrx; A resource contention and multi-task processing test module (60) tests the responsiveness of the circuit board under resource contention by simulating a multi-task parallel environment, collects multi-task processing data, and constructs a resource contention index Zyzq; The abnormal state and self-repair capability test module (70) tests the self-detection and repair capability of the circuit board by intentionally introducing abnormal states such as power failure, signal interference and communication interruption, collects abnormal processing data, and constructs a self-repair index Xfz; A first evaluation module (80) is used to perform dimensionless processing on a power stability index Dyw, a power switching fault tolerance index Qhrc, a thermal stress index Ryl, a mechanical stress index Mvz, an electromagnetic compatibility index Jrx, a resource contention index Zyzq and a self-repair index Xfz, and obtain a comprehensive evaluation index Pc by weighted calculation; when the comprehensive evaluation index Pc is lower than a first performance threshold X, it indicates that the comprehensive evaluation result is unqualified, and a second multi-level evaluation command is generated; The second multi-level evaluation module (90) receives the second multi-level evaluation command, and performs hierarchical evaluation on the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz under the comprehensive evaluation index Pc, obtains corresponding test evaluation results, and generates corresponding debugging instructions.

2. The online fully automatic detection system for printed circuit board defects based on data analysis according to claim 1 is characterized in that: The power supply stability test module (10) comprises a high-frequency voltage fluctuation simulation unit (11), a power supply noise injection unit (12), a transient voltage drop simulation unit (13) and a transient voltage spike simulation unit (14); The high-frequency voltage fluctuation simulation unit (11) is used to use a fluctuation generator to simulate the voltage fluctuation of the power supply at high frequency, test the adaptability of the circuit board to frequent voltage changes, monitor the voltage response of the circuit board through a data recorder, and calculate the high-frequency voltage fluctuation response index V by the following formula fluct : Where Pg i represents the task processing speed of the circuit board at the ith measurement point in a high-frequency voltage environment, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V ferq Indicates the amplitude of high-frequency voltage fluctuation, V Z Indicates the reference voltage value under normal conditions; n indicates the total number of measurements; The power supply noise injection unit (12) is used to use a noise generator and a noise injector to simulate power supply noise signals of different frequencies and intensities and inject them into the power supply input terminal to monitor the noise response change of the circuit board under noise interference, and obtain the noise interference response index R by calculating the following formula: noise : In the formula, Pz i represents the task processing speed of the circuit board at the ith measurement point under noise conditions, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V noise Indicates the voltage amplitude including the noise signal; The transient voltage drop simulation unit (13) is used to generate a sudden voltage drop signal using a signal generator and inject it into the power input terminal of the circuit board, and monitor the response of the circuit board under the voltage drop condition, and obtain the transient voltage drop response index R by calculating the following formula: voltage，drop : Where Pd i represents the task processing speed of the circuit board at the ith measurement point under transient voltage drop conditions, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V drop Indicates the voltage value of transient voltage drop; The transient voltage spike simulation unit (14) is used to inject a transient voltage spike signal from a signal generator into the power input terminal of the circuit board, and monitor the response of the circuit board under the voltage spike condition, and obtain the transient voltage spike response index R by calculating the following formula: voltage，spike : Where Ps i represents the task processing speed of the circuit board at the ith measurement point under transient voltage spike conditions, P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, V drop Indicates the voltage value of a transient voltage spike.

3. The online fully automatic detection system for printed circuit board defects based on data analysis according to claim 2 is characterized in that: The power supply stability test module (10) further comprises a power supply comprehensive analysis unit (15), wherein the power supply comprehensive analysis unit (15) is used to calculate a high-frequency voltage fluctuation response index V fluct , Noise interference response index R noise , transient voltage drop response index R voltage，drop and transient voltage spike response index R voltage，spike The power stability index Dyw is obtained by associating the following formula: Where a1, a2, a3 and a4 are high frequency voltage fluctuation response index V fluct , Noise interference response index R noise , transient voltage drop response index R voltage，drop and transient voltage spike response index R voltage，spike The weight value of , C1 is represented as the first correction constant.

4. The online fully automatic detection system for printed circuit board defects based on data analysis according to claim 1 is characterized in that: The power switching fault tolerance index Qhrc is calculated by the following formula: In the formula, ΔI surge,i It represents the sudden change value of the circuit board power supply current during the jth circuit board power supply switching process, I norm Indicates the reference current value under normal working conditions, ΔV fluc,j Indicates the voltage fluctuation of the circuit board power supply during the jth circuit board power supply switching process, through ΔV fluc,j =|VV Z |Calculate and obtain, V Z Indicates the reference voltage value under normal conditions; Frt j Frt is the time required for the circuit board to resume normal operation after the jth power switch. max Indicates the maximum allowable recovery time of the preset circuit board; w1, w2 and w3 represent weight coefficients.

5. The online fully automatic detection system for printed circuit board defects based on data analysis according to claim 1 is characterized in that: The thermal stress index Ryl is calculated by the following formula: Where P Z Indicates the processing speed of the circuit board benchmark task under normal conditions, P T1 Indicates the processing speed of the circuit board at an ambient temperature of -20°C, P T2 It indicates the task processing speed of the circuit board at an ambient temperature of -10℃, P T3 It indicates the task processing speed of the circuit board at an ambient temperature of 20°C, P T4 It indicates the task processing speed of the circuit board at 40℃ ambient temperature, P T5 It indicates the task processing speed of the circuit board under 60℃ air temperature condition. The thermal stress index Ryl comprehensively summarizes the performance fluctuation of the task processing speed under different temperature conditions relative to the reference temperature.

6. The online fully automatic detection system for printed circuit board defects based on data analysis according to claim 1 is characterized in that: The mechanical stress index Mvz is calculated by the following formula: Where PcJ represents the task processing speed of the circuit board under the jth vibration or shock condition, P1 represents the benchmark task processing speed without vibration or shock, and k represents the total number of vibration or shock tests; The electromagnetic compatibility index Jrx is calculated by the following formula: Where, PE j represents the circuit board task processing speed under the jth electromagnetic interference condition, P2 represents the benchmark task processing speed under the condition of no electromagnetic interference, and y represents the total number of electromagnetic interference tests; the resource contention index Zyzq is calculated and obtained by the following formula; Where, Pmt j represents the circuit board task processing speed in the jth multitasking environment, P3 represents the circuit board task processing speed in the benchmark environment without task contention, Rrc j represents the resource contention rate in the j-th multitasking environment, R ref represents the resource contention rate in the benchmark environment, which is set to 1, and q is the total number of tests in different multi-tasking environments; The self-repair index Xfz is calculated by the following formula: Where Rfix j Rerr represents the repair success rate under the jth abnormal state. j Trep represents the error rate under the jth abnormal state. j represents the time required to repair the jth abnormal state, T max Indicates the maximum allowed repair time, and B indicates the total number of abnormal state tests.

7. The online fully automatic detection system for printed circuit board defects based on data analysis according to claim 1 is characterized in that: The first evaluation module (80) comprises a first associating unit (81) and a first evaluating unit (82); The first associating unit (81) is used to perform dimensionless processing on the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz, and obtain the comprehensive evaluation index Pc through the following associating calculation: Among them, b1, b2, b3, b4, b5, b6 and b7 represent the weight values ​​of the power supply stability index Dyw, the power supply switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repairing index Xfz, respectively, which are adjusted and set by the user; The first evaluation unit (82) is used to preset a first performance threshold value X, and compare the comprehensive evaluation index Pc with the first performance threshold value X, including a first evaluation result, including: When the comprehensive evaluation index Pc ≥ the first performance threshold X, it means that the circuit board has passed the comprehensive performance test, and the circuit board is a qualified product and is allowed to circulate in the market; When the comprehensive evaluation index Pc is less than the first performance threshold value X, it indicates that the comprehensive performance test fails, a performance failure label is marked, and a second multi-level evaluation command is generated.

8. The online fully automatic detection system for printed circuit board defects based on data analysis according to claim 7 is characterized in that: The second multi-level evaluation module (90) comprises a hierarchical evaluation unit (91) and a strategy unit (92); The hierarchical evaluation unit (91) is used to perform hierarchical evaluation on the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz under the comprehensive evaluation index Pc after receiving the second multi-level evaluation command, and obtain corresponding test evaluation results, including: When the power supply stability index Dyw ≥ the first stability threshold X1, it indicates that the performance of the circuit board under the conditions of high-frequency voltage fluctuation, noise interference, transient voltage drop and transient voltage spike is qualified, and a first qualified label is generated; When the power supply stability index Dyw is less than the first stability threshold value X1, it indicates that the circuit board power supply stability is unqualified, and the strategy unit (92) generates a first debugging instruction, including: optimizing the circuit board power supply voltage stabilization module or optimizing the power supply design, with the target improvement range being to improve the power supply stability by 10%; When the power switching fault tolerance index Qhrc ≥ the second stable threshold value X2, it indicates that the circuit board has qualified performance in the power switching, and the recovery time after the power switching is within the expected range, and a second qualified label is generated; When the power switching fault tolerance index Qhrc is less than the second stable threshold value X2, it indicates that the circuit board fails to meet the performance requirements during power switching, and the strategy unit (92) generates a second debugging instruction, including: improving the power design, with the goal of reducing the recovery time after power switching by 15%; When the thermal stress index Ryl ≥ the third stable threshold value X3, it indicates that the performance of the circuit board is qualified under temperature changes, and a third qualified label is generated; When the thermal stress index Ryl is less than the third stable threshold value X3, it indicates that the performance of the circuit board is unqualified under temperature changes, and the strategy unit (92) generates a third debugging instruction, including: optimizing the heat dissipation design, with the goal of improving the heat dissipation efficiency by 20%; When the mechanical stress index Mvz ≥ the fourth stable threshold value X4, it indicates that the circuit board has qualified performance under mechanical vibration and impact conditions, and a fourth qualified label is generated; When the mechanical stress index Mvz is less than the fourth stable threshold value X4, it indicates that the performance of the circuit board under mechanical vibration and impact conditions is unqualified, and the strategy unit (92) generates a fourth debugging instruction, including: optimizing the installation and fixing design, adding 10% of shockproof materials, with the goal of improving the impact resistance by 20%; When the electromagnetic compatibility index Jrx ≥ the fifth stable threshold value X5, it indicates that the performance of the circuit board under the electromagnetic interference condition is qualified; the fifth qualified label is generated; When the electromagnetic compatibility index Jrx is less than the fifth stable threshold value X5, it indicates that the performance of the circuit board under electromagnetic interference conditions is unqualified; the strategy unit (92) generates a fifth debugging instruction, including: adding 10% electromagnetic shielding material, the goal is to reduce the impact of interference signals on performance by 20%; When the resource contention index Zyzq ≥ the sixth stable threshold value X6, it indicates that the performance of the circuit board in the multi-tasking and resource contention environment is qualified, and a sixth qualified label is generated; When the resource contention index Zyzq is less than the sixth stable threshold value X6, it indicates that the performance of the circuit board under the multi-tasking and resource contention environment is unqualified, and the strategy unit (92) generates a sixth debugging instruction, including: increasing 10% of spare processor memory and storage, using data parallel processing technology, dividing data into multiple parts and processing them simultaneously on different processing units; When the self-repair index Xfz ≥ the seventh stable threshold value X7, it indicates that the self-repair capability of the circuit board in an abnormal state is qualified, and the seventh qualified label is generated; When the self-repair index Xfz is less than the seventh stable threshold value X7, it indicates that the circuit board fails to meet the performance requirements under abnormal conditions. The strategy unit (92) generates a seventh debugging instruction, including: strengthening fault detection and repair capabilities, shortening repair time, and aiming to reduce repair time by 20%.

9. A method for online fully automatic detection of printed circuit board defects based on data analysis, applied to an online fully automatic detection system for printed circuit board defects based on data analysis according to any one of claims 1 to 8, characterized in that: The following steps are involved: Step 1: configure the test system, install and debug the power stability test module (10), the dynamic power switching and fault tolerance test module (20), the thermal stress and thermal fatigue test module (30), the impact test module (40), the electromagnetic compatibility and immunity test module (50), the resource contention and multi-tasking test module (60) and the abnormal state and self-repair capability test module (70), and set the test conditions; Step 2: Collect the circuit board test working status data under each simulation environment in step 1, analyze and calculate to obtain the power stability index Dyw, power switching fault tolerance index Qhrc, thermal stress index Ryl, mechanical stress index Mvz, electromagnetic compatibility index Jrx, resource contention index Zyzq and self-repair index Xfz; Step 3: After the power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz are dimensionlessly processed, a comprehensive evaluation index Pc is obtained through an associated calculation formula and evaluated to obtain a first evaluation result; Step 4: When the comprehensive evaluation index Pc is lower than the first performance threshold value X, it indicates that the comprehensive performance test is unqualified, a performance unqualified label is marked, and a second multi-level evaluation command is generated. The power stability index Dyw, the power switching fault tolerance index Qhrc, the thermal stress index Ryl, the mechanical stress index Mvz, the electromagnetic compatibility index Jrx, the resource contention index Zyzq and the self-repair index Xfz are graded and evaluated through the second multi-level evaluation module (90) to obtain corresponding test evaluation results and generate corresponding debugging instructions.

Images