Circuit board conductive film quality detection method, system, and storage medium

Abstract

The present invention discloses a quality detection method, system, and storage medium for a conductive film of a circuit board, the method comprising obtaining temperature distribution data, microstructure image, and thermal resistance value of the conductive film; performing calculations based on the temperature distribution data to obtain heat flux density distribution data; performing microstructure analysis of the conductive film based on the microstructure image to obtain microstructure parameters; performing thermal conductivity performance analysis to obtain the influence of microstructure parameters on the thermal conductivity performance of the conductive film; performing aging prediction analysis of the conductive film to obtain an aging prediction index; when the aging prediction index is greater than a preset index threshold, performing thermal area analysis on the temperature distribution data to obtain a thermal distribution abnormality report; performing performance degradation risk assessment to obtain a performance assessment result, and performing local repair and replacement operations on the conductive film based on the performance assessment result. The method can comprehensively evaluate the thermal performance of the conductive film under working conditions and its local overheating conditions, thereby improving the accuracy and reliability of quality detection.

Description

Technical Field

[0001] The present invention relates to the technical field of quality detection of conductive films of circuit boards, and in particular to a method, system and storage medium for quality detection of conductive films of circuit boards. Background Art

[0002] With the continuous development of electronic equipment, the conductive film of the circuit board plays a vital role in electrical connection and signal transmission. In the process of quality inspection of the conductive film, obtaining the thermal distribution is a key issue. The conductive film will generate a lot of heat in the working state, and its thermal distribution is crucial for analyzing the thermal conductivity and local overheating of the conductive film. However, the heat generated by the conductive film during operation is affected by many factors, such as the conductive film material, structure and process, which makes the thermal distribution complicated and difficult to obtain accurately.

[0003] At present, in the process of testing the quality of the conductive film of the circuit board, obtaining the heat distribution is crucial for analyzing the thermal conductivity and local overheating of the conductive film. However, the existing technology mainly relies on temperature sensor arrays or infrared imaging technology to obtain the thermal distribution data on the surface of the conductive film. These methods can only provide surface temperature information and cannot fully reflect the heat flux density distribution inside the conductive film.

[0004] The existing technology is unable to comprehensively evaluate the thermal performance of the conductive film under working conditions and its local overheating conditions, which affects the accuracy and reliability of quality inspection. Summary of the invention

[0005] The present invention provides a method, system and storage medium for detecting the quality of a conductive film of a circuit board, so as to comprehensively evaluate the thermal performance of the conductive film in a working state and its local overheating condition, and improve the accuracy and reliability of quality detection.

[0006] In a first aspect, in order to solve the above technical problems, the present invention provides a method for detecting the quality of a conductive film of a circuit board, comprising:

[0007] Obtain temperature distribution data, microstructure images and thermal resistance values ​​of conductive films;

[0008] According to the temperature distribution data, heat flux density distribution calculation is performed to obtain heat flux density distribution data of different areas of the conductive film;

[0009] According to the microstructure image, a microstructure analysis of the conductive film is performed to obtain microstructure parameters;

[0010] Based on a multivariate regression algorithm combined with the microstructure parameters and the heat flux density distribution data, a thermal conductivity performance analysis is performed to obtain the influence of the microstructure parameters on the thermal conductivity of the conductive film;

[0011] Based on the time series analysis method combined with the influencing law and the thermal resistance value, the conductive film aging prediction analysis is performed to obtain the aging prediction index of the conductive film thermal resistance changing with time;

[0012] When the aging prediction index is greater than a preset index threshold, a thermal region analysis is performed on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report;

[0013] A performance degradation risk assessment is performed based on the thermal distribution abnormality report, the aging prediction index and the microstructure parameters to obtain a performance assessment result.

[0014] Preferably, performing heat flux density distribution calculation according to the temperature distribution data to obtain heat flux density distribution data of different regions of the conductive film includes:

[0015] Based on the finite element analysis method, the conductive film is divided into a plurality of discrete units to obtain the conductive film discrete unit;

[0016] According to the temperature distribution data of the conductive film and the discrete units of the conductive film, a temperature field analysis is performed on the surface of the conductive film to obtain approximate temperature distribution data of the discrete units of the conductive film;

[0017] Based on Fourier's heat conduction law and the approximate temperature distribution data, the heat flux density vector of the discrete unit of the conductive film is calculated to obtain the internal heat flux density data of different regions of the conductive film;

[0018] According to the internal heat flux density data, interpolation and smoothing are performed to obtain a continuous and smooth heat flux density distribution surface;

[0019] According to the gradient corresponding to the heat flux density distribution surface, regional feature extraction is performed to obtain the heat flux direction and heat flux intensity of different regions of the conductive film;

[0020] A cluster analysis is performed according to the heat flow direction and the heat flow intensity to obtain heat flux density distribution data of different areas of the conductive film, wherein the heat flux density distribution data is a division of the heat flux density area of ​​the conductive film, and can distinguish the main channels and key areas of heat transfer inside the conductive film.

[0021] Preferably, the conducting film microstructure analysis based on the microstructure image to obtain microstructure parameters includes:

[0022] Digitally processing the microstructure image to obtain a grain region on the surface of the conductive film;

[0023] Based on the image segmentation algorithm, the grain area is calculated by pixel scale calibration to obtain the size distribution data of each grain;

[0024] Based on the edge detection algorithm combined with the size distribution data, the grain boundary characteristics are calculated to obtain the grain boundary length, grain boundary orientation angle and total grain boundary length;

[0025] Performing density calculation according to the grain boundary length, the grain boundary direction angle and the total length of the grain boundary to obtain the grain boundary density;

[0026] Based on an image classification algorithm, the defect area of ​​the microstructure image is identified to obtain the defect type and defect density;

[0027] Establishing a correlation with the performance of the conductive film according to the size distribution data, the grain boundary density, the defect type and the defect density, and obtaining microstructure parameters related to the performance of the conductive film;

[0028] Wherein, the grain boundary density is calculated according to the following formula:

[0029]

[0030] in, The grain boundary in a specific direction The grain boundary density on is the total length of the grain boundary, For the The length of the grain boundary, For the Segment grain boundary orientation angle, is the number of grain boundary segments, is the Dirac delta function.

[0031] Preferably, the thermal conductivity performance analysis is performed based on the multivariate regression algorithm in combination with the microstructure parameters and the heat flux density distribution data to obtain the influence of the microstructure parameters on the thermal conductivity performance of the conductive film, including:

[0032] According to the microstructure parameters and the heat flux density distribution data, data cleaning and normalization are performed to obtain a standardized data set;

[0033] According to the standardized data set, a quantitative relationship between the microstructure of the conductive film and the thermal conductivity performance is established and calculated based on a multivariate regression algorithm to obtain a predicted value of the thermal conductivity of the conductive film;

[0034] Based on the predicted value of thermal conductivity, a sensitivity analysis is performed to obtain the influence of the microstructure parameters on the thermal conductivity performance of the conductive film.

[0035] Preferably, the time series analysis method is combined with the influencing law and the thermal resistance value to perform conductive film aging prediction analysis to obtain an aging prediction index of the conductive film thermal resistance over time, including:

[0036] Based on the time series analysis method and the influence law, the future thermal resistance value of the conductive film is predicted to obtain the predicted thermal resistance value, the predicted time and the predicted end time;

[0037] According to the predicted time, the predicted end time, the thermal resistance value, the predicted thermal resistance value and the preset thermal resistance threshold, a conductive film aging prediction calculation is performed to obtain an aging prediction index of the conductive film thermal resistance changing with time.

[0038] Preferably, the calculation formula of the aging prediction index is:

[0039]

[0040] In the formula, is the aging prediction index, is the total number of prediction points, For the The predicted thermal resistance value of each prediction point is is the thermal resistance value; is the preset thermal resistance threshold, is the time decay factor, is the predicted end time; For the The prediction time of each prediction point.

[0041] Preferably, when the aging prediction index is greater than a preset index threshold, a thermal region analysis is performed on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report, including:

[0042] When the aging prediction index is greater than a preset index threshold, removing outliers and noise from the temperature distribution data to obtain preprocessed temperature distribution data;

[0043] Extracting temperature distribution features based on the pre-processed temperature distribution data to obtain temperature feature distribution data;

[0044] Based on the support vector machine algorithm combined with the cross-validation method, the temperature characteristic distribution data is subjected to thermal area analysis to obtain the coordinates of the abnormal area;

[0045] Extracting the minimum temperature, maximum temperature and average temperature corresponding to the coordinates of the abnormal area;

[0046] A temperature gradient report generation operation is performed according to the minimum temperature, the maximum temperature and the average temperature to obtain a thermal distribution anomaly report.

[0047] Preferably, the performing of a performance degradation risk assessment based on the abnormal heat distribution report, the aging prediction index and the microstructure parameter to obtain a performance assessment result, and performing a local repair and replacement operation on the conductive film based on the performance assessment result, includes:

[0048] According to the abnormal heat distribution report, a temperature analysis is performed on a local area of ​​the conductive film to obtain a first high temperature area that exceeds the temperature limit of the conductive film material;

[0049] According to the aging prediction index and the microstructure parameter, a change analysis of the thermal resistance of the conductive film is performed to obtain a second high temperature region where the heat dissipation efficiency decreases due to an increase in the thermal resistance of the conductive film;

[0050] The conductive films corresponding to the first high-temperature region and the second high-temperature region are partially repaired and replaced.

[0051] In a second aspect, the present invention provides a circuit board conductive film quality detection system, comprising:

[0052] A data acquisition module, used to obtain temperature distribution data, microstructure images and thermal resistance values ​​of the conductive film;

[0053] A heat flux density module is used to calculate the heat flux density distribution according to the temperature distribution data to obtain the heat flux density distribution data of different areas of the conductive film;

[0054] A microstructure module, used to perform microstructure analysis of the conductive film according to the microstructure image to obtain microstructure parameters;

[0055] An influence law module is used to analyze the thermal conductivity performance based on a multivariate regression algorithm in combination with the microstructure parameters and the heat flux density distribution data, and obtain the influence law of the microstructure parameters on the thermal conductivity performance of the conductive film;

[0056] A prediction index module, used to perform conductive film aging prediction analysis based on a time series analysis method combined with the influencing law and the thermal resistance value, and obtain an aging prediction index of the conductive film thermal resistance changing with time;

[0057] An abnormality analysis module, used for performing a thermal area analysis on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report when the aging prediction index is greater than a preset index threshold;

[0058] The performance evaluation module is used to perform performance degradation risk evaluation according to the abnormal heat distribution report, the aging prediction index and the microstructure parameter to obtain a performance evaluation result.

[0059] In a third aspect, the present invention further provides an electronic device comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements any one of the above-mentioned methods for detecting the quality of conductive film of a circuit board when executing the computer program.

[0060] In a fourth aspect, the present invention further provides a computer-readable storage medium, the computer-readable storage medium comprising a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute any one of the above-mentioned methods for detecting the quality of conductive films of circuit boards.

[0061] Compared with the prior art, the present invention has the following beneficial effects:

[0062] The present invention discloses a method, system and storage medium for detecting the quality of a conductive film of a circuit board, which can evaluate the thermal performance of the conductive film in a working state and its local overheating condition, and solve the problem in the prior art that the internal heat flux density distribution of the conductive film and its aging process cannot be accurately evaluated.

[0063] The present invention obtains the temperature distribution data, microstructure image and thermal resistance value data of the conductive film, and combines the multivariate regression algorithm to analyze the quantitative relationship between the microstructure of the conductive film and the thermal conductivity performance, and reveals the influence of the microstructure parameters on the thermal conductivity and thermal resistance of the conductive film. The present invention introduces a time series analysis method to predict the trend of the thermal resistance of the conductive film over time, and combines the thermal resistance aging prediction index to perform aging evaluation. By predicting the law of thermal resistance change over time, the present invention can effectively identify the aging process of the conductive film, timely warn of the performance degradation problems that occur, and ensure the reliability and stability of the conductive film under long-term working conditions. When the aging prediction index exceeds the preset threshold, the present invention further combines the support vector machine algorithm to perform thermal area analysis on the temperature distribution data of the conductive film, and can automatically identify the overheating area of ​​the conductive film. Through the abnormal thermal distribution report, the potential overheating problem area is located to avoid local damage or failure problems caused by heat accumulation. Finally, the present invention combines heat flux density distribution data, microstructure parameters and aging prediction index to conduct risk assessment of conductive film performance degradation. Through a comprehensive risk assessment method, the present invention can take timely measures to repair potential problems in local areas of the conductive film, thereby extending the service life of the conductive film and ensuring the stability of the overall function of the circuit board.

[0064] In summary, the present invention can comprehensively evaluate the thermal performance of the conductive film in a working state and its local overheating condition, thereby improving the accuracy and reliability of quality detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] Figure 1It is a schematic diagram of the process flow of the circuit board conductive film quality detection method provided by the first embodiment of the present invention;

[0066] Figure 2 It is a schematic diagram of the structure of a circuit board conductive film quality detection system provided by the second embodiment of the present invention. DETAILED DESCRIPTION

[0067] 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.

[0068] Reference Figure 1 The first embodiment of the present invention provides a method for detecting the quality of a conductive film of a circuit board, comprising the following steps:

[0069] S11, obtaining temperature distribution data, microstructure image and thermal resistance value of the conductive film;

[0070] S12, performing heat flux density distribution calculation according to the temperature distribution data to obtain heat flux density distribution data of different regions of the conductive film;

[0071] S13, performing microstructure analysis of the conductive film according to the microstructure image to obtain microstructure parameters;

[0072] S14, performing a thermal conductivity analysis based on a multivariate regression algorithm in combination with the microstructure parameters and the heat flux density distribution data to obtain the influence of the microstructure parameters on the thermal conductivity of the conductive film;

[0073] S15, performing conductive film aging prediction analysis based on a time series analysis method in combination with the influencing law and the thermal resistance value, and obtaining an aging prediction index of the conductive film thermal resistance changing with time;

[0074] S16, when the aging prediction index is greater than a preset index threshold, performing a thermal region analysis on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report;

[0075] S17, performing a performance degradation risk assessment based on the abnormal thermal distribution report, the aging prediction index and the microstructure parameter to obtain a performance assessment result.

[0076] In step S11, it is necessary to obtain the temperature distribution data, microstructure image and thermal resistance value of the conductive film, including:

[0077] In a specific embodiment, a combination of two methods is used to obtain temperature distribution data. On the one hand, a high-precision temperature sensor array is arranged on the surface of the conductive film, and the sensors are evenly distributed to cover the key areas of the conductive film. The measurement accuracy of the sensor is better than ±0.1°C, and the sampling frequency is set to once per second. The sensor array is connected to the computer through a data acquisition module to record the surface temperature data of the conductive film in the working state in real time. On the other hand, a high-resolution infrared thermal imager is used to scan the surface of the conductive film and generate a temperature distribution map in real time. The resolution of the infrared thermal imager must reach 640×480 pixels or higher, and the sensitivity range must be adjusted to adapt to the working temperature of the conductive film (for example, 0°C to 100°C). After the infrared image data is processed by special software, a complete surface temperature distribution map of the conductive film can be formed.

[0078] In a specific embodiment, the microstructure of the conductive film is observed and analyzed by scanning electron microscopy (SEM). First, a representative small-size sample (no more than 1 cm²) is cut from the conductive film, and its surface is ground and polished to meet the SEM imaging requirements. Then, the sample is placed in the SEM, the acceleration voltage is adjusted to 5~15kV, and the backscattered electron mode is used to obtain high-resolution images of the grain area, grain boundary and defect characteristics of the conductive film. The magnification is set to 5000× to 10000× to clearly display the microstructural details. The collected microstructure images are digitally analyzed by image processing technology, the grain area is extracted using an image segmentation algorithm, and key parameters such as grain size, grain boundary density and defect distribution are calculated by edge detection and morphological processing. Combined with the image classification algorithm, different defect types (such as holes, cracks, etc.) can also be identified to further evaluate the impact of the microstructure of the conductive film on its performance.

[0079] In a specific embodiment, the present invention adopts a thermal resistance measurement method under constant power conditions. In the experiment, a heat source with a constant power (such as 10W to 100W) is applied to the conductive film through a temperature control device, and a temperature sensor and a thermocouple are used to record the temperature of the hot end and the cold end of the conductive film respectively. The thermal resistance value is calculated by the formula Calculation is performed, where Th and Tc are the hot end and cold end temperatures, respectively, and P is the input power. To further improve the measurement efficiency and accuracy, the present invention introduces an automated measurement system to record the curve of thermal resistance changing with temperature through real-time monitoring and data acquisition software. In addition, considering that the conductive film is affected by environmental noise and heat loss under actual working conditions, the present invention ensures the accuracy and stability of the measurement results by optimizing experimental conditions (such as stable temperature difference setting and thermal isolation design).

[0080] In step S12, it is necessary to perform heat flux density distribution calculation according to the temperature distribution data to obtain heat flux density distribution data of different regions of the conductive film, including:

[0081] First, based on the finite element analysis method, the conductive film is divided into multiple discrete units to obtain the conductive film discrete units. By establishing a geometric model of the conductive film, the overall area of ​​the conductive film is divided into multiple small finite element units, such as triangular or quadrilateral units, so as to perform discretization calculations of the temperature field and heat flux density. During the division process, it is preferred to use an automatic mesh generation algorithm to make the unit distribution more uniform, while focusing on strengthening the mesh density of key areas (such as areas with large temperature gradients) to ensure calculation accuracy. In practical applications, the conductive film can be divided into 1000 to 5000 finite element units according to its size and performance requirements, and the mesh size is controlled at the micron level. The specific division process uses commercial finite element software such as ANSYS or COMSOL Multiphysics.

[0082] In a specific embodiment, the conductive film surface temperature field analysis is performed based on the conductive film temperature distribution data and the conductive film discrete unit to obtain the approximate temperature distribution data of the conductive film discrete unit. Within each discrete unit, it is assumed that the temperature distribution changes linearly, and the discrete unit is interpolated and analyzed using the temperature distribution data to calculate the temperature value of each node in the unit. The surface temperature data obtained by the temperature sensor array or infrared thermal imaging is used as the boundary condition for the temperature field analysis. With the boundary temperature of the conductive film as a known condition, a finite element solution method (such as a weighted residual method or a least squares method) is used to construct a temperature field model on the surface of the conductive film, and the approximate temperature distribution data in each unit is obtained by calculation.

[0083] Specifically, based on Fourier's law of heat conduction combined with the approximate temperature distribution data, the heat flux density vector of the discrete unit of the conductive film is calculated to obtain the internal heat flux density data of different regions of the conductive film. According to Fourier's law of heat conduction, the heat flux density vector can be expressed by the formula Calculate, where q is the heat flux vector, k is the thermal conductivity of the conductive film, is the temperature gradient. For each discrete unit, the temperature gradient is calculated using its node temperature value, and the heat flux density vector data in the unit is calculated in combination with the thermal conductivity parameter of the material. The direction of the heat flux vector is determined by the temperature gradient, and its size reflects the heat flow rate per unit area.

[0084] In a specific embodiment, interpolation and smoothing are performed according to the internal heat flux density data to obtain a continuous and smooth heat flux density distribution surface. For the calculated discrete heat flux density data, local discontinuities will be introduced due to unit division, and a continuous distribution surface needs to be generated by interpolation and smoothing. Preferably, the heat flux density data is interpolated by cubic spline interpolation method, and combined with Gaussian smoothing filter processing to reduce noise and calculation errors. After processing, the heat flux density distribution surface achieves a smooth transition between units, and clearly shows the overall change trend of the heat flux density inside the conductive film.

[0085] Specifically, according to the gradient corresponding to the heat flux density distribution surface, regional feature extraction is performed to obtain the heat flow direction and heat flux intensity of different regions of the conductive film. By calculating the gradient information of the heat flux density distribution surface, the main direction and intensity of heat transfer in the conductive film are extracted. The gradient direction indicates the transfer direction of the heat flow, and the gradient size reflects the strength of the heat flow. The feature extraction algorithm is used to analyze the gradient data to identify the areas where heat transfer exists in the conductive film, such as the heat source gathering area or the heat dissipation path.

[0086] In a specific embodiment, cluster analysis is performed according to the heat flow direction and the heat flow intensity to obtain heat flux density distribution data of different areas of the conductive film. The heat flux density data is classified using a clustering algorithm (such as K-means clustering), and areas with similar heat flux density distribution characteristics are grouped together to achieve the division of heat flux density areas of the conductive film. This division can effectively distinguish the main channels and key areas of heat transfer inside the conductive film, providing an important basis for optimizing the thermal performance of the conductive film. For example, areas with higher heat flux density can be identified as the main paths of heat transfer, which need to be monitored or improved, while areas with lower heat flux density need to enhance thermal conductivity to evenly distribute heat.

[0087] In step S13, it is necessary to perform microstructure analysis of the conductive film according to the microstructure image to obtain microstructure parameters, including:

[0088] First, the microstructure image is digitized to obtain the grain region on the surface of the conductive film. A high-resolution image of the surface of the conductive film is obtained by scanning electron microscopy (SEM) or transmission electron microscopy (TEM). Assume that a SEM device is used with an accelerating voltage of 10 kV to obtain a high-resolution image with a magnification of 2000×. During the image processing, image enhancement techniques (such as contrast stretching and noise filtering) are used to improve the image quality. Subsequently, an image segmentation algorithm (such as the Otsu method or the K-means algorithm) is applied to separate the grain region in the image from the background. This process converts the microstructure image of the conductive film into digital image data that can be further analyzed. For example, through the image segmentation algorithm, the number of grain regions is identified as 2000, and the boundaries and morphological characteristics of the grains are successfully extracted.

[0089] In a specific embodiment, based on the image segmentation algorithm, the grain area is calculated by pixel scale calibration to obtain the size distribution data of each grain. After the extraction of the grain area is completed, pixel scale calibration is required to convert the number of pixels in the image into the actual size. For example, assuming that 1000 pixels correspond to 1 micron, the actual size of the grain can be inferred by calculating the number of pixels occupied by each grain in the image. If a grain occupies 500 pixels in the image, its actual size is 0.5 microns. In this way, the size distribution data of all grains can be calculated and statistical information of the grain size can be generated. For example, the calculated average grain size is 0.75 microns, the maximum size is 2 microns, the minimum size is 0.3 microns, and the standard deviation is 0.2 microns. These data provide a basis for subsequent grain boundary feature analysis.

[0090] In a specific embodiment, the grain boundary characteristics are calculated based on the edge detection algorithm combined with the size distribution data to obtain the grain boundary length, grain boundary orientation angle and total grain boundary length. The grain image is processed using an edge detection algorithm (such as the Canny algorithm or the Sobel operator) to identify the boundaries between the grains. By refining the edges, the length of each grain boundary can be accurately obtained. Assume that in the processed image, the total length of the grain boundary is calculated to be 500 microns. Combined with the size distribution data, the orientation angle of the grain boundary is further calculated (for example, a grain boundary orientation angle of 45° indicates that the grain boundary is at an angle of 45° to the horizontal line), and the distribution range of the grain boundary orientation angle is calculated. Through these calculations, the distribution characteristics of the grain boundary in the conductive film can be obtained, thereby providing support for the thermal performance analysis of the conductive film.

[0091] Specifically, density calculation is then performed based on the grain boundary length, the grain boundary direction angle, and the total length of the grain boundary to obtain the grain boundary density. The grain boundary density is calculated according to the following formula:

[0092]

[0093] in, The grain boundary in a specific direction The grain boundary density on is the total length of the grain boundary, For the The length of the grain boundary, For the Segment grain boundary orientation angle, is the number of grain boundary segments, is the Dirac delta function.

[0094] It should be noted that the Dirac delta function is a mathematical function with the following properties: It takes on an infinite value at one position and zero at other positions. Its function is to "select" a particular value , that is, only in It affects other values ​​when it is in use, and is usually used to describe discrete characteristics in direction or position. In grain boundary density calculation, it is used to extract the distribution characteristics of grain boundaries in a specific direction. By calculating the length, direction angle and corresponding temperature response of each grain boundary segment, the distribution density of grain boundaries in each direction is finally obtained. For example, the grain boundary density in a certain direction is 0.2 , indicating that there are more grain boundaries per unit length in this direction.

[0095] In a specific embodiment, based on an image classification algorithm, the defect area of ​​the microstructure image is identified to obtain the defect type and defect density. An image classification algorithm (such as a support vector machine (SVM) or a convolutional neural network (CNN)) is used to identify the defect area of ​​the microstructure image of the conductive film. By training the classification model, the defect types in the image, such as holes, cracks, oxygen vacancies, etc., can be accurately identified. The number of each defect type can be obtained by counting the number of pixels in each defect area, and the defect density per unit area is calculated. For example, if 5 holes and 3 microcracks are found in an area of ​​100 µm², the hole density is 0.05 / µm² and the crack density is 0.03 / µm².

[0096] In a specific embodiment, the size distribution data, the grain boundary density, the defect type and the defect density are correlated with the performance of the conductive film to obtain microstructural parameters related to the performance of the conductive film. Specifically, first, by obtaining the microstructural parameters of the conductive film, such as grain size distribution, grain boundary density and defect density, a large data set containing these parameters is compiled. For example, assume that data such as grain size, grain boundary density and defect density are extracted from 500 samples and statistically analyzed.

[0097] Specifically, multiple regression analysis, neural networks or other machine learning algorithms are used to establish a quantitative relationship model between the microstructure of the conductive film and performance parameters such as thermal conductivity and thermal resistance. Taking multiple linear regression as an example, by taking grain size, grain boundary density and defect density as independent variables and thermal conductivity and thermal resistance as dependent variables, regression analysis is performed to establish a mathematical model. After training on a large amount of experimental data, the model can reveal the relationship between microstructure parameters and thermal conductivity and thermal resistance of the conductive film.

[0098] For example, the experimental results show that: the grain size is negatively correlated with thermal conductivity, that is, the larger the grain size, the lower the thermal conductivity; the grain boundary density is positively correlated with thermal resistance, that is, the larger the grain boundary density, the higher the thermal resistance value; there is also a certain negative correlation between defect density and thermal conductivity, the more defects there are, the lower the thermal conductivity. Through this quantitative relationship model, the thermal conductivity performance of the conductive film under different microstructure conditions can be accurately predicted, which helps optimize the design and manufacturing process of the conductive film material and improve its thermal management performance.

[0099] In step S14, it is necessary to perform a thermal conductivity analysis based on a multivariate regression algorithm in combination with the microstructure parameters and the heat flux density distribution data to obtain the influence of the microstructure parameters on the thermal conductivity of the conductive film, including:

[0100] First, according to the microstructure parameters and the heat flux distribution data, data cleaning and normalization are performed to obtain a standardized data set. Before performing thermal conductivity performance analysis, the collected data needs to be preprocessed. Data cleaning includes removing missing values ​​and outliers. For example, suppose that some grain sizes or heat flux values ​​in the sample data set are abnormal, caused by measurement errors. By setting reasonable thresholds, these abnormal data points are removed to ensure the reliability of the analysis results. Next, data normalization is performed to map all data to a unified standard range (for example, all microstructure parameters and heat flux data are normalized to the [0,1] interval). Normalization can effectively avoid the impact of data of different dimensions on the analysis results, making the contribution of each variable to the regression model more balanced. This step ensures the stability and accuracy of subsequent multivariate regression analysis.

[0101] In a specific embodiment, according to the standardized data set, a quantitative relationship between the microstructure of the conductive film and the thermal conductivity is established and calculated based on a multiple regression algorithm to obtain a predicted value of the thermal conductivity of the conductive film. After data cleaning and normalization, a multiple regression algorithm (such as multiple linear regression or ridge regression) is used to establish a quantitative relationship between the microstructure parameters of the conductive film and the thermal conductivity. The input of the regression analysis includes microstructure parameters such as grain size, grain boundary density, defect density, and heat flux density distribution data (such as heat flux intensity in each region). For example, assuming that grain size, grain boundary density, and defect density are used as independent variables, and thermal conductivity is used as a dependent variable, the least squares method is used to solve the regression coefficient. Through the fitted regression equation, the predicted value of thermal conductivity under different microstructure parameter conditions can be obtained. This regression model provides a basis for predicting the thermal conductivity of different conductive films under different microstructures. For example, in the experiment, after regression analysis, a model for predicting thermal conductivity is obtained: thermal conductivity = a (Grain size)+b (grain boundary density)+c (defect density), where a, b, and c are coefficients obtained through regression analysis. By substituting the experimental data into this equation, the predicted values ​​of thermal conductivity under different microstructural conditions can be obtained.

[0102] In a specific embodiment, according to the predicted value of thermal conductivity, a sensitivity analysis is performed to obtain the influence of the microstructure parameters on the thermal conductivity of the conductive film. After the thermal conductivity prediction model is established, it is necessary to perform a sensitivity analysis on the model to determine the degree of influence of different microstructure parameters on thermal conductivity. Sensitivity analysis can be performed by calculating the contribution of each independent variable (such as grain size, grain boundary density, defect density, etc.) to the predicted change in thermal conductivity. Common methods include single factor sensitivity analysis or Monte Carlo simulation. Taking single factor sensitivity analysis as an example, the values ​​of grain size, grain boundary density or defect density are changed one by one to observe the change in the predicted value of thermal conductivity. For example, it is found that when the grain size increases, the thermal conductivity decreases, and when the grain boundary density increases, the thermal conductivity increases, which indicates that the grain size and thermal conductivity are negatively correlated, while the grain boundary density and thermal conductivity are positively correlated. Through these analysis results, it can be clarified which microstructure parameters have the greatest impact on the thermal conductivity of the conductive film, providing a theoretical basis for subsequent optimization of the conductive film. For example, after sensitivity analysis, the following conclusions were obtained: grain size has an effect on thermal conductivity, the larger the grain size, the lower the thermal conductivity; grain boundary density has a strong positive effect on thermal conductivity, the larger the grain boundary density, the higher the thermal conductivity; defect density is negatively correlated with thermal conductivity, the larger the defect density, the lower the thermal conductivity. These findings can help design more efficient conductive films and optimize thermal conductivity performance by adjusting microstructural parameters (such as grain size, grain boundary density, etc.).

[0103] In step S15, it is necessary to perform conductive film aging prediction analysis based on a time series analysis method in combination with the influencing law and the thermal resistance value to obtain an aging prediction index of the conductive film thermal resistance changing with time, including:

[0104] First, based on the time series analysis method combined with the above-mentioned influencing rules, the future thermal resistance value of the conductive film is predicted to obtain the predicted thermal resistance value, prediction time and prediction end time. The time series is extracted from the historical thermal resistance data of the conductive film, usually by regularly monitoring the change value of the thermal resistance of the conductive film at different working times to form a complete thermal resistance time series data set. For example, the hourly thermal resistance data is obtained from 100 hours of experimental data, and the thermal resistance data is modeled using the ARIMA model (autoregressive integral moving average model) or LSTM (long short-term memory network) to predict the future thermal resistance change trend. Through these analyses, the system can predict the thermal resistance value in the future period of time and determine the start and end time of the prediction. Assuming that the prediction time is 100 hours and the end time is 120 hours, the system can give the predicted thermal resistance value at each time point, thereby providing a basis for subsequent aging prediction.

[0105] In a specific embodiment, the conductive film aging prediction calculation is performed according to the predicted time, the predicted end time, the thermal resistance value, the predicted thermal resistance value and the preset thermal resistance threshold, and an aging prediction index of the conductive film thermal resistance over time is obtained. The obtained predicted thermal resistance value will be used as input, and the aging prediction calculation is performed in combination with the thermal resistance threshold of the conductive film (such as the maximum allowable thermal resistance threshold Rthreshold). The preset thermal resistance threshold indicates that when the thermal resistance value exceeds a certain level, the thermal conductivity of the conductive film has degraded and maintenance measures need to be taken. The calculation formula of the aging prediction index is:

[0106]

[0107] In the formula, is the aging prediction index, is the total number of prediction points, For the The predicted thermal resistance value of each prediction point is is the thermal resistance value; is the preset thermal resistance threshold, is the time decay factor, is the predicted end time; For the The prediction time of each prediction point.

[0108] It should be noted that the aging prediction index calculated by this formula is , which can reflect the trend of the thermal resistance of the conductive film changing over time. When the value is large, it means that the thermal resistance of the conductive film has approached or exceeded the preset threshold, indicating that the thermal conductivity of the conductive film has been seriously degraded, affecting its normal use and needs to be repaired or replaced. If the thermal conductivity of the conductive film is small, it means that the thermal conductivity of the conductive film is still within an acceptable range. Assume that in the prediction process, the predicted thermal resistance value of the conductive film in the next 100 hours is obtained through time series analysis (for example, the predicted thermal resistance values ​​are R1=1.2, R2=1.3,..., R10=1.8). If the initial thermal resistance value R0=1.0, the preset thermal resistance threshold Rthreshold=2.0, the time attenuation factor λ=0.0, and the prediction end time tend=120 hours, the calculated aging prediction index can be used to evaluate the aging degree of the conductive film. If the final If the value is greater than the set warning threshold (for example, assuming the set warning threshold is 0.6), it means that the thermal resistance of the conductive film is close to the threshold and local repair or replacement is required.

[0109] In a specific embodiment, by combining the thermal resistance prediction value and the preset thermal resistance threshold, the aging trend of the conductive film is evaluated and the remaining service life thereof is predicted. , the system can further analyze the remaining service life of the conductive film and provide a comprehensive aging risk assessment based on the microstructural parameters of the conductive film, such as grain size, grain boundary density, defect density, etc. Exceeding a preset threshold (e.g. >0.6), the system will automatically sound an alarm to remind the user that preventive measures need to be taken, such as local repair or replacement of the conductive film, to prevent thermal failure or overheating.

[0110] In step S16, when the aging prediction index is greater than a preset index threshold, a thermal region analysis is performed on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report, including:

[0111] First, when the aging prediction index is greater than the preset index threshold, the temperature distribution data is subjected to an outlier and noise removal operation to obtain preprocessed temperature distribution data. In the thermal performance test of the conductive film, the temperature distribution data contains outliers or noise caused by environmental noise or sensor errors, which will affect the accuracy of subsequent analysis. Therefore, it is first necessary to perform denoising and outlier processing on the temperature distribution data. Common denoising methods include Gaussian filtering, mean filtering or median filtering. For example, a 3x3 Gaussian filter is used to smooth the original temperature data to reduce the temperature data deviation caused by environmental fluctuations or measurement errors. For outlier detection, the out-of-range abnormal data can be eliminated or corrected by setting upper and lower limits (such as a maximum temperature not exceeding 150°C). The data after removing outliers and noise can more accurately reflect the actual temperature distribution of the conductive film, providing a more reliable data basis for subsequent analysis.

[0112] In a specific embodiment, temperature distribution feature extraction is performed based on the preprocessed temperature distribution data to obtain temperature feature distribution data. After the preprocessing is completed, the next step is to extract feature information from the temperature distribution data. The purpose of temperature distribution feature extraction is to convert temperature distribution information into numerical data that can be analyzed, usually through statistical analysis or pattern recognition methods. For example, the extracted features may include the mean, variance, area and shape of the hot spot area, etc. of the temperature. The mean and variance of the temperature distribution can reflect the thermal stability of the conductive film, while the area and shape of the hot spot area help to identify local overheating. By extracting these features, temperature feature distribution data can be obtained, providing the necessary input for subsequent thermal area analysis.

[0113] In a specific embodiment, based on the support vector machine algorithm combined with the cross-validation method, the temperature feature distribution data is subjected to a thermal region analysis to obtain the coordinates of the abnormal region. The extracted temperature feature distribution data is subjected to a thermal region analysis using the support vector machine (SVM) algorithm. Support vector machine is an effective classification algorithm that can learn the relationship between temperature features and thermal regions based on labeled training data. By using normal thermal distribution regions and abnormal thermal regions in historical data sets as training data, SVM can learn and classify different temperature distribution features. For example, assuming that the temperature values ​​of certain regions exceed a certain threshold and the distribution range is large, they are marked as abnormal regions. The accuracy of the SVM classification model is evaluated using a cross-validation method to select the optimal model parameters (such as penalty factor C and kernel function parameters). The trained SVM model can be used to predict which regions in the new data have thermal anomalies and output the coordinates of these regions.

[0114] In a specific embodiment, the minimum temperature, maximum temperature and average temperature corresponding to the coordinates of the abnormal area are extracted. Once the coordinates of the abnormal area are determined, the next step is to analyze these areas in detail and extract the temperature characteristics of each abnormal area. Specifically, it is necessary to extract the minimum temperature, maximum temperature and average temperature values ​​of each abnormal area. For example, the minimum temperature of a certain abnormal area is 120°C, the maximum temperature is 150°C, and the average temperature is 135°C. These temperature characteristics will help to further evaluate the thermal performance of the area and determine whether the critical value for repair or replacement has been reached.

[0115] In a specific embodiment, a temperature gradient report generation operation is performed according to the minimum temperature, the maximum temperature and the average temperature to obtain a thermal distribution abnormality report. A temperature gradient report is generated by analyzing the temperature characteristics of the abnormal area. A temperature gradient report usually includes a changing trend of the temperature distribution, a detailed description of the overheated area, and a comparison with a preset standard. For example, if the temperature of an area exceeds a preset threshold, the area will be marked as a "high temperature area" and a detailed report will be generated, including the minimum, maximum and average temperature, temperature gradient value, heat flux intensity and other information of the area. The report can also provide recommended measures, such as suggesting to increase the heat dissipation design or replace the conductive film in the overheated area. The temperature gradient report can not only help engineers quickly identify problems, but also provide a scientific basis for optimizing the conductive film design.

[0116] For example, assuming that in the surface temperature distribution of the conductive film, the system collected temperature data at 100 different locations through the temperature sensor, and after denoising, obtained smooth temperature distribution data. Using the SVM algorithm, the system successfully identified 5 abnormal heat areas, and by extracting temperature features, it was found that the temperature of a certain abnormal area reached 160°C, far exceeding the normal operating temperature of 100°C, and the temperature gradient of this area was 50°C / cm. The final generated thermal distribution abnormality report shows that the heat transfer in this area is not smooth, and it is recommended to increase the heat dissipation design or partially replace the conductive film to prevent further damage.

[0117] It should be noted that the setting of the preset index threshold is based on the thermal resistance characteristics and material properties of the conductive film, and is determined by experimental data and aging prediction models. Specifically, the initial thermal resistance value of the conductive film is set to R0=1.0Ω cm 2 , thermal resistance threshold Rthreshold = 2.0Ω cm 2Indicates performance degradation. When the aging prediction index P exceeds 0.6, it means that the thermal conductivity of the conductive film begins to degrade and thermal area analysis is required. Specifically, when P>0.6, the system triggers the analysis, checks the thermal distribution anomaly, and generates a thermal distribution anomaly report. If the P value exceeds 0.8, it means that the conductive film has been severely degraded and needs maintenance or replacement. This threshold can be dynamically adjusted according to the working environment of the conductive film to ensure its stability under different working conditions.

[0118] In step S17, it is necessary to perform a performance degradation risk assessment based on the abnormal thermal distribution report, the aging prediction index and the microstructure parameter to obtain a performance assessment result, including:

[0119] First, according to the abnormal thermal distribution report, the temperature analysis of the local area of ​​the conductive film is performed to obtain the first high temperature area that exceeds the temperature limit of the conductive film material. The abnormal thermal distribution report provides detailed distribution information of the surface temperature of the conductive film, especially the abnormally high temperature area. In this step, the temperature data in the abnormal thermal distribution report is first parsed to identify the area that exceeds the temperature limit of the conductive film material. Assuming that the temperature limit of the conductive film is 150°C, the system will mark all areas where the temperature exceeds this value. These areas are considered "first high temperature areas" and they are potential sources of thermal failure. For example, the temperature of a certain area reaches 160°C, so it is identified as the first high temperature area.

[0120] In a specific embodiment, the change in the thermal resistance of the conductive film is analyzed according to the aging prediction index combined with the microstructure parameters, and a second high temperature area is obtained where the thermal resistance of the conductive film increases and the heat dissipation efficiency decreases. The change in the thermal resistance of the conductive film is analyzed by combining the aging prediction index (for example, P=0.7) and microstructure parameters (such as grain size, grain boundary density, etc.). As the thermal resistance increases, the heat dissipation efficiency of the conductive film decreases, and some areas are locally overheated. By analyzing the change in thermal resistance, the system can identify the "second high temperature area" where the heat dissipation efficiency decreases. For example, when the aging prediction index is large, the thermal resistance of a certain area increases to 2.5Ω cm², which results in the ineffective transfer of heat in this area and the temperature rises to 155°C, which is identified as the second high temperature area.

[0121] For example, suppose that in a certain conductive film quality inspection, the thermal distribution abnormality report shows that the temperature of a certain area on the conductive film surface has reached 160°C, exceeding the material's tolerance temperature limit, and the system identifies this area as the first high temperature area. At the same time, based on the aging prediction index P=0.7 and microstructure parameters (such as larger grain size and higher grain boundary density), the system analysis found that the thermal resistance value of this area has increased to 2.3 Ω cm², resulting in reduced heat dissipation efficiency and a temperature increase of 155°C, which was identified as the second high temperature area. Ultimately, the system triggered a local repair operation, installing a heat sink in the first high temperature area and replacing the conductive film material in the second high temperature area to improve the overall heat dissipation performance and prevent overheating and potential failure.

[0122] In summary, the present invention discloses a method for detecting the quality of a conductive film of a circuit board. By acquiring the temperature distribution data, microstructure image and thermal resistance value data of the conductive film, and combining the multivariate regression algorithm, the quantitative relationship between the microstructure of the conductive film and the thermal conductivity performance is analyzed, and the influence of the microstructure parameters on the thermal conductivity and thermal resistance of the conductive film is revealed. The present invention introduces a time series analysis method to predict the trend of the thermal resistance of the conductive film over time, and combines the thermal resistance aging prediction index to perform aging evaluation. By predicting the law of the change of thermal resistance over time, the present invention can effectively identify the aging process of the conductive film, timely warn of the performance degradation problem, and ensure the reliability and stability of the conductive film under long-term working conditions. When the aging prediction index exceeds the preset threshold, the present invention further combines the support vector machine algorithm to perform thermal area analysis on the temperature distribution data of the conductive film, and can automatically identify the overheating area of ​​the conductive film. Through the abnormal report of the thermal distribution, the potential overheating problem area is located, avoiding the problem of local damage or failure caused by heat accumulation. Finally, the present invention combines the heat flux density distribution data, microstructure parameters and aging prediction index to perform the risk assessment of the performance degradation of the conductive film. Through a comprehensive risk assessment method, the present invention can take timely measures to repair potential problems in a local area of ​​the conductive film, thereby extending the service life of the conductive film and ensuring the stability of the overall function of the circuit board. The present invention can comprehensively evaluate the thermal performance of the conductive film in a working state and its local overheating conditions, thereby improving the accuracy and reliability of quality inspection.

[0123] Reference Figure 2 The second embodiment of the present invention provides a circuit board conductive film quality detection system, comprising:

[0124] A data acquisition module, used to obtain temperature distribution data, microstructure images and thermal resistance values ​​of the conductive film;

[0125] A heat flux density module is used to calculate the heat flux density distribution according to the temperature distribution data to obtain the heat flux density distribution data of different areas of the conductive film;

[0126] A microstructure module, used to perform microstructure analysis of the conductive film according to the microstructure image to obtain microstructure parameters;

[0127] An influence law module is used to analyze the thermal conductivity performance based on a multivariate regression algorithm in combination with the microstructure parameters and the heat flux density distribution data, and obtain the influence law of the microstructure parameters on the thermal conductivity performance of the conductive film;

[0128] A prediction index module, used to perform conductive film aging prediction analysis based on a time series analysis method combined with the influencing law and the thermal resistance value, and obtain an aging prediction index of the conductive film thermal resistance changing with time;

[0129] An abnormality analysis module, used for performing a thermal area analysis on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report when the aging prediction index is greater than a preset index threshold;

[0130] The performance evaluation module is used to perform performance degradation risk evaluation according to the abnormal heat distribution report, the aging prediction index and the microstructure parameter to obtain a performance evaluation result.

[0131] Preferably, the data acquisition module is used to acquire temperature distribution data, microstructure image and thermal resistance value of the conductive film;

[0132] Preferably, the heat flux density module is used to calculate the heat flux density distribution according to the temperature distribution data to obtain the heat flux density distribution data of different areas of the conductive film, including:

[0133] The heat flux density distribution is calculated based on the temperature distribution data to obtain the heat flux density distribution data of different areas of the conductive film, including:

[0134] Based on the finite element analysis method, the conductive film is divided into a plurality of discrete units to obtain the conductive film discrete unit;

[0135] According to the temperature distribution data of the conductive film and the discrete units of the conductive film, a temperature field analysis is performed on the surface of the conductive film to obtain approximate temperature distribution data of the discrete units of the conductive film;

[0136] Based on Fourier's heat conduction law and the approximate temperature distribution data, the heat flux density vector of the discrete unit of the conductive film is calculated to obtain the internal heat flux density data of different regions of the conductive film;

[0137] According to the internal heat flux density data, interpolation and smoothing are performed to obtain a continuous and smooth heat flux density distribution surface;

[0138] According to the gradient corresponding to the heat flux density distribution surface, regional feature extraction is performed to obtain the heat flux direction and heat flux intensity of different regions of the conductive film;

[0139] A cluster analysis is performed according to the heat flow direction and the heat flow intensity to obtain heat flux density distribution data of different areas of the conductive film, wherein the heat flux density distribution data is a division of the heat flux density area of ​​the conductive film, and can distinguish the main channels and key areas of heat transfer inside the conductive film.

[0140] Preferably, the microstructure module is used to perform microstructure analysis of the conductive film according to the microstructure image to obtain microstructure parameters, including:

[0141] The step of performing a microstructure analysis of the conductive film according to the microstructure image to obtain microstructure parameters includes:

[0142] Digitally processing the microstructure image to obtain a grain region on the surface of the conductive film;

[0143] Based on the image segmentation algorithm, the grain area is calculated by pixel scale calibration to obtain the size distribution data of each grain;

[0144] Based on the edge detection algorithm combined with the size distribution data, the grain boundary characteristics are calculated to obtain the grain boundary length, grain boundary orientation angle and total grain boundary length;

[0145] Performing density calculation according to the grain boundary length, the grain boundary direction angle and the total length of the grain boundary to obtain the grain boundary density;

[0146] Based on an image classification algorithm, the defect area of ​​the microstructure image is identified to obtain the defect type and defect density;

[0147] Establishing a correlation with the performance of the conductive film according to the size distribution data, the grain boundary density, the defect type and the defect density, and obtaining microstructure parameters related to the performance of the conductive film;

[0148] Wherein, the grain boundary density is calculated according to the following formula:

[0149]

[0150] in, The grain boundary in a specific direction The grain boundary density on is the total length of the grain boundary, For the The length of the grain boundary, For the Segment grain boundary orientation angle, is the number of grain boundary segments, is the Dirac delta function.

[0151] Preferably, the influence law module is used to perform thermal conductivity performance analysis based on a multivariate regression algorithm in combination with the microstructure parameters and the heat flux density distribution data to obtain the influence law of the microstructure parameters on the thermal conductivity performance of the conductive film, including:

[0152] The heat conduction performance analysis is performed based on the multivariate regression algorithm in combination with the microstructure parameters and the heat flux density distribution data to obtain the influence of the microstructure parameters on the heat conduction performance of the conductive film, including:

[0153] According to the microstructure parameters and the heat flux density distribution data, data cleaning and normalization are performed to obtain a standardized data set;

[0154] According to the standardized data set, a quantitative relationship between the microstructure of the conductive film and the thermal conductivity performance is established and calculated based on a multivariate regression algorithm to obtain a predicted value of the thermal conductivity of the conductive film;

[0155] Based on the predicted value of thermal conductivity, a sensitivity analysis is performed to obtain the influence of the microstructure parameters on the thermal conductivity performance of the conductive film.

[0156] Preferably, the prediction index module is used to perform conductive film aging prediction analysis based on a time series analysis method in combination with the influencing law and the thermal resistance value to obtain an aging prediction index of the conductive film thermal resistance changing with time, including:

[0157] The time series analysis method is combined with the influencing law and the thermal resistance value to perform conductive film aging prediction analysis to obtain an aging prediction index of the conductive film thermal resistance over time, including:

[0158] Based on the time series analysis method and the influence law, the future thermal resistance value of the conductive film is predicted to obtain the predicted thermal resistance value, the predicted time and the predicted end time;

[0159] According to the predicted time, the predicted end time, the thermal resistance value, the predicted thermal resistance value and the preset thermal resistance threshold, a conductive film aging prediction calculation is performed to obtain an aging prediction index of the conductive film thermal resistance changing with time.

[0160] The calculation formula of the aging prediction index is:

[0161]

[0162] In the formula, is the aging prediction index, is the total number of prediction points, For the The predicted thermal resistance value of each prediction point is is the thermal resistance value; is the preset thermal resistance threshold, is the time decay factor, is the predicted end time; For the The prediction time of each prediction point.

[0163] Preferably, the abnormality analysis module is used to perform thermal area analysis on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report when the aging prediction index is greater than a preset index threshold, including:

[0164] When the aging prediction index is greater than a preset index threshold, a thermal region analysis is performed on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report, including:

[0165] When the aging prediction index is greater than a preset index threshold, removing outliers and noise from the temperature distribution data to obtain preprocessed temperature distribution data;

[0166] Extracting temperature distribution features based on the pre-processed temperature distribution data to obtain temperature feature distribution data;

[0167] Based on the support vector machine algorithm combined with the cross-validation method, the temperature characteristic distribution data is subjected to thermal area analysis to obtain the coordinates of the abnormal area;

[0168] Extracting the minimum temperature, maximum temperature and average temperature corresponding to the coordinates of the abnormal area;

[0169] A temperature gradient report generation operation is performed according to the minimum temperature, the maximum temperature and the average temperature to obtain a thermal distribution anomaly report.

[0170] Preferably, the performance evaluation module is used to perform a performance degradation risk evaluation based on the thermal distribution abnormality report, the aging prediction index and the microstructure parameter to obtain a performance evaluation result, and perform a local repair and replacement operation on the conductive film based on the performance evaluation result, including:

[0171] The method of performing a performance degradation risk assessment based on the abnormal heat distribution report, the aging prediction index and the microstructure parameter to obtain a performance assessment result, and performing a local repair and replacement operation on the conductive film based on the performance assessment result, includes:

[0172] According to the abnormal heat distribution report, a temperature analysis is performed on a local area of ​​the conductive film to obtain a first high temperature area that exceeds the temperature limit of the conductive film material;

[0173] According to the aging prediction index and the microstructure parameter, a change analysis of the thermal resistance of the conductive film is performed to obtain a second high temperature region where the heat dissipation efficiency decreases due to an increase in the thermal resistance of the conductive film;

[0174] The conductive films corresponding to the first high-temperature region and the second high-temperature region are partially repaired and replaced.

[0175] It should be noted that a circuit board conductive film quality detection system provided in an embodiment of the present invention is used to execute all process steps of a circuit board conductive film quality detection method in the above embodiment. The working principles and beneficial effects of the two correspond one to one, and thus will not be described in detail.

[0176] The embodiment of the present invention further provides an electronic device. The electronic device includes: a processor, a memory, and a computer program stored in the memory and executable on the processor, such as a calculation program for an aging prediction index. When the processor executes the computer program, the steps in the above-mentioned circuit board conductive film quality detection method embodiments are implemented, such as Figure 1 Alternatively, when the processor executes the computer program, the functions of each module / unit in the above-mentioned device embodiments are implemented, such as the data acquisition module.

[0177] Exemplarily, the computer program may be divided into one or more modules / units, which are stored in the memory and executed by the processor to implement the present invention. The one or more modules / units may be a series of computer program instruction segments capable of implementing specific functions, which are used to describe the execution process of the computer program in the electronic device.

[0178] The electronic device may be a computing device such as a desktop computer, a notebook, a PDA, and a smart tablet. The electronic device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the above components are merely examples of electronic devices and do not constitute a limitation on the electronic device. The electronic device may include more or fewer components than the above components, or may combine certain components, or different components. For example, the electronic device may also include input and output devices, network access devices, buses, etc.

[0179] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the electronic device, and uses various interfaces and lines to connect various parts of the entire electronic device.

[0180] The memory can be used to store the computer program and / or module, and the processor realizes various functions of the electronic device by running or executing the computer program and / or module stored in the memory, and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Media Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0181] Wherein, if the module / unit integrated in the electronic device is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Wherein, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying the computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), electric carrier signal, telecommunication signal and software distribution medium. It should be noted that the content contained in the computer-readable medium can be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer-readable media do not include electric carrier signals and telecommunication signals.

[0182] It should be noted that the device embodiments described above are merely schematic, wherein the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, they may be located in one place, or they may be distributed on multiple network units. Some or all of the modules may be selected according to actual needs to achieve the purpose of the scheme of this embodiment. In addition, in the accompanying drawings of the device embodiments provided by the present invention, the connection relationship between the modules indicates that there is a communication connection between them, which may be specifically implemented as one or more communication buses or signal lines. A person of ordinary skill in the art may understand and implement it without paying any creative effort.

[0183] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.

Claims

1. A method for detecting the quality of a conductive film on a circuit board, characterized in that include: Obtain temperature distribution data, microstructure images and thermal resistance values ​​of conductive films; According to the temperature distribution data, heat flux density distribution calculation is performed to obtain heat flux density distribution data of different areas of the conductive film; According to the microstructure image, a microstructure analysis of the conductive film is performed to obtain microstructure parameters; Based on a multivariate regression algorithm combined with the microstructure parameters and the heat flux density distribution data, a thermal conductivity performance analysis is performed to obtain the influence of the microstructure parameters on the thermal conductivity of the conductive film; Based on the time series analysis method combined with the influencing law and the thermal resistance value, the conductive film aging prediction analysis is performed to obtain the aging prediction index of the conductive film thermal resistance changing with time; When the aging prediction index is greater than a preset index threshold, a thermal region analysis is performed on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report; A performance degradation risk assessment is performed based on the thermal distribution abnormality report, the aging prediction index and the microstructure parameters to obtain a performance assessment result.

2. The method for detecting the quality of the conductive film of a circuit board according to claim 1, characterized in that: The heat flux density distribution is calculated based on the temperature distribution data to obtain the heat flux density distribution data of different areas of the conductive film, including: Based on the finite element analysis method, the conductive film is divided into a plurality of discrete units to obtain the conductive film discrete unit; According to the temperature distribution data of the conductive film and the discrete units of the conductive film, a temperature field analysis is performed on the surface of the conductive film to obtain approximate temperature distribution data of the discrete units of the conductive film; Based on Fourier's heat conduction law and the approximate temperature distribution data, the heat flux density vector of the discrete unit of the conductive film is calculated to obtain the internal heat flux density data of different regions of the conductive film; According to the internal heat flux density data, interpolation and smoothing are performed to obtain a continuous and smooth heat flux density distribution surface; According to the gradient corresponding to the heat flux density distribution surface, regional feature extraction is performed to obtain the heat flux direction and heat flux intensity of different regions of the conductive film; A cluster analysis is performed according to the heat flow direction and the heat flow intensity to obtain heat flux density distribution data of different areas of the conductive film, wherein the heat flux density distribution data is a division of the heat flux density area of ​​the conductive film, and can distinguish the main channels and key areas of heat transfer inside the conductive film.

3. The method for detecting the quality of the conductive film of a circuit board according to claim 1, characterized in that: The step of performing a microstructure analysis of the conductive film according to the microstructure image to obtain microstructure parameters includes: Digitally processing the microstructure image to obtain a grain region on the surface of the conductive film; Based on the image segmentation algorithm, the grain area is calculated by pixel scale calibration to obtain the size distribution data of each grain; Based on the edge detection algorithm combined with the size distribution data, the grain boundary characteristics are calculated to obtain the grain boundary length, grain boundary orientation angle and total grain boundary length; Performing density calculation according to the grain boundary length, the grain boundary direction angle and the total length of the grain boundary to obtain the grain boundary density; Based on an image classification algorithm, the defect area of ​​the microstructure image is identified to obtain the defect type and defect density; Establishing a correlation with the performance of the conductive film according to the size distribution data, the grain boundary density, the defect type and the defect density, and obtaining microstructure parameters related to the performance of the conductive film; Wherein, the grain boundary density is calculated according to the following formula: in, The grain boundary in a specific direction The grain boundary density on is the total length of the grain boundary, For the The length of the grain boundary, For the Segment grain boundary orientation angle, is the number of grain boundary segments, is the Dirac delta function.

4. The method for detecting the quality of the conductive film of a circuit board according to claim 1, characterized in that: The heat conduction performance analysis is performed based on the multivariate regression algorithm in combination with the microstructure parameters and the heat flux density distribution data to obtain the influence of the microstructure parameters on the heat conduction performance of the conductive film, including: According to the microstructure parameters and the heat flux density distribution data, data cleaning and normalization are performed to obtain a standardized data set; According to the standardized data set, a quantitative relationship between the microstructure of the conductive film and the thermal conductivity performance is established and calculated based on a multivariate regression algorithm to obtain a predicted value of the thermal conductivity of the conductive film; Based on the predicted value of thermal conductivity, a sensitivity analysis is performed to obtain the influence of the microstructure parameters on the thermal conductivity performance of the conductive film.

5. The method for detecting the quality of the conductive film of a circuit board according to claim 1, characterized in that: The time series analysis method is combined with the influencing law and the thermal resistance value to perform conductive film aging prediction analysis to obtain an aging prediction index of the conductive film thermal resistance over time, including: Based on the time series analysis method and the influence law, the future thermal resistance value of the conductive film is predicted to obtain the predicted thermal resistance value, the predicted time and the predicted end time; According to the predicted time, the predicted end time, the thermal resistance value, the predicted thermal resistance value and the preset thermal resistance threshold, a conductive film aging prediction calculation is performed to obtain an aging prediction index of the conductive film thermal resistance changing with time.

6. The method for detecting the quality of the conductive film of a circuit board according to claim 5, characterized in that: The calculation formula of the aging prediction index is: In the formula, is the aging prediction index, is the total number of prediction points, For the The predicted thermal resistance value of each prediction point is is the thermal resistance value; is the preset thermal resistance threshold, is the time decay factor, is the predicted end time; For the The prediction time of each prediction point.

7. The method for detecting the quality of the conductive film of a circuit board according to claim 1, characterized in that: When the aging prediction index is greater than a preset index threshold, a thermal region analysis is performed on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report, including: When the aging prediction index is greater than a preset index threshold, removing outliers and noise from the temperature distribution data to obtain preprocessed temperature distribution data; Extracting temperature distribution features based on the pre-processed temperature distribution data to obtain temperature feature distribution data; Based on the support vector machine algorithm combined with the cross-validation method, the temperature characteristic distribution data is subjected to thermal area analysis to obtain the coordinates of the abnormal area; Extracting the minimum temperature, maximum temperature and average temperature corresponding to the coordinates of the abnormal area; A temperature gradient report generation operation is performed according to the minimum temperature, the maximum temperature and the average temperature to obtain a thermal distribution anomaly report.

8. The method for detecting the quality of the conductive film of a circuit board according to claim 1, characterized in that: The method of performing a performance degradation risk assessment based on the abnormal heat distribution report, the aging prediction index and the microstructure parameter to obtain a performance assessment result, and performing a local repair and replacement operation on the conductive film based on the performance assessment result, includes: According to the abnormal heat distribution report, a temperature analysis is performed on a local area of ​​the conductive film to obtain a first high temperature area that exceeds the temperature limit of the conductive film material; According to the aging prediction index and the microstructure parameter, a change analysis of the thermal resistance of the conductive film is performed to obtain a second high temperature region where the heat dissipation efficiency decreases due to an increase in the thermal resistance of the conductive film; The conductive films corresponding to the first high-temperature region and the second high-temperature region are partially repaired and replaced.

9. A circuit board conductive film quality detection system, characterized in that: include: A data acquisition module, used to obtain temperature distribution data, microstructure images and thermal resistance values ​​of the conductive film; A heat flux density module is used to calculate the heat flux density distribution according to the temperature distribution data to obtain the heat flux density distribution data of different areas of the conductive film; A microstructure module, used to perform microstructure analysis of the conductive film according to the microstructure image to obtain microstructure parameters; An influence law module is used to analyze the thermal conductivity performance based on a multivariate regression algorithm in combination with the microstructure parameters and the heat flux density distribution data, and obtain the influence law of the microstructure parameters on the thermal conductivity performance of the conductive film; A prediction index module, used to perform conductive film aging prediction analysis based on a time series analysis method combined with the influencing law and the thermal resistance value, and obtain an aging prediction index of the conductive film thermal resistance changing with time; An abnormality analysis module, used for performing a thermal area analysis on the temperature distribution data based on a support vector machine algorithm to obtain a thermal distribution abnormality report when the aging prediction index is greater than a preset index threshold; The performance evaluation module is used to perform performance degradation risk evaluation according to the abnormal heat distribution report, the aging prediction index and the microstructure parameter to obtain a performance evaluation result.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is executed, the device where the computer-readable storage medium is located is controlled to execute the method for detecting the quality of the conductive film of the circuit board according to any one of claims 1 to 8.

Images