Process optimization method and system for realizing bonding of silver wires

Abstract

The invention relates to the technical field of silver wire bonding, and discloses a process optimization method and system for realizing silver wire bonding, and the method comprises the steps: analyzing the material bonding attribute of a silver wire preparation material in the preparation process; calculating a loss coefficient corresponding to the silver wire bonding equipment, and evaluating the bonding performance corresponding to the silver wire bonding equipment; collecting a silver wire working condition parameter corresponding to the bonding silver wire in each process of the bonding process, and calculating the bonding precision corresponding to the bonding silver wire; carrying out product sampling treatment on the bonding silver wire in each process of the bonding process to obtain a process bonding sample, carrying out quality evaluation on the process bonding sample to obtain multi-dimensional quality data, and evaluating the process bonding quality corresponding to the bonding process; and in combination with material bonding properties, bonding precision, bonding performance and process bonding quality, carrying out process optimization treatment on the bonded silver wire to obtain an optimization result. The invention mainly aims to solve the problem of low optimization efficiency of a bonding silver wire process.

Description

Technical Field

[0001] The invention relates to a process optimization method and system for realizing silver wire bonding, and belongs to the technical field of silver wire bonding. Background Art

[0002] Silver bonding wire is widely used in the electronics industry and other fields, and its quality is crucial to the performance and reliability of electronic equipment. With the rapid development of electronic technology, the quality requirements for silver bonding wire are increasing day by day, and the market demand for silver bonding wire is also increasing. In order to meet the needs of high-quality and large-scale production, the optimization of the production process of silver bonding wire has become the key.

[0003] At present, the process optimization of silver bonding wire is mainly based on traditional process control methods, mainly the doping element adjustment optimization method, which adds various doping elements such as palladium, nickel, cerium, calcium, aluminum, chromium, tin, etc. to the silver-based material in specific proportions to form a diversified solid solution structure, thereby improving the mechanical properties, processing properties and chemical stability of the silver bonding wire. However, this method requires a lot of research and experiments to determine the best types and ratios of doping elements, with high R&D costs and long cycles. In addition, if the proportion of doping elements is not properly controlled, the performance of the silver bonding wire may be unstable or even quality problems may occur. For example, if too much nickel is added, the toughness of the silver bonding wire may be reduced, affecting its reliability in the wire bonding process, and further affecting the overall quality and production efficiency of the silver bonding wire, resulting in low efficiency in the process optimization of the silver bonding wire. Summary of the invention

[0004] The present invention provides a method and system for realizing process optimization of silver wire bonding, the main purpose of which is to solve the problem of low efficiency of process optimization of silver wire bonding.

[0005] To achieve the above object, the present invention provides a process optimization method for bonding silver wires, comprising:

[0006] Acquire a silver wire preparation material corresponding to the bonding silver wire, collect material property data corresponding to the silver wire preparation material, and analyze the material bonding properties of the silver wire preparation material during the preparation process based on the material property data;

[0007] Querying the silver wire bonding equipment and bonding process in the preparation process of the bonding silver wire, collecting the equipment parameter set corresponding to the silver wire bonding equipment, and taking the equipment tool image corresponding to the silver wire bonding equipment, calculating the loss coefficient corresponding to the silver wire bonding equipment based on the equipment tool image, and evaluating the bonding performance corresponding to the silver wire bonding equipment in combination with the equipment parameter set and the loss coefficient;

[0008] Collecting silver wire working condition parameters corresponding to the bonding silver wire in each process of the bonding process, and calculating the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters;

[0009] Performing product sampling processing on the bonding silver wires in each process of the bonding process to obtain process bonding samples, performing quality evaluation on the process bonding samples to obtain multi-dimensional quality data, and evaluating the process bonding quality corresponding to the bonding process according to the multi-dimensional quality data;

[0010] In combination with the material bonding properties, the bonding precision, the bonding performance and the process bonding quality, a process optimization process for the bonding silver wire is performed to obtain an optimization result.

[0011] Optionally, analyzing the material bonding properties of the silver wire preparation material during the preparation process based on the material property data includes:

[0012] Performing data filtering processing on the material attribute data to obtain target attribute data;

[0013] Analyzing the attribute categories corresponding to the target attribute data, and extracting key attribute categories from the attribute categories;

[0014] Based on the key attribute categories, performing data clustering processing on the target attribute data to obtain key attribute data;

[0015] Analyzing the attribute correlation between the key attribute data;

[0016] In combination with the attribute correlation and the key attribute data, the material bonding properties of the silver wire preparation material during the preparation process are analyzed.

[0017] Optionally, analyzing the attribute correlation between the key attribute data includes:

[0018] Sorting the key attribute data to obtain a data rank;

[0019] According to the data rank, the data level difference between the key attribute data is calculated, and the number of data corresponding to the key attribute data is counted;

[0020] Combining the data level difference and the data quantity, the data correlation coefficient between the key attribute data can be calculated by the following formula:

[0021]

[0022] Among them, A represents the data correlation coefficient, B aIt represents the data level difference corresponding to the a-th data in the key attribute data, a represents the data sequence number corresponding to the key attribute data, and q represents the number of data;

[0023] The attribute correlation between the key attribute data is analyzed according to the data correlation coefficient.

[0024] Optionally, calculating the loss coefficient corresponding to the silver wire bonding device based on the device tool image includes:

[0025] Extracting tool geometric features and tool visual features from the equipment tool image;

[0026] Querying a standard device image corresponding to the silver wire bonding device, and extracting standard geometric features and standard visual features corresponding to the device tool from the standard device image;

[0027] Calculating the geometric deviation between the tool geometric feature and the standard geometric feature;

[0028] Calculating the visual deviation between the tool visual feature and the standard visual feature;

[0029] The loss coefficient corresponding to the silver wire bonding device is calculated by combining the geometric deviation and the visual deviation.

[0030] Optionally, extracting the tool geometric features and tool visual features in the equipment tool image includes:

[0031] Performing image noise reduction processing on the equipment tool image to obtain a noise-reduced tool image;

[0032] Performing a main body extraction process on the noise reduction tool image to obtain a tool main body image;

[0033] Detecting a tool contour in the tool body image, performing feature extraction on the tool contour, and obtaining tool geometric features in the tool body image;

[0034] The visual features of the tool body image are extracted using a preset visual feature extraction model to obtain the tool visual features.

[0035] Optionally, the calculating the geometric deviation between the tool geometric feature and the standard geometric feature comprises:

[0036] Performing feature quantification processing on the tool geometric feature and the standard geometric feature respectively to obtain tool geometric feature value and standard geometric feature value;

[0037] Calculating geometric feature weights corresponding to the tool geometric features;

[0038] Combining the geometric feature weight, the tool geometric feature value and the standard geometric feature value, the geometric deviation between the tool geometric feature and the standard geometric feature can be calculated by the following formula:

[0039]

[0040] Where D represents the geometric deviation between the tool geometric features and the standard geometric features, β b The geometric feature weight corresponding to the bth feature of the tool geometric feature, E b The tool geometric feature value corresponding to the bth feature of the tool geometric feature, F b It represents the standard geometric feature value corresponding to the bth feature of the tool geometric feature in the standard geometric feature, b represents the feature serial number corresponding to the tool geometric feature, and t represents the number of features corresponding to the tool geometric feature.

[0041] Optionally, the calculating the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameter includes:

[0042] Performing parameter calibration processing on the silver wire working condition parameters to obtain calibration working condition parameters;

[0043] Standardizing the calibration working condition parameters to obtain standard working condition parameters, and identifying working condition indicators corresponding to the silver wire working condition parameters;

[0044] The precision contribution coefficient corresponding to the working condition index is calculated, and the bonding precision corresponding to the bonding silver wire is calculated by combining the standard working condition parameters and the precision contribution coefficient.

[0045] Optionally, evaluating the process bonding quality corresponding to the bonding process according to the multidimensional quality data includes:

[0046] Identify a data identifier corresponding to the multidimensional quality data, and analyze the quality attribute dimension corresponding to the multidimensional quality data based on the data identifier;

[0047] Based on the multidimensional quality data, calculating the dimensional interaction strength between the quality attribute dimensions;

[0048] Combining the dimension interaction strength and the quality attribute dimension, clustering the multidimensional quality data to obtain clustered quality data;

[0049] Based on the clustering quality data, calculating a dimension quality score corresponding to the quality attribute dimension;

[0050] Combined with the dimension quality scores, the process bonding quality corresponding to the bonding process is evaluated.

[0051] Optionally, the calculating, based on the multidimensional quality data, the dimensional interaction strength between the quality attribute dimensions includes:

[0052] Extracting data quantization values ​​corresponding to the multidimensional quality data, and calculating average measurement values ​​corresponding to the quality attribute dimensions according to the data quantization values;

[0053] Calculate the dimension discrete value corresponding to the quality attribute dimension by combining the average metric value and the data quantization value;

[0054] Combining the data quantization value, the average metric value and the dimension discrete value, the dimensional interaction intensity between the quality attribute dimensions can be calculated by the following formula:

[0055]

[0056] Among them, G represents the dimensional interaction strength between quality attribute dimensions, Q i Represents the data quantification value of the i-th data in the Q-th dimension of the quality attribute dimension, represents the average measurement value of the Qth dimension in the quality attribute dimension, M i The data quantification value of the i-th data in the M-th dimension of the quality attribute dimension, represents the average measurement value of the Mth dimension in the quality attribute dimension, i represents the data sequence number of the multidimensional quality data, p represents the number of multidimensional quality data corresponding to the quality attribute dimension, δ Q and δ M They respectively represent the dimensional discrete values ​​corresponding to the Qth dimension and the Mth dimension in the quality attribute dimension.

[0057] A process optimization system for realizing silver wire bonding, characterized in that the system comprises:

[0058] A bonding property analysis module, used to obtain the silver wire preparation material corresponding to the bonding silver wire, collect material property data corresponding to the silver wire preparation material, and analyze the material bonding properties of the silver wire preparation material during the preparation process based on the material property data;

[0059] A bonding performance evaluation module, used to query the silver wire bonding equipment and bonding process in the preparation process of the bonding silver wire, collect the equipment parameter set corresponding to the silver wire bonding equipment, and take the equipment tool image corresponding to the silver wire bonding equipment, calculate the loss coefficient corresponding to the silver wire bonding equipment based on the equipment tool image, and evaluate the bonding performance corresponding to the silver wire bonding equipment in combination with the equipment parameter set and the loss coefficient;

[0060] A bonding precision calculation module, used for collecting silver wire working condition parameters corresponding to the bonding silver wire in each process of the bonding process, and calculating the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters;

[0061] A process bonding quality evaluation module is used to perform product sampling processing on the bonding silver wire in each process of the bonding process to obtain process bonding samples, perform quality evaluation on the process bonding samples to obtain multi-dimensional quality data, and evaluate the process bonding quality corresponding to the bonding process according to the multi-dimensional quality data;

[0062] The process optimization module is used to perform process optimization processing on the bonding silver wire in combination with the material bonding properties, the bonding precision, the bonding performance and the process bonding quality to obtain an optimization result.

[0063] Compared with the problems described in the background technology, the present invention analyzes the material bonding properties of the silver wire preparation material during the preparation process based on the material property data, so as to understand the compatibility of the silver wire preparation material during preparation and avoid the formation of other metal compounds during the bonding process, thereby affecting the bonding quality. The present invention calculates the loss coefficient corresponding to the silver wire bonding equipment based on the equipment tool image, and the loss degree corresponding to the silver wire bonding equipment can be understood through the loss coefficient, thereby facilitating the subsequent improvement of the evaluation of the bonding performance corresponding to the silver wire bonding equipment. The present invention calculates the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters, and the bonding precision corresponding to the bonding silver wire can be understood through the bonding precision. Quality, thereby providing a basis for the formulation of the subsequent process optimization strategy corresponding to the bonding silver wire. The present invention can obtain the production samples of the bonding silver wire by performing product sampling processing on the bonding silver wire in each process of the bonding process. By performing quality evaluation on the bonding samples of the process, multiple aspects of quality test data of the bonding samples of the process can be obtained, thereby improving the evaluation accuracy of the process bonding quality corresponding to the subsequent bonding process. The present invention combines the bonding properties of the material, the bonding precision, the bonding performance and the bonding quality of the process to perform process optimization processing on the bonding silver wire, which can improve the efficiency of the process optimization of the bonding silver wire and thereby improve the subsequent processing quality of the bonding silver wire. Therefore, the present invention proposes a method and system for realizing process optimization of bonding silver wires, thereby improving the efficiency of process optimization of bonding silver wires. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] Figure 1 A schematic flow chart of a process optimization method for implementing silver wire bonding provided by an embodiment of the present invention;

[0065] Figure 2A functional module diagram of a process optimization system for bonding silver wires provided in accordance with an embodiment of the present invention.

[0066] The purpose, features and advantages of the present invention will be further described with reference to the accompanying drawings in conjunction with the embodiments. DETAILED DESCRIPTION

[0067] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0068] The embodiment of the present application provides a process optimization method for bonding silver wires. The execution subject of the process optimization method for bonding silver wires includes but is not limited to at least one of the electronic devices such as a server and a terminal that can be configured to execute the method provided by the embodiment of the present application. In other words, the process optimization method for bonding silver wires can be executed by software or hardware installed in a terminal device or a server device. The server includes but is not limited to: a single server, a server cluster, a cloud server or a cloud server cluster, etc.

[0069] Embodiment 1:

[0070] Reference Figure 1 FIG. 1 is a flow chart of a process optimization method for realizing silver wire bonding provided by an embodiment of the present invention. In this embodiment, the process optimization method for realizing silver wire bonding includes:

[0071] S1. Obtain a silver wire preparation material corresponding to the bonding silver wire, collect material property data corresponding to the silver wire preparation material, and analyze the material bonding properties of the silver wire preparation material during the preparation process based on the material property data.

[0072] The present invention analyzes the material bonding properties of the silver wire preparation material during the preparation process based on the material property data, so as to understand the compatibility of the silver wire preparation material during preparation and avoid the formation of other metal compounds during the bonding process, thereby affecting the bonding quality. It should be explained that the silver wire preparation material is the raw material for processing the bonding silver wire, and the material property data is the description data of various properties corresponding to the silver wire preparation material, and the material bonding property is the property and behavior performance of the silver wire preparation material during the preparation process; further, the collection of the material property data corresponding to the silver wire preparation material can be achieved through collection tools, such as a precision electronic balance for measuring the mass of the silver wire preparation material to determine the content of its components, and a particle size analyzer for analyzing the particle size distribution of the material to understand the size of the particles.

[0073] In detail, the analyzing the material bonding properties of the silver wire preparation material during the preparation process based on the material property data includes:

[0074] Performing data filtering processing on the material attribute data to obtain target attribute data;

[0075] Analyzing the attribute categories corresponding to the target attribute data, and extracting key attribute categories from the attribute categories;

[0076] Based on the key attribute categories, performing data clustering processing on the target attribute data to obtain key attribute data;

[0077] Analyzing the attribute correlation between the key attribute data;

[0078] In combination with the attribute correlation and the key attribute data, the material bonding properties of the silver wire preparation material during the preparation process are analyzed.

[0079] It should be explained that the target attribute data is a specific data subset obtained after screening the material attribute data; the attribute category is the result of classifying the target attribute data based on certain rules; the key attribute category is a category selected from the attribute category that has an important impact on the material bonding property; the attribute correlation represents the mutual correlation relationship between the key attribute data.

[0080] Furthermore, the data filtering processing of the material attribute data can be achieved by setting reasonable data screening rules and thresholds; the attribute categories corresponding to the target attribute data can be analyzed through data analysis algorithms and professional knowledge, and the categories that have a key impact on the material bonding attributes can be extracted as key attribute categories; based on the key attribute categories, the data clustering processing of the target attribute data can be achieved through a clustering algorithm; in combination with the attribute correlation and the key attribute data, the material bonding properties of the silver wire preparation material during the preparation process are analyzed. For example, from the perspective of the chemical composition in the key attribute data, if the purity of silver is high and the impurity content is low, Combined with its correlation with electrical properties, it can be seen that it may have better conductivity during the bonding process, which is beneficial to the transmission of electrons at the bonding interface, thereby improving the bonding effect. Considering the key attribute data of hardness, if the hardness is moderate and is correlated with the deformation ability, it can deform moderately under the bonding pressure, better fill the tiny gaps at the bonding interface, and enhance the mechanical bonding effect. According to the key attribute data related to the crystal structure, such as the face-centered cubic structure and its correlation with atomic diffusion, there may be channels in specific directions that are more conducive to atomic diffusion, which promotes the diffusion of atoms in the metallurgical bonding process, improves the bonding quality, and then analyzes the material bonding properties.

[0081] Further, as an optional embodiment of the present invention, the analyzing the attribute correlation between the key attribute data includes:

[0082] Sorting the key attribute data to obtain a data rank;

[0083] According to the data rank, the data level difference between the key attribute data is calculated, and the number of data corresponding to the key attribute data is counted;

[0084] Combining the data level difference and the data quantity, the data correlation coefficient between the key attribute data can be calculated by the following formula:

[0085]

[0086] Among them, A represents the data correlation coefficient, B a It represents the data level difference corresponding to the a-th data in the key attribute data, a represents the data sequence number corresponding to the key attribute data, and q represents the number of data;

[0087] The attribute correlation between the key attribute data is analyzed according to the data correlation coefficient.

[0088] It should be explained that the data rank is the data sorting of the key attribute data, the data level difference is the difference in the corresponding data ranks between the key attribute data, and the data correlation coefficient represents the degree of monotonic correlation between the key attribute data.

[0089] Furthermore, the sorting process of the key attribute data can be sorted according to the characteristics of the data, such as nominal data, for example, different material categories (such as pure silver, silver alloy A, silver alloy B, etc.), can be sorted in a custom order. This order is usually determined based on research purposes or professional knowledge. If it is text data, it can be sorted according to the corresponding semantic meaning, which may require professional knowledge and a deep understanding of the meaning of the data. For example, for the text data "Rough", "Smooth", and "Polished" that describe the surface state of the material, it can be sorted in order from high to low surface roughness, that is, the rank of "Rough" is 1, "Smooth" is 2, and "Polished" is 3.

[0090] S2. Query the silver wire bonding equipment and bonding process during the preparation of the bonding silver wire, collect the equipment parameter set corresponding to the silver wire bonding equipment, and take the equipment tool image corresponding to the silver wire bonding equipment, calculate the loss coefficient corresponding to the silver wire bonding equipment based on the equipment tool image, and evaluate the bonding performance corresponding to the silver wire bonding equipment in combination with the equipment parameter set and the loss coefficient.

[0091] The present invention calculates the loss coefficient corresponding to the silver wire bonding equipment based on the equipment tool image, and the loss coefficient can be used to understand the loss degree corresponding to the silver wire bonding equipment, thereby facilitating the subsequent improvement of the evaluation of the bonding performance corresponding to the silver wire bonding equipment. It should be explained that the silver wire bonding equipment is the equipment used in the preparation process of the bonding silver wire, the bonding process is the operation process of the bonding silver wire in the preparation process, the equipment parameter set is the various indicator information of the equipment corresponding to the silver wire bonding equipment, the equipment tool image is the image corresponding to the silver wire bonding equipment, and the loss coefficient represents the loss degree corresponding to the silver wire bonding equipment. Furthermore, the query of the silver wire bonding equipment and the bonding process in the preparation process of the bonding silver wire can be achieved through the enterprise production management system, the collection of the equipment parameter set corresponding to the silver wire bonding equipment can be achieved through the equipment control system, and the shooting of the equipment tool image corresponding to the silver wire bonding equipment can be achieved through an industrial camera.

[0092] Specifically, the loss coefficient corresponding to the silver wire bonding device is calculated based on the device tool image, including:

[0093] Extracting tool geometric features and tool visual features from the equipment tool image;

[0094] Querying a standard device image corresponding to the silver wire bonding device, and extracting standard geometric features and standard visual features corresponding to the device tool from the standard device image;

[0095] Calculating the geometric deviation between the tool geometric feature and the standard geometric feature;

[0096] Calculating the visual deviation between the tool visual feature and the standard visual feature;

[0097] The loss coefficient corresponding to the silver wire bonding device is calculated by combining the geometric deviation and the visual deviation.

[0098] It should be explained that the tool geometric features and the tool visual features are respectively the geometric shape related features and visual appearance related features of the tool body image; the standard equipment image is the image of the silver wire bonding equipment in the standard state; the standard geometric features and the standard visual features are respectively the geometric shape features and visual appearance features corresponding to the equipment tools extracted from the standard equipment image; the geometric deviation degree indicates the degree of difference between the tool geometric features and the standard geometric features; the visual deviation degree indicates the degree of difference between the tool visual features and the standard visual features.

[0099] Furthermore, the standard equipment image corresponding to the silver wire bonding equipment can be obtained by querying from the official website of the equipment manufacturer through human-computer interaction; the standard geometric features and standard visual features corresponding to the equipment tools can be extracted from the standard equipment image through the above-mentioned convolutional neural network; the calculation principle of the visual deviation is the same as that of the geometric deviation, which will not be elaborated here; the geometric deviation and the visual deviation are weightedly summed to obtain the loss coefficient corresponding to the silver wire bonding equipment, and the corresponding weight can be allocated through the hierarchical analysis method.

[0100] Further, as an optional embodiment of the present invention, the extracting the tool geometric features and tool visual features in the equipment tool image includes:

[0101] Performing image noise reduction processing on the equipment tool image to obtain a noise-reduced tool image;

[0102] Performing a main body extraction process on the noise reduction tool image to obtain a tool main body image;

[0103] Detecting a tool contour in the tool body image, performing feature extraction on the tool contour, and obtaining tool geometric features in the tool body image;

[0104] The visual features of the tool body image are extracted using a preset visual feature extraction model to obtain the tool visual features.

[0105] It should be explained that the denoised tool image is an image obtained after denoising the equipment tool image; the tool body image is an image of the equipment tool body part extracted from the denoised tool image; the tool contour is the shape and appearance corresponding to the tool in the tool body image; the visual feature extraction model is a model for extracting features corresponding to image color and texture, such as the VGG (Visual Geometry Group) model.

[0106] Furthermore, the image denoising processing of the equipment tool image can be achieved by a low-pass filter; the main body extraction processing of the denoised tool image can be achieved by a threshold segmentation method; the tool contour in the tool main body image can be detected by a Sobel operator; the tool contour can be feature extracted by a Fourier descriptor to obtain the tool geometric features in the tool main body image, such as converting the tool contour to the frequency domain and obtaining a series of coefficients through Fourier transform, which constitute a Fourier descriptor. The Fourier descriptor can reflect the shape information of the contour and has rotation, translation and scale invariance.

[0107] Further, as an optional embodiment of the present invention, the calculating the geometric deviation between the tool geometric feature and the standard geometric feature includes:

[0108] Performing feature quantification processing on the tool geometric feature and the standard geometric feature respectively to obtain tool geometric feature value and standard geometric feature value;

[0109] Calculating geometric feature weights corresponding to the tool geometric features;

[0110] Combining the geometric feature weight, the tool geometric feature value and the standard geometric feature value, the geometric deviation between the tool geometric feature and the standard geometric feature can be calculated by the following formula:

[0111]

[0112] Where D represents the geometric deviation between the tool geometric features and the standard geometric features, β b The geometric feature weight corresponding to the bth feature of the tool geometric feature, E b The tool geometric feature value corresponding to the bth feature of the tool geometric feature, F b It represents the standard geometric feature value corresponding to the bth feature of the tool geometric feature in the standard geometric feature, b represents the feature serial number corresponding to the tool geometric feature, and t represents the number of features corresponding to the tool geometric feature.

[0113] It should be explained that the tool geometric feature value and the standard geometric feature value are respectively the expression values ​​corresponding to the tool geometric features and the standard geometric features, and the geometric feature weight represents the importance of the tool geometric features. Furthermore, the feature quantization processing of the tool geometric features and the standard geometric features can be achieved through feature encoding, such as one-hot encoding; the calculation of the geometric feature weight corresponding to the tool geometric features can be achieved through the entropy weight method.

[0114] The present invention evaluates the bonding performance corresponding to the silver wire bonding device by combining the device parameter set and the loss coefficient, so as to understand the quality of the bonding work of the silver wire bonding device in the current state. It should be explained that the bonding performance is an expression of the quality of the bonding work corresponding to the silver wire bonding device. Furthermore, the parameter difference between the parameters in the device parameter set and the normal parameters is calculated, and the bonding performance corresponding to the silver wire bonding device is evaluated according to the parameter difference and the loss coefficient. If the parameter difference is less than the preset threshold and the loss coefficient is between 0.1, it indicates that the bonding performance is good.

[0115] S3. Collecting the silver wire working condition parameters corresponding to the bonding silver wire in each process of the bonding process, and calculating the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters.

[0116] The present invention calculates the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters, and the bonding quality corresponding to the bonding silver wire can be understood through the bonding precision, thereby providing a basis for the subsequent formulation of the process optimization strategy corresponding to the bonding silver wire. It should be explained that the silver wire working condition parameters are parameters related to the working state corresponding to the bonding silver wire in each process of the bonding process, and the bonding precision indicates the accuracy of the performance indicators corresponding to the bonding silver wire. Furthermore, the collection of the silver wire working condition parameters corresponding to the bonding silver wire in each process of the bonding process can be achieved through sensors, such as pressure sensors, temperature sensors and displacement sensors.

[0117] In detail, the calculating the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters includes:

[0118] Performing parameter calibration processing on the silver wire working condition parameters to obtain calibration working condition parameters;

[0119] Standardizing the calibration working condition parameters to obtain standard working condition parameters, and identifying working condition indicators corresponding to the silver wire working condition parameters;

[0120] The precision contribution coefficient corresponding to the working condition index is calculated, and the bonding precision corresponding to the bonding silver wire is calculated by combining the standard working condition parameters and the precision contribution coefficient.

[0121] It should be explained that the calibration operating condition parameters are parameters obtained after calibration of the erroneous parameters or abnormal parameters in the silver wire operating condition parameters, the standard operating condition parameters are parameters obtained after eliminating the differences between the calibration operating condition parameters, the operating condition indicators are the operating condition types corresponding to the silver wire operating condition parameters, such as pressure-related indicators and deformation indicators, etc., and the precision contribution coefficient represents the degree of influence of the operating condition indicators on the precision performance.

[0122] Furthermore, the parameter calibration processing of the silver wire working condition parameters can be achieved through an adaptive control algorithm; the standardization processing of the calibration working condition parameters can be achieved through a Z-score standardization method; the identification of the working condition indicators corresponding to the silver wire working condition parameters can be achieved through an OCR identification method; the precision contribution coefficient corresponding to the working condition indicator can be calculated by combining the bonding production history data and using a single factor experimental method; the standard working condition parameters and the corresponding precision contribution coefficient are multiplied and summed to obtain the bonding precision corresponding to the bonding silver wire.

[0123] S4. Sampling the bonding silver wires in each bonding process to obtain process bonding samples, performing quality assessment on the process bonding samples to obtain multi-dimensional quality data, and assessing the process bonding quality corresponding to the bonding process based on the multi-dimensional quality data.

[0124] The present invention can obtain production samples of the bonding silver wires by performing product sampling processing on the bonding silver wires in each process of the bonding process, and can obtain multiple aspects of quality test data of the process bonding samples by performing quality evaluation on the process bonding samples, thereby improving the evaluation accuracy of the process bonding quality corresponding to the subsequent bonding process. It should be explained that the process bonding samples are samples of the bonding silver wires in mass production in each process of the bonding process, and the multi-dimensional quality data are quality test record data obtained after a series of tests on the process bonding samples, and the number of data of the quality attribute dimensions corresponding to different dimensions is the same, which is convenient for subsequent related processing and analysis. Furthermore, the product sampling processing of the bonding silver wires in each process of the bonding process can be achieved by a random sampling method; the quality evaluation of the process bonding samples can be achieved by professional testing instruments and equipment, such as tensile testing machines, hardness testers and scanning electron microscopes.

[0125] The present invention evaluates the process bonding quality corresponding to the bonding process based on the multi-dimensional quality data, so as to understand the overall quality level of the bonding silver wire processed by the bonding process, thereby facilitating the subsequent formulation of the process optimization strategy corresponding to the bonding silver wire. It should be explained that the process bonding quality represents the processing quality corresponding to the bonding process.

[0126] In detail, the step of evaluating the bonding quality of the bonding process corresponding to the bonding process according to the multidimensional quality data includes:

[0127] Identify a data identifier corresponding to the multidimensional quality data, and analyze the quality attribute dimension corresponding to the multidimensional quality data based on the data identifier;

[0128] Based on the multidimensional quality data, calculating the dimensional interaction strength between the quality attribute dimensions;

[0129] Combining the dimension interaction strength and the quality attribute dimension, clustering the multidimensional quality data to obtain clustered quality data;

[0130] Based on the clustering quality data, calculating a dimension quality score corresponding to the quality attribute dimension;

[0131] Combined with the dimension quality scores, the process bonding quality corresponding to the bonding process is evaluated.

[0132] It should be explained that the data identifier is the label introduction corresponding to the multidimensional quality data, the quality attribute dimension is the quality feature category divided from different properties (such as mechanics, microstructure, electricity, etc.) corresponding to the multidimensional quality data; the dimension interaction strength represents the degree of mutual correlation between the quality attribute dimensions; the clustering quality data is the data obtained after clustering the multidimensional quality data in combination with the dimension interaction strength and the quality attribute dimension; the dimension quality score represents the quantitative score of the quality performance corresponding to the quality attribute dimension.

[0133] Furthermore, the identification of the data identifier corresponding to the multidimensional quality data can be achieved through the above-mentioned OCR recognition method; extracting the key characters in the data identifier, and analyzing the quality attribute dimension corresponding to the multidimensional quality data according to the key characters; clustering the multidimensional quality data in combination with the dimension interaction intensity and the quality attribute dimension to obtain clustered quality data; converting the clustered quality data into corresponding scores according to a preset score conversion rule, summing the scores, and obtaining a dimension quality score corresponding to the quality attribute dimension, and the preset score conversion rule can be set according to expert experience; evaluating the process bonding quality corresponding to the bonding process in combination with the dimension quality score and a preset quality grade standard, if the dimension quality score is in a high grade interval, the process bonding quality is high.

[0134] Further, as an optional embodiment of the present invention, the calculating the dimensional interaction strength between the quality attribute dimensions based on the multidimensional quality data includes:

[0135] Extracting data quantization values ​​corresponding to the multidimensional quality data, and calculating average measurement values ​​corresponding to the quality attribute dimensions according to the data quantization values;

[0136] Calculate the dimension discrete value corresponding to the quality attribute dimension by combining the average metric value and the data quantization value;

[0137] Combining the data quantization value, the average metric value and the dimension discrete value, the dimensional interaction intensity between the quality attribute dimensions can be calculated by the following formula:

[0138]

[0139] Among them, G represents the dimensional interaction strength between quality attribute dimensions, Q i Represents the data quantification value of the i-th data in the Q-th dimension of the quality attribute dimension, represents the average measurement value of the Qth dimension in the quality attribute dimension, M iThe data quantification value of the i-th data in the M-th dimension of the quality attribute dimension, represents the average measurement value of the Mth dimension in the quality attribute dimension, i represents the data sequence number of the multidimensional quality data, p represents the number of multidimensional quality data corresponding to the quality attribute dimension, δ Q and δ M They respectively represent the dimensional discrete values ​​corresponding to the Qth dimension and the Mth dimension in the quality attribute dimension.

[0140] Furthermore, the extraction of data quantization values ​​corresponding to the multidimensional quality data can be achieved through an extraction function, which is compiled by a scripting language; the average measurement value corresponding to the quality attribute dimension can be achieved through an average function; combining the average measurement value and the data quantization value, the corresponding dimensional discrete value can be obtained by calculating the standard deviation corresponding to the quality attribute dimension.

[0141] S5. In combination with the material bonding properties, the bonding precision, the bonding performance and the process bonding quality, process optimization processing is performed on the bonding silver wire to obtain an optimization result.

[0142] The present invention performs process optimization processing on the bonding silver wire by combining the bonding properties of the material, the bonding precision, the bonding performance and the bonding quality of the process, so as to improve the efficiency of the process optimization of the bonding silver wire, and thus improve the subsequent processing quality of the bonding silver wire. Further, the process optimization processing of the bonding silver wire is performed in combination with the bonding properties of the material, the bonding precision, the bonding performance and the bonding quality of the process to obtain an optimization result. If the bonding property of the material is general, the bonding precision is medium, the bonding performance is good, and the bonding quality of the process is medium, then the corresponding optimization result is: according to the material characteristics, the process parameters are adjusted, and the pressure is appropriately reduced to reduce the poor bonding situation that may be caused by high hardness; for the precision problem, the calibration procedure of the bonding equipment is optimized to reduce the deviation of the position, size and angle of the bonding point; in terms of bonding performance, special treatment of the surface of the bonding silver wire is considered to improve its stability in a complex environment, so as to improve the overall process bonding quality, and then realize process optimization processing. The setting of specific optimization parameters can be adjusted in combination with the actual application scenario.

[0143] Compared with the problems described in the background technology, the present invention analyzes the material bonding properties of the silver wire preparation material during the preparation process based on the material property data, so as to understand the compatibility of the silver wire preparation material during preparation and avoid the formation of other metal compounds during the bonding process, thereby affecting the bonding quality. The present invention calculates the loss coefficient corresponding to the silver wire bonding equipment based on the equipment tool image, and the loss degree corresponding to the silver wire bonding equipment can be understood through the loss coefficient, thereby facilitating the subsequent improvement of the evaluation of the bonding performance corresponding to the silver wire bonding equipment. The present invention calculates the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters, and the bonding precision corresponding to the bonding silver wire can be understood through the bonding precision. Quality, and thus provide a basis for the formulation of the subsequent process optimization strategy corresponding to the bonding silver wire. The present invention can obtain the production sample of the bonding silver wire by performing product sampling processing on the bonding silver wire in each process of the bonding process. By performing quality evaluation on the bonding samples of the process, multiple aspects of quality test data of the bonding samples of the process can be obtained, thereby improving the evaluation accuracy of the process bonding quality corresponding to the subsequent bonding process. The present invention combines the bonding properties of the material, the bonding precision, the bonding performance and the bonding quality of the process to perform process optimization processing on the bonding silver wire, which can improve the efficiency of the process optimization of the bonding silver wire and thereby improve the subsequent processing quality of the bonding silver wire. Therefore, the present invention proposes a process optimization method for realizing bonding silver wire, thereby improving the process optimization efficiency of bonding silver wire.

[0144] Embodiment 2:

[0145] like Figure 2 1 is a functional module diagram of a process optimization system for bonding silver wires provided by an embodiment of the present invention.

[0146] The process optimization system 100 for realizing silver wire bonding described in the present invention can be installed in an electronic device. According to the functions to be realized, the process optimization system 100 for realizing silver wire bonding can include a bonding property analysis module 101, a bonding performance evaluation module 102, a bonding precision calculation module 103, a process bonding quality evaluation module 104 and a process optimization module 105. The module described in the present invention can also be called a unit, which refers to a series of computer program segments that can be executed by an electronic device processor and can complete fixed functions, which are stored in the memory of the electronic device.

[0147] In this embodiment, the functions of each module / unit are as follows:

[0148] The bonding property analysis module 101 is used to obtain the silver wire preparation material corresponding to the bonding silver wire, collect material property data corresponding to the silver wire preparation material, and analyze the material bonding property of the silver wire preparation material during the preparation process based on the material property data;

[0149] The bonding performance evaluation module 102 is used to query the silver wire bonding equipment and bonding process in the preparation process of the bonding silver wire, collect the equipment parameter set corresponding to the silver wire bonding equipment, and take the equipment tool image corresponding to the silver wire bonding equipment, calculate the loss coefficient corresponding to the silver wire bonding equipment based on the equipment tool image, and evaluate the bonding performance corresponding to the silver wire bonding equipment in combination with the equipment parameter set and the loss coefficient;

[0150] The bonding precision calculation module 103 is used to collect the silver wire working condition parameters corresponding to the bonding silver wire in each process of the bonding process, and calculate the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters;

[0151] The process bonding quality evaluation module 104 is used to perform product sampling processing on the bonding silver wire in each process of the bonding process to obtain process bonding samples, perform quality evaluation on the process bonding samples to obtain multi-dimensional quality data, and evaluate the process bonding quality corresponding to the bonding process according to the multi-dimensional quality data;

[0152] The process optimization module 105 is used to perform process optimization processing on the bonding silver wire in combination with the material bonding properties, the bonding precision, the bonding performance and the process bonding quality to obtain an optimization result.

[0153] In detail, each module described in the process optimization system 100 for realizing silver wire bonding described in the embodiment of the present application is used in the same manner as described above. Figure 1 The same technical means as described in the process optimization method for bonding silver wires can produce the same technical effects, which will not be repeated here.

[0154] It is obvious to those skilled in the art that the present invention is not limited to the details of the above exemplary embodiments, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0155] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the spirit and scope of the technical solution of the present invention.

Claims

1. A process optimization method for realizing silver wire bonding, characterized in that: The method comprises: Acquire a silver wire preparation material corresponding to the bonding silver wire, collect material property data corresponding to the silver wire preparation material, and analyze the material bonding properties of the silver wire preparation material during the preparation process based on the material property data; Querying the silver wire bonding equipment and bonding process in the preparation process of the bonding silver wire, collecting the equipment parameter set corresponding to the silver wire bonding equipment, and taking the equipment tool image corresponding to the silver wire bonding equipment, calculating the loss coefficient corresponding to the silver wire bonding equipment based on the equipment tool image, and evaluating the bonding performance corresponding to the silver wire bonding equipment in combination with the equipment parameter set and the loss coefficient; Collecting silver wire working condition parameters corresponding to the bonding silver wire in each process of the bonding process, and calculating the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters; Performing product sampling processing on the bonding silver wires in each process of the bonding process to obtain process bonding samples, performing quality evaluation on the process bonding samples to obtain multi-dimensional quality data, and evaluating the process bonding quality corresponding to the bonding process according to the multi-dimensional quality data; In combination with the material bonding properties, the bonding precision, the bonding performance and the process bonding quality, a process optimization process for the bonding silver wire is performed to obtain an optimization result.

2. The process optimization method for realizing silver wire bonding according to claim 1, characterized in that: The step of analyzing the material bonding properties of the silver wire preparation material during the preparation process based on the material property data includes: Performing data filtering processing on the material attribute data to obtain target attribute data; Analyzing the attribute categories corresponding to the target attribute data, and extracting key attribute categories from the attribute categories; Based on the key attribute categories, performing data clustering processing on the target attribute data to obtain key attribute data; Analyzing the attribute correlation between the key attribute data; In combination with the attribute correlation and the key attribute data, the material bonding properties of the silver wire preparation material during the preparation process are analyzed.

3. The process optimization method for realizing silver wire bonding according to claim 2, characterized in that: The analyzing the attribute correlation between the key attribute data includes: Sorting the key attribute data to obtain a data rank; According to the data rank, the data level difference between the key attribute data is calculated, and the number of data corresponding to the key attribute data is counted; Combining the data level difference and the data quantity, the data correlation coefficient between the key attribute data can be calculated by the following formula: Among them, A represents the data correlation coefficient, B a It represents the data level difference corresponding to the a-th data in the key attribute data, a represents the data sequence number corresponding to the key attribute data, and q represents the number of data; The attribute correlation between the key attribute data is analyzed according to the data correlation coefficient.

4. The process optimization method for realizing silver wire bonding according to claim 1, characterized in that: The calculating the loss coefficient corresponding to the silver wire bonding device based on the device tool image includes: Extracting tool geometric features and tool visual features from the equipment tool image; Querying a standard device image corresponding to the silver wire bonding device, and extracting standard geometric features and standard visual features corresponding to the device tool from the standard device image; Calculating the geometric deviation between the tool geometric feature and the standard geometric feature; Calculating the visual deviation between the tool visual feature and the standard visual feature; The loss coefficient corresponding to the silver wire bonding device is calculated by combining the geometric deviation and the visual deviation.

5. The process optimization method for bonding silver wires according to claim 4, characterized in that: The extracting of tool geometric features and tool visual features from the equipment tool image includes: Performing image noise reduction processing on the equipment tool image to obtain a noise-reduced tool image; Performing a main body extraction process on the noise reduction tool image to obtain a tool main body image; Detecting a tool contour in the tool body image, performing feature extraction on the tool contour, and obtaining tool geometric features in the tool body image; The visual features of the tool body image are extracted using a preset visual feature extraction model to obtain the tool visual features.

6. The process optimization method for realizing silver wire bonding according to claim 4, characterized in that: The calculating of the geometric deviation between the tool geometric feature and the standard geometric feature comprises: Performing feature quantification processing on the tool geometric feature and the standard geometric feature respectively to obtain tool geometric feature value and standard geometric feature value; Calculating geometric feature weights corresponding to the tool geometric features; Combining the geometric feature weight, the tool geometric feature value and the standard geometric feature value, the geometric deviation between the tool geometric feature and the standard geometric feature can be calculated by the following formula: Where D represents the geometric deviation between the tool geometric features and the standard geometric features, β b The geometric feature weight corresponding to the bth feature of the tool geometric feature, E b The tool geometric feature value corresponding to the bth feature of the tool geometric feature, F b It represents the standard geometric feature value corresponding to the bth feature of the tool geometric feature in the standard geometric feature, b represents the feature serial number corresponding to the tool geometric feature, and t represents the number of features corresponding to the tool geometric feature.

7. The process optimization method for bonding silver wires according to claim 1, characterized in that: The step of calculating the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters includes: Performing parameter calibration processing on the silver wire working condition parameters to obtain calibration working condition parameters; Standardizing the calibration working condition parameters to obtain standard working condition parameters, and identifying working condition indicators corresponding to the silver wire working condition parameters; The precision contribution coefficient corresponding to the working condition index is calculated, and the bonding precision corresponding to the bonding silver wire is calculated by combining the standard working condition parameters and the precision contribution coefficient.

8. The process optimization method for realizing silver wire bonding according to claim 1, characterized in that: The step of evaluating the bonding quality of the bonding process corresponding to the bonding process according to the multi-dimensional quality data includes: Identify a data identifier corresponding to the multidimensional quality data, and analyze the quality attribute dimension corresponding to the multidimensional quality data based on the data identifier; Based on the multidimensional quality data, calculating the dimensional interaction strength between the quality attribute dimensions; Combining the dimension interaction strength and the quality attribute dimension, clustering the multidimensional quality data to obtain clustered quality data; Based on the clustering quality data, calculating a dimension quality score corresponding to the quality attribute dimension; Combined with the dimension quality scores, the process bonding quality corresponding to the bonding process is evaluated.

9. The process optimization method for realizing silver wire bonding according to claim 8, characterized in that: The calculating, based on the multidimensional quality data, the dimensional interaction strength between the quality attribute dimensions comprises: Extracting data quantization values ​​corresponding to the multidimensional quality data, and calculating average measurement values ​​corresponding to the quality attribute dimensions according to the data quantization values; Calculate the dimension discrete value corresponding to the quality attribute dimension by combining the average metric value and the data quantization value; Combining the data quantization value, the average metric value and the dimension discrete value, the dimensional interaction intensity between the quality attribute dimensions can be calculated by the following formula: Among them, G represents the dimensional interaction strength between quality attribute dimensions, Q i Represents the data quantification value of the i-th data in the Q-th dimension of the quality attribute dimension, represents the average measurement value of the Qth dimension in the quality attribute dimension, M i The data quantification value of the i-th data in the M-th dimension of the quality attribute dimension, represents the average measurement value of the Mth dimension in the quality attribute dimension, i represents the data sequence number of the multidimensional quality data, p represents the number of multidimensional quality data corresponding to the quality attribute dimension, δ Q and δ M They respectively represent the dimensional discrete values ​​corresponding to the Qth dimension and the Mth dimension in the quality attribute dimension.

10. A process optimization system for realizing silver wire bonding, characterized in that: The system comprises: A bonding property analysis module, used to obtain the silver wire preparation material corresponding to the bonding silver wire, collect material property data corresponding to the silver wire preparation material, and analyze the material bonding properties of the silver wire preparation material during the preparation process based on the material property data; A bonding performance evaluation module, used to query the silver wire bonding equipment and bonding process in the preparation process of the bonding silver wire, collect the equipment parameter set corresponding to the silver wire bonding equipment, and take the equipment tool image corresponding to the silver wire bonding equipment, calculate the loss coefficient corresponding to the silver wire bonding equipment based on the equipment tool image, and evaluate the bonding performance corresponding to the silver wire bonding equipment in combination with the equipment parameter set and the loss coefficient; A bonding precision calculation module, used for collecting silver wire working condition parameters corresponding to the bonding silver wire in each process of the bonding process, and calculating the bonding precision corresponding to the bonding silver wire based on the silver wire working condition parameters; A process bonding quality evaluation module is used to perform product sampling processing on the bonding silver wire in each process of the bonding process to obtain process bonding samples, perform quality evaluation on the process bonding samples to obtain multi-dimensional quality data, and evaluate the process bonding quality corresponding to the bonding process according to the multi-dimensional quality data; The process optimization module is used to perform process optimization processing on the bonding silver wire in combination with the material bonding properties, the bonding precision, the bonding performance and the process bonding quality to obtain an optimization result.

Images