Intelligent mounting welding control method and system for integrated circuit mainboard

The welding paste quantity information is obtained through image recognition technology and the welding parameters are optimized, which solves the problem of insufficient accuracy of traditional mounting welding control and improves welding accuracy and production efficiency.

CN118742009BActive Publication Date: 2024-11-15JIANGSU TUOERQI PHOTOELECTRIC TECH CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
CN202411178426.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-08-27
Publication Date
2024-11-15
Estimated Expiration
2044-08-27

AI Technical Summary

Technical Problem

The traditional mounting welding control method lacks precise control of mounting welding parameters, resulting in poor welding accuracy, stability and accuracy, thereby reducing welding quality and production efficiency.

Method used

By collecting images after the solder paste is applied to multiple solder joints on the circuit motherboard, combining image recognition technology to obtain solder paste quantity information, calculate solder paste accuracy information, and optimize the mounting pressure, heating temperature, and welding time based on this information, to achieve Accurate control of the welding process.

Benefits of technology

It improves the accuracy, stability and accuracy of welding, ensures a good fit between the solder paste and the solder joints, and improves the welding quality and production efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118742009B_ABST
    Figure CN118742009B_ABST
Patent Text Reader

Abstract

The present invention provides an intelligent mounting welding control method and system for an integrated circuit mainboard, which relates to the field of welding control technology, including: collecting images of multiple solder joints on a circuit mainboard after solder paste is applied to obtain a mainboard image; performing identification to obtain multiple solder paste quantity information, combining solder joint graphics, and calculating to obtain multiple solder paste accuracy information; optimizing the mounting pressure to obtain the optimal mounting pressure, and performing mounting control; after the mounting is completed, the mounting image of the mounting components of the circuit mainboard is collected, and the mounting distance information is obtained; the heating temperature and welding time of the heated solder paste for welding are optimized to obtain the optimal heating temperature and the optimal welding time, and welding control is performed. The present invention solves the technical problem that the traditional mounting welding control method lacks precise control of the mounting welding parameters, resulting in poor accuracy, stability and precision, and further leads to reduced welding quality and production efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of welding control, and in particular to an intelligent mounting welding control method and system for an integrated circuit mainboard. Background Art

[0002] Mounting welding is to pre-coat the solder paste, such as tin, on the solder points on the pads, then mount the components on the pads of the PCB board, and use a soldering iron or a heating furnace to melt the solder to achieve welding. At present, the mounting welding of integrated circuit motherboards is generally carried out automatically, and is fixedly executed through a pre-set automated program. When errors occur in the integrated circuit motherboard, problems such as leaking solder joints and poor contact of solder joints are prone to occur. Summary of the invention

[0003] The present application provides an intelligent mounting and welding control method for an integrated circuit mainboard, aiming to solve the technical problem that the traditional mounting and welding control method lacks precise control of mounting and welding parameters, resulting in poor accuracy, stability and precision, and further leading to reduced welding quality and production efficiency.

[0004] In view of the above problems, the present application provides a method and system for controlling intelligent mounting and welding of an integrated circuit mainboard.

[0005] In a first aspect disclosed in the present application, a smart mounting and welding control method for an integrated circuit mainboard is provided, the method comprising: collecting an image of a plurality of solder joints on a circuit mainboard to be mounted and welded after solder paste is applied thereto, to obtain a mainboard image; identifying the mainboard image, obtaining a plurality of solder paste quantity information, and calculating and obtaining a plurality of solder paste accuracy information in combination with the solder joint graphics of the plurality of solder joints; optimizing the mounting pressure of mounting components mounted and welded on the circuit mainboard according to the plurality of solder paste accuracy information, obtaining an optimal mounting pressure, and performing mounting control on the mounting components; after mounting is completed, collecting a mounting image of the mounting components mounted on the circuit mainboard, performing identification, and obtaining mounting distance information; optimizing the heating temperature and welding time for heating the solder paste for welding according to the mounting distance information and the plurality of solder paste accuracy information, obtaining an optimal heating temperature and an optimal welding time, and performing welding control.

[0006] The second aspect disclosed in the present application provides an intelligent mounting and welding control system for an integrated circuit motherboard, the system being used for the intelligent mounting and welding control method for the integrated circuit motherboard mentioned above, the system comprising: an image acquisition module, the image acquisition module being used to acquire images of a plurality of solder joints on a circuit motherboard to be mounted and welded after solder paste is applied, to obtain a motherboard image; an image recognition module, the image recognition module being used to recognize the motherboard image, obtain a plurality of solder paste amount information, and calculate and obtain a plurality of solder paste accuracy information in combination with the solder joint graphics of the plurality of solder joints; a pressure optimization module, the pressure optimization module being used to calculate and obtain a plurality of solder paste accuracy information according to the plurality of solder joints; The module comprises a plurality of solder paste accuracy information, and optimizes the mounting pressure of the mounting components mounted on the circuit mainboard to obtain the optimal mounting pressure and to control the mounting of the mounting components; a mounting distance acquisition module, wherein the mounting distance acquisition module is used to collect the mounting image of the mounting components mounted on the circuit mainboard after the mounting is completed, identify the mounting distance information and obtain the mounting distance information; a welding control module, wherein the welding control module is used to optimize the heating temperature and welding time of the heated solder paste for welding according to the mounting distance information and the plurality of solder paste accuracy information, obtain the optimal heating temperature and the optimal welding time and to perform welding control.

[0007] According to a third aspect disclosed in the present application, a computer device is provided, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, any step of the first aspect disclosed in the present application is implemented.

[0008] The fourth aspect disclosed in the present application provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements any step of the first aspect disclosed in the present application.

[0009] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0010] By collecting images of multiple solder joints on the circuit board after solder paste is applied, combined with image recognition technology, multiple solder paste amount information is obtained, and multiple solder paste accuracy information is calculated, so that the coating of solder paste and the position of solder joints can be effectively identified, and the accuracy of welding is improved; according to multiple solder paste accuracy information, the mounting pressure is optimized to obtain the optimal mounting pressure, thereby realizing the mounting control of the mounted components, which can ensure the good fit between the solder paste and the solder joints, and improve the stability and reliability of welding; after the mounting is completed, the mounting image of the mounted components is collected, identified, and the mounting distance information is obtained, so that the distance between the mounted components and the circuit board can be accurately grasped, which helps to improve the accuracy and stability of welding; according to the mounting distance information and multiple solder paste accuracy information, the heating temperature and welding time of the heated solder paste are optimized to obtain the optimal heating temperature and welding time, so that the solder paste can be fully melted and fully fitted with the solder joints, and the quality and stability of welding are improved. In summary, this method can effectively solve the problems of accuracy, stability and precision in the traditional mounting welding process, thereby improving the welding quality and production efficiency.

[0011] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0012] Figure 1 A schematic flow chart of a smart mounting and welding control method for an integrated circuit mainboard provided in an embodiment of the present application;

[0013] Figure 2 A schematic diagram of the structure of an intelligent mounting and welding control system for an integrated circuit mainboard provided in an embodiment of the present application;

[0014] Figure 3 This is a diagram of the internal structure of a computer device provided in an embodiment of the present application.

[0015] Explanation of the reference numerals: image acquisition module 10 , image recognition module 20 , pressure optimization module 30 , mounting distance acquisition module 40 , welding control module 50 . DETAILED DESCRIPTION

[0016] The embodiment of the present application solves the technical problem that the traditional mounting welding control method lacks precise control of mounting welding parameters, resulting in poor accuracy, stability and precision, and further leads to reduced welding quality and production efficiency by providing an intelligent mounting welding control method for an integrated circuit mainboard.

[0017] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0018] like Figure 1 As shown, the embodiment of the present application provides a smart mounting and welding control method for an integrated circuit mainboard, the method comprising:

[0019] Collect images of multiple solder joints on a circuit mainboard to be mounted and soldered after solder paste is applied to obtain a mainboard image;

[0020] Prepare a circuit board to be mounted and soldered, as well as a device for capturing images, such as a high-definition camera, and apply solder paste to multiple solder joints on the circuit board to be soldered, ensuring that each solder joint is covered with an appropriate amount of solder paste. After coating, use a camera to capture images of the circuit board, ensuring that the image clarity is high and the details of the solder paste are clearly visible. Save the resulting image as a motherboard image file for subsequent image recognition and analysis.

[0021] Recognize the motherboard image, obtain multiple solder paste quantity information, and calculate and obtain multiple solder paste accuracy information in combination with the solder joint graphics of the multiple solder joints;

[0022] Based on the convolutional neural network, a solder paste quantity identifier is constructed, and the identifier is used to identify the solder paste area in the motherboard image to obtain the solder paste quantity information on each solder joint. Graphic recognition is performed on multiple solder joints to identify the position and shape of each solder joint. The solder paste quantity information and solder joint graphic information are combined to calculate the accuracy information of the solder paste on each solder joint. For example, the accuracy of the solder paste can be evaluated by comparing the difference between the actual position of the solder paste and the ideal position, and the difference between the actual shape of the solder paste and the expected shape. The solder paste accuracy can be expressed in a numerical way, such as using a percentage.

[0023] Furthermore, the motherboard image is identified to obtain multiple solder paste amount information, including:

[0024] Preprocessing the mainboard image;

[0025] Based on the mounting soldering history data, a sample motherboard image set is obtained, and the solder paste amount of multiple solder joints in the sample main body image is identified and marked to obtain multiple sample solder paste amount information sets;

[0026] The sample motherboard image set and multiple sample solder paste quantity information sets are used to construct a solder paste quantity identifier based on a convolutional neural network, and the preprocessed motherboard image is identified to obtain the multiple solder paste quantity information.

[0027] The purpose of preprocessing the motherboard image is to improve accuracy and efficiency in the subsequent image processing and analysis process. The preprocessing operation includes denoising, such as removing noise in the image by applying a digital filter to reduce interference in subsequent processing steps; adjusting image brightness and contrast to enhance the visibility of key features such as solder joints and solder paste in the image; using smoothing techniques such as Gaussian blur or mean filtering to smooth the image to reduce details and noise in the image, making solder joints and solder paste easier to identify. In this way, noise and interference in the image can be reduced, key features such as solder joints and solder paste can be highlighted, and a basis for subsequent image recognition and analysis can be provided.

[0028] A series of sample motherboard images are collected from the historical data of mounting welding, which cover various welding situations to ensure the diversity of samples. The solder paste quantity information of the solder joint area in each sample motherboard image can be identified by using image processing and computer vision techniques, such as threshold segmentation and edge detection, to identify the location of the solder joint and mark the solder paste quantity, which can be expressed as the volume of the solder paste. The solder paste quantity information of multiple solder joints in each sample motherboard image is organized into a sample solder paste quantity information set, and each sample solder paste quantity information corresponds to a sample motherboard image for subsequent model training.

[0029] The sample motherboard image set and the corresponding sample solder paste amount information set are integrated into the construction data and divided into a training set and a validation set. A convolutional neural network (CNN) model suitable for image recognition tasks is designed. The CNN includes components such as convolutional layers, pooling layers, and fully connected layers. The input layer matches the sample motherboard image, and the output layer matches the sample solder paste amount information.

[0030] The designed CNN model is trained using the training set. During the training process, optimization is performed based on the sample motherboard image and the corresponding solder paste amount information to minimize the error between the predicted value and the actual value. The performance of the trained model is evaluated using the validation set. The generalization ability and recognition performance of the model are evaluated by calculating indicators such as the loss value and accuracy of the model on the validation set. Finally, a model that meets the preset accuracy is obtained, which is the solder paste amount identifier.

[0031] The solder paste quantity identifier is used to identify the preprocessed motherboard image. The identifier outputs multiple solder paste quantity information according to the learned pattern, and the identified solder paste quantity information is associated with the solder point position on the motherboard image to obtain the solder paste quantity information of each solder point.

[0032] Furthermore, combining the solder joint patterns of the plurality of solder joints, a plurality of solder paste accuracy information is calculated and obtained, including:

[0033] According to the solder joint patterns of the plurality of solder joints on the solder pad, index and obtain a plurality of standard solder paste quantity information;

[0034] The plurality of solder paste accuracy information is obtained by calculation according to the plurality of standard solder paste amount information and the plurality of solder paste amount information.

[0035] Establish a standard solder paste quantity information index. Specifically, obtain a large number of different solder joints and their corresponding standard solder paste quantity information from known welding standards or welding quality certification standards, associate these standard solder paste quantity information with the corresponding solder joint graphics, and establish a database to record the standard solder paste quantity information of each solder joint.

[0036] During the mounting process, the shape of the solder joint on the pad is obtained through image processing technology, the shape of the actual solder joint is matched with the established database, the corresponding standard solder paste quantity information is found, and through indexing, multiple standard solder paste quantity information matching the multiple solder joints is obtained.

[0037] According to the plurality of solder paste accuracy information, the mounting pressure of the mounting components mounted and soldered on the circuit mainboard is optimized to obtain the optimal mounting pressure, and the mounting of the mounting components is controlled;

[0038] According to the calculated accuracy information of multiple solder pastes, the amount of solder paste on each solder joint is analyzed. Among them, the solder joints with less solder paste need to increase the mounting pressure to ensure the fit of the solder joints, and vice versa, the solder joints with more solder paste need to reduce the mounting pressure to avoid solder paste overflow. Based on the analysis results of the solder paste amount information, a mounting pressure optimization strategy is formulated to obtain the optimal mounting pressure. According to the optimized mounting pressure, the mounting components on the circuit board are controlled, such as by adjusting the parameters of the mounting equipment to achieve the required mounting pressure, and the precise control of the mounting components is achieved, thereby ensuring the welding quality and stability.

[0039] Furthermore, according to the plurality of solder paste accuracy information, the mounting pressure of the mounting components mounted and soldered on the circuit main board is optimized, including:

[0040] According to the plurality of solder paste accuracy information, a mounting pressure function is constructed to optimize the mounting pressure of mounting components on the circuit mainboard, as shown in the following formula:

[0041] ;

[0042] Where, pre is the pressure adaptability, M is the number of multiple welding points, is the weight of the ith solder joint assigned according to the size of multiple solder paste accuracy information. The size of the solder paste accuracy information is positively correlated with the weight. is the extent of mismatch between solder paste and solder joint pattern after the i-th solder joint is mounted under mounting pressure;

[0043] A first mounting pressure is randomly generated, and the amplitude information of the mismatch between the solder paste and the solder joint image after the mounting of the multiple solder joints is predicted in combination with the multiple solder paste accuracy information. A first pressure fitness is calculated based on the mounting pressure function, wherein a solder paste mounting deviation identifier is constructed by using a sample mounting pressure set, multiple sample solder paste accuracy information sets, and multiple sample deviation amplitude information sets to predict the amplitude of the mismatch between the solder paste and the solder joint image after mounting;

[0044] Continue to randomly generate and optimize the mounting pressure until convergence, and obtain the mounting pressure with the maximum pressure adaptability as the optimal mounting pressure.

[0045] Specifically, the mounting pressure function is as follows:

[0046] ;

[0047] The purpose of this function is to adjust the mounting pressure according to the accuracy information of the solder paste to minimize the extent of the inconsistency between the solder paste and the solder joint pattern. Among them, pre represents the pressure adaptability, and the greater the adaptability, the more appropriate the mounting pressure; M represents the number of multiple solder joints; represents the weight of the ith solder joint assigned according to the size of multiple solder paste accuracy information. This weight is positively correlated with the size of the solder paste accuracy information, that is, the higher the accuracy, the greater the weight; It indicates the extent of mismatch between solder paste and solder joint pattern after the i-th solder joint is mounted under mounting pressure. The smaller this value is, the higher the matching degree between solder paste and solder joint pattern under mounting pressure is, and the higher the adaptability is.

[0048] The goal of the placement pressure function is to adjust the weights , so that the solder paste and solder joint pattern do not conform to the extent As small as possible, so as to maximize the pressure fitness pre. Such an optimization process can be achieved through various optimization algorithms, such as genetic algorithm, simulated annealing algorithm, particle swarm optimization algorithm, etc., to optimize the mounting pressure and thus improve the welding quality and stability.

[0049] A first mounting pressure is randomly generated from a preset pressure range as an initial value, and a constructed solder paste mounting deviation identifier is used to predict multiple solder paste accuracy information and the randomly generated first mounting pressure, and output the amplitude information of the solder paste and solder joint image not matching after multiple solder joints are mounted. The solder paste mounting deviation identifier can be obtained by training a model with sample data, and the input of the model includes solder paste accuracy information and mounting pressure, and the output is the amplitude information of the solder paste and solder joint image not matching.

[0050] According to the mounting pressure function, the amplitude information of the mismatch between the solder paste and the solder spot image after the predicted multiple solder spots are mounted and the corresponding solder paste accuracy information are substituted for calculation to obtain a first pressure adaptability.

[0051] The next mounting pressure is randomly generated within the preset pressure range as the second mounting pressure, and the second pressure fitness is calculated. An optimization algorithm, such as a genetic algorithm, a simulated annealing algorithm, a particle swarm optimization algorithm, etc., is used to adjust the mounting pressure according to the current pressure fitness and the selected optimization goal, that is, maximizing the fitness. The optimization algorithm updates the value of the mounting pressure according to the larger value of the current second pressure fitness and the first pressure fitness, in the hope of achieving a better fitness.

[0052] Determine whether the optimization process has converged, that is, whether the conditions for stopping optimization have been met, such as whether the fitness change for several consecutive iterations is less than a certain threshold, or whether the preset maximum number of iterations has been reached. When the optimization process converges, the optimal mounting pressure is obtained, and its corresponding fitness is the maximum pressure fitness. This optimal mounting pressure can be used as the mounting pressure parameter in the subsequent mounting process to ensure that the mounting pressure can adapt to the characteristics of the solder paste and solder joint image to the greatest extent, thereby improving the welding quality and stability.

[0053] After the mounting is completed, the mounting image of the mounting component mounted on the circuit mainboard is collected, and identification is performed to obtain mounting distance information;

[0054] After the mounting is completed, a camera is used to capture images of the mounted components on the circuit main board in a direction parallel to the circuit main board to obtain a mounting image, and a mounting distance identifier is used to identify and locate the mounted components in the mounting image to obtain the mounting distance, that is, the distance between the mounted components and the surface of the circuit main board.

[0055] Furthermore, collecting the mounting image of the mounting component mounted on the circuit mainboard, performing identification, and obtaining mounting distance information includes:

[0056] In a direction parallel to the circuit mainboard, collecting a mounting image of the mounting component mounted on the circuit mainboard;

[0057] Preprocessing the mounting image;

[0058] According to the mounting welding history data, a sample mounting image set is collected, and the distance between the circuit board and the mounting components in the sample mounting image is identified and marked to obtain a sample mounting distance information set;

[0059] The sample mounting image set and the sample mounting distance information set are used to construct a mounting distance identifier based on a convolutional neural network, and the pre-processed mounting images are identified to obtain the mounting distance information.

[0060] Place the camera device in a direction parallel to the circuit board and close enough to the surface of the board, and adjust the angle and direction of the camera to make it perpendicular to the surface of the circuit board to maximize the distance information between the circuit board and the mounted components. Use the arranged camera device to capture images of the circuit board with mounted components to obtain mounting images.

[0061] The collected mounting images are preprocessed. The goal of preprocessing is to improve image quality, reduce noise, and enhance features to facilitate subsequent image recognition, distance measurement and other operations, including grayscale processing, converting color images into grayscale images to simplify the complexity of image processing; using filters such as Gaussian filters, median filters, etc., to denoise the images to reduce the impact of noise in the images.

[0062] The historical data generated during the mounting and welding process, including mounting images, are collected, a part of the collected historical data is selected as samples, and a sample mounting image set is established, which covers different types of mounting components and mounting positions.

[0063] The sample mounting images are processed, and the distance between the circuit board and the mounting components is identified using image processing and computer vision technology. For example, edge detection, feature extraction and other technologies can be used to determine the position of the mounting components and the contour of the mainboard surface, and then calculate the distance between them. For each sample mounting image, the distance information between the circuit board and the mounting component is recorded, and the sample mounting images and their corresponding distance information are combined into a sample mounting distance information set.

[0064] The sample mounting image set is used as training data, and the sample mounting distance information set is used as the corresponding label. The convolutional neural network architecture is designed, including convolutional layer, pooling layer, fully connected layer, etc. The input and output of the network are determined. The input is the mounting image, and the output is the mounting distance information.

[0065] The convolutional neural network is trained using a set of sample mounting images and a set of sample mounting distance information. During the training process, the network parameters are optimized by minimizing the loss function so that the network can accurately predict the mounting distance. The trained model is verified using a validation set to evaluate the performance of the model. The model's hyperparameters, such as learning rate and network structure, are adjusted based on the validation results to improve the generalization and accuracy of the model. The trained mounting distance identifier is used to perform distance recognition on the preprocessed mounting images and output the mounting distance information.

[0066] According to the mounting distance information and the multiple solder paste accuracy information, the heating temperature and welding time of the heated solder paste for welding are optimized to obtain the optimal heating temperature and the optimal welding time, and welding control is performed.

[0067] According to the placement distance information and solder paste accuracy information, an optimization strategy for heating temperature and welding time is formulated, including that for solder joints with large placement distances, the heating temperature and welding time need to be increased to ensure that the solder paste can fully melt and fill the gap between the solder joint and the component; for solder joints with inaccurate solder paste coating, the heating temperature and welding time need to be adjusted to compensate for the impact of insufficient or excessive solder paste coating; at the same time, the component should not be heated too much to cause damage. Based on the above optimization strategy, the optimal heating temperature and optimal welding time are determined through the optimization algorithm.

[0068] Based on the calculated optimal heating temperature and optimal welding time, welding control is performed, including adjusting the parameters of the welding equipment to ensure that the solder paste melts and flows at the optimal heating temperature, complete the welding process, improve the welding quality and stability, and ensure the reliable connection between the mounted components and the circuit board.

[0069] Furthermore, according to the mounting distance information and the plurality of solder paste accuracy information, the heating temperature and the soldering time for soldering by heating the solder paste are optimized, including:

[0070] According to the plurality of solder paste accuracy information, a preset optimization step length for optimizing the heating temperature and the soldering time is corrected and calculated to obtain the optimization step length;

[0071] Construct a welding optimization function to optimize the heating temperature and welding time of the heated solder paste for welding, as shown in the following formula:

[0072] ;

[0073] Among them, wel is welding fitness, and are the melting weights and component weights, and The sum of is 1, is the melting rate of solder paste at the ith solder joint under the welding parameters, K is the mounting accuracy coefficient, is the temperature rise of the mounted component at the ith solder joint after welding according to the welding parameters, is the mounting distance information, It is the standard mounting distance between mounting components and circuit boards;

[0074] randomly generating a first heating temperature and a first welding time as first welding parameters;

[0075] Using the first welding parameter, combined with the plurality of solder paste amount information and mounting distance information, analyzing and obtaining the melting rate of the solder paste in the plurality of solder joints after welding and the rising temperature of the mounted components, and calculating and obtaining the first welding fitness based on the welding optimization function;

[0076] The welding heating temperature and welding time are continuously optimized until convergence, and welding parameters with the maximum welding fitness are output to obtain the optimal heating temperature and optimal welding time.

[0077] Set an initial optimization step value, for example, set an initial value based on experience as the preset optimization step. For each solder joint, adjust the optimization step according to its corresponding solder paste accuracy information, including: if the solder paste accuracy of a solder joint is high, it means that it is more sensitive to the adjustment of heating temperature and welding time, and the optimization step of the solder joint can be appropriately increased; if the solder paste accuracy of a solder joint is low, it means that it has little effect on the adjustment of heating temperature and welding time, and the optimization step of the solder joint can be appropriately reduced. According to the solder paste accuracy information of each solder joint, the optimization step of each solder joint is corrected and calculated, so as to better guide the optimization process of heating temperature and welding time.

[0078] The welding optimization function is as follows:

[0079] ;

[0080] Among them, wel is welding fitness, which is an evaluation index of welding quality. The larger the index is, the better the welding quality is. and They are melting weight and component weight, and their sum is 1. These two weights are used to adjust the influence of different factors on welding suitability; is the melting rate of the solder paste at the ith solder joint under the welding parameters, indicating the degree to which the solder paste is melted under the given welding parameters; K is the mounting accuracy coefficient, which is a value calculated based on the welding parameters and reflects the rising temperature of the mounted component at the ith solder joint after welding. The calculation of K is based on the mounting distance information. And the standard mounting distance between the mounting components and the circuit board The goal of the welding optimization function is to maximize the welding fitness wel by adjusting the heating temperature and welding time.

[0081] First, determine the reasonable range of heating temperature and welding time. This range can be determined based on actual conditions and previous experience to ensure that the parameters are within an acceptable range. After determining the range of the parameters, use a random number generator to uniformly randomly select a value from the parameter range as the initial parameter, and use the randomly generated heating temperature and welding time as the initial parameters for the first welding.

[0082] The welding process is simulated in a simulation environment using randomly generated first welding parameters. During the simulation, the melting rate of solder paste and the rising temperature of mounted components are recorded and substituted into the welding optimization function to calculate the first welding fitness, which is used as an evaluation index for the first welding to evaluate the quality of the first welding.

[0083] Set an iteration termination condition to determine when to stop the optimization process, which can be when a certain number of iterations is reached, or when the change in fitness is less than a threshold, or when the fitness reaches a preset target value. According to the welding fitness and optimization algorithm, such as gradient descent method, genetic algorithm, etc., the welding parameters, i.e., heating temperature and welding time, are updated. The updated parameters are used for the next iteration. In each iteration, it is determined whether the convergence condition is met. If so, the optimization process is stopped and the optimal welding parameters and corresponding fitness are output. This ensures that the welding process achieves the best quality and improves the efficiency and stability of welding.

[0084] Further, the first welding parameter is used, combined with the plurality of solder paste amount information and mounting distance information, to analyze and obtain the melting rate of the solder paste in the plurality of solder joints after welding and the rising temperature of the mounted components, including:

[0085] According to the mounting welding history data, a sample welding parameter set, a sample solder paste amount information set, a sample mounting distance information set, a sample melting rate information set and a sample rise temperature set are obtained;

[0086] The sample welding parameter set, the sample solder paste amount information set, and the sample mounting distance information set are used as sample inputs, and the sample melting rate information set and the sample rising temperature set are used as sample outputs to train and obtain a welding simulation channel;

[0087] Based on the welding simulation channel, a welding simulation analysis is performed on the first welding parameter in combination with the multiple solder paste amount information and mounting distance information to obtain multiple melting rates of the solder paste in the multiple solder joints and multiple rising temperatures of the mounted components.

[0088] Collect historical data of mounting welding, including sample welding parameter sets under different welding parameters, sample solder paste amount information sets, sample mounting distance information sets, and corresponding sample melting rate information sets and sample rise temperature sets. These historical data will be used as input and labels for training models to build welding simulation channels.

[0089] Among them, the sample welding parameter set includes parameter combinations of different heating temperatures and welding times used in the historical welding process; the sample solder paste amount information set records the solder paste coating amount under each sample welding parameter; the sample mounting distance information set includes the distance between the mounting components and the circuit board under each sample welding parameter.

[0090] The sample melting rate information set records the melting rate of the solder paste under each sample welding parameter, which can be statistically obtained based on the actual welding conditions or the simulated welding results; the sample rising temperature set records the rising temperature of the mounted components under each sample welding parameter, and these data can also be obtained through actual measurement or simulation.

[0091] The sample welding parameters, solder paste amount and mounting distance are used as input features, and the sample melting rate and rising temperature are used as output labels to construct a training set, and select a suitable machine learning model or deep learning model, such as a neural network, to model the relationship between the melting rate and rising temperature during the welding process.

[0092] The model is trained using the training set, and the model parameters are optimized through the back propagation algorithm so that the model can predict the melting rate of solder paste and the temperature rise of mounted components as accurately as possible. The validation set is used to evaluate the performance of the model to check whether the model can accurately predict the melting rate and temperature rise during the welding process. The model is tuned according to the evaluation results, such as adjusting the model structure and hyperparameters, to improve the performance of the model.

[0093] Through such a training process, an accurate welding simulation channel can be obtained, which can accurately predict the melting rate of solder paste and the rising temperature of mounted components based on given welding parameters, solder paste amount and mounting distance.

[0094] The first welding parameter, multiple solder paste amount information and mounting distance information are taken as input and input into the trained welding simulation channel. The welding simulation channel is used to perform simulation analysis on each solder joint, and the melting rate of the solder paste in each solder joint and the rising temperature of the mounted component are predicted. The predicted results of the melting rate of the solder paste in each solder joint and the rising temperature of the mounted component are obtained as the simulation analysis results under this welding parameter.

[0095] In summary, the intelligent mounting and welding control method of an integrated circuit mainboard provided by the embodiment of the present application has the following technical effects:

[0096] 1. By collecting images of multiple solder joints on the circuit board after solder paste is applied, combined with image recognition technology, multiple solder paste quantity information is obtained, and multiple solder paste accuracy information is calculated, so that the solder paste coating condition and the position of the solder joints can be effectively identified, thereby improving the accuracy of welding;

[0097] 2. According to the accuracy information of multiple solder pastes, the mounting pressure is optimized to obtain the optimal mounting pressure, thereby realizing the mounting control of the mounted components, which can ensure the good fit between the solder paste and the solder joints and improve the stability and reliability of welding;

[0098] 3. After the mounting is completed, the mounting image of the mounted component is collected, identified, and the mounting distance information is obtained. This can accurately grasp the distance between the mounted component and the circuit board, which helps to improve the accuracy and stability of welding;

[0099] 4. According to the mounting distance information and multiple solder paste accuracy information, the heating temperature and welding time of the heated solder paste are optimized to obtain the optimal heating temperature and welding time, which can ensure that the solder paste is fully melted and fully fits the solder joint, thereby improving the quality and stability of welding.

[0100] In summary, this method can effectively solve the problems of accuracy, stability and precision in the traditional mounting welding process, thereby improving the welding quality and production efficiency.

[0101] Based on the same inventive concept as the intelligent mounting welding control method of an integrated circuit motherboard in the aforementioned embodiment, Figure 2 As shown, the present application provides an intelligent mounting and welding control system for an integrated circuit mainboard, the system comprising:

[0102] Image acquisition module 10, the image acquisition module 0 is used to acquire images of a plurality of solder joints on a circuit mainboard to be mounted and soldered after solder paste is applied, to obtain a mainboard image;

[0103] An image recognition module 20, the image recognition module 20 is used to recognize the motherboard image, obtain multiple solder paste quantity information, and calculate multiple solder paste accuracy information in combination with the solder joint graphics of the multiple solder joints;

[0104] A pressure optimization module 30, the pressure optimization module 30 is used to optimize the mounting pressure of the mounting components mounted and soldered on the circuit mainboard according to the multiple solder paste accuracy information, obtain the optimal mounting pressure, and perform mounting control on the mounting components;

[0105] A mounting distance acquisition module 40, which is used to collect a mounting image of the mounting component mounted on the circuit mainboard after the mounting is completed, perform identification, and obtain mounting distance information;

[0106] The welding control module 50 is used to optimize the heating temperature and welding time of the heated solder paste for welding according to the mounting distance information and the multiple solder paste accuracy information, obtain the optimal heating temperature and the optimal welding time, and perform welding control.

[0107] Furthermore, the system further includes a solder paste quantity information acquisition module to perform the following operation steps:

[0108] Preprocessing the mainboard image;

[0109] Based on the mounting soldering history data, a sample motherboard image set is obtained, and the solder paste amount of multiple solder joints in the sample main body image is identified and marked to obtain multiple sample solder paste amount information sets;

[0110] The sample motherboard image set and multiple sample solder paste quantity information sets are used to construct a solder paste quantity identifier based on a convolutional neural network, and the preprocessed motherboard image is identified to obtain the multiple solder paste quantity information.

[0111] Furthermore, the system further includes a solder paste accuracy information acquisition module to perform the following operation steps:

[0112] According to the solder joint patterns of the plurality of solder joints on the solder pad, index and obtain a plurality of standard solder paste quantity information;

[0113] The plurality of solder paste accuracy information is obtained by calculation according to the plurality of standard solder paste amount information and the plurality of solder paste amount information.

[0114] Furthermore, the system further includes an optimal mounting pressure acquisition module to perform the following operation steps:

[0115] According to the plurality of solder paste accuracy information, a mounting pressure function is constructed to optimize the mounting pressure of mounting components on the circuit mainboard, as shown in the following formula:

[0116] ;

[0117] Where, pre is the pressure adaptability, M is the number of multiple welding points, is the weight of the ith solder joint assigned according to the size of multiple solder paste accuracy information. The size of the solder paste accuracy information is positively correlated with the weight. is the extent of mismatch between solder paste and solder joint pattern after the i-th solder joint is mounted under mounting pressure;

[0118] A first mounting pressure is randomly generated, and the amplitude information of the mismatch between the solder paste and the solder joint image after the mounting of the multiple solder joints is predicted in combination with the multiple solder paste accuracy information. A first pressure fitness is calculated based on the mounting pressure function, wherein a solder paste mounting deviation identifier is constructed by using a sample mounting pressure set, multiple sample solder paste accuracy information sets, and multiple sample deviation amplitude information sets to predict the amplitude of the mismatch between the solder paste and the solder joint image after mounting;

[0119] Continue to randomly generate and optimize the mounting pressure until convergence, and obtain the mounting pressure with the maximum pressure adaptability as the optimal mounting pressure.

[0120] Furthermore, the system further includes a mounting distance information acquisition module to perform the following operation steps:

[0121] In a direction parallel to the circuit mainboard, collecting a mounting image of the mounting component mounted on the circuit mainboard;

[0122] Preprocessing the mounting image;

[0123] According to the mounting welding history data, a sample mounting image set is collected, and the distance between the circuit board and the mounting components in the sample mounting image is identified and marked to obtain a sample mounting distance information set;

[0124] The sample mounting image set and the sample mounting distance information set are used to construct a mounting distance identifier based on a convolutional neural network, and the pre-processed mounting images are identified to obtain the mounting distance information.

[0125] Furthermore, the system further includes an optimal welding parameter acquisition module to perform the following operation steps:

[0126] According to the plurality of solder paste accuracy information, a preset optimization step length for optimizing the heating temperature and the soldering time is corrected and calculated to obtain the optimization step length;

[0127] Construct a welding optimization function to optimize the heating temperature and welding time of the heated solder paste for welding, as shown in the following formula:

[0128] ;

[0129] Among them, wel is welding fitness, and are the melting weights and component weights, and The sum of is 1, is the melting rate of solder paste at the ith solder joint under the welding parameters, K is the mounting accuracy coefficient, is the temperature rise of the mounted component at the ith solder joint after welding according to the welding parameters, is the mounting distance information, It is the standard mounting distance between mounting components and circuit boards;

[0130] randomly generating a first heating temperature and a first welding time as first welding parameters;

[0131] Using the first welding parameter, combined with the plurality of solder paste amount information and mounting distance information, analyzing and obtaining the melting rate of the solder paste in the plurality of solder joints after welding and the rising temperature of the mounted components, and calculating and obtaining the first welding fitness based on the welding optimization function;

[0132] The welding heating temperature and welding time are continuously optimized until convergence, and welding parameters with the maximum welding fitness are output to obtain the optimal heating temperature and optimal welding time.

[0133] Furthermore, the system also includes a welding simulation analysis module to perform the following operation steps:

[0134] According to the mounting welding history data, a sample welding parameter set, a sample solder paste amount information set, a sample mounting distance information set, a sample melting rate information set and a sample rise temperature set are obtained;

[0135] The sample welding parameter set, the sample solder paste amount information set, and the sample mounting distance information set are used as sample inputs, and the sample melting rate information set and the sample rising temperature set are used as sample outputs to train and obtain a welding simulation channel;

[0136] Based on the welding simulation channel, a welding simulation analysis is performed on the first welding parameter in combination with the multiple solder paste amount information and mounting distance information to obtain multiple melting rates of the solder paste in the multiple solder joints and multiple rising temperatures of the mounted components.

[0137] Through the detailed description of the intelligent mounting and welding control method of an integrated circuit mainboard in the foregoing specification, those skilled in the art can clearly understand the intelligent mounting and welding control system of an integrated circuit mainboard in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the method part description.

[0138] In one embodiment, a computer device is provided. The computer device may be a server, and its internal structure diagram may be as follows: Figure 3 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus, wherein the processor of the computer device is used to provide computing and control capabilities; the memory of the computer device includes a non-volatile storage medium and an internal memory, wherein the non-volatile storage medium stores an operating system, a computer program and a database, and the internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium; the network interface of the computer device is used to communicate with an external terminal through a network connection. The computer program is executed by the processor to implement an intelligent mounting and welding control method for an integrated circuit motherboard.

[0139] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0140] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0141] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. An intelligent mounting and welding control method for an integrated circuit mainboard, It is characterized in that The method comprises: Collect images of multiple solder joints on a circuit mainboard to be mounted and soldered after solder paste is applied to obtain a mainboard image; Recognize the motherboard image, obtain multiple solder paste quantity information, and calculate and obtain multiple solder paste accuracy information in combination with the solder joint graphics of the multiple solder joints; According to the plurality of solder paste accuracy information, the mounting pressure of the mounting components mounted and soldered on the circuit mainboard is optimized to obtain the optimal mounting pressure, and the mounting of the mounting components is controlled; After the mounting is completed, the mounting image of the mounting component mounted on the circuit mainboard is collected, and identification is performed to obtain mounting distance information; According to the mounting distance information and the multiple solder paste accuracy information, the heating temperature and welding time of the heated solder paste for welding are optimized to obtain the optimal heating temperature and the optimal welding time, and perform welding control; Among them, according to the plurality of solder paste accuracy information, the mounting pressure of the mounting components mounted and soldered on the circuit mainboard is optimized, including: According to the plurality of solder paste accuracy information, a mounting pressure function is constructed to optimize the mounting pressure of mounting components on the circuit mainboard, as shown in the following formula: Where, pre is the pressure adaptability, M is the number of multiple welding points, is the weight of the ith solder joint assigned according to the size of multiple solder paste accuracy information. The size of the solder paste accuracy information is positively correlated with the weight. is the extent of mismatch between solder paste and solder joint pattern after the i-th solder joint is mounted under mounting pressure; A first mounting pressure is randomly generated, and the amplitude information of the mismatch between the solder paste and the solder joint image after the mounting of the multiple solder joints is predicted in combination with the multiple solder paste accuracy information. A first pressure fitness is calculated based on the mounting pressure function, wherein a solder paste mounting deviation identifier is constructed by using a sample mounting pressure set, multiple sample solder paste accuracy information sets, and multiple sample deviation amplitude information sets to predict the amplitude of the mismatch between the solder paste and the solder joint image after mounting; Continue to randomly generate and optimize the mounting pressure until convergence, and obtain the mounting pressure with the maximum pressure adaptability as the optimal mounting pressure; According to the mounting distance information and the plurality of solder paste accuracy information, the heating temperature and the soldering time for soldering the heated solder paste are optimized, including: According to the plurality of solder paste accuracy information, a preset optimization step length for optimizing the heating temperature and the soldering time is corrected and calculated to obtain the optimization step length; Construct a welding optimization function to optimize the heating temperature and welding time of the heated solder paste for welding, as shown in the following formula: Among them, wel is welding fitness, and are the melting weights and component weights, and The sum of is 1, is the melting rate of solder paste at the ith solder joint under the welding parameters, K is the mounting accuracy coefficient, is the temperature rise of the mounted component at the ith solder joint after welding according to the welding parameters, is the mounting distance information, It is the standard mounting distance between mounting components and circuit boards; randomly generating a first heating temperature and a first welding time as first welding parameters; Using the first welding parameter, combined with the plurality of solder paste amount information and mounting distance information, analyzing and obtaining the melting rate of the solder paste in the plurality of solder joints after welding and the rising temperature of the mounted components, and calculating and obtaining the first welding fitness based on the welding optimization function; The welding heating temperature and welding time are continuously optimized until convergence, and welding parameters with the maximum welding fitness are output to obtain the optimal heating temperature and optimal welding time.

2. The method according to claim 1, It is characterized in that The motherboard image is identified to obtain multiple solder paste amount information, including: Preprocessing the mainboard image; Based on the mounting soldering history data, a sample motherboard image set is obtained, and the solder paste amount of multiple solder joints in the sample main body image is identified and marked to obtain multiple sample solder paste amount information sets; The sample motherboard image set and multiple sample solder paste quantity information sets are used to construct a solder paste quantity identifier based on a convolutional neural network, and the preprocessed motherboard image is identified to obtain the multiple solder paste quantity information.

3. The method according to claim 1, It is characterized in that Combined with the solder joint patterns of the plurality of solder joints, a plurality of solder paste accuracy information is calculated and obtained, including: According to the solder joint patterns of the plurality of solder joints on the solder pad, index and obtain a plurality of standard solder paste quantity information; The plurality of solder paste accuracy information is obtained by calculation according to the plurality of standard solder paste amount information and the plurality of solder paste amount information.

4. The method according to claim 1, It is characterized in that Collecting the mounting image of the mounting component mounted on the circuit mainboard, identifying it, and obtaining mounting distance information, including: In a direction parallel to the circuit mainboard, collecting a mounting image of the mounting component mounted on the circuit mainboard; Preprocessing the mounting image; According to the mounting welding history data, a sample mounting image set is collected, and the distance between the circuit board and the mounting components in the sample mounting image is identified and marked to obtain a sample mounting distance information set; The sample mounting image set and the sample mounting distance information set are used to construct a mounting distance identifier based on a convolutional neural network, and the pre-processed mounting images are identified to obtain the mounting distance information.

5. The method according to claim 1, It is characterized in that The first welding parameter is used, combined with the plurality of solder paste amount information and mounting distance information, to analyze and obtain the melting rate of the solder paste in the plurality of solder joints after welding and the rising temperature of the mounted components, including: According to the mounting welding history data, a sample welding parameter set, a sample solder paste amount information set, a sample mounting distance information set, a sample melting rate information set and a sample rise temperature set are obtained; The sample welding parameter set, the sample solder paste amount information set, and the sample mounting distance information set are used as sample inputs, and the sample melting rate information set and the sample rising temperature set are used as sample outputs to train and obtain a welding simulation channel; Based on the welding simulation channel, a welding simulation analysis is performed on the first welding parameter in combination with the multiple solder paste amount information and mounting distance information to obtain multiple melting rates of the solder paste in the multiple solder joints and multiple rising temperatures of the mounted components.

6. An intelligent mounting and welding control system for integrated circuit motherboards, It is characterized in that A method for controlling intelligent mounting and welding of an integrated circuit mainboard according to any one of claims 1 to 5, the system comprising: An image acquisition module, the image acquisition module is used to acquire images of a plurality of solder joints on a circuit mainboard to be mounted and soldered after solder paste is applied thereto, to obtain an image of the mainboard; An image recognition module, the image recognition module is used to recognize the motherboard image, obtain multiple solder paste quantity information, and calculate multiple solder paste accuracy information in combination with the solder joint graphics of the multiple solder joints; A pressure optimization module, the pressure optimization module is used to optimize the mounting pressure of the mounting components mounted and soldered on the circuit mainboard according to the multiple solder paste accuracy information, obtain the optimal mounting pressure, and perform mounting control on the mounting components; A mounting distance acquisition module, which is used to collect a mounting image of the mounting component mounted on the circuit mainboard after the mounting is completed, perform identification, and obtain mounting distance information; A welding control module, wherein the welding control module is used to optimize the heating temperature and welding time of the heated solder paste for welding according to the mounting distance information and the accuracy information of the plurality of solder pastes, obtain an optimal heating temperature and an optimal welding time, and perform welding control; The system further includes an optimal mounting pressure acquisition module to perform the following operation steps: According to the plurality of solder paste accuracy information, a mounting pressure function is constructed to optimize the mounting pressure of mounting components on the circuit mainboard, as shown in the following formula: Where, pre is the pressure adaptability, M is the number of multiple welding points, is the weight of the ith solder joint assigned according to the size of multiple solder paste accuracy information. The size of the solder paste accuracy information is positively correlated with the weight. is the extent of mismatch between solder paste and solder joint pattern after the i-th solder joint is mounted under mounting pressure; A first mounting pressure is randomly generated, and the amplitude information of the mismatch between the solder paste and the solder joint image after the mounting of the multiple solder joints is predicted in combination with the multiple solder paste accuracy information. A first pressure fitness is calculated based on the mounting pressure function, wherein a solder paste mounting deviation identifier is constructed by using a sample mounting pressure set, multiple sample solder paste accuracy information sets, and multiple sample deviation amplitude information sets to predict the amplitude of the mismatch between the solder paste and the solder joint image after mounting; Continue to randomly generate and optimize the mounting pressure until convergence, and obtain the mounting pressure with the maximum pressure adaptability as the optimal mounting pressure; The system further includes an optimal welding parameter acquisition module to perform the following operation steps: According to the plurality of solder paste accuracy information, a preset optimization step length for optimizing the heating temperature and the soldering time is corrected and calculated to obtain the optimization step length; Construct a welding optimization function to optimize the heating temperature and welding time of the heated solder paste for welding, as shown in the following formula: Among them, wel is welding fitness, and are the melting weights and component weights, and The sum of is 1, is the melting rate of solder paste at the ith solder joint under the welding parameters, K is the mounting accuracy coefficient, is the temperature rise of the mounted component at the ith solder joint after welding according to the welding parameters, is the mounting distance information, It is the standard mounting distance between mounting components and circuit boards; randomly generating a first heating temperature and a first welding time as first welding parameters; Using the first welding parameter, combined with the plurality of solder paste amount information and mounting distance information, analyzing and obtaining the melting rate of the solder paste in the plurality of solder joints after welding and the rising temperature of the mounted components, and calculating and obtaining the first welding fitness based on the welding optimization function; The welding heating temperature and welding time are continuously optimized until convergence, and welding parameters with the maximum welding fitness are output to obtain the optimal heating temperature and optimal welding time.

7. A computer device comprising a memory and a processor, wherein the memory stores a computer program. It is characterized in that When the processor executes the computer program, the steps of the intelligent mounting and welding control method of an integrated circuit mainboard according to any one of claims 1 to 5 are implemented.

8. A computer-readable storage medium having a computer program stored thereon, It is characterized in that When the computer program is executed by a processor, the steps of the intelligent mounting and welding control method for an integrated circuit mainboard according to any one of claims 1 to 5 are implemented.

Citation Information

Patent Citations

  • Camera module production management monitoring analysis system based on machine vision

    CN116206381A

  • Optimization control method and system for SMT (Surface Mount Technology)

    CN117202532A

  • Intelligent wearable FPC continuous tin prevention production process

    CN117615568A

  • High-precision IGBT (Insulated Gate Bipolar Translator) mounting method, equipment and medium

    CN118488702A