Welding control method and apparatus

Abstract

The application discloses a welding control method and device, relates to the technical field of welding, and comprises the following steps: through solder paste dynamic compensation operation, the printing dynamics parameters are adjusted in real time, the reflow soldering parameters are optimized through a multi-factor coupling model, source control and process optimization of welding are realized, adaptive adjustment is performed on key process parameters of welding, welding defect rate is reduced, and welding consistency and stability are improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of welding, in particular to a welding control method and device. BACKGROUND

[0002] The welding quality of printed circuit board assemblies of high-density and high-reliability electronic products directly affects the stability and reliability of the products. With the development of products towards miniaturization and integration, the welding difficulty is greatly increased, and the probability of welding defects is larger.

[0003] At present, welding technology mainly relies on manual experience and semi-automatic means, and fixed parameters are usually used for solder paste printing, and the reflow soldering temperature curve is set based on empirical formula or historical data. However, the related technology uses static parameters for welding, which cannot adapt to the variable fluctuations in the actual production process, resulting in high welding defect rate, poor welding quality and poor welding reliability. SUMMARY

[0004] The present application provides a welding control method and device to at least solve the problem of high welding defect rate, poor welding quality and poor welding reliability in related technology.

[0005] The present application provides a welding control method, comprising:

[0006] obtaining a solder paste volume measurement value of a current pad;

[0007] According to the solder paste volume measurement value, a solder paste dynamic compensation operation is performed to determine the printing kinetics parameters of the subsequent pad according to the operation result; wherein the printing kinetics parameters include at least one of the template lifting angle, the squeegee pressure or the printing speed;

[0008] According to the printing kinetics parameters, the solder paste printing process is performed on the subsequent pad;

[0009] obtaining a thermal process influence parameter of the subsequent pad; wherein the thermal process influence parameter includes at least one of the solder paste type, the environment temperature and humidity, the printed circuit board thickness, the component distribution density and the previous thermal response feedback information;

[0010] According to the thermal process influence parameter and the preset multi-factor coupling model, the reflow soldering parameters of the subsequent pad are determined;

[0011] According to the reflow soldering parameters, the reflow soldering process is performed on the subsequent pad to obtain the completed pad.

[0012] The present application also provides an electronic device, comprising: a memory for storing a computer program; a processor for executing the computer program to implement the steps of any one of the above welding control methods.

[0013] Through the application, due to the solder paste dynamic compensation operation, the printing dynamics parameters are adjusted in real time, the reflow soldering parameters are optimized through the multi-factor coupling model, the source control and process optimization of welding are realized, the key process parameters of welding are adaptively adjusted, the welding defect rate is reduced, and the welding consistency and stability are improved. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to more clearly illustrate the embodiments of the present application, the drawings needed in the embodiments will be briefly introduced as follows. Obviously, the drawings in the following description are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0015] Figure 1 The application scenario diagram of the welding control method provided by the embodiments of the present application is shown.

[0016] Figure 2 The flowchart of the welding control method provided by the embodiments of the present application is shown.

[0017] Figure 3 The schematic diagram of another welding control method provided by the embodiments of the present application is shown.

[0018] Figure 4 The process schematic diagram of a multi-factor coupling model optimizing reflow soldering parameters provided by the embodiments of the present application is shown.

[0019] Figure 5 The schematic diagram of a defect propagation model prediction provided by the embodiments of the present application is shown.

[0020] Figure 6 The schematic diagram of a cross-process closed-loop control architecture provided by the embodiments of the present application is shown.

[0021] Figure 7 The structural schematic diagram of the welding control device provided by the embodiments of the present application is shown.

[0022] Figure 8 The structural schematic diagram of the electronic device provided by the embodiments of the present application is shown. DETAILED DESCRIPTION

[0023] The technical solutions in the embodiments of the present application will be described clearly and completely in combination with the drawings in the embodiments of the present application. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the protection scope of the present application.

[0024] It should be noted that in the description of the present application, the terms "comprising", "containing" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed or inherent to such a process, method, article or device. The terms "first", "second" and the like in the present application are used to distinguish similar objects, not to describe a specific order or sequence.

[0025] The inventor starts from the root problem of generating welding defects in related technologies, first introduces a dynamic compensation operation of solder paste, realizes the consistency of solder paste volume by adjusting the printing dynamics parameters in real time. Subsequently, a multi-factor coupling model is introduced to optimize the reflow soldering parameters, realizing the source control and process optimization of welding, and the key process parameters of welding are adaptively adjusted, which reduces the welding defect rate and improves the welding consistency and stability.

[0026] In order to enable those skilled in the art to better understand the present application, the present application will be further described in detail below in combination with the drawings and specific embodiments.

[0027] In combination with the specific application environment architecture or specific hardware architecture on which the welding control method is executed, the specific application environment architecture or specific hardware architecture is described here. Reference Figure 1 , Figure 1 The application scenario of the welding control method provided by the embodiments of the present application is shown in the figure. As Figure 1 shown, the scenario is a computer device, which includes a receiving device 101, a processor 102 and a display device 103.

[0028] It can be understood that the structure shown in the embodiments of the present application does not constitute a specific limitation on the welding control method. In other possible embodiments of the present application, the above-mentioned architecture can include more or fewer components than the figure, or combine certain components, or split certain components, or different component arrangement, which can be determined according to actual application scenario, which is not limited here. Figure 1 The components shown can be realized in hardware, software, or a combination of software and hardware.

[0029] In the specific implementation process, the receiving device 101 can be an input / output interface or a communication interface, which can obtain the solder paste volume measurement value of the current pad and the thermal process influence parameter of the subsequent pad.

[0030] The processor 102 can process the solder paste volume measurement value of the current pad to determine the printing dynamics parameters.

[0031] The processor 102 can process the thermal process influence parameter of the subsequent pad to determine the reflow soldering parameters.

[0032] The display device 103 can be used to display the printing dynamics parameters and reflow soldering parameters, etc.

[0033] The display device can also be a touch display screen, which is used to receive user instructions while displaying the above-mentioned content, so as to realize the interaction with the user.

[0034] It should be understood that the above-mentioned processor can be realized by reading the instructions in the memory by the processor and executing the instructions.

[0035] In addition, the network architecture and business scenarios described in the embodiments of the present application are used to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided by the embodiments of the present application. It can be known by those skilled in the art that, with the evolution of network architecture and the appearance of new business scenarios, the technical solutions provided by the embodiments of the present application are also applicable to similar technical problems.

[0036] Figure 2 The flowchart of the welding control method provided by the embodiments of the present application is shown in FIG. 1. Figure 2 The embodiments of the present application provide a welding control method, which is described in detail as follows.

[0037] S201: Obtain the solder paste volume measurement value of the current solder pad.

[0038] The solder pad is a metal area on a printed circuit board for welding elements, and is the target area of solder paste printing.

[0039] Optionally, the laser scanning is used to measure the solder paste volume to obtain the solder paste volume measurement value.

[0040] S202: Perform solder paste dynamic compensation operation according to the solder paste volume measurement value, so as to determine the printing dynamics parameters of the subsequent solder pad according to the operation result; wherein the printing dynamics parameters include at least one of the stencil lifting angle, the squeegee pressure or the printing speed.

[0041] The stencil lifting angle is the angle at which the solder paste printing stencil is lifted after printing, which affects the solder paste transfer efficiency.

[0042] The squeegee pressure is the pressure applied by the squeegee to the stencil during printing, which affects the solder paste volume transfer amount.

[0043] The printing speed is the speed at which the squeegee moves on the stencil, which affects the solder paste spreading uniformity.

[0044] Optionally, the step S202 comprises: obtaining a preset standard volume value; calculating a deviation amount according to the preset standard volume value and the solder paste volume measurement value; performing weighting processing on the deviation amount according to a preset proportional gain coefficient to obtain a volume adjustment amount; determining a target solder paste volume of the subsequent solder pad according to the volume adjustment amount and the preset standard volume value; and determining the printing dynamics parameter of the subsequent solder pad according to the target solder paste volume.

[0045] The proportional gain coefficient is used to amplify the influence of the deviation amount.

[0046] The deviation calculation between the volume measurement value and the standard volume value and the weighting processing by using the proportional gain coefficient can finally output the target solder paste volume to guide the parameter adjustment, thereby improving the accuracy of the parameter adjustment, reducing the welding defect rate, and improving the welding quality and welding reliability.

[0047] Optionally, the calculation formula of the target solder paste volume is as follows:

[0048]

[0049] In the formula, the target solder paste volume is represented by Vtarget; the preset standard volume value is represented by Vstandard; the solder paste volume measurement value is represented by Vmeasured; the preset proportional gain coefficient is represented by K.

[0050] Optionally, before the deviation amount is calculated according to the preset standard volume value and the solder paste volume measurement value, the method further comprises: obtaining historical good product data according to historical production data, and determining a solder paste volume measurement sequence and a printing dynamics parameter corresponding to the historical good product data; and performing learning training on the solder paste volume measurement sequence and the printing dynamics parameter corresponding to the historical good product data with the optimization target of minimizing the solder paste volume variance to obtain the preset proportional gain coefficient.

[0051] Optionally, the historical good product data can be historical good product data of different product models.

[0052] The preset proportional gain coefficient is obtained by training the historical good product data, which can be adaptively adjusted according to different product models to ensure that the control response is not overshoot or hysteresis, and to improve the printing consistency and process fault tolerance capability.

[0053] Optionally, after the learning training is performed on the solder paste volume measurement sequence and the printing dynamics parameter corresponding to the historical good product data with the optimization target of minimizing the solder paste volume variance to obtain the preset proportional gain coefficient, the method further comprises: obtaining viscosity data of the solder paste used to perform the solder paste printing process; and adjusting the preset proportional gain coefficient according to the viscosity data.

[0054] Optionally, the viscosity data of the solder paste used for performing the solder paste printing process is acquired by collecting the viscosity data of the solder paste on-line through a micro rheometer sensor installed in a solder paste tank.

[0055] By adaptively adjusting the proportional gain coefficient according to the real-time change of the viscosity data of the solder paste, the printing quality fluctuation caused by the change of the environmental temperature, the volatilization of the solder paste flux, etc. can be effectively compensated, and the printing consistency and the process fault tolerance capability are improved.

[0056] S203: performing the solder paste printing process on the subsequent pads according to the printing kinetics parameters.

[0057] Optionally, the real-time printing kinetics parameters during the execution of the solder paste printing process on the subsequent pads are monitored in real time, and the real-time printing kinetics parameters are adjusted according to the printing kinetics parameters to complete the solder paste printing process on the subsequent pads.

[0058] By monitoring and adjusting the printing kinetics parameters in real time during the printing process, the printing consistency and the process fault tolerance capability are further improved.

[0059] Optionally, the printing kinetics parameters are input to a printer controller, and the solder paste printing process is performed on the subsequent pads by the printer controller.

[0060] S204: acquiring a thermal process influence parameter of the subsequent pads; wherein the thermal process influence parameter includes at least one of a solder paste type, an environmental temperature and humidity, a printed circuit board thickness, a component distribution density, and a previous thermal response feedback information.

[0061] The solder paste type is distinguished according to the alloy composition of the solder paste.

[0062] Optionally, the method for acquiring the component distribution density includes: acquiring a computer-aided design (CAD) file; and counting the number of components in each region in the computer-aided design file to obtain the component distribution density.

[0063] Optionally, the environmental temperature and humidity can be collected by a workshop temperature and humidity sensor.

[0064] Optionally, the previous thermal response feedback information can include the peak temperature and the cooling rate of the solder joints of the current pads, etc.

[0065] It should be noted that the printed circuit board thickness is directly related to the heat conduction performance of the printed circuit board, and the printed circuit board thickness can be used to represent the heat conduction performance of the printed circuit board.

[0066] S205: determining the reflow soldering parameters of the subsequent pads according to the thermal process influence parameters and a preset multi-factor coupling model.

[0067] Optionally, the step S205 comprises: obtaining a preset multi-factor coupling model; wherein the preset multi-factor coupling model is trained by historical production data corresponding thermal process influence parameter samples and reflow soldering temperature sequence samples; inputting the thermal process influence parameter into the preset multi-factor coupling model to output a target reflow soldering temperature sequence according to the preset multi-factor coupling model; determining the reflow soldering parameter of the subsequent pad according to the target reflow soldering temperature sequence; wherein the reflow soldering parameter is a target temperature sequence set value.

[0068] The multi-factor coupling optimization of the thermal process influence parameter by the multi-factor coupling model can obtain the reflow soldering parameter, which can dynamically generate optimal temperature rising, holding and cooling curves for different products and different environments, and improve the reliability and consistency of the soldering.

[0069] Optionally, the preset multi-factor coupling model is a nonlinear mapping function, and its expression is:

[0070]

[0071] In the formula, M represents the type of solder paste; H represents the environmental temperature and humidity; E represents the thickness of the printed circuit board; D represents the component distribution density; and R(t) represents the previous thermal response feedback information. represents the target reflow soldering temperature sequence.

[0072] Optionally, the expression of the target reflow soldering temperature sequence is:

[0073]

[0074] In the formula, n is the number of temperature zones of the reflow soldering furnace; and T represents the target temperature of the i-th temperature zone. represents the target temperature of the i-th temperature zone.

[0075] Optionally, the determination of the reflow soldering parameter of the subsequent pad according to the target reflow soldering temperature sequence comprises: determining a target temperature sequence set value of the subsequent pad according to the target reflow soldering temperature sequence to determine the reflow soldering parameter.

[0076] Optionally, the target temperature sequence set value is a temperature sequence, which is one-to-one corresponding to the target reflow soldering temperature sequence.

[0077] Optionally, before obtaining the preset multi-factor coupling model, the method further comprises: obtaining an original multi-factor coupling model; wherein the original multi-factor coupling model comprises an input layer, a hidden layer and an output layer, the input layer comprises a plurality of nodes corresponding to the thermal process influence parameter; obtaining the historical production data corresponding thermal process influence parameter samples and reflow soldering temperature sequence samples; training the original multi-factor coupling model according to the historical production data corresponding thermal process influence parameter samples and reflow soldering temperature sequence samples to obtain the preset multi-factor coupling model.

[0078] Optionally, after obtaining the sample of the thermal process influence parameter corresponding to the historical production data and the sample of the reflow soldering temperature sequence, the method further comprises: obtaining a soldering quality result corresponding to the historical production data, wherein the soldering quality result comprises an Automated Optical Inspection (AOI) yield and a repair rate.

[0079] Correspondingly, the original multi-factor coupling model is trained according to the sample of the thermal process influence parameter corresponding to the historical production data and the sample of the reflow soldering temperature sequence to obtain the preset multi-factor coupling model, comprising: training the original multi-factor coupling model according to the sample of the thermal process influence parameter corresponding to the historical production data, the sample of the reflow soldering temperature sequence and the soldering quality result to obtain the preset multi-factor coupling model.

[0080] The original multi-factor coupling model is trained through the sample of the thermal process influence parameter corresponding to the historical production data and the sample of the reflow soldering temperature sequence, which can adaptively adjust according to different product models and improve the adaptability of the preset multi-factor coupling model.

[0081] Optionally, the original multi-factor coupling model can be a neural network structure, comprising an input layer, a hidden layer and an output layer; the input layer comprises a plurality of nodes corresponding to the thermal process influence parameters; the hidden layer comprises at least two full connection layers using a Rectified Linear Unit (ReLU) activation function; and the output layer is used to output a target temperature sequence, the target temperature sequence comprising target temperatures of each temperature zone.

[0082] It should be noted that the multi-factor coupling model is constructed by using a neural network structure, has nonlinear modeling capability and continuous learning capability, and can adapt to the influence caused by changes in the welding environment and updates of materials; the input of the input layer includes static parameters (printed circuit board thickness and component distribution density) and dynamic parameters (environmental temperature and humidity and previous thermal response feedback information), ensuring that the model output result has real-time and forward-looking characteristics; after the target temperature sequence is output, the reflow soldering furnace controller performs secondary adjustment according to the current furnace thermal inertia and actual temperature feedback to form a complete closed-loop control link, which can improve the controllability of the overall welding process.

[0083] S206: performing a reflow soldering process on the subsequent pads according to the reflow soldering parameters to obtain the soldered pads.

[0084] Optionally, step S206 comprises: inputting the reflow soldering parameters to a reflow soldering furnace controller to guide the heating strategy of the reflow soldering furnace controller; and performing the reflow soldering process on the subsequent pads by the reflow soldering furnace controller to obtain the soldered pads.

[0085] Optionally, the heating strategy of the reflow soldering furnace controller comprises: reflow soldering furnace temperature zone heating rate setting, peak temperature holding time adjustment, cooling section air speed control, and actual temperature feedback closed-loop adjustment.

[0086] Optionally, step S206 comprises: performing reflow soldering on the subsequent solder pad according to the reflow soldering parameters; collecting temperature change data of the preset area on the subsequent solder pad in real time; calculating thermal stress of the preset area on the subsequent solder pad according to the temperature change data; and adjusting the cooling rate and / or heating rate of the reflow soldering process according to the thermal stress until the soldering is completed on the solder pad.

[0087] Optionally, collecting the temperature change data of the preset area on the subsequent solder pad in real time comprises: synchronously collecting the temperature change data of the preset area at multiple preset monitoring points by the infrared temperature measurement module.

[0088] It should be noted that the installation position of the infrared temperature measurement module needs to be optimized in combination with the printed circuit board design layout and the distribution of the heat-sensitive area to ensure that the collected data is representative.

[0089] The thermal stress is a stress value of the solder pad area caused by temperature change, which may cause cracks or peeling.

[0090] By calculating the thermal stress of the preset area in real time and dynamically adjusting the cooling and / or heating rate, the stability of the reflow soldering process is improved.

[0091] Optionally, calculating the thermal stress of the preset area on the subsequent solder pad according to the temperature change data comprises: obtaining the solder paste type; matching the solder paste type from a preset database configuration file to obtain the solder joint material parameters; calculating the temperature difference of the preset area according to the temperature change data; and calculating the thermal stress of the preset area on the subsequent solder pad according to the temperature difference of the preset area and the solder joint material parameters.

[0092] By automatically matching the database configuration file according to the solder paste type, the thermal stress calculation adaptation under different alloy components is realized.

[0093] The solder joint material parameters comprise the elastic modulus of the solder paste and the thermal expansion coefficient of the solder paste.

[0094] Optionally, the formula for calculating the thermal stress of the preset area on the subsequent solder pad is:

[0095]

[0096] In the formula, represents the thermal stress of the preset area at time t; E represents the elastic modulus of the solder paste; and a represents the thermal expansion coefficient of the solder paste. represents the temperature difference of the preset area at time t.

[0097] Optionally, the temperature difference of the preset region at the preset time can be obtained by subtracting the maximum temperature from the minimum temperature in the temperature change data of the preset region.

[0098] Optionally, the cooling rate and / or the heating rate of the reflow soldering process are adjusted according to the thermal stress until a soldering completed pad is obtained, including: obtaining a material yield limit; comparing the material yield limit with the thermal stress; and if the thermal stress exceeds the material yield limit, reducing the heating rate and / or increasing the cooling rate.

[0099] Optionally, the material yield limit can be a yield limit value of the solder paste.

[0100] It should be noted that the thermal stress threshold is derived from a material manual or laboratory test data, and can also be set by the user, which can meet the product demand of different reliability levels.

[0101] Optionally, before calculating the thermal stress of the preset region on the subsequent pad according to the temperature change data, the method further includes: detecting whether the trend of the temperature change data is a temperature rising trend; if the trend is a temperature rising trend, obtaining temperature rising data of a first preset time period; inputting the temperature rising data of the first preset time period into a preset time series prediction model to output a peak thermal stress of a second preset time period; obtaining a material yield limit; and if the peak thermal stress exceeds the material yield limit, reducing the heating rate and / or increasing the cooling rate.

[0102] By using the preset time series prediction model, the control of the heating rate and the cooling rate can be started in advance, the thermal stress is inhibited in advance, and the soldering stability is improved.

[0103] The preset time series prediction model can be a time series prediction model based on a long short-term memory network.

[0104] Optionally, before inputting the temperature rising data of the first preset time period into the preset time series prediction model to output the peak thermal stress of the second preset time period, the method further includes: obtaining an original time series prediction model; the original time series prediction model includes an input layer, a hidden layer, and an output layer, the hidden layer includes a plurality of long short-term memory network units and a random inactivation layer; obtaining temperature rising data samples of the first preset time period and peak thermal stress samples of the second time period corresponding to historical production data; and training the original time series prediction model according to the temperature rising data samples of the first preset time period and the peak thermal stress samples of the second time period corresponding to the historical production data to obtain the preset time series prediction model.

[0105] Optionally, after step S206, the method further comprises: performing defect identification processing on the soldered pads to obtain a defect image; predicting a defect occurrence probability of a next batch of pads after the soldered pads according to the defect image; and triggering a reverse adjustment mechanism to adjust printing dynamics parameters and / or reflow parameters of the next batch of pads if the defect occurrence probability is greater than a preset defect probability threshold.

[0106] It should be noted that the preset defect probability threshold can be based on historical yield rate statistics and process maturity evaluation, or can be set by the user according to different product quality levels.

[0107] By predicting the defect occurrence probability of the next batch of pads after the soldered pads and the reverse adjustment mechanism, the defect diffusion risk is reduced, and the reliability of the soldering is improved.

[0108] Optionally, predicting the defect occurrence probability of the next batch of pads after the soldered pads according to the defect image comprises: performing feature extraction processing on the defect image to obtain defect features; wherein the feature extraction processing comprises at least one of edge detection processing, grayscale analysis processing, and morphological processing, and the defect features comprise at least one of bridging features, empty solder features, and offset features; and performing prediction processing according to the defect features and a preset defect propagation model to obtain the defect occurrence probability of the next batch of pads; wherein the preset defect propagation model is obtained by training historical defect features and defect occurrence probability labels of the next batch of pads corresponding to the historical defect features.

[0109] By performing feature extraction processing on the defect image to obtain defect features, and combining the preset defect propagation model to perform prediction processing, the defect occurrence probability of different types of defects can be obtained, providing fine data for the subsequent reverse adjustment mechanism.

[0110] Optionally, the edge detection processing can use a Canny edge detection algorithm, the grayscale analysis processing can use an Otsu grayscale segmentation algorithm, and the morphological processing can use morphological opening and closing operations.

[0111] Optionally, performing feature extraction processing on the defect image to obtain defect features comprises: performing feature extraction processing on the defect image to obtain bridging defects, empty solder defects, and offset defects; performing weighted statistical processing on the bridging defects, the empty solder defects, and the offset defects respectively to obtain a bridging weight coefficient, an empty solder weight coefficient, and an offset weight coefficient; obtaining a weighted statistical feature vector according to the bridging weight coefficient, the empty solder weight coefficient, and the offset weight coefficient; and determining the defect features according to the weighted statistical feature vector.

[0112] The bridging defect represents a short circuit of the solder paste between adjacent pads; the empty defect represents that the volume of the solder paste is insufficient to form a solder joint; and the offset defect represents that the position of the solder paste deviates from the center of the pad.

[0113] Optionally, the expression of the preset defect propagation model is:

[0114]

[0115] In the formula, represents the defect occurrence probability of the next batch of pads; represents the historical defect occurrence probability; represents the defect feature, that is, a weighted statistical feature vector; γ is a historical defect influence weight coefficient, and β is a defect feature influence weight coefficient.

[0116] It should be noted that the preset defect propagation model introduces a weighted fusion mechanism of the historical defect occurrence probability and the current defect feature, so that the prediction result can reflect both long-term trends and sudden abnormalities, and the early warning sensitivity and prediction accuracy of the defect propagation model are improved.

[0117] Optionally, the weighted statistical feature vector can be a three-dimensional vector, and the components respectively represent weight coefficients corresponding to the bridging defect, the empty defect, and the offset defect.

[0118] It should be noted that the weighted statistical feature vector can also be set according to industry standards or according to user-specific requirements, and has high flexibility.

[0119] It should be noted that the preset defect propagation model is used to determine whether the subsequent batch has a risk of rising defect trend, and serves as a trigger basis for process parameter adjustment.

[0120] Optionally, after predicting the defect occurrence probability of the next batch of pads after the pads are welded according to the defect image, the method further includes: dividing the printed circuit board according to a preset spatial division rule to obtain a plurality of grids; counting the number of defects in the plurality of grids; calculating Shannon entropy according to the number of defects in the plurality of grids; and identifying local process abnormalities according to the Shannon entropy.

[0121] Optionally, dividing the printed circuit board according to the preset spatial division rule includes: dividing the printed circuit board into M×N grids.

[0122] Optionally, the formula for calculating the Shannon entropy is:

[0123]

[0124] In the formula, represents the proportion of the number of defects in the (i, j) grid to the total number of defects of the printed circuit board; represents the Shannon entropy; b can be 2 or e.

[0125] Optionally, the reverse adjustment mechanism comprises: re-performing the dynamic compensation operation of the solder paste according to the solder paste volume measurement value to obtain the printing kinetics parameters; re-calculating the reflow soldering parameters by using a preset multi-factor coupling model; adjusting the cooling strategy according to the reflow soldering parameters. Performing the solder paste printing process according to the printing kinetics parameters; performing the reflow soldering process according to the reflow soldering parameters and the cooling strategy. Recording the comparison data of the printing kinetics parameters and the reflow soldering parameters before and after this adjustment, for subsequent optimization and process window updating.

[0126] By preferentially adjusting the printing kinetics parameters, the reflow soldering parameters and the cooling strategy that have the most significant impact on defects, and by cross-process linkage control closed loop, the defects can be quickly responded and inhibited from spreading.

[0127] Optionally, the reverse adjustment mechanism further comprises: parameter rollback suppression logic, wherein the parameter rollback suppression logic comprises: if the predicted defect probability fails to fall below the preset defect probability threshold after two consecutive reverse adjustments, switching to a conservative control mode based on the historical optimal process window, and triggering an artificial intervention alarm.

[0128] Optionally, the reverse adjustment mechanism further comprises: obtaining historical good product data of the historical process window; performing dimension reduction processing on the historical good product data by using a principal component analysis algorithm to obtain key influencing factors; updating the reference process window according to the key influencing factors to obtain the latest process window, wherein the reference process window is obtained by performing filtering processing, statistical calibration processing and hard constraint correction processing on the historical process window through the key influencing factors.

[0129] By combining the principal component analysis algorithm with the process window updating mechanism, the process can be kept stable while having certain tolerance adjustment capability, thereby improving the adaptability to small-batch multi-variety production and the process migration efficiency.

[0130] Optionally, the statistical calibration processing comprises: calculating the parameter average value and the standard deviation of the key influencing factors of the historical good product data; and determining the tolerance range of the reference process window according to the parameter average value and the standard deviation.

[0131] Optionally, updating the reference process window according to the key influencing factors to obtain the latest process window comprises: obtaining a historical good product data set according to the key influencing factors, wherein the historical good product data set contains historical qualified parameter values of the key factors, corresponding yield data and process fluctuation records; and updating the reference process window according to the historical good product data set to obtain the latest process window.

[0132] Optionally, the update formula of the reference process window is:

[0133]

[0134] wherein, represents the latest process window; represents the reference process window; represents the historical good product data set; and δ represents an expansion coefficient, which is adaptively adjusted according to real-time good product rate fluctuations; represents a machine learning-based feature quantization function, which is used to analyze the process stability (such as the fluctuation standard deviation of a key parameter, the correlation between the good product rate and the parameter) of the historical good product data set, and outputs a dimensionless adjustment basis value.

[0135] The welding control method provided by the embodiments of the present application realizes source control and process optimization of welding by means of dynamic compensation operation of solder paste, real-time adjustment of printing kinetics parameters, and optimization of a multi-factor coupling model for reflow soldering parameters, and adaptively adjusts key process parameters of welding, thereby reducing the defect rate of welding and improving the consistency and stability of welding.

[0136] Figure 3 Another schematic diagram of a welding control method provided by the embodiments of the present application is shown in FIG. 6, and the method comprises the following steps. Figure 3

[0137] A solder paste volume measurement value of a current pad is obtained.

[0138] According to the solder paste volume measurement value, a dynamic compensation operation of the solder paste is performed to determine printing kinetics parameters of a subsequent pad according to an operation result. According to the printing kinetics parameters, a solder paste printing process is performed on the subsequent pad.

[0139] A thermal process influence parameter of the subsequent pad is obtained. According to the thermal process influence parameter and a preset multi-factor coupling model, a multi-factor coupling reflow soldering optimization process is performed to determine reflow soldering parameters of the subsequent pad.

[0140] According to the reflow soldering parameters, a reflow soldering process is performed on the subsequent pad. The thermal stress of a preset area on the subsequent pad is monitored in real time. According to the thermal stress, the cooling rate and / or the heating rate of the reflow soldering process are adjusted until a welding completed pad is obtained.

[0141] Optionally, the real-time monitoring of the thermal stress of the preset area on the subsequent pad comprises: real-time acquisition of temperature change data of the preset area on the subsequent pad; and calculation of the thermal stress of the preset area on the subsequent pad according to the temperature change data.

[0142] Defect identification processing is performed on the welding completed pad to obtain a defect image. Feature extraction processing is performed on the defect image to obtain defect features.

[0143] According to the defect features and a preset defect propagation model, prediction processing is performed to obtain a defect occurrence probability of a next batch of pads. ​

[0144] If the defect occurrence probability is greater than the preset defect probability threshold, a reverse adjustment mechanism is triggered to adjust the printing kinetics parameters and / or reflow soldering parameters of the next batch of pads.

[0145] Figure 4 A process schematic diagram of a multi-factor coupling model for optimizing reflow soldering parameters provided by an embodiment of the present application is shown in FIG. 1. As shown in FIG. 1, the thermal process influence parameters are input into a preset multi-factor coupling model to output a target reflow soldering temperature sequence according to the preset multi-factor coupling model; the reflow soldering parameters of the subsequent pads are determined according to the target reflow soldering temperature sequence; and the reflow soldering parameters are input into a reflow soldering furnace controller. Figure 4

[0146] Optionally, the thermal process influence parameters include at least one of a solder paste type, an ambient temperature and humidity, a printed circuit board thickness, a component distribution density, and previous thermal response feedback information.

[0147] Figure 5 A schematic diagram of a defect propagation model prediction provided by an embodiment of the present application is shown in FIG. 2. As shown in FIG. 2, the details are as follows: Figure 5

[0148] The pads after soldering are subjected to defect identification processing to obtain a defect image.

[0149] The defect image is subjected to feature extraction processing to obtain defect features.

[0150] The defect features and the historical defect occurrence probability are input into a preset defect propagation model for prediction processing to obtain a defect occurrence probability of the next batch of pads.

[0151] If the defect occurrence probability is greater than the preset defect probability threshold, a reverse adjustment mechanism is triggered to adjust the printing kinetics parameters and / or reflow soldering parameters of the next batch of pads.

[0152] Figure 6 A schematic diagram of a cross-process closed-loop control architecture provided by an embodiment of the present application is shown in FIG. 3. As shown in FIG. 3, the architecture includes a solder paste volume detection module, a printer controller, a reflow soldering furnace controller, a defect identification module, and a trend prediction module. Figure 6

[0153] The cross-process closed-loop control process includes:

[0154] The solder paste volume measurement value of the current pads is obtained by the solder paste volume detection module.

[0155] The solder paste dynamic compensation operation is performed according to the solder paste volume measurement value to determine the printing kinetics parameters of the subsequent pads according to the operation result.

[0156] ​​​inputting the printing kinetics parameters into a printer controller; performing a solder paste printing process on the subsequent pads by the printer controller.

[0157] determining reflow soldering parameters of the subsequent pads according to the thermal process influence parameters and a preset multi-factor coupling model; inputting the reflow soldering parameters into a reflow soldering furnace controller for guiding a heating strategy of the reflow soldering furnace controller.

[0158] collecting temperature variation data of a preset region on the subsequent pads in real time; calculating thermal stress according to the temperature variation data; adjusting a cooling rate and / or a heating rate of the reflow soldering process according to the thermal stress until a pad with completed soldering is obtained.

[0159] performing defect identification processing on the pad with completed soldering by a defect identification module to obtain a defect image; performing feature extraction processing on the defect image to obtain defect features.

[0160] performing prediction processing according to the defect features and a preset defect propagation model by a trend prediction module to obtain a defect occurrence probability of the next batch of pads.

[0161] if the defect occurrence probability is greater than a preset defect probability threshold, triggering a reverse adjustment mechanism to adjust the printing kinetics parameters of the printer controller and / or the reflow soldering parameters of the reflow soldering furnace controller of the next batch of pads.

[0162] Embodiments of the present application provide a soldering control system, which comprises a printing compensation module, a temperature optimization module, a thermal stress monitoring module, a defect identification module, a trend prediction module and a parameter adjustment module.

[0163] The printing compensation module is configured to obtain a solder paste volume measurement value on the pads, compare the volume measurement value with a standard volume value, calculate a target solder paste volume of the subsequent pads, and adjust at least one of a stencil lift angle, a squeegee pressure or a printing speed according to the target solder paste volume, so as to realize dynamic compensation of the solder paste printing quality.

[0164] The temperature optimization module is configured to input a solder paste type, an ambient temperature and humidity, a printed circuit board thickness, a component distribution density and a previous thermal response feedback information into a preset multi-factor coupling model after the solder paste printing is completed, output reflow soldering parameters of the subsequent pads, and send the reflow soldering parameters to a reflow soldering furnace controller for guiding a heating strategy of each temperature zone in the reflow soldering process.

[0165] The thermal stress monitoring module is configured to collect temperature variation data of a preset region during the reflow soldering process, calculate thermal stress of the preset region in combination with a solder joint material parameter, and automatically adjust a cooling rate and / or a heating rate when the thermal stress value exceeds a material yield limit, so as to prevent occurrence of solder joint cracks or structural damage.

[0166] The defect identification module is configured to, after obtaining the completed soldering pads, perform defect identification processing on the completed soldering pads to obtain defect images, and perform feature extraction processing on the defect images to extract a weighted statistical feature vector of defects such as bridging, missing soldering and offset, as a data basis for subsequent defect trend analysis.

[0167] The trend prediction module is configured to construct a defect propagation model by combining the historical defect occurrence probability and the weighted statistical feature vector to predict a defect occurrence probability of the next batch of soldering pads, and trigger a reverse adjustment mechanism when the predicted defect occurrence probability of the next batch of soldering pads exceeds a preset defect probability threshold.

[0168] The parameter adjustment module is configured to execute the reverse adjustment mechanism.

[0169] Through the description of the above embodiments, those skilled in the art can clearly understand that the method according to the above embodiments can be realized by means of software and necessary general hardware platforms, of course, it can also be realized by hardware, but in many cases, the former is a better embodiment.

[0170] Figure 7 The structure schematic diagram of the welding control device provided by the embodiments of the present application is shown in FIG. 1. Figure 7 As shown in FIG. 1, the embodiments of the present application further provide a welding control device, which comprises a first acquisition module 701, an operation module 702, a first execution module 703, a second acquisition module 704, a determination module 705 and a second execution module 706.

[0171] The first acquisition module 701 is configured to acquire a solder paste volume measurement value of a current soldering pad.

[0172] The operation module 702 is configured to perform solder paste dynamic compensation operation according to the solder paste volume measurement value, so as to determine printing dynamics parameters of a subsequent soldering pad according to an operation result; wherein the printing dynamics parameters comprise at least one of a stencil lifting angle, a squeegee pressure or a printing speed.

[0173] The first execution module 703 is configured to perform a solder paste printing process on the subsequent soldering pad according to the printing dynamics parameters.

[0174] The second acquisition module 704 is configured to acquire a thermal process influence parameter of the subsequent soldering pad; wherein the thermal process influence parameter comprises at least one of a solder paste type, an environment temperature and humidity, a printed circuit board thickness, a component distribution density and a previous thermal response feedback information.

[0175] The determination module 705 is configured to determine reflow soldering parameters of the subsequent soldering pad according to the thermal process influence parameter and a preset multi-factor coupling model.

[0176] The second execution module 706 is configured to perform the reflow soldering process on the subsequent pad according to the reflow soldering parameter, so as to obtain the pad with completed soldering.

[0177] In a possible implementation, the operation module 702 is specifically configured to: obtain a preset standard volume value; calculate a deviation amount according to the preset standard volume value and the solder paste volume measurement value; perform weighting processing on the deviation amount according to a preset proportional gain coefficient, so as to obtain a volume adjustment amount; determine a target solder paste volume of the subsequent pad according to the volume adjustment amount and the preset standard volume value; and determine the printing kinetics parameter of the subsequent pad according to the target solder paste volume.

[0178] In a possible implementation, the soldering control device further includes:

[0179] The training module is configured to: obtain historical good product data according to historical production data, and determine a solder paste volume measurement sequence and a printing kinetics parameter corresponding to the historical good product data; and perform learning training on the solder paste volume measurement sequence and the printing kinetics parameter corresponding to the historical good product data, so as to obtain the preset proportional gain coefficient, with minimization of a solder paste volume variance as an optimization target.

[0180] In a possible implementation, the soldering control device further includes:

[0181] The adjustment module is configured to: obtain viscosity data of the solder paste used to perform the solder paste printing process; and adjust the preset proportional gain coefficient according to the viscosity data.

[0182] In a possible implementation, the determination module 705 is specifically configured to: obtain a preset multi-factor coupling model; wherein the preset multi-factor coupling model is obtained by training a thermal process influence parameter sample and a reflow soldering temperature sequence sample corresponding to historical production data; input the thermal process influence parameter into the preset multi-factor coupling model, so as to output a target reflow soldering temperature sequence according to the preset multi-factor coupling model; and determine the reflow soldering parameter of the subsequent pad according to the target reflow soldering temperature sequence; wherein the reflow soldering parameter is a target temperature sequence setting value.

[0183] In a possible implementation, the determination module 705 is specifically configured to: obtain an original multi-factor coupling model; wherein the original multi-factor coupling model includes an input layer, a hidden layer and an output layer, the input layer includes a plurality of nodes corresponding to the thermal process influence parameter; obtain a thermal process influence parameter sample and a reflow soldering temperature sequence sample corresponding to the historical production data; and train the original multi-factor coupling model according to the thermal process influence parameter sample and the reflow soldering temperature sequence sample corresponding to the historical production data, so as to obtain the preset multi-factor coupling model.

[0184] In a possible implementation, the second execution module 706 is specifically configured to: perform a reflow soldering process on the subsequent pads according to the reflow soldering parameters; collect temperature change data of the preset area on the subsequent pads in real time; calculate thermal stress of the preset area on the subsequent pads according to the temperature change data; and adjust a cooling rate and / or a heating rate of the reflow soldering process according to the thermal stress until the pads are completely soldered.

[0185] In a possible implementation, the welding control device further includes:

[0186] The prediction module is configured to perform defect identification processing on the pads that are completely soldered to obtain defect images, predict a defect occurrence probability of the next batch of pads after the pads that are completely soldered according to the defect images, and trigger a reverse adjustment mechanism to adjust the printing kinetic parameters and / or the reflow soldering parameters of the next batch of pads if the defect occurrence probability is greater than a preset defect probability threshold.

[0187] In a possible implementation, when predicting the defect occurrence probability of the next batch of pads after the pads that are completely soldered according to the defect images, the prediction module is specifically configured to: perform feature extraction processing on the defect images to obtain defect features, wherein the feature extraction processing includes at least one of edge detection processing, gray scale analysis processing, and morphological processing, and the defect features include at least one of a bridging feature, a missing solder feature, and a misalignment feature; and perform prediction processing according to the defect features and a preset defect propagation model to obtain the defect occurrence probability of the next batch of pads, wherein the preset defect propagation model is obtained by training historical defect features and defect occurrence probability labels of the next batch of pads corresponding to the historical defect features.

[0188] The features of the embodiments of the welding control device can be referred to the related descriptions of the embodiments of the welding control method, which will not be repeated here.

[0189] Figure 8 The structure schematic diagram of the electronic device provided in the embodiments of the present application is shown in FIG. 1. As shown in FIG. 1, the electronic device provided in the embodiments of the present application includes at least one processor 801 and a memory 802. Optionally, the electronic device further includes a communication component 803. The processor 801, the memory 802, and the communication component 803 are connected through a bus 804. Figure 8

[0190] In the specific implementation process, the at least one processor 801 executes the computer execution instructions stored in the memory 802, so that the at least one processor 801 performs the welding control method embodiments described above.

[0191] The specific implementation process of the processor 801 can be referred to the method embodiments described above, which has similar implementation principles and technical effects, and will not be repeated here. ​

[0192] In the above embodiments, it should be understood that the processor can be a central processing unit (CPU), and can also be other general-purpose processors, digital signal processors (DSP), application specific integrated circuits (ASIC), etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor and the like. The steps of the method disclosed in combination with the application can be directly embodied as hardware processor execution, or executed by a combination of hardware and software modules in the processor.

[0193] The memory can include a random access memory (RAM), and can also include a non-volatile memory (NVM), such as at least one disk memory.

[0194] The bus can be an industry standard architecture (ISA) bus, a peripheral component (PCI) bus, an extended industry standard architecture (EISA) bus, or the like. The bus can be divided into an address bus, a data bus, a control bus, and the like. For ease of representation, the bus in the drawings of the present application does not limit to only one bus or one type of bus.

[0195] The embodiments of the present application also provide a computer readable storage medium, which stores a computer program, wherein the computer program is configured to execute the steps in any of the above welding control method embodiments when running.

[0196] In an exemplary embodiment, the above computer readable storage medium can include, but is not limited to, a U disk, a read-only memory (ROM), a random access memory (RAM), a mobile hard disk, a magnetic disk or an optical disk, and various media that can store computer programs.

[0197] The embodiments of the present application also provide a computer program product, which includes a computer program, and the computer program is executed by a processor to implement the steps in any of the above welding control method embodiments.

[0198] The embodiment of the present application further provides another computer program product, comprising a nonvolatile computer readable storage medium, the nonvolatile computer readable storage medium stores a computer program, the computer program is executed by a processor to implement the steps in any of the above welding control method embodiments.

[0199] Those skilled in the art will further appreciate that the functions of the examples described herein-based units and algorithm steps can be implemented using electronic hardware, computer software, or any combination thereof. To clearly illustrate this interchangeability of hardware and software, various examples have been described herein in terms of their general functionality. Whether such functionality is implemented as hardware or software depends on the particular application and design constraints imposed on the overall system. Skilled artisans can implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.

[0200] The above describes in detail a welding control method and device provided by the present application. The principles and implementation manners of the present application are described herein by applying specific examples, and the above description of the examples is only for helping to understand the method of the present application and its core idea. It should be pointed out that, for those skilled in the art, some improvements and modifications can be made to the present application without departing from the principles of the present application, and these improvements and modifications also fall within the protection scope of the claims of the present application.

Claims

1. A welding control method characterized by, The method comprises the following steps: obtaining a solder paste volume measurement value of a current pad; performing a solder paste dynamic compensation operation according to the solder paste volume measurement value to determine a printing dynamics parameter of a subsequent pad according to an operation result, comprising: obtaining a preset standard volume value; calculating a deviation amount according to the preset standard volume value and the solder paste volume measurement value; performing weighted processing on the deviation amount according to a preset proportional gain coefficient to obtain a volume adjustment amount; determining a target solder paste volume of the subsequent pad according to the volume adjustment amount and the preset standard volume value; determining the printing dynamics parameter of the subsequent pad according to the target solder paste volume; wherein the printing dynamics parameter comprises at least one of a stencil lifting angle, a squeegee pressure or a printing speed; performing a solder paste printing process on the subsequent pad according to the printing dynamics parameter; obtaining a thermal process influence parameter of the subsequent pad; wherein the thermal process influence parameter comprises at least one of a solder paste type, an environmental temperature and humidity, a printed circuit board thickness, a component distribution density and a previous thermal response feedback information; determining a reflow soldering parameter of the subsequent pad according to the thermal process influence parameter and a preset multi-factor coupling model, comprising: obtaining a preset multi-factor coupling model; wherein the preset multi-factor coupling model is trained by a thermal process influence parameter sample corresponding to historical production data and a reflow soldering temperature sequence sample; inputting the thermal process influence parameter into the preset multi-factor coupling model to output a target reflow soldering temperature sequence according to the preset multi-factor coupling model; determining the reflow soldering parameter of the subsequent pad according to the target reflow soldering temperature sequence; wherein the reflow soldering parameter is a target temperature sequence setting value; performing a reflow soldering process on the subsequent pad according to the reflow soldering parameter to obtain a soldered pad, comprising: performing a reflow soldering process on the subsequent pad according to the reflow soldering parameter; collecting temperature change data of a preset area on the subsequent pad in real time; calculating a thermal stress of the preset area on the subsequent pad according to the temperature change data; adjusting a cooling rate and / or a heating rate of the reflow soldering process according to the thermal stress until a soldered pad is obtained.

2. The method of claim 1, wherein, Before the step of performing weighted processing on the deviation amount according to a preset proportional gain coefficient to obtain a volume adjustment amount, the method further comprises the following steps: obtaining historical good product data from historical production data and determining a solder paste volume measurement sequence and a printing dynamics parameter corresponding to the historical good product data; learning and training the solder paste volume measurement sequence and the printing dynamics parameter corresponding to the historical good product data as an optimization target of minimizing a solder paste volume variance to obtain the preset proportional gain coefficient.

3. The method of claim 2, wherein, After the step of learning and training the solder paste volume measurement sequence and the printing dynamics parameter corresponding to the historical good product data as an optimization target of minimizing a solder paste volume variance to obtain the preset proportional gain coefficient, the method further comprises the following steps: obtaining viscosity data of a solder paste used for performing a solder paste printing process; adjusting the preset proportional gain coefficient according to the viscosity data.

4. The method of claim 1, wherein, Before the step of obtaining a preset multi-factor coupling model, the method further comprises the following steps: obtain an original multi-factor coupling model; wherein the original multi-factor coupling model comprises an input layer, a hidden layer and an output layer, the input layer comprises a plurality of nodes corresponding to the thermal process influence parameters; obtain the historical production data corresponding to the thermal process influence parameter samples and the reflow soldering temperature sequence samples; train the original multi-factor coupling model according to the historical production data corresponding to the thermal process influence parameter samples and the reflow soldering temperature sequence samples, to obtain a preset multi-factor coupling model.

5. The method according to any one of claims 1 to 4, characterized in that, After the reflow soldering process is performed on the subsequent pads according to the reflow soldering parameters to obtain the soldered pads, the method further comprises: performing defect identification processing on the soldered pads to obtain a defect image; predicting a defect occurrence probability of a next batch of pads after the soldered pads according to the defect image; if the defect occurrence probability is greater than a preset defect probability threshold, triggering a reverse adjustment mechanism to adjust the printing kinetics parameters and / or the reflow soldering parameters of the next batch of pads.

6. The method of claim 5, wherein, The method of predicting the defect occurrence probability of the next batch of pads after the soldered pads according to the defect image comprises: performing feature extraction processing on the defect image to obtain defect features; wherein the feature extraction processing comprises at least one of edge detection processing, gray scale analysis processing and morphological processing, and the defect features comprise at least one of a bridging feature, a missing solder feature and a misalignment feature; performing prediction processing according to the defect features and a preset defect propagation model to obtain the defect occurrence probability of the next batch of pads; wherein the preset defect propagation model is trained by historical defect features and defect occurrence probability labels of the next batch of pads corresponding to the historical defect features.

7. An electronic device, comprising: comprise: a memory for storing a computer program; a processor for executing the computer program to implement the steps of the soldering control method according to any one of claims 1 to 6.

Images