PCB reflow soldering temperature field prediction method and device and computing equipment

Abstract

The invention discloses a PCB reflow soldering temperature field prediction method and device and computing equipment. The method comprises the following steps: acquiring the temperature and environmental parameters of a key position of a PCB in a reflow oven to obtain real-time data; constructing a mechanism model according to the real-time data; solving the mechanism model to obtain a prediction result; and outputting a prediction result. By implementing the method provided by the invention, efficient prediction and optimization of the temperature field in the PCB reflow soldering process can be realized, and the efficiency and accuracy of soldering quality control are improved.

Description

TECHNICAL FIELD

[0001] The application relates to a temperature field prediction method, in particular to a PCB reflow soldering temperature field prediction method, device and computing equipment. BACKGROUND

[0002] At present, in the electronic manufacturing industry, the temperature curve in the PCB (Printed Circuit Board) soldering process is mainly obtained by means of repeated experimental testing. Specifically, the product is soldered for multiple times under different conditions, and a temperature meter is used to record the temperature change, so as to obtain the corresponding temperature curve. This method is time-consuming and laborious, and is difficult to adapt to the modern rapid change of production demand and high-precision control requirements. In addition, due to the complex and changeable actual production environment, the data obtained by simply relying on experimental testing often cannot comprehensively and accurately reflect all possible situations, thereby affecting the quality and reliability of the final product.

[0003] Therefore, it is necessary to design a new method to realize efficient prediction and optimization of the temperature field in the PCB reflow soldering process, and improve the efficiency and accuracy of the soldering quality control. SUMMARY

[0004] The application aims to overcome the defects of the prior art and provide a PCB reflow soldering temperature field prediction method, device and computing equipment.

[0005] To achieve the above-mentioned purpose, the application adopts the following technical scheme: a PCB reflow soldering temperature field prediction method, comprising:

[0006] Obtaining the temperature of the key position of the PCB in the reflow oven and the environmental parameters to obtain real-time data;

[0007] Constructing a mechanism model according to the real-time data;

[0008] Solving the mechanism model to obtain a prediction result;

[0009] Outputting the prediction result.

[0010] The further technical scheme is that the key position includes a soldering point, a BGA bottom and a position between the PCB and a thermal sensor, and the distance between the PCB and the thermal sensor meets the requirements.

[0011] The further technical scheme is that the environmental parameters include a multi-dimensional input matrix composed of equipment state data and external environment data; the equipment state data includes a conveyor belt speed and a heating zone power; and the external environment data includes humidity and workshop temperature.

[0012] The further technical scheme is that the mechanism model is constructed according to the real-time data, comprising:

[0013] The length of the reflow furnace is equally divided to obtain a plurality of nodes, and the temperature of each node is determined according to the real-time data to obtain a spatial discretization result;

[0014] Based on the spatial discretization result, each node is discretized in time using an implicit format to obtain a discrete equation;

[0015] A three-diagonal linear equation group is generated according to the discrete equation to obtain a mechanism model.

[0016] Further technical solutions thereof are as follows: the mechanism model is solved to obtain a prediction result, comprising:

[0017] The mechanism model is solved by applying a Thomas algorithm to obtain a prediction result.

[0018] Further technical solutions thereof are as follows: after the mechanism model is solved by applying the Thomas algorithm to obtain the prediction result, further comprising:

[0019] The temperature field is updated by iteration until a steady state is reached or a predetermined time endpoint is met.

[0020] Further technical solutions thereof are as follows: the mechanism model is dynamically calibrated by machine learning.

[0021] The application further provides a PCB reflow soldering temperature field prediction device, comprising:

[0022] An acquisition unit is configured to acquire the temperature of a key position of a PCB in a reflow furnace and an environmental parameter to obtain real-time data;

[0023] A mechanism model construction unit is configured to construct a mechanism model according to the real-time data;

[0024] A solving unit is configured to solve the mechanism model to obtain a prediction result;

[0025] An output unit is configured to output the prediction result.

[0026] The application further provides a computer device, which comprises a memory and a processor, the memory stores a computer program, and the processor implements the above method when executing the computer program.

[0027] The application further provides a storage medium, which stores a computer program, and the computer program is executed by a processor to implement the above method.

[0028] Compared with the prior art, the present application has the beneficial effects that: the present application realizes efficient monitoring and optimization of the PCB reflow soldering process by acquiring the temperature and environmental parameters of the key positions of the PCB in the reflow soldering furnace in real time, constructing a mechanism model based on these data, and accurately solving the model to predict the temperature field distribution in the soldering process; this method not only improves the accuracy and immediacy of temperature prediction, but also allows dynamic adjustment of soldering process parameters according to real-time feedback, greatly improving the efficiency and reliability of soldering quality control, ensuring the stability and consistency of the soldering link in the electronic product manufacturing process; ultimately, this data-driven intelligent optimization strategy provides strong technical support for improving production efficiency and reducing defect rate.

[0029] The present application will be further described below in conjunction with the accompanying drawings and specific embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0030] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are some embodiments of the present application, and other drawings can also be obtained by those skilled in the art without creative labor on the basis of these drawings.

[0031] Figure 1 The application scenario diagram of the PCB reflow soldering temperature field prediction method provided by the embodiment of the present application;

[0032] Figure 2 The flowchart of the PCB reflow soldering temperature field prediction method provided by the embodiment of the present application;

[0033] Figure 3 The sub-flowchart of the PCB reflow soldering temperature field prediction method provided by the embodiment of the present application;

[0034] Figure 4 The schematic block diagram of the PCB reflow soldering temperature field prediction device provided by the embodiment of the present application;

[0035] Figure 5 The schematic block diagram of the solving unit of the PCB reflow soldering temperature field prediction device provided by the embodiment of the present application;

[0036] Figure 6 The schematic block diagram of the computer device provided by the embodiment of the present application. DETAILED DESCRIPTION

[0037] With reference to the drawings of the embodiments of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described. Obviously, the described embodiments are some of the embodiments of the present application but not all the embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts should fall within the scope of the present application.

[0038] It should be understood that the terms "comprising" and "including" as used in the specification and the appended claims indicate the presence of the described features, integers, steps, operations, elements, and / or components but do not preclude the presence or addition of one or more other features, integers, steps, operations, elements, components, and / or groups thereof.

[0039] It should also be understood that the terms used in the present application specification are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the present application specification and the appended claims, the singular forms "a", "an" and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0040] It should be further understood that the term "and / or" as used in the present application specification and the appended claims means any combination of one or more of the associated listed items and all possible combinations thereof.

[0041] Please refer to Figure 1 and Figure 2 , Figure 1 The application scenario diagram of the PCB reflow soldering temperature field prediction method provided by the embodiments of the present application. Figure 2 The schematic flowchart of the PCB reflow soldering temperature field prediction method provided by the embodiments of the present application. The PCB reflow soldering temperature field prediction method is applied to a server, which interacts with sensors and terminals for data, acquires temperatures and environmental parameters including conveyor belt speed, heating zone power, humidity, and workshop temperature at key positions such as solder joints and BGA bottoms in real time, constructs a mechanism model based on spatial and temporal discretization, and solves the model using an implicit format to generate a three-diagonal linear equation system to predict temperature field distribution. Further, Thomas algorithm is used to solve the equation system and iteratively update until a stable state or a predetermined time endpoint is reached, ensuring the accuracy and stability of temperature field prediction. In addition, the method also combines machine learning technology to dynamically calibrate the mechanism model, effectively adapting to complex and variable actual working conditions, thereby significantly improving the efficiency and accuracy of welding quality control. This method not only enhances the understanding and control ability of the PCB reflow soldering process, but also provides solid technical support for intelligent production.

[0042] Figure 2is a flowchart of a PCB reflow soldering temperature field prediction method provided by an embodiment of the present application. As shown in Figure 2 the method comprises the following steps S110-S140.

[0043] S110, acquire the temperature of the key positions of the PCB in the reflow oven and the environmental parameters to obtain real-time data.

[0044] In this embodiment, the real-time data refers to the temperature of the key positions and the environmental parameters.

[0045] The key positions include solder joints, BGA bottoms, and positions between the PCB and the thermal sensors that meet the requirements.

[0046] Solder joints: As the most vulnerable places to thermal stress damage during soldering, temperature monitoring at solder joints is crucial. By deploying high-precision thermocouples or infrared sensors in this area, temperature changes during soldering can be monitored and recorded in real time, ensuring that solder joints can reach the ideal melting state without overheating.

[0047] BGA bottom: Ball grid array (BGA) packaging may encounter poor heat dissipation problems in the part connected to the PCB during reflow soldering due to its unique structure. Therefore, setting sensors at these positions helps to accurately control the soldering temperature and avoid soldering defects caused by overheating or uneven cooling.

[0048] Positions meeting requirements: In addition to the above specific parts, a number of positions between the PCB and the thermal sensors that meet certain standards are also selected for monitoring. The selection of these positions is based on the understanding of the overall thermal distribution of the PCB, aiming to capture the temperature change trend of the entire board surface and ensure that all key components are within the optimal soldering temperature range.

[0049] The environmental parameters include a multi-dimensional input matrix composed of device state data and external environmental data; the device state data includes conveyor belt speed and heating zone power; the external environmental data includes humidity and workshop temperature. The method of this embodiment generates a soldering temperature curve to guide the soldering of the product. In addition to the inherent factors, the state of the equipment and the state of the external environment will also affect the soldering. Not only the position of the product in the reflow oven, but also the external environmental factors have a smaller impact.

[0050] Device state data: including information such as conveyor belt speed and heating zone power. Conveyor belt speed determines the length of time the PCB stays in each temperature zone, directly affecting the soldering quality; while heating zone power is related to whether enough heat can be provided to complete the soldering task. The two work together to form an important factor affecting the soldering effect of the PCB.

[0051] External environmental data: This consists of humidity and workshop temperature. Humidity can affect the air's thermal conductivity, thus interfering with the actual soldering temperature. Workshop temperature is directly related to the initial setpoint temperature and is crucial for maintaining a stable production environment. These two data points are typically integrated into a multidimensional input matrix and used as model inputs along with other internal parameters to more accurately simulate and predict the temperature distribution of the PCB during the reflow process.

[0052] By comprehensively considering all of the above information, this step provides detailed and accurate data support for the subsequent establishment of the mechanism model, which is a prerequisite for achieving efficient and precise welding quality control. In addition, this comprehensive data collection method also lays the foundation for a data-driven approach to dynamically calibrate the mechanism model, enabling the system to better adapt to complex actual working conditions.

[0053] S120: Constructing a mechanism model according to the real-time data.

[0054] In this embodiment, a mechanistic model refers to a mathematical model based on fundamental scientific principles such as physics and chemistry. In this particular example, the mechanistic model is based on the fundamental principles of heat transfer, specifically considering the one-dimensional, unsteady-state heat conduction process of a PCB in a reflow oven. It attempts to describe how temperature varies over time and space through mathematical equations, thereby predicting the temperature distribution at different locations on the PCB.

[0055] In one embodiment, see Figure 3 , the above-mentioned step S120 may include steps S121 to S123.

[0056] S121. Divide the length of the reflow furnace into equal parts to obtain a plurality of nodes, and determine the temperature of each node according to the real-time data to obtain a spatial discretization result.

[0057] In this embodiment, the spatial discretization result refers to the process and result of dividing the continuous spatial domain (i.e., the length of the reflow oven) into a number of discrete points or nodes. In this case, this means dividing the reflow oven into N equal segments along its length, each segment representing a node. Each node corresponds to a specific position, and it is assumed that the temperature at this position is consistent. This approach simplifies the computational complexity, allowing the originally complex partial differential equations to be solved by numerical methods.

[0058] For example, if the length of the reflow oven is L meters and is divided into N equal parts, the length (step length) of each part is Δx = L / N. In this way, the entire oven can be represented as a series of nodes Where i represents the spatial position index and n represents the time step index.

[0059] Specifically, in step S121, the length of the reflow oven is divided into multiple nodes at equal intervals, and each node represents a temperature measurement point at a specific location. This operation is to simplify the calculation and improve the simulation accuracy. Through this spatial discretization method, the continuous spatial problem can be converted into a series of discrete points for processing. For each node, the initial temperature value is determined using the data obtained from real-time monitoring of multiple sensors. This forms the spatial discretization result, providing a basis for subsequent time discretization.

[0060] S122, based on the spatial discretization result, using implicit format to each node in time discretization, to get discrete equation;

[0061] In this embodiment, the discrete equation is an algebraic equation obtained by spatial and temporal discretization of the continuous partial differential equation. In this example, the one-dimensional unsteady heat conduction equation is discretized in time using the implicit format, obtaining the time evolution equation for each node. These equations take into account material properties (such as density ρ, specific heat capacity Cp, thermal conductivity k), boundary conditions (such as ambient temperature, heating zone set temperature) and initial conditions.

[0062] The specific form is as follows:

[0063] P refers to the density of the PCB material; c p refers to the specific heat capacity; h refers to the equivalent convective heat transfer coefficient; T oven (iΔx) refers to the set temperature of the oven body at position x; t refers to the sampling time; K refers to the thermal conductivity (calibrated by repeated experimental tests).

[0064] Ti(n) represents the temperature value of node i at time step n. This equation describes the temperature update rule of each node over time and is the basis for constructing a three-diagonal linear equation system.

[0065] Specifically, in step S122, based on the previously obtained spatial discretization result, each node is discretized in time using the implicit format. The advantage of the implicit format is that it can provide better numerical stability, especially when solving non-steady-state heat conduction problems. Through this method, we can establish a discrete equation for each node. These equations take into account material properties (such as density ρ, specific heat capacity Cp and thermal conductivity k), hot air circulation conditions (such as equivalent convective heat transfer coefficient h) and external environmental factors (such as workshop temperature and humidity).

[0066] S123, generating a three-diagonal linear equation system according to the discrete equation to obtain a mechanism model.

[0067] Finally, in step S123, based on the above established discrete equations, a three-diagonal linear equation system AT (n+1) = B is arranged for all nodes. Here, the coefficient matrix A and the right-hand side B contain information about the thermal conductivity k, the time step Δt, the source term, etc. This structure is particularly suitable for solving using the Thomas algorithm, which is an efficient direct solution method specifically designed for three-diagonal matrix problems. By iteratively updating the temperature field until a steady state is reached or a predetermined time endpoint is met, the construction of the mechanism model is completed.

[0068] In summary, through these three steps - spatial discretization, time discretization and discrete equation establishment, and generation of a three-diagonal linear equation system, an accurate mechanism model describing the temperature distribution of PCB in the reflow oven can be constructed. This not only helps to understand the temperature variation law during welding, but also provides a theoretical basis for subsequent data-driven correction.

[0069] S130, solving the mechanism model to obtain a prediction result.

[0070] In this embodiment, the prediction result refers to the temperature distribution of the PCB (Printed Circuit Board) in the reflow oven obtained by applying the Thomas algorithm to solve the mechanism model. Specifically:

[0071] Temperature field prediction: By solving the one-dimensional unsteady heat conduction model, the temperature values of the PCB at different time points and positions are predicted. This means that the temperature variation curves of various key positions on the PCB (such as solder joints, BGA bottoms, and the vicinity of thermal sensors, etc.) during the entire heating process can be obtained.

[0072] Reaching a steady state or meeting a predetermined time endpoint: The prediction process iteratively updates the temperature field until the system reaches a state of thermal equilibrium (i.e., the temperature no longer changes significantly with time), or reaches a pre-set simulation time endpoint. This allows us to understand how the PCB is heated to the required temperature for welding and maintains that temperature for a period of time to complete the welding process.

[0073] Provide basic data support: These prediction results not only help to understand the thermal behavior of the PCB in the reflow oven, but also provide necessary input for subsequent data-driven correction steps. For example, based on the differences between actual measurement data and model prediction results, machine learning methods such as random forests, LSTM networks or gradient boosting trees can be used to dynamically calibrate the model, thereby improving the prediction accuracy in complex scenarios.

[0074] In summary, the prediction result refers to the temperature distribution of the PCB at a specific time and spatial location after precise calculation, which is of great significance for optimizing reflow soldering process parameters and ensuring product quality.

[0075] The mechanism model is solved using the Thomas algorithm to obtain the prediction result.

[0076] The temperature field is iteratively updated until a steady state is reached or a predetermined time endpoint is met.

[0077] In this embodiment, the Thomas algorithm, also known as the Tridiagonal Matrix Algorithm (TDMA), is a highly efficient method for solving tridiagonal linear equations. Such equations often arise in numerical simulation problems after discretizing partial differential equations, such as the heat conduction model in this example. By applying implicit time and spatial discretization methods, an equation group of the form AT n+1 = B is obtained, where A is a tridiagonal coefficient matrix, T n+1 represents the temperature distribution vector at the future time step, and B contains information at the current time step and other source terms.

[0078] First, T 0 needs to be initialized according to the given boundary conditions and initial conditions.

[0079] Construct the coefficient matrix A and the right-hand side B: based on the spatial and temporal discretization formulas mentioned earlier, combined with PCB material properties, environmental parameters, and other information, calculate the values of each item in the coefficient matrix A and the right-hand side B.

[0080] Iterative update: use the Thomas algorithm to gradually solve the temperature field T n+1 at each time step until the predetermined time endpoint is reached or the system enters a steady state (i.e., the temperature no longer changes significantly over time).

[0081] To accurately capture the temperature changes over time and space, the above solving process needs to be repeated, with each update based on the results of the previous step. Specifically:

[0082] Time stepping: starting from the initial time, proceed forward according to the set time step Δt, and update the temperature field using the latest obtained data at each time step.

[0083] Check convergence conditions: at the end of each iteration, check whether the preset stopping criteria are met, such as whether the steady state is reached or the specified final time is reached. If not, continue to the next iteration.

[0084] Reaching a steady state or meeting a predetermined time endpoint, where a steady state is a state in which the temperature distribution of the system gradually tends to a state that does not change with time, at which point the system is considered to have reached thermal equilibrium or steady state. In this case, if the temperature change between consecutive iterations is less than a certain threshold, it can be considered to have reached a steady state.

[0085] Predetermined time endpoint: For some transient analysis, more attention may be paid to the temperature change trend within a certain period of time rather than the final steady state. At this time, as long as the simulation reaches a predetermined length of time, even if it has not completely reached a steady state, the calculation can be stopped.

[0086] In summary, in step S130, by accurately solving the mechanism model and using efficient numerical methods such as Thomas algorithm, the temperature distribution of PCB in reflow oven can be effectively predicted, providing basic data support for subsequent data-driven correction. This not only helps to improve production efficiency, but also ensures the consistency and reliability of product quality.

[0087] S140, output the prediction result.

[0088] The predicted result is output to the terminal.

[0089] In an embodiment, the mechanism model is dynamically calibrated by machine learning.

[0090] Specifically, first, a large amount of experimental data is needed as training set and validation set. These data include but are not limited to:

[0091] Mechanism model predicted temperature (TMODEL): the temperature of PCB at different time points and positions calculated based on existing one-dimensional non-steady heat conduction model.

[0092] Oven position coordinates (x): the specific position coordinates of PCB in reflow oven, because different positions may have different heating effects.

[0093] Conveyor belt speed (v): affects the speed of PCB through the reflow oven, and in turn affects the heating process.

[0094] Component density (D): the layout density of components on PCB, high-density areas may affect local heat distribution.

[0095] At the same time, the actual measured temperature value (Treal) is also needed, that is, the temperature data in the real environment, as the output label.

[0096] To solve the above problems, the following machine learning models can be selected for dynamic calibration:

[0097] Random Forest: Suitable for handling large numbers of feature inputs and effectively avoiding overfitting.

[0098] Long Short-Term Memory Networks (LSTM): Particularly suitable for processing time series data, capable of capturing trends in temperature changes over time.

[0099] XGBoost (eXtreme Gradient Boosting): An efficient ensemble learning method with high accuracy and good interpretability.

[0100] Perform necessary preprocessing on the collected data, mainly including:

[0101] Standardization / Normalization: Scale different input features to the same scale to facilitate subsequent model training.

[0102] Missing value processing: Check and fill or delete data points containing missing values.

[0103] Feature engineering: Create new feature variables according to actual conditions, such as considering the distance between components and local thermal resistance.

[0104] Use the selected machine learning model to dynamically calibrate the original mechanism model, the main steps are as follows:

[0105] Feed the preprocessed input features (TMODEL, x, v, D) into the machine learning model, the goal is to predict results close to the actual measured temperature Treal.

[0106] Adjust model parameters during training to make model predictions as close as possible to actual measurements.

[0107] Use the validation set to evaluate model performance and ensure good generalization ability to perform stably on unseen data.

[0108] After sufficient training, the dynamic calibration model can be used to correct the prediction results of the mechanism model in real time, especially in complex scenarios such as multi-layer boards and high-density component layouts, which can significantly improve prediction accuracy. In addition, the model can be continuously updated and optimized based on new data from the actual production process, forming a closed-loop feedback mechanism to further enhance the adaptability and robustness of the system.

[0109] In summary, advanced machine learning techniques are used to compensate for the possible shortcomings of traditional mechanism models in complex situations, providing more accurate and reliable temperature field prediction services.

[0110] In addition, environmental parameters play a crucial role in the dynamic calibration process. They not only affect the input features of the model but also directly relate to the accuracy and reliability of the model's prediction results. Here is a detailed explanation of how environmental parameters are applied in specific steps:

[0111] In this phase, in addition to collecting basic parameters such as mechanism model predicted temperature (TMODEL), furnace position coordinates (x), conveyor speed (v), and component density (D), special attention should be paid to collecting environment-related parameters. These environmental parameters may include but are not limited to:

[0112] Workshop temperature fluctuations: Average temperature changes at different time periods within the workshop.

[0113] Equipment aging status: Information such as the service life of reflow soldering equipment, maintenance records, etc., indirectly reflects heating uniformity.

[0114] External climate conditions: Such as humidity, air pressure, etc., which have potential effects on heating results, especially in non-enclosed environments.

[0115] For the collected environmental parameters, the following points need to be considered during data preprocessing:

[0116] Standardization: Ensure that all environmental parameters are within the same scale range, avoiding certain parameters dominating the model training process due to large differences in magnitude.

[0117] Correlation analysis: Through correlation analysis or principal component analysis (PCA) and other means, identify which environmental parameters have a significant impact on the final temperature prediction, and decide whether to include them as part of the model input.

[0118] Missing value filling: For possible missing environmental parameters, use appropriate strategies (such as mean filling, interpolation method, etc.) for completion.

[0119] In the actual dynamic calibration process, the application of environmental parameters is reflected in the following aspects:

[0120] Model input enhancement: Add pre-processed environmental parameters to the original input feature set, such as using workshop temperature fluctuations as a new input dimension, so that the model can consider the influence of external environmental changes on the internal temperature field distribution.

[0121] Scenario simulation and adjustment: Use the complete input vector containing environmental parameters to simulate temperature responses under different working conditions, help understand the behavior patterns of the system under specific environmental conditions, and adjust model parameters to optimize prediction performance accordingly.

[0122] Real-time feedback mechanism: When deploying the dynamic calibration model on the production line, changes in environmental parameters can be monitored in real time, and the model prediction results can be updated in a timely manner to ensure that the prediction accuracy remains high even when environmental conditions change.

[0123] With environmental parameters taken into account, the adaptability and flexibility of the entire system have been significantly improved. Especially in the face of changing external conditions, this dynamic calibration method based on environmental perception can more accurately capture various abnormal conditions that may occur in the actual working environment, providing strong support for subsequent quality control and process optimization. In addition, by continuously tracking and analyzing the influence of environmental parameters on product quality, scientific basis can be provided for further improving the production process.

[0124] Specifically, in the PCB reflow soldering process, due to the involvement of complex scenarios such as multi-layer boards and high-density component layout, traditional mechanism models may not fully and accurately reflect the actual temperature field distribution. To overcome these challenges, machine learning techniques can be used to dynamically calibrate the mechanism model to improve the adaptability and prediction accuracy of the model for complex environments.

[0125] Consideration of three-dimensional effects: Traditional models often simplify the heat conduction process, ignoring the three-dimensional structural characteristics of PCBs and components. By introducing machine learning algorithms, the temperature variation patterns under different layouts can be learned from a large amount of experimental data, thus making up for this deficiency.

[0126] Capture of local thermal fusion differences: Different component materials and designs will result in different thermal response characteristics during the soldering process. Machine learning can be used to identify and quantify these differences, adjusting model parameters to more accurately simulate actual conditions.

[0127] Optimization of thermal conductivity k and convective heat transfer coefficient h: These two parameters directly affect the calculation of heat transfer efficiency. Machine learning models trained based on historical data can help automatically adjust these parameter values to make them more close to the actual situation.

[0128] Influence of workshop environmental factors: Temperature fluctuations within the workshop will indirectly affect the working conditions inside the reflow oven. By monitoring the external environment in real time and combining it with the internal model, the prediction results can be dynamically adjusted to reduce errors caused by external interference.

[0129] Compensation for equipment aging effects: Over time, heating devices may experience performance degradation. By applying machine learning techniques to analyze long-term operation data, changes in device status trends can be identified and used to correct model predictions, ensuring that even in the case of device aging, high soldering quality standards can be maintained.

[0130] In summary, dynamic calibration of mechanism models with the aid of machine learning techniques not only effectively addresses the error issues caused by physical model simplification, but also provides more accurate and reliable temperature field prediction capabilities in the face of increasingly complex electronic product manufacturing demands, thereby improving the quality control level of the entire welding process. This method embodies the development direction of modern industrial automation and intelligentization, providing strong support for achieving efficient and stable electronic assembly processes.

[0131] The PCB reflow soldering temperature field prediction method described above, by real-time acquisition of the temperature of the key positions of the PCB in the reflow soldering furnace and the environmental parameters, constructs a mechanism model based on these data, and accurately solves the model to predict the temperature field distribution in the welding process, thereby realizing efficient monitoring and optimization of the PCB reflow soldering process; this method not only improves the accuracy and immediacy of temperature prediction, but also allows dynamic adjustment of welding process parameters according to real-time feedback, greatly improving the efficiency and reliability of welding quality control, ensuring the stability and consistency of the welding link in the electronic product manufacturing process; ultimately, this data-driven intelligent optimization strategy provides strong technical support for improving production efficiency and reducing defect rates.

[0132] Figure 4 is a schematic block diagram of a PCB reflow soldering temperature field prediction device 300 provided by an embodiment of the present application. As Figure 4 indicated, corresponding to the above PCB reflow soldering temperature field prediction method, the present application also provides a PCB reflow soldering temperature field prediction device 300. The PCB reflow soldering temperature field prediction device 300 includes units for executing the above PCB reflow soldering temperature field prediction method, and the device can be configured in a server. Specifically, please refer to Figure 4 , the PCB reflow soldering temperature field prediction device 300 includes an acquisition unit 301, a mechanism model construction unit 302, a solving unit 303, and an output unit 304.

[0133] The acquisition unit 301 is configured to acquire the temperature of the key positions of the PCB in the reflow furnace and the environmental parameters to obtain real-time data; the mechanism model construction unit 302 is configured to construct a mechanism model according to the real-time data; the solving unit 303 is configured to solve the mechanism model to obtain a prediction result; and the output unit 304 is configured to output the prediction result.

[0134] In an embodiment, as Figure 5 indicated, the mechanism model construction unit 302 includes a spatial discretization subunit 3021, a temporal discretization subunit 3022, and a generation subunit 3023.

[0135] The space discretization subunit 3021 is configured to divide the length of the reflow furnace equally to obtain a plurality of nodes, and determine the temperature of each node according to the real-time data, so as to obtain a space discretization result. The time discretization subunit 3022 is configured to discretize each node in time based on the space discretization result by using an implicit format, so as to obtain a discrete equation. The generation subunit 3023 is configured to generate a three-diagonal linear equation group according to the discrete equation, so as to obtain a mechanism model.

[0136] In an embodiment, the solving unit 303 is configured to apply a Thomas algorithm to solve the mechanism model, so as to obtain a prediction result.

[0137] In an embodiment, the solving unit 303 is further configured to update the temperature field by iteration until a steady state is reached or a predetermined time endpoint is met.

[0138] It should be noted that the specific implementation process of the PCB reflow soldering temperature field prediction apparatus 300 and each unit can be clearly understood by those skilled in the art, and can refer to the corresponding description in the foregoing method embodiments. For the convenience and brevity of description, details are not repeated here.

[0139] The PCB reflow soldering temperature field prediction apparatus 300 can be implemented in the form of a computer program, which can run on a computer device as shown in the computer device. Figure 6

[0140] Please refer to Figure 6 , Figure 6 is a schematic block diagram of a computer device provided by an embodiment of the present application. The computer device 500 can be a server, wherein the server can be a stand-alone server or a server cluster composed of multiple servers.

[0141] Referring to Figure 6 , the computer device 500 includes a processor 502, a memory, and a network interface 505 connected through a system bus 501, wherein the memory can include a non-volatile storage medium 503 and an internal memory 504.

[0142] The non-volatile storage medium 503 can store an operating system 5031 and a computer program 5032. The computer program 5032 includes program instructions, which, when executed, can cause the processor 502 to perform a PCB reflow soldering temperature field prediction method.

[0143] The processor 502 is configured to provide computing and control capabilities to support the operation of the entire computer device 500.

[0144] ​The memory 504 provides an environment for running the computer program 5032 in the nonvolatile storage medium 503, which, when executed by the processor 502, can cause the processor 502 to perform a PCB reflow solder temperature field prediction method.

[0145] The network interface 505 is used for network communication with other devices. Figure 6 The structure shown in the figure is only a block diagram of part of the structure related to the scheme of the present application, and does not constitute a limitation on the computer device 500 to which the scheme of the present application is applied. The specific computer device 500 can include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0146] The processor 502 is configured to run the computer program 5032 stored in the memory to implement the following steps:

[0147] Obtain the temperature of the key position of the PCB in the reflow oven and the environmental parameters to obtain real-time data; construct a mechanism model according to the real-time data; solve the mechanism model to obtain a prediction result; and output the prediction result.

[0148] The key position includes a solder joint, a BGA bottom, and a position between the PCB and a thermal sensor that meets the requirements.

[0149] The environmental parameters include a multi-dimensional input matrix composed of device state data and external environmental data; the device state data includes conveyor belt speed and heating zone power; and the external environmental data includes humidity and workshop temperature.

[0150] The mechanism model is dynamically calibrated through machine learning.

[0151] In an embodiment, when implementing the step of constructing a mechanism model according to the real-time data, the processor 502 specifically implements the following steps:

[0152] The length of the reflow oven is equally divided to obtain a plurality of nodes, and the temperature of each node is determined according to the real-time data to obtain a spatial discretization result; based on the spatial discretization result, each node is discretized in time using an implicit format to obtain a discrete equation; and a three-diagonal linear equation group is generated according to the discrete equation to obtain a mechanism model.

[0153] In an embodiment, when implementing the step of solving the mechanism model to obtain a prediction result, the processor 502 specifically implements the following steps:

[0154] The mechanism model is solved by applying the Thomas algorithm to obtain a prediction result.

[0155] In an embodiment, the processor 502, after implementing the step of applying the Thomas algorithm to solve the mechanism model to obtain a prediction result, further implements the following steps:

[0156] updating the temperature field by iteration until a steady state is reached or a predetermined time endpoint is met.

[0157] It should be understood that, in the embodiments of the present application, the processor 502 can be a central processing unit (CPU), and the processor 502 can also be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor.

[0158] It can be understood by those skilled in the art that all or part of the processes in the method of implementing the above embodiments can be completed by a computer program instructing related hardware. The computer program includes program instructions, and the computer program can be stored in a storage medium, which is a computer-readable storage medium. The program instructions are executed by at least one processor in the computer system to implement the process steps of the above-mentioned embodiments of the method.

[0159] Therefore, the present application also provides a storage medium. The storage medium can be a computer-readable storage medium. The storage medium stores a computer program, wherein the computer program is executed by a processor to make the processor perform the following steps:

[0160] obtaining temperatures of key positions of a PCB in a reflow oven and environmental parameters to obtain real-time data; constructing a mechanism model according to the real-time data; solving the mechanism model to obtain a prediction result; and outputting the prediction result.

[0161] The key positions include solder joints, BGA bottoms, and positions between the PCB and the thermal sensor that meet the requirements.

[0162] The environmental parameters include a multi-dimensional input matrix composed of device state data and external environment data; the device state data includes conveyor belt speed and heating zone power; and the external environment data includes humidity and workshop temperature.

[0163] The mechanism model is dynamically calibrated by machine learning.

[0164] In an embodiment, the processor, when implementing the step of constructing a mechanism model according to the real-time data by executing the computer program, specifically implements the following steps:

[0165] The length of the reflow furnace is equally divided to obtain a plurality of nodes, and the temperature of each node is determined according to the real-time data to obtain a spatial discretization result; based on the spatial discretization result, each node is discretized in time using an implicit format to obtain a discrete equation; and a three-diagonal linear equation group is generated according to the discrete equation to obtain a mechanism model.

[0166] In an embodiment, the processor, when implementing the step of solving the mechanism model to obtain a prediction result by executing the computer program, specifically implements the following steps:

[0167] The mechanism model is solved by applying a Thomas algorithm to obtain a prediction result.

[0168] In an embodiment, the processor, after implementing the step of solving the mechanism model by applying a Thomas algorithm to obtain a prediction result by executing the computer program, further implements the following steps:

[0169] The temperature field is iteratively updated until a steady state is reached or a predetermined time endpoint is met.

[0170] The storage medium can be a U disk, a mobile hard disk, a read-only memory (ROM), a magnetic disk or an optical disk, and various computer readable storage media that can store program codes.

[0171] Those skilled in the art can realize that the units and algorithm steps of the examples described in combination with the embodiments disclosed herein can be realized by electronic hardware, computer software or a combination of both. In order to clearly illustrate the interchangeability of hardware and software, the components and steps of the examples have been described in general terms in the above description. Whether the functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. A person skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of the present application.

[0172] In several embodiments provided by the present application, it should be understood that the disclosed apparatus and method can be implemented in other manners. For example, the embodiments of the apparatus described above are merely schematic. For example, the division of the units is merely a logical function division. For example, a plurality of units or components can be combined or integrated into another system, or some features can be ignored or not executed. In this way, the inventive idea can be implemented.

[0173] The steps in the method embodiments of the present application can be adjusted, combined and deleted in sequence according to actual needs. The units in the apparatus embodiments of the present application can be combined, divided and deleted according to actual needs. In addition, each functional unit in each embodiment of the present application can be integrated in one processing unit, or each unit can exist physically, or two or more units can be integrated in one unit.

[0174] The integrated unit, if realized in the form of a software functional unit and sold or used as an independent product, can be stored in a storage medium. Based on such understanding, the technical solutions of the present application essentially or the part that contributes to the prior art, or the whole or part of the technical solutions can be embodied in the form of a software product. The computer software product is stored in a storage medium, and includes a plurality of instructions for causing a computer device (which can be a personal computer, a terminal, or a network device, etc.) to execute all or part of the steps of the methods described in the embodiments of the present application.

[0175] The above describes only specific embodiments of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical range disclosed by the present application, and these modifications or replacements should be covered in the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claims.

Claims

1. PCB reflow temperature field prediction method, characterized in that: include: Obtain the temperature and environmental parameters of key locations of the PCB in the reflow oven to obtain real-time data; constructing a mechanism model based on the real-time data; Solving the mechanism model to obtain prediction results; The prediction result is output.

2. The PCB reflow temperature field prediction method according to claim 1, characterized in that: The key positions include solder joints, the bottom of the BGA, and positions where the distance between the PCB and the thermal sensor meets the requirements.

3. The PCB reflow temperature field prediction method according to claim 1, characterized in that: The environmental parameters include a multi-dimensional input matrix composed of equipment status data and external environment data; the equipment status data includes conveyor belt speed and heating zone power; the external environment data includes humidity and workshop temperature.

4. The PCB reflow temperature field prediction method according to claim 1, characterized in that: The constructing of a mechanism model according to the real-time data comprises: Dividing the length of the reflow furnace into equal parts to obtain a plurality of nodes, and determining the temperature of each node according to the real-time data to obtain a spatial discretization result; Based on the spatial discretization result, each node is discretized in time using an implicit format to obtain a discrete equation; A tridiagonal linear equation system is generated according to the discrete equation to obtain a mechanism model.

5. The PCB reflow temperature field prediction method according to claim 1, wherein: Solving the mechanism model to obtain a prediction result includes: The Thomas algorithm is applied to solve the mechanism model to obtain prediction results.

6. The PCB reflow temperature field prediction method according to claim 5, characterized in that: After applying the Thomas algorithm to solve the mechanism model to obtain the prediction result, the method further includes: The temperature field is updated iteratively until a steady state is reached or a predetermined time end is met.

7. The PCB reflow temperature field prediction method according to claim 1, characterized in that: The mechanistic model is dynamically calibrated through machine learning.

8. PCB reflow temperature field prediction device, characterized in that, include: An acquisition unit is used to obtain the temperature and environmental parameters of key locations of the PCB in the reflow oven to obtain real-time data; A mechanism model building unit, used to build a mechanism model according to the real-time data; A solving unit, used for solving the mechanism model to obtain a prediction result; An output unit is used to output the prediction result.

9. A computer device, characterized in that: The computer device includes a memory and a processor, the memory stores a computer program, and the processor implements the method according to any one of claims 1 to 7 when executing the computer program.

10. A storage medium, characterized in that: The storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 7 is implemented.

Images