Method for controlling high-frequency interference and improving welding quality in production of electronic control board

Abstract

The invention relates to a high-frequency interference control and welding quality improvement method in electronic control board production in the technical field of electronic control board manufacturing, which comprises the following steps of: acquiring high-frequency interference data in the electronic control board production, starting from the structural design of a substrate, adjusting the parameters of the substrate through a partition isolation technology and a multilayer board layout algorithm, an optimized substrate structure design scheme is obtained; extracting real-time monitoring data from a welding process flow adapted to various components, recording the state and defect characteristics of a welding point by adopting a full-process detection tracing system, and judging a process stability index of welding quality improvement; production line adjustment data during remodeling are obtained, the remodeling process is optimized through the production flexible adjustment technology, and an equipment configuration scheme for optimizing the remodeling efficiency is determined; and obtaining the production response time of the multi-variety order according to the equipment configuration scheme of model changing efficiency optimization, and adjusting the production line layout through a modular production capacity evaluation algorithm to obtain a multi-variety quick response production scheduling scheme.

Description

Technical Field

[0001] The present invention relates to the technical field of electronic control board manufacturing, and particularly to a method for high-frequency interference control and welding quality improvement in the production of electronic control boards. Background Art

[0002] The technical field of electronic control board manufacturing occupies a crucial position in modern industry. Especially in the production of core components such as variable-frequency speed controllers, its performance directly affects the stability and energy efficiency of the equipment. With the increasing demand for small-batch and multi-variety orders, efficient and low-defect production methods have become an inevitable trend in the industry's development. However, the current production methods have shown significant deficiencies in coping with this trend and urgently require technological innovation to break through the bottleneck. Existing methods have obvious limitations in solving problems such as high-frequency interference, welding quality, and production flexibility. Traditional single-layer PCB designs are difficult to effectively suppress electromagnetic interference, resulting in impaired stability; manual inspection and fixed process parameters lead to frequent welding defects and difficulty in adapting to diverse components; in addition, frequent line changeovers and debugging take too long, restricting the rapid response ability to multi-variety orders. These limitations make it difficult for enterprises to achieve a balance in cost, quality, and delivery cycle. Focusing on the core challenges, high-frequency interference, welding defect rate, and changeover efficiency have become the key technical factors restricting production yield. High-frequency interference stems from the simplicity of substrate design and the lack of partition isolation, leading to a decline in signal integrity; the high welding defect rate is due to the lack of dynamic adaptive process control and comprehensive inspection means; the low changeover efficiency reflects the insufficient flexibility of the production line to quickly adjust to different product specifications. These unresolved technical factors together have caused unique problems such as insufficient stability, difficulty in defect traceability, and low efficiency in the production process. Therefore, how to improve welding quality and achieve rapid multi-variety changeovers while ensuring the anti-interference ability of the control board has become the key issue in improving the modular production yield of electronic control boards for variable-frequency controllers. Solving this problem requires starting from optimizing the substrate structure, intelligent process adaptation, and full-process inspection and traceability to meet the complex requirements in the small-batch and multi-variety scenarios. Summary of the Invention

[0003] The present invention provides a method for high-frequency interference control and welding quality improvement in the production of electronic control boards, and the method includes the following steps:

[0004] Obtain high-frequency interference data in the production of electronic control boards. Starting from the substrate structure design, adjust the substrate parameters through partition isolation technology and multi-layer board layout algorithms to obtain an optimized substrate structure design scheme;

[0005] For the optimized substrate structure design scheme, use electromagnetic simulation technology to analyze signal integrity, and adjust impedance matching and shielding layer configuration under the requirements of high-frequency interference control to determine the substrate parameters for ensuring signal integrity;

[0006] Obtain the temperature curve and component distribution data during the soldering process based on the substrate parameters for signal integrity guarantee, and adjust the soldering parameters through a process dynamic adaptation algorithm to obtain a soldering process flow suitable for multiple types of components;

[0007] Extract real-time monitoring data from the soldering process flow suitable for multiple types of components, use a full-process inspection and traceability system to record the soldering point status and defect characteristics, and judge the process stability index for improving soldering quality;

[0008] For the process stability index for improving soldering quality, obtain the production line adjustment data during model changeover, optimize the model changeover process through production flexible adjustment technology, and determine the equipment configuration plan for optimizing model changeover efficiency;

[0009] According to the equipment configuration plan for optimizing model changeover efficiency, obtain the production response time for multiple types of orders, and adjust the production line layout through a modular production capacity evaluation algorithm to obtain a production scheduling plan for rapid response to multiple types of products;

[0010] Extract the defect rate data from the production scheduling plan for rapid response to multiple types of products, use defect rate reduction technology to analyze the failure distribution caused by soldering defects and interference, and judge the comprehensive improvement effect of high-frequency interference control and soldering quality improvement;

[0011] For the comprehensive improvement effect, obtain the historical data stored in the full-process inspection and traceability system, analyze the correlation between substrate structure design and process dynamic adaptation through machine learning algorithms, and determine the optimization direction of modular production capacity;

[0012] According to the optimization direction of modular production capacity, obtain the real-time interference signals and soldering quality data during the production process, and adjust the shielding strategy and process parameters through electromagnetic interference suppression technology to obtain the final production yield improvement plan.

[0013] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0014] The present invention discloses a method for high-frequency interference control and welding quality improvement in the production of electronic control boards. By analyzing the correlation between substrate structure design and process dynamic adaptation, the modular production capacity is optimized. First, the substrate parameters are adjusted by using the partition isolation technology and the multilayer board layout algorithm, and the signal integrity is analyzed by using the electromagnetic simulation technology. Then, the welding parameters are adjusted by the process dynamic adaptation algorithm to optimize the welding process flow of various components. The present invention also adopts a full-process detection and traceability system to record the state of welding points and defect characteristics, and optimizes the changeover process by using the production flexible adjustment technology. Finally, the historical data is analyzed by using the machine learning algorithm, and the shielding strategy and process parameters are adjusted in combination with the electromagnetic interference suppression technology to significantly improve the production yield. This method effectively solves the problems of high-frequency interference and welding quality in the production of electronic control boards, and improves the production efficiency and product quality. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 It is a flowchart of the method for high-frequency interference control and welding quality improvement in the production of the electronic control board of the present invention.

[0016] Figure 2 It is a schematic diagram of the method for high-frequency interference control and welding quality improvement in the production of the electronic control board of the present invention.

[0017] Figure 3 It is another schematic diagram of the method for high-frequency interference control and welding quality improvement in the production of the electronic control board of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] In order to make the objectives, technical solutions and advantages of the present invention clearer, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] As Figures 1-3 , the method for high-frequency interference control and welding quality improvement in the production of the electronic control board in this embodiment may specifically include:

[0020] S101. Obtain the high-frequency interference data in the production of the electronic control board, start from the substrate structure design, and adjust the substrate parameters by using the partition isolation technology and the multilayer board layout algorithm to obtain an optimized substrate structure design scheme.

[0021] Obtain high-frequency interference data through sensors, and determine the interference distribution characteristics of the high-frequency interference data; process the interference distribution characteristics using zone isolation technology to obtain a preliminary isolation layout; optimize the preliminary isolation layout through a multi-layer board layout algorithm to obtain adjusted substrate parameters; if the adjusted substrate parameters exceed a preset threshold, correct the adjusted substrate parameters through iterative calculation to obtain stable substrate parameters; generate an improved substrate structure plan according to the stable substrate parameters; judge the degree of high-frequency interference suppression according to the improved substrate structure plan to obtain an optimized design plan; verify the optimized design plan through simulation to determine the final substrate structure.

[0022] Exemplarily, in the production of electronic control boards, the acquisition of high-frequency interference data can be measured by an electromagnetic compatibility tester. For example, in the frequency spectrum ranging from 1 MHz to 1 GHz, the amplitude of the interference signal is measured to be -30 dBm to -20 dBm. To solve the high-frequency interference problem starting from the substrate structure design, first, the zone isolation technology is adopted to divide the substrate into a digital signal area, an analog signal area, and a power supply area. The distance between each area is set to 5 mm, and an isolation band with a width of 0.5 mm is added between each area. The isolation band is made of copper foil material with a thickness of 35 μm to effectively reduce signal crosstalk. Then, the substrate parameters are adjusted through a multi-layer board layout algorithm. A 6-layer board design is adopted, where the first layer and the sixth layer are signal layers, the second layer and the fifth layer are power supply layers, and the third layer and the fourth layer are ground plane layers. The dielectric thickness between each layer is 0.2 mm, and the dielectric constant is 4.5. In the signal layer, differential pair wiring is adopted, with a line width of 0.15 mm and a line spacing of 0.3 mm to reduce electromagnetic radiation. The optimized substrate structure is simulated and analyzed through an electromagnetic field simulation software. The results show that in the frequency range of 1 MHz to 1 GHz, the amplitude of the interference signal is reduced to -40 dBm to -30 dBm, meeting the electromagnetic compatibility requirements. Finally, the optimized substrate structure design plan is generated through an automated design software and can be directly used for production and manufacturing.

[0023] S102. For the optimized substrate structure design plan, use electromagnetic simulation technology to analyze signal integrity, adjust impedance matching and shielding layer configuration under the requirements of high-frequency interference control, and determine the substrate parameters for ensuring signal integrity.

[0024] The substrate structure is analyzed using electromagnetic simulation technology to obtain preliminary data on signal integrity; the distribution characteristics of high-frequency interference are judged based on the preliminary data, and the adjustment direction of the shielding layer configuration is determined; impedance matching is optimized for the adjustment direction to obtain optimized substrate parameters; if the optimized substrate parameters meet the control requirements, signal integrity is verified using electromagnetic simulation technology to obtain verification data; the shielding layer configuration is adjusted according to the verification data to determine the final substrate parameters; a machine learning algorithm is used to model the relationship between the final substrate parameters and signal integrity to obtain a prediction model; the impact of different design schemes is analyzed through the prediction model to determine the substrate parameter configuration for signal integrity guarantee.

[0025] Exemplarily, when optimizing the substrate structure design scheme, first, electromagnetic simulation technology is used to analyze signal integrity. By establishing a three-dimensional electromagnetic field model, setting the frequency range from 1 GHz to 10 GHz, and using the finite element method (FEM) for simulation calculation, the electromagnetic field distribution on the signal transmission path is obtained. According to the simulation results, the main source of high-frequency interference is identified as the coupling effect between the signal line and the ground plane, and its coupling coefficient is 0.85. To control high-frequency interference, impedance matching is adjusted. A microstrip line structure is adopted, and by adjusting the line width and the dielectric layer thickness, the characteristic impedance is optimized from 50 Ω to 75 Ω. Using transmission line theory, the calculated line width is 0.15 mm, and the dielectric layer thickness is 0.2 mm. At the same time, to enhance the shielding effect, copper foil shielding layers with a thickness of 0.035 mm are added on both sides of the substrate, and the shielding effectiveness is verified through electromagnetic simulation. The results show that at a frequency of 10 GHz, the shielding effectiveness is increased to 40 dB. Finally, the substrate parameters are determined as: signal line width 0.15 mm, dielectric layer thickness 0.2 mm, shielding layer thickness 0.035 mm, ensuring that signal integrity is effectively guaranteed under the control requirements of high-frequency interference.

[0026] S103. According to the substrate parameters for signal integrity guarantee, obtain the temperature curve and component distribution data during the welding process, and adjust the welding parameters through a process dynamic adaptation algorithm to obtain a welding process flow suitable for multi-variety components.

[0027] Obtain the signal integrity data in the substrate parameters, use a preset threshold to judge whether the signal integrity meets the requirements, and obtain the initial analysis result. Extract the temperature curve characteristics from the initial analysis result, and use the data analysis method to determine the change trend of the temperature curve. According to the change trend of the temperature curve, combined with the component distribution data, judge the adjustment direction of the welding parameters to obtain a preliminary adjustment plan. For the preliminary adjustment plan, use a dynamic adaptation algorithm to process the matching relationship between the welding parameters and the component distribution to obtain an optimized set of welding parameters. Obtain the execution order of the process flow from the optimized set of welding parameters, and determine the adaptation plan for multi-variety components through information processing technology. Adjust the temperature control logic in the process flow according to the adaptation plan, and use the linear regression algorithm to predict the stability after process optimization to obtain the final process flow. Verify the consistency between the component distribution and the signal integrity through the final process flow, and obtain the adjusted process execution data.

[0028] Exemplarily, in the substrate parameters for signal integrity guarantee, by analyzing the dielectric constant (such as 4.5) and the loss tangent value (such as 0.02) of the substrate, combined with the transmission line impedance (such as 50 ohms) and the signal propagation speed (such as 1.5×10^8 m / s), the electrical characteristics of the substrate can be determined. During the welding process, use infrared thermal imaging technology to obtain the temperature curve. For example, when the welding peak temperature reaches 250°C, the holding time is controlled within 10 seconds to ensure that the components are not damaged by heat. By collecting the distribution data of the components, such as different package sizes (such as 0402, 0603) and heat capacity values (such as 0.1 J / °C), combined with the process dynamic adaptation algorithm, use the gradient descent method to optimize the welding parameters, such as adjusting the welding temperature from 230°C to 240°C and extending the welding time from 8 seconds to 9 seconds to achieve the adaptation of different varieties of components. By analyzing the electrical performance after welding, such as the signal reflection coefficient (such as -20 dB) and the insertion loss (such as 0.5 dB), verify the reliability of the welding process and further optimize the algorithm parameters to ensure the stability and consistency of the welding process flow.

[0029] S104. Extract the real-time monitoring data from the welding process flow for adapting multi-variety components, use the full-process detection and traceability system to record the welding point status and defect characteristics, and judge the process stability index for improving the welding quality.

[0030] Real-time monitoring data is obtained from the welding process through sensors and stored in a database to obtain an original data set. A traceability system is used to classify the original data set, extract the welding point status and defect features, and obtain a status feature set. If the defect features in the status feature set exceed a preset threshold, the defect features are grouped through a clustering algorithm to obtain a defect classification result. Based on the comparison and analysis of the defect classification result and the welding point status, a statistical method is used to calculate the process stability index to obtain a stability quantification value. The stability quantification value is compared with historical data for trend comparison to judge the quality change trend and obtain quality trend data. The welding process parameters are adjusted through the quality trend data, and the real-time monitoring data is updated to obtain an optimized process data set. If the stability quantification value in the optimized process data set reaches the preset standard, the status feature set is recorded to obtain the final process optimization result.

[0031] Exemplarily, in the welding process flow for adapting to multiple types of components, real-time monitoring data is collected by high-precision sensors. For example, a laser displacement sensor is used to measure the height of the solder joint with a resolution of 0.01 mm, and at the same time, an infrared thermal imager records the welding temperature curve at a sampling frequency of 30 frames per second, with a temperature control accuracy of ±2°C. The collected time-series data is denoised through the Kalman filter algorithm. For example, when performing state estimation on the temperature signal, the process noise covariance is set to 0.1, and the observation noise covariance is set to 0.5 to achieve data smoothing. The full-process detection and traceability system uses a defect classification model based on deep learning, with the ResNet50 network structure. The solder joint image of 256×256 pixels is input, and through transfer learning, it is trained on 100,000 labeled samples, and the defect recognition accuracy reaches 98.7%. The system records the feature vectors of each solder joint, including 18-dimensional parameters such as area (e.g., 1.2±0.3 square millimeters), wetting angle (the qualified range is 35° - 50°), etc. The DBSCAN clustering algorithm (neighborhood radius ε = 0.8, minimum number of samples minPts = 15) is used to automatically divide the batches with abnormal process parameters. The process stability index is calculated using the Six Sigma method. For example, the CPK value of a batch of 2000 solder joints is increased from 1.25 to 1.68, and the process capability index is increased by 34.4%. At the same time, based on the EWMA control chart (λ = 0.2, control limits ±3σ), the drift of key parameters is monitored in real time, and when 5 consecutive points exceed the 2σ warning line, the process parameter adaptive adjustment is triggered. All data is stored in a time-series database with a sampling interval of 100 milliseconds, supporting the analysis of the solder joint defect rate trend per hour based on SQL window functions. For example, the window size for calculating the moving average is 30 minutes, and the standard deviation threshold is set to 0.15.

[0032] S105. For the process stability index for improving welding quality, obtain the production line adjustment data during model changeover, optimize the model changeover process through production flexible adjustment technology, and determine the equipment configuration plan for optimizing the model changeover efficiency.

[0033] Collect the production line adjustment data during changeover through sensors and store them in a database to obtain a changeover data set. Use data analysis methods to process the changeover data set, extract features related to process stability, and determine process stability indicators. If the process stability indicator is lower than a preset threshold, classify the changeover data set through a clustering algorithm to identify abnormal points in the changeover process. According to the analysis results of the abnormal points, adjust the production flexibility parameters to obtain an optimized changeover process plan. Verify the optimized changeover process plan through simulation technology to obtain a changeover efficiency evaluation value. If the changeover efficiency evaluation value is higher than the preset threshold, generate an equipment configuration plan according to the optimized changeover process plan. Use a database comparison method to match the equipment configuration plan with the existing equipment parameters to determine the final equipment configuration adjustment plan.

[0034] Exemplarily, in the process stability indicators for improving welding quality, first obtain the production line adjustment data during changeover through a data acquisition system, such as changeover time, equipment downtime, number of changeovers, etc. The specific values are, for example, the average changeover time is 30 minutes, the downtime is 15 minutes, and the number of changeovers is 10 times per month. Utilize these data and optimize them using production flexibility adjustment technology. Analyze the bottleneck links in the changeover process through algorithms. For example, use a genetic algorithm to optimize the changeover sequence, reducing the changeover time to 20 minutes and the downtime to 10 minutes. Further, through the optimization of the equipment configuration plan, determine the optimal equipment combination. For example, adopt a modular equipment design to increase the changeover efficiency by 15%, specifically manifested as the changeover time being reduced to 17 minutes and the downtime being reduced to 8 minutes. Throughout the process, dynamically adjust the changeover process through a real-time monitoring system to ensure the continuous improvement of process stability and production efficiency.

[0035] For example, by introducing an intelligent scheduling system, reduce the number of changeovers to 8 times per month, further improving the overall efficiency of the production line. Through these technical means, the process stability of welding quality has been significantly improved, and the optimization effect of the changeover process has also been fully verified.

[0036] S106. According to the equipment configuration plan optimized for changeover efficiency, obtain the production response time for multi-variety orders, and adjust the production line layout through a modular production capacity evaluation algorithm to obtain a production scheduling plan for multi-variety rapid response.

[0037] Obtain multi-variety order data, use a classification algorithm to determine the order priority, and obtain an order processing sequence. Obtain the production task allocation from the order processing sequence, combine it with a modular production capacity evaluation algorithm, judge the equipment configuration plan, and obtain the production resource allocation result. If the production resource allocation result meets the preset response time threshold, adjust the equipment position parameters through the production line layout data to obtain an optimized production line layout plan. According to the optimized production line layout plan, adopt a dynamic adjustment mechanism to generate a real-time scheduling sequence for production tasks, and obtain a fast response scheduling plan. Extract the production response time from the fast response scheduling plan, combine it with the order processing efficiency data, judge whether it meets the response time optimization goal for multi-variety orders, and obtain the final production scheduling plan. If the final production scheduling plan does not meet the response time optimization goal, adjust the equipment configuration plan through the production resource allocation result to obtain a new production scheduling plan. According to the new production scheduling plan, obtain the production response time, determine the fast response ability for multi-variety orders, and obtain an optimized production operation sequence.

[0038] Exemplarily, in the optimization of changeover efficiency, first collect the historical data of the production line (such as the average changeover time of 45 minutes and the standard deviation of 8 minutes), and use the genetic algorithm to perform multi-objective optimization on the equipment configuration. The objective function is to minimize the changeover time and maximize the equipment utilization rate, and the constraint conditions include that the cost of a single changeover does not exceed 2000 yuan. For example, for a certain automotive parts production line, after algorithm iteration, the optimal configuration is to increase the tool magazine capacity of 3 CNC machine tools from 12 to 18, reducing the average changeover time to 32 minutes. Then, based on the order data (such as the monthly demand fluctuation range of 300 - 800 pieces for 5 types of products), use discrete event simulation to simulate the production response time. The input parameters include the equipment failure rate (MTBF = 120 hours) and the process cycle time (45 seconds / piece). Through AnyLogic simulation, the average order delivery time in the current layout is 5.2 days. To improve the response speed, use a modular capacity evaluation algorithm to quantitatively analyze the flexibility index of the production line (such as the equipment versatility score is increased from 0.6 to 0.8), and combine it with the process route similarity matrix (such as the process overlap degree of product A and B reaches 75%), and adjust the original parallel layout to a U-shaped cell line, reducing the product changeover time by 22%. Finally, use the improved NSGA-II algorithm for multi-objective scheduling, with the number of changeovers (≤3 times / shift) and the rate of overdue orders (≤5%) as constraints, and output the optimal scheduling plan: when processing orders X (500 pieces) and Y (300 pieces) simultaneously, adopt a mixed flow line mode, and balance the cycle time difference through dynamic buffer inventory (the safety inventory is set to 25 pieces), so that the comprehensive production efficiency is increased by 18%, and the order delivery cycle is compressed to 3.8 days. During the whole process, use digital twin technology to calibrate the model parameters in real time to ensure that the deviation rate between the algorithm output and the physical production line is controlled within 3%.

[0039] S107. Extract the defect rate data from the production scheduling plan for rapid response to multiple product varieties, and use defect rate reduction technology to analyze the failure distribution caused by welding defects and interference, and judge the comprehensive improvement effect of high-frequency interference control and welding quality improvement.

[0040] Obtain the defect rate data in the production scheduling plan for multiple product varieties, calculate the defect rate distribution of each product variety using statistical methods, and obtain the preliminary defect rate statistical results. For the preliminary defect rate statistical results, use cluster analysis to divide the welding defect and interference failure categories, and determine the distribution patterns of defects and interference. Extract the fault characteristics related to high-frequency interference from the distribution patterns. If the fault characteristics exceed the preset threshold, it is judged as dominated by high-frequency interference, and the interference influence range is obtained. Screen the welding defect data through the interference influence range, use regression analysis to quantify the correlation between high-frequency interference and welding defects, and obtain the correlation coefficient distribution. Adjust the production scheduling parameters according to the correlation coefficient distribution, obtain the optimized defect rate data, and judge the quality improvement amplitude. Extract the improvement effect indicators from the optimized defect rate data, and determine the comprehensive improvement effect by comparing the preliminary defect rate statistical results. For the comprehensive improvement effect, obtain the updated production scheduling plan for each product variety, and judge the improvement amplitude of the rapid response ability.

[0041] Exemplarily, in the production scheduling plan for rapid response to multiple product varieties, by collecting the defect rate data of the welding production line in real time. For example, a certain production line welds 5,000 solder joints every day, and the number of defective solder joints is 75, and the initial defect rate is 1.5%. Use a defect classification method based on the K-means clustering algorithm, set the number of clusters k = 3, and divide the defects into three categories: pores (accounting for 45%), lack of fusion (accounting for 30%), and cracks (accounting for 25%). Collect the high-frequency interference signals during the welding process through a spectrum analyzer, and find that the interference frequency is concentrated in the range of 10 kHz to 50 kHz, and the interference amplitude near 30 kHz is the highest, reaching 12 mV. Use the wavelet transform algorithm to denoise the interference signal, select the db4 wavelet basis function for 5-layer decomposition, and the reconstructed interference amplitude is reduced to 3 mV. Combine the defect distribution data and the interference signal characteristics to establish a BP neural network model. The input layer includes 6 parameters such as interference frequency and amplitude, the hidden layer is set with 8 neurons, and the output layer is the probability of defect type, and the accuracy of the training set reaches 92%. After implementing the interference suppression measures, the defect rate drops to 0.8%, and the pore defects are reduced by 60%, verifying the effectiveness of high-frequency interference control in improving welding quality. Through variance analysis, it is calculated that the contribution degree of the interference suppression measures to the defect rate reduction is 65%, which is significantly higher than the influence of other process parameters.

[0042] S108. For the comprehensive improvement effect, obtain the historical data stored in the full-process inspection and traceability system, analyze the correlation between the substrate structure design and the dynamic adaptation of the process through machine learning algorithms, and determine the optimization direction of the modular production capacity.

[0043] Obtain historical data from the full-process inspection and traceability system, analyze the correlation between the substrate structure design and the dynamic adaptation of the process using the random forest algorithm to obtain a preliminary correlation model. Extract the change characteristics of the process dynamics from the preliminary correlation model, perform clustering analysis on the change characteristics to determine the adaptation mode of the substrate structure. Obtain key parameters from the adaptation mode, judge the process adjustment range through dynamic analysis to obtain the boundary conditions of the modular production capacity. If the boundary conditions exceed the preset threshold, analyze the mapping relationship between the historical data and the optimization direction through the linear regression algorithm to determine the adjusted production capacity range. According to the adjusted production capacity range, obtain the matching degree between the process dynamics and the substrate structure, and judge the optimization space of the modular design. For the size of the optimization space, use clustering analysis to separate high-adaptability modules to obtain the final optimization direction. Extract the key points for improving the production capacity from the final optimization direction, and verify the stability of the key points through the historical data to determine the improvement plan for the modular production capacity.

[0044] Exemplarily, in the comprehensive improvement effect analysis, first extract historical data from the full-process inspection and traceability system. For example, collect the production data of a certain type of substrate in the past 6 months, including design parameters (such as line width 35μm, dielectric constant 4.2), process parameters (such as etching time 120s, temperature 185°C), and quality indicators (such as yield 92.3%). Through the calculation of the Pearson correlation coefficient, it is found that the correlation coefficient between the line width and the etching time reaches 0.78, indicating that the design parameters significantly affect the process adaptability. Then use the random forest algorithm to construct a prediction model. The input layer includes 8 design features and 5 process features. Determine the optimal hyperparameters (n_estimators = 200, max_depth = 10) through grid search, and the model reaches a prediction accuracy of R 2 = 0.91 on the test set. The analysis using SHAP values shows that the contribution weight of the dielectric constant to the yield is 23.7%. Based on this, an optimization plan is proposed to adjust the dielectric constant tolerance from ±0.5 to ±0.3. Finally, based on K-means clustering (k = 3), divide the production data into three modules of high, medium, and low efficiency. It is found that the process window fluctuation range of the high-efficiency module is narrower (temperature fluctuation ±2°C vs. conventional ±5°C). Based on this, it is recommended to preferentially use the dynamic temperature control algorithm (PID parameters Kp = 1.2, Ki = 0.8, Kd = 0.5) in the modular production line to achieve precise process control.

[0045] S109. According to the optimization direction of modular production capacity, obtain the real-time interference signals and welding quality data during the production process, adjust the shielding strategy and process parameters through electromagnetic interference suppression technology, and obtain the final production yield improvement plan.

[0046] Collect real-time interference signals and welding quality data through sensors to obtain a preliminary production process dataset. Use signal processing technology to filter the real-time interference signals and determine the characteristic parameters of electromagnetic interference. If the electromagnetic interference characteristic parameters exceed the preset threshold, adjust the shielding strategy through electromagnetic interference suppression technology to obtain an updated shielding configuration. Adjust the process parameters according to the updated shielding configuration to obtain optimized production control parameters. Analyze the welding quality data and the optimized production control parameters through a machine learning algorithm to judge the change trend of production yield. According to the production yield change trend, use the support vector machine algorithm to predict the modular production capacity and obtain the final optimization plan. For the final optimization plan, verify the production yield improvement amplitude through real-time data and determine the adjusted production parameter configuration.

[0047] Exemplarily, during the optimization process of modular production capacity, first, collect the electromagnetic interference signals and welding quality data in the production process in real time through a sensor network. For example, during the welding process, it is detected that the intensity of the electromagnetic interference signal with a frequency of 50 Hz reaches 80 dB, and at the same time, the defect rate of the welded joint is 5%. Use the fast Fourier transform algorithm to perform spectral analysis on the interference signal to identify that the main interference source is the nearby high-voltage transmission line. Based on this, use the adaptive filtering algorithm to suppress the interference signal and reduce the interference intensity to below 40 dB. Combine the welding quality data and establish a relationship model between the welding parameters and the quality through multiple linear regression analysis. It is found that when the welding current is in the range of 150 A to 180 A, the welding quality is the best. According to the analysis results, adjust the electromagnetic shielding strategy, adopt a double-layer copper mesh shielding structure, and improve the shielding effectiveness to 60 dB. At the same time, optimize the welding process parameters, stabilize the welding current at 165 A, and control the welding speed at 1.2 m / min. Continuously track the production data through the real-time monitoring system. After one week of operation, the welding defect rate is reduced to 1.5%, and the production yield is increased from the original 92% to 96.8%. The entire optimization process realizes the closed-loop control of data collection, analysis, decision-making, and execution, ensuring the stability of the production process and the continuous improvement of product quality.

[0048] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. Any modifications, equivalent replacements, and improvements made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. Method for controlling high-frequency interference and improving welding quality in the production of electronic control boards, characterized in that, The method includes the following steps: S101. Obtain the high-frequency interference data in the production of the electronic control board. Starting from the substrate structure design, adjust the substrate parameters through the partition isolation technology and the multi-layer board layout algorithm to obtain an optimized substrate structure design scheme; S102. For the optimized substrate structure design scheme, use electromagnetic simulation technology to analyze the signal integrity, and adjust the impedance matching and shielding layer configuration under the high-frequency interference control requirements to determine the substrate parameters for ensuring signal integrity; S103. According to the substrate parameters for ensuring signal integrity, obtain the temperature curve and component distribution data during the welding process, and adjust the welding parameters through the process dynamic adaptation algorithm to obtain a welding process flow suitable for multiple types of components; S104. Extract the real-time monitoring data from the welding process flow suitable for multiple types of components, use the full-process detection and traceability system to record the welding point status and defect characteristics, and judge the process stability index for improving welding quality; S105. For the process stability index for improving welding quality, obtain the production line adjustment data during changeover, and optimize the changeover process through the production flexibility adjustment technology to determine the equipment configuration scheme for optimizing changeover efficiency; S106. According to the equipment configuration scheme for optimizing changeover efficiency, obtain the production response time of multi-variety orders, and adjust the production line layout through the modular production capacity evaluation algorithm to obtain a production scheduling scheme for rapid response to multiple varieties; S107. Extract the defect rate data from the production scheduling scheme for rapid response to multiple varieties, use the defect rate reduction technology to analyze the failure distribution caused by welding defects and interference, and judge the comprehensive improvement effect of high-frequency interference control and welding quality improvement; S108. For the comprehensive improvement effect, obtain the historical data stored in the full-process detection and traceability system, and analyze the correlation between the substrate structure design and the process dynamic adaptation through the machine learning algorithm to determine the optimization direction of the modular production capacity; S109. According to the optimization direction of the modular production capacity, obtain the real-time interference signals and welding quality data during the production process, and adjust the shielding strategy and process parameters through the electromagnetic interference suppression technology to obtain the final production yield improvement scheme.

2. The method for high-frequency interference control and welding quality improvement in the production of electronic control boards according to claim 1, wherein S101 includes: Obtain the high-frequency interference data through sensors, and determine the interference distribution characteristics of the high-frequency interference data; Use the partition isolation technology to process the interference distribution characteristics to obtain a preliminary isolation layout; Optimize the preliminary isolation layout through the multi-layer board layout algorithm to obtain the adjusted substrate parameters; If the adjusted substrate parameters exceed the preset threshold, correct the adjusted substrate parameters through iterative calculation to obtain stable substrate parameters; Generate an improved substrate structure scheme according to the stable substrate parameters; Judge the degree of high-frequency interference suppression according to the improved substrate structure scheme to obtain an optimized design scheme; Verify the optimized design scheme through simulation to determine the final substrate structure.

3. The method for high-frequency interference control and welding quality improvement in the production of electronic control boards according to claim 1, wherein The S102 includes: Use electromagnetic simulation technology to analyze the substrate structure and obtain preliminary data on signal integrity; Judge the interference distribution characteristics of high-frequency interference according to the preliminary data, and determine the adjustment direction of the shielding layer configuration; Optimize the impedance matching for the adjustment direction to obtain the optimized substrate parameters; If the optimized substrate parameters meet the control requirements, verify the signal integrity through electromagnetic simulation technology to obtain verification data; Adjust the shielding layer configuration according to the verification data to determine the final substrate parameters; Use a machine learning algorithm to model the relationship between the final substrate parameters and signal integrity to obtain a prediction model; Analyze the influence of different design schemes through the prediction model to determine the substrate parameter configuration for ensuring signal integrity.

4. The method for high-frequency interference control and welding quality improvement in the production of electronic control boards according to any one of claims 1-3, characterized in that, The S103 includes: Obtain the signal integrity data in the substrate parameters, use a preset threshold to judge whether the signal integrity meets the requirements, and obtain the initial analysis result; Extract the temperature curve characteristics from the initial analysis result, and use a data analysis method to determine the change trend of the temperature curve; According to the change trend of the temperature curve, combined with the component distribution data, judge the adjustment direction of the welding parameters to obtain a preliminary adjustment plan; For the preliminary adjustment plan, use a dynamic adaptation algorithm to process the matching relationship between the welding parameters and the component distribution to obtain an optimized set of welding parameters; Obtain the execution order of the process flow from the optimized set of welding parameters, and determine the adaptation plan for multi-variety components through information processing technology; Adjust the temperature control logic in the process flow according to the adaptation plan, and use a linear regression algorithm to predict the stability after process optimization to obtain the final process flow; Verify the consistency between the component distribution and the signal integrity through the final process flow to obtain the adjusted process execution data.

5. The method for high-frequency interference control and welding quality improvement in the production of electronic control boards according to any one of claims 1-3, characterized in that, The S104 includes: Obtain real-time monitoring data from the welding process through sensors and store it in the database to obtain the original data set; Use a traceability system to classify the original data set, extract the welding point status and defect characteristics to obtain the status feature set; If the defect characteristics in the status feature set exceed the preset threshold, group the defect characteristics through a clustering algorithm to obtain the defect classification result; Compare and analyze the defect classification result with the welding point status, and use a statistical method to calculate the process stability index to obtain the stability quantification value; Compare the trend of the stability quantification value with historical data to judge the quality change trend and obtain the quality trend data; Adjust the welding process parameters through the quality trend data, update the real-time monitoring data to obtain the optimized process data set; If the stability quantification value in the optimized process data set reaches the preset standard, record the status feature set to obtain the final process optimization result.

6. The method for high-frequency interference control and welding quality improvement in the production of electronic control boards according to any one of claims 1-3, characterized in that, The S105 includes: Collect the production line adjustment data during tool change through sensors and store it in the database to obtain the tool change data set; Use a data analysis method to process the tool change data set, extract the characteristics related to process stability, and determine the process stability index; If the process stability index is lower than the preset threshold, classify the tool change data set through a clustering algorithm to judge the abnormal points in the tool change process; According to the analysis result of the abnormal points, adjust the production flexibility parameters to obtain the optimized tool change process plan; Verify the optimized changeover process plan through simulation technology to obtain a changeover efficiency evaluation value; If the changeover efficiency evaluation value is higher than the preset threshold, generate an equipment configuration plan according to the optimized changeover process plan; Adopt a database comparison method to match the equipment configuration plan with the existing equipment parameters to determine the final equipment configuration adjustment plan.

7. The method for high-frequency interference control and welding quality improvement in the production of electronic control boards according to any one of claims 1 to 3, characterized in that The S106 includes: Obtain multi-variety order data, use a classification algorithm to determine the order priority, and obtain an order processing sequence; Obtain production task assignments from the order processing sequence, combine with a modular production capacity evaluation algorithm, judge the equipment configuration plan, and obtain a production resource allocation result; If the production resource allocation result meets the preset response time threshold, adjust the equipment position parameters through the production line layout data to obtain an optimized production line layout plan; According to the optimized production line layout plan, adopt a dynamic adjustment mechanism to generate a real-time scheduling sequence for production tasks to obtain a fast response scheduling plan; Extract the production response time from the fast response scheduling plan, combine with the order processing efficiency data, judge whether it meets the response time optimization goal for multi-variety orders, and obtain the final production scheduling plan; If the final production scheduling plan does not meet the response time optimization goal, adjust the equipment configuration plan through the production resource allocation result to obtain a new production scheduling plan; According to the new production scheduling plan, obtain the production response time, determine the fast response ability for multi-variety orders, and obtain an optimized production operation sequence.

8. The method for high-frequency interference control and welding quality improvement in the production of electronic control boards according to any one of claims 1-3, characterized in that, The S107 includes: Obtain the defect rate data in the multi-variety production scheduling plan, use a statistical method to calculate the defect rate distribution of each variety, and obtain a preliminary defect rate statistical result; For the preliminary defect rate statistical result, use cluster analysis to divide welding defects and interference fault categories, and determine the distribution pattern of defects and interference; Extract the fault characteristics related to high-frequency interference from the distribution pattern. If the fault characteristics exceed the preset threshold, judge it as dominated by high-frequency interference to obtain the interference influence range; Screen the welding defect data through the interference influence range, use regression analysis to quantify the correlation between high-frequency interference and welding defects, and obtain the correlation coefficient distribution; Adjust the production scheduling parameters according to the correlation coefficient distribution, obtain the optimized defect rate data, and judge the quality improvement amplitude; Extract the improvement effect index from the optimized defect rate data, and determine the comprehensive improvement effect by comparing the preliminary defect rate statistical result; For the comprehensive improvement effect, obtain the production scheduling update plan for each variety and judge the improvement amplitude of the fast response ability.

9. The method for high-frequency interference control and welding quality improvement in the production of electronic control boards according to any one of claims 1-3, characterized in that, The S108 includes: Obtain historical data from the full-process inspection and traceability system, use a random forest algorithm to analyze the correlation between substrate structure design and process dynamic adaptation, and obtain a preliminary correlation model; Extract the change characteristics of process dynamics from the preliminary correlation model, conduct cluster analysis on the change characteristics, and determine the adaptation mode of the substrate structure; Obtain the key parameters from the adaptation mode, and judge the process adjustment amplitude through dynamic analysis to obtain the boundary conditions of modular production capacity; If the boundary condition exceeds the preset threshold, analyze the mapping relationship between the historical data and the optimization direction through a linear regression algorithm to determine the adjusted production capacity range; According to the adjusted production capacity range, obtain the matching degree between the process dynamics and the substrate structure, and judge the optimization space of the modular design; For the size of the optimization space, use cluster analysis to separate the highly adaptable modules to obtain the final optimization direction; Extract the key points for improving production capacity from the final optimization direction, verify the stability of the key points through the historical data, and determine the improvement plan for modular production capacity.

Images