Precision stamping forming method for radio frequency springs

Abstract

This invention relates to a precision stamping method for radio frequency (RF) spring sheets in the field of precision metal stamping. The method includes: embedding an optical fiber sensor array inside a mold to monitor displacement deviation and stamping pressure changes during the stamping process in real time, acquiring dynamic data of the mold-material contact surface; calculating the stress distribution of material flow based on the displacement deviation and stamping pressure data collected by the optical fiber sensors, and determining whether local stress concentration exists; calculating the displacement deviation caused by mold wear according to a mold wear compensation model, dynamically adjusting the mold position and pressure parameters, and compensating for wear-induced errors in real time; applying the real-time adjusted displacement control parameters to a multi-stage stamping control system to ensure that the displacement accuracy of each stamping stage reaches the micrometer level, meeting the high-frequency transmission performance requirements of the RF spring sheet; and updating the multi-stage stamping displacement control model and mold wear compensation model based on the final test results to continuously optimize the forming accuracy and production efficiency of the RF spring sheet.

Description

Technical Field

[0001] This invention relates to the field of information technology, specifically to the field of precision metal stamping, and particularly to a method for precision stamping and forming of radio frequency springs. Background Technology

[0002] In the high-precision stamping process of radio frequency (RF) spring sheets, the core technical problem faced by traditional processes is the contradiction between precise control of stamping displacement and real-time compensation for mold wear. Because RF spring sheets have extremely high requirements for dimensional accuracy (±0.01mm) and need to maintain stable contact performance during high-frequency signal transmission, displacement control during the stamping process must achieve micron-level precision.

[0003] However, traditional stamping processes employ a single stamping stage, resulting in relatively coarse displacement control that cannot meet the forming requirements of complex-shaped springs. Particularly in the fine blanking stage, the complex flow behavior of material within the die easily leads to localized stress concentrations, causing accumulated displacement deviations and ultimately affecting the overall accuracy of the spring. Furthermore, dies inevitably wear down after prolonged use, and traditional offline detection methods cannot capture minute changes in the die in real time, resulting in a gradual decline in stamping accuracy.

[0004] To address this issue, a multi-stage adaptive displacement control system needs to be developed. This system divides the stamping process into multiple micro-stages, setting independent displacement rate and pressure parameters for each stage. Simultaneously, fiber optic sensors embedded in the mold need to monitor stamping force and displacement deviations in real time, dynamically adjusting the mold's position and pressure based on the detection results to compensate for errors caused by mold wear. Solving this technical problem directly impacts the molding accuracy and production efficiency of RF spring sheets, and is crucial for improving the quality of mass production of 5G high-frequency spring sheets. Summary of the Invention

[0005] This invention provides a method for precision stamping of radio frequency springs, comprising the following steps:

[0006] S1. A multi-stage stamping division strategy is adopted. For radio frequency springs of different shapes, displacement control models of multiple micro-stamping stages are pre-established according to material properties and forming requirements. Each stage is set with independent displacement rate and pressure parameters to ensure uniform material flow.

[0007] S2. An array of fiber optic sensors is embedded inside the mold to monitor displacement deviation and stamping force changes in real time during the stamping process and obtain dynamic data of the contact surface between the mold and the material.

[0008] S3. Based on the displacement deviation and impact force data collected by the fiber optic sensor, calculate the stress distribution of the material flow and determine whether there is a local stress concentration phenomenon.

[0009] S4. If local stress concentration is detected, dynamically adjust the displacement rate and pressure parameters of the current stamping stage, and use an adaptive control algorithm to optimize the material flow behavior and reduce the impact of stress concentration on forming accuracy.

[0010] S5. By collecting displacement and pressure data through fiber optic sensors, analyze the subtle trends in mold wear, establish a data-driven real-time compensation model for mold wear, and predict the impact of mold wear on displacement control.

[0011] S6. Based on the mold wear compensation model, calculate the displacement deviation caused by mold wear, dynamically adjust the mold position and pressure parameters, and compensate for the error caused by wear in real time.

[0012] S7. Apply the real-time adjusted displacement control parameters to the multi-stage stamping control system to ensure that the displacement accuracy of each stamping stage reaches the micrometer level, meeting the high-frequency transmission performance requirements of the RF spring sheet.

[0013] S8. After the stamping is completed, the overall displacement accuracy of the spring sheet is detected by the fiber optic sensor array to determine whether the overall size meets the accuracy requirement of ±0.01mm. If it does not meet the requirement, the multi-stage stamping parameters are re-optimized.

[0014] S9. Based on the final test results, update the multi-stage stamping displacement control model and the mold wear compensation model to continuously optimize the forming accuracy and production efficiency of the RF spring sheet.

[0015] The technical solutions provided by the embodiments of the present invention may include the following beneficial effects:

[0016] This invention discloses a precision stamping method for radio frequency (RF) spring sheets. The method employs a multi-stage stamping strategy, using an embedded fiber optic sensor array within the die to monitor displacement deviations and stamping force changes in real time. Based on the collected data, the invention calculates the stress distribution of material flow and dynamically adjusts stamping parameters to optimize material flow behavior. Simultaneously, the invention establishes a data-driven real-time die wear compensation model to predict and compensate for the impact of die wear on displacement control. Through these innovative technologies, this invention achieves micron-level displacement accuracy control during the RF spring sheet stamping process, effectively meeting the requirements of high-frequency transmission performance. Ultimately, by continuously optimizing the multi-stage stamping displacement control model and the die wear compensation model, this invention significantly improves the forming accuracy and production efficiency of RF spring sheets. Detailed Implementation

[0017] The technical solution of the present invention will be clearly and completely described below with reference to the embodiments. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] The precision stamping method for radio frequency springs in this embodiment may specifically include:

[0019] Step S1: A multi-stage stamping division strategy is adopted. For radio frequency springs of different shapes, displacement control models of multiple micro-stamping stages are pre-established according to material properties and forming requirements. Independent displacement rate and pressure parameters are set for each stage to ensure uniform material flow.

[0020] The shape characteristics of the RF spring are acquired, and the forming requirements are determined based on material properties. Multiple micro-stamping stages are pre-divided. For each stamping stage, an independent displacement control model is established, and displacement rate and pressure parameters are set. Based on the material flow state, it is determined whether the uniform distribution condition is met. If not, the displacement rate and pressure parameters are adjusted. The adjusted parameters are obtained, the displacement control model is updated, and the material flow state is recalculated. If the material flow is uniform, the current stage parameters are saved, and the process proceeds to the next stamping stage. The above steps are repeated to complete the parameter setting for all stamping stages, resulting in the final displacement control model. A machine learning algorithm is used to optimize the displacement rate and pressure parameters based on historical data to improve the uniformity of material flow.

[0021] Specifically, the shape characteristics of RF springs are mainly reflected in surface finish, bending angle, and precision. The mechanical properties of commonly used materials such as beryllium copper and phosphor bronze must be fully considered during manufacturing. Taking a common spring contact as an example, the surface roughness requirement is 0.4 micrometers, and the material's springback characteristics need to be considered during forming. Determining forming requirements involves several key parameters, such as spring thickness tolerance, bending angle, and contact force. For RF springs with dimensions of 10 mm x 15 mm, the deformation is typically controlled within 0.05 mm. Considering material hardness and elastic modulus, the entire stamping process is divided into three main stages: rough stamping, fine stamping, and shaping. When modeling displacement control, the stress-strain relationship of the material at different deformation stages should be the focus. Taking the fine stamping stage as an example, the displacement rate is set at 0.5 mm per second, and the pressure parameter is controlled at 200 MPa. The material flow state is analyzed using stress distribution cloud maps to ensure that the stress difference in each region is within 5%. The judgment of material flow uniformity is mainly based on the stress distribution in the deformation zone. For beryllium copper springs, when local stress concentration exceeds 300 MPa, the displacement rate needs to be reduced to 0.3 mm / s, while the pressure is adjusted to 180 MPa to prevent local cracking. Through iterative optimization, the stress distribution deviation is reduced to within 3%. When updating the displacement control model, a piecewise linear interpolation method is used to establish the correspondence between pressure and displacement. At the bending point of the spring edge, the displacement rate needs to be reduced to 0.2 mm / s to ensure a smooth corner transition. Parameters for each stage are stored in a data table format, including key information such as time nodes, displacement amounts, and pressure values. Machine learning optimization uses support vector regression, based on historical production data, to establish a mapping relationship between displacement rate, pressure parameters, and finished product qualification rate. By analyzing thousands of sets of historical data, the optimal parameter combination is extracted. For example, in the fine blanking stage, the system automatically optimizes the displacement rate to 0.4 mm / s, increasing the product qualification rate to over 98%. During parameter optimization, factors such as mold temperature and material batches also need to be considered, establishing a multivariate optimization model.

[0022] Step S2: An optical fiber sensor array is embedded inside the mold to monitor the displacement deviation and stamping force changes during the stamping process in real time, and to obtain dynamic data of the contact surface between the mold and the material.

[0023] Displacement deviation data and stamping force change data collected by the fiber optic sensor array inside the mold are acquired. These data represent dynamic monitoring data of the mold-material contact surface during the stamping process. A preset threshold is used to determine if the displacement deviation data exceeds this threshold range. If so, a stamping process adjustment mechanism is triggered to generate optimized parameters for the mold-material contact state. A preset range is used to determine if the stamping force change data exceeds this range. If so, a stamping parameter correction module is activated to recalculate the stamping force distribution parameters. Based on the correlation between the displacement deviation data and the stamping force change data, a stamping process optimization model is established. The displacement deviation data, stamping force change data, mold-material contact state parameters, and stamping force distribution parameters are input into the stamping process optimization model. A machine learning algorithm is used to train the model to obtain the optimal parameter configuration for the mold-material contact surface. Based on the optimal parameter configuration, a stamping process parameter adjustment scheme is generated to adaptively optimize and control the stamping process.

[0024] Specifically, fiber optic sensor arrays are deployed at key stress and deformation locations inside the mold, achieving high-precision measurement through the Bragg grating principle. Multiple fiber optic probes, spaced 0.5 mm apart, form an array network at the front end of the mold, with each probe simultaneously monitoring displacement and pressure changes. The sensors convert optical signals into electrical signals, enabling real-time monitoring of millimeter-level displacement and kilopascal-level pressure. Taking the stamping of radio frequency spring sheets as an example, the fiber optic probes collect data at a frequency of 1,000 times per second to ensure the acquisition of transient change information. During normal stamping, displacement deviation is controlled within 0.02 mm, and stamping pressure fluctuation does not exceed 5% of the nominal value. If a displacement deviation of 0.03 mm is detected, the system automatically adjusts the upper mold height and pressure distribution. For sudden changes in stamping pressure, such as exceeding 10% of the nominal value, the stamping speed is immediately reduced to prevent material breakage. Correlation analysis is used to establish a mapping relationship between displacement deviation and stamping pressure changes. Historical data analysis reveals that increased displacement deviation is often accompanied by increased stamping pressure fluctuations. When the displacement deviation reaches 0.025 mm, the stamping pressure fluctuation amplitude increases to approximately 7%. Based on this correlation, the system establishes an early warning mechanism to intervene and adjust the system when the displacement deviation reaches 0.02 mm. The optimization model employs a support vector machine algorithm, with input features including multi-dimensional data such as displacement time series and pressure change trends. The model training samples are derived from multiple batches of successful stamping data, establishing a correspondence between optimal parameters and sensor data. Taking the side forming of a spring sheet as an example, when an increase in displacement deviation is detected, the system automatically reduces the stamping speed from 60 mm / s to 40 mm / s, while simultaneously adjusting the pressure distribution to ensure that the edge pressure is slightly lower than the center pressure. For spring sheets of different shapes and sizes, the system stores corresponding parameter configuration templates. For example, for straight spring sheets, displacement control is more stringent, with a deviation threshold set at 0.015 mm. Curved spring sheets allow for larger displacement changes, and the threshold can be relaxed to 0.025 mm. The system dynamically adjusts these parameters based on real-time monitoring data to achieve intelligent control. Through continuous optimization and self-learning, the system can adapt to different working conditions and maintain stable stamping quality.

[0025] Step S3: Based on the displacement deviation and impact force data collected by the fiber optic sensor, calculate the stress distribution of the material flow and determine whether there is a local stress concentration phenomenon.

[0026] The process involves acquiring displacement and impact force data from fiber optic sensors, combining them with preset material parameters, and generating stress values ​​for material flow dynamics using a preset calculation method. A stress distribution map of the material flow dynamics is then plotted based on these stress values. Stress concentration zones are identified within the stress distribution map. The stress values ​​of these concentration zones are then acquired; if the stress values ​​exceed a preset threshold, local stress concentration is identified within the material flow dynamics. A preset machine learning algorithm is used to extract features from the stress values ​​of the concentration zones, obtaining stress change characteristic information for the local region. Based on the stress change characteristic information, the degree of stress concentration in the local region is determined. A local stress concentration analysis report is generated, including the degree of stress concentration in the local region. Historical stress value data from a preset data source is acquired, and a preset machine learning model is trained using this historical stress value data to obtain a trained machine learning model. The trained machine learning model is then used to optimize the calculation accuracy of the stress distribution for the material flow dynamics. Based on the optimized stress distribution map and the local stress concentration analysis report, the material flow stress analysis results are output.

[0027] Specifically, fiber optic sensors can acquire displacement and stamping pressure data in molds using the Bragg grating principle. Taking the stamping of an automobile door panel as an example, a fiber optic sensor is arranged every 200 millimeters along the length of the mold cavity surface, forming a sensor array network. When the material deforms, the signal captured by the fiber optic sensor will experience wavelength drift. The optical signal is converted into an electrical signal by a demodulator, thereby obtaining the specific displacement and pressure values. In practical applications, assuming that during the stamping process of an automobile door panel, the sensor detects a displacement of 3 millimeters and a pressure of 400 MPa in the middle region of the workpiece, combined with the yield strength, elastic modulus, and other material parameters of the high-strength steel plate used in the door panel, the equivalent stress value of this region can be obtained using the finite element method: 500 MPa. By interpolating the stress values ​​at multiple measuring points, a stress distribution cloud map of the entire workpiece can be generated. In the stress distribution map, if the stress value at the corner of the door panel reaches 700 MPa, significantly higher than the 400 to 500 MPa of the surrounding area, and exceeds the preset threshold of 600 MPa, it can be determined that there is a stress concentration phenomenon at that location. For such stress concentration areas, convolutional neural networks can be used to extract their features, including stress gradient and stress peak location. By analyzing data from multiple consecutive stamping cycles, the stress variation over time in this area can be obtained. Localized stress concentration areas can be classified using different evaluation criteria. For example, stress concentration can be divided into three levels: slight, moderate, and severe, with a severe level defined as stress exceeding 80% of the material's yield strength. The generated stress concentration report records detailed information such as the location, extent, maximum stress value, and duration of each concentration area. A stress prediction model is established by collecting extensive historical production data. For example, a support vector machine algorithm can be used, with displacement, pressure, and material parameters as input features and stress as the output feature, to train the model and improve stress calculation accuracy. The optimized model can control the stress prediction error to within 5%, providing a reliable basis for real-time monitoring of the material forming process. The optimized stress distribution map, combined with the stress concentration report, can intuitively reflect the stress state and potential risks of the material during the forming process. This information is of great guiding significance for timely adjustment of process parameters, prevention of mold damage, and improvement of product quality. By establishing a material flow stress analysis system, precise control and optimization of the stamping process can be achieved.

[0028] Step S4: If local stress concentration is detected, dynamically adjust the displacement rate and pressure parameters of the current stamping stage, and use an adaptive control algorithm to optimize the material flow behavior and reduce the impact of stress concentration on forming accuracy.

[0029] The process involves acquiring stress distribution data during the stamping process and determining whether the stress distribution data exceeds a preset threshold. If the stress distribution data exceeds the preset threshold, the current displacement velocity parameters and pressure parameters are acquired, and adjustment values ​​are calculated. An adaptive control algorithm is used to dynamically adjust the displacement velocity parameters and pressure parameters based on the adjustment values, resulting in adjusted displacement velocity parameters and pressure parameters. Based on the adjusted displacement velocity parameters and pressure parameters, new stamping parameters are generated, and the operating status of the stamping equipment is updated. The updated stress distribution data is acquired, and it is determined whether the updated stress distribution data meets the optimization conditions. If the updated stress distribution data does not meet the optimization conditions, the adjustment values ​​are recalculated, and the dynamic adjustment steps are repeated until the updated stress distribution data meets the optimization conditions. If the updated stress distribution data meets the optimization conditions, the dynamic adjustment process is terminated, and the final displacement velocity parameters and pressure parameters are recorded. Based on the stress distribution data, the influence value of the stress set on the stamping forming accuracy is determined, and a stamping parameter optimization report is generated.

[0030] Specifically, real-time monitoring of stress distribution during the stamping process is crucial for ensuring molding quality. Taking automotive door panel stamping as an example, stress data is acquired through fiber optic sensors positioned at the four corners of the mold, with a stress threshold set at 85% of the material's yield strength. When the stress value at the right front corner reaches 320 MPa, exceeding the threshold of 300 MPa, the system immediately extracts the current displacement velocity of 20 mm / s and the pressure parameter of 400 kN. To adapt to changes in material flow, the system calculates adjustment values ​​based on the stress exceeding the limit. For example, the displacement velocity needs to be reduced by 15%, and the pressure needs to be reduced to 360 kN. The adaptive control algorithm dynamically adjusts the process parameters based on these adjustment values. The algorithm uses fuzzy control rules to map stress deviations to parameter adjustment amounts, achieving a smooth transition. Taking air conditioner panel stamping as an example, the parameter optimization process is demonstrated. Initially, slight wrinkles were found on the right side of the product, with stress concentration reaching 280 MPa. The system extracts the current displacement velocity of 15 mm / s and adjusts the velocity to 12 mm / s through iterative calculation. After three iterations of optimization, the stress value is reduced to 240 MPa, and the wrinkling phenomenon is significantly improved. The stamping machine control system updates its operating status in a timely manner based on optimized parameters. Taking the instrument panel bracket as an example, when stress concentration exceeds the threshold, the system updates the parameters within 0.3 seconds. By monitoring the workpiece forming status in real time, the system determines whether the material flow meets the optimization requirements. If not, data is re-collected and adjustment values ​​are calculated. The engine cover stamping process is used as a typical case to demonstrate the optimization effect evaluation process. The system records the product accuracy deviation caused by stress concentration, with an initial deviation of 0.8 mm. Through dynamic parameter adjustment, the deviation is reduced to 0.3 mm, meeting the process requirements. The system generates an optimization report, detailing the parameter adjustment process and its effects, providing a reference for subsequent similar products. In the door sill mold optimization case, by introducing a temperature sensor in conjunction with stress monitoring, it was discovered that uneven material flow was related to local overheating. Based on this, the system adjusted the pressure distribution and stamping rhythm, increasing the product qualification rate by 15%. This demonstrates the important role of multi-sensor collaboration and adaptive control in improving forming quality.

[0031] Step S5: Analyze the minute trends of mold wear using displacement and pressure data collected by fiber optic sensors, establish a data-driven real-time compensation model for mold wear, and predict the impact of mold wear on displacement control.

[0032] The process involves acquiring the displacement and pressure values ​​of the mold to obtain basic data on its operating status; calculating the trend of mold wear based on the changes in displacement and pressure values, and establishing a wear trend line; inputting the wear trend line into a pre-established data-driven compensation model to calculate the real-time compensation amount for mold wear; if the real-time compensation amount exceeds a preset threshold, a compensation mechanism is triggered to adjust the displacement control amount and reduce the impact of mold wear on displacement control; acquiring the real-time pressure value of the mold to determine its stress state; updating the parameters of the compensation model based on the stress state of the mold to optimize the compensation amount calculation results; applying the optimized compensation amount to the displacement control system to achieve real-time compensation for mold wear; continuously monitoring the displacement control amount, evaluating the compensation effect, and forming a closed-loop control.

[0033] Specifically, mold wear monitoring and compensation is a key technology in the precision machining field. Real-time monitoring can be achieved by collecting data through fiber optic sensors. Taking stamping dies as an example, a fiber optic grating sensor array is installed on the die surface. When the die moves up and down, the sensors can capture the displacement changes. If long-term monitoring shows that the displacement at a certain point gradually increases from the standard value of 0.05 mm to 0.1 mm, it indicates that wear may have occurred at that point. Time series analysis methods can help predict wear trends. By analyzing the collected data and plotting wear trend graphs, for example, after 10,000 stamping cycles, the wear on the die working surface shows a linear growth trend, with an increase of 0.02 mm for every additional 1,000 cycles. This trend analysis helps to detect abnormal wear in a timely manner. In establishing the compensation model, the wear characteristics under different working conditions need to be considered. For example, the wear rate of the die is relatively stable during stamping at room temperature, while the wear rate may accelerate under high temperature conditions. By establishing a compensation model that includes factors such as temperature, pressure, and speed, the wear amount can be predicted more accurately. The trigger threshold setting of the real-time compensation mechanism requires a trade-off between accuracy and efficiency. If the mold wear exceeds 0.08 mm, compensation will be initiated, and the system will automatically increase the displacement to offset the wear. Simultaneously, the compensation parameters are dynamically adjusted based on force data from the pressure sensor to ensure accuracy. When applying compensation in the displacement control system, the smoothness of the compensation process must be considered. For example, if a wear of 0.1 mm is detected, compensation should not be completed all at once, but rather in stages, compensating 0.02 mm each time, for a total of five stages, to avoid impact during the compensation process. The closed-loop control system continuously monitors the displacement to evaluate the compensation effect. If the measured displacement after compensation still deviates from the target value, the system will automatically fine-tune the compensation parameters. For example, if the actual displacement after compensation is 0.095 mm, differing from the target value of 0.1 mm by 0.005 mm, the system will increase the compensation in the next cycle. This dynamic feedback mechanism ensures the accuracy of the compensation, keeping the mold in optimal working condition.

[0034] Step S6: Based on the mold wear compensation model, calculate the displacement deviation caused by mold wear, dynamically adjust the mold position and pressure parameters, and compensate for the error caused by wear in real time.

[0035] The process involves: acquiring mold wear values ​​to characterize the degree of mold wear; using a pre-established mold wear compensation model to calculate the displacement value caused by mold wear; determining the mold deviation value if the displacement value exceeds a preset threshold; dynamically adjusting the mold position and pressure values ​​based on the deviation value; updating the mold compensation value using a real-time compensation algorithm based on the mold position and pressure values; outputting the mold adjustment value through a dynamic adjustment module; updating the mold's real-time parameter values ​​based on the adjustment value; and recalculating the mold model value using a model optimization algorithm based on the mold's real-time parameter values.

[0036] Specifically, mold wear values ​​are acquired using high-precision displacement sensors. These sensors sample at a frequency of up to 1,000 times per second, accurately capturing wear changes down to the micrometer level. For example, in a stamping die, 0.03 millimeters of wear may occur after eight hours of continuous operation; this minute change can lead to a decrease in product precision. The wear compensation model is built upon historical data, predicting wear trends by analyzing the relationship between mold usage time and wear amount. For instance, under normal operating conditions, a certain injection mold may experience approximately 0.1 millimeters of wear every 1,000 hours of operation, allowing for the establishment of a linear compensation model. When the detected wear reaches 0.05 millimeters, the displacement compensation mechanism is triggered. Mold deviation values ​​are determined by comparing the ideal position with the actual position. In practical applications, such as when a die-casting mold is operating, if the mold closing position is detected to deviate by 0.08 millimeters from the initial value and exceeds the preset threshold of 0.06 millimeters, position compensation is required. The dynamic adjustment of position and pressure values ​​employs a closed-loop control system. Taking a stamping die as an example, when a deviation of 0.04 millimeters in the upper mold position is detected, the system automatically increases air pressure compensation and simultaneously fine-tunes the upper mold position to bring the actual position back to the ideal range. The real-time compensation algorithm adaptively adjusts based on changes in working conditions. For example, in high-temperature environments, the thermal expansion of the mold can affect positional accuracy; the algorithm dynamically adjusts the compensation amount based on temperature. When the mold's operating temperature rises from room temperature to 200 degrees Celsius, the compensation amount needs to be increased by 0.02 millimeters. The adjustment value output by the dynamic adjustment module needs to consider multiple parameters. Taking injection molds as an example, not only the positional compensation amount but also factors such as injection pressure and holding time need to be considered. When the positional compensation amount is 0.05 millimeters, the injection pressure may need to be adjusted by 2%. Real-time parameter updates are based on sensor feedback data. During continuous stamping, the system updates positional parameters, pressure parameters, and other data every minute to ensure the accuracy of compensation. If the pressure value fluctuation exceeds 5% of the set value, the parameter optimization program is triggered. The model optimization algorithm adopts an adaptive approach, continuously optimizing model parameters based on actual operating data. For example, after the mold has run for 2,000 hours, analysis of accumulated data may reveal that the actual wear curve exhibits non-linear changes; in this case, the calculation method of the compensation model needs to be adjusted accordingly to improve prediction accuracy.

[0037] Step S7: Apply the real-time adjusted displacement control parameters to the multi-stage stamping control system to ensure that the displacement accuracy of each stamping stage reaches the micrometer level, meeting the high-frequency transmission performance requirements of the RF spring sheet.

[0038] The displacement parameters of the multi-stage stamping control system are obtained, and it is determined whether the displacement parameters are within a preset micrometer-level accuracy range. If the displacement parameters exceed the micrometer-level accuracy range, a real-time adjustment algorithm is used to correct the displacement parameters to obtain corrected displacement parameters. Based on the corrected displacement parameters, the control commands of the multi-stage stamping control system are updated. A correlation model between the displacement parameters of the RF spring and the high-frequency transmission performance is established. The correlation model is used to optimize the displacement parameters to obtain optimized displacement parameters that meet the preset high-frequency transmission performance requirements. Based on the optimized displacement parameters, the control parameters of the multi-stage stamping control system are reconfigured. The real-time displacement parameters of the multi-stage stamping control system are obtained; it is determined whether the real-time displacement parameters are stable within the micrometer-level accuracy range. If the real-time displacement parameters exceed the micrometer-level accuracy range, the steps of correcting and optimizing the displacement parameters are repeated.

[0039] Specifically, the multi-stage stamping control system acquires displacement parameters through a precision data acquisition device to achieve micron-level precision control. In practical applications, the system pre-sets an acceptable displacement accuracy range, typically ±2 microns. For example, when the displacement value at a stamping station reaches 3.5 microns, exceeding the preset accuracy range, the system immediately activates a real-time adjustment algorithm. The adjustment algorithm analyzes the direction and magnitude of the displacement deviation, calculates a compensation value, and corrects the displacement parameters to a reasonable range. For the high-frequency transmission performance of RF springs, displacement parameter control is particularly critical. Taking a 4G communication spring as an example, its operating frequency range is between 2.4GHz and 2.5GHz, requiring that the displacement error at each contact point of the spring not exceed 1 micron; otherwise, it will lead to increased signal transmission loss. By establishing a correlation model between displacement parameters and transmission performance, it was found that for every 0.5 microns increase in displacement error, the signal transmission loss increases by approximately 0.2dB. When optimizing displacement parameters, the system comprehensively considers multiple factors. Taking the stamping of a certain type of 5G RF spring as an example, the system first analyzes the structural characteristics of the spring and determines the key dimensional control points. Analysis of experimental data revealed that displacement control in the contact area at the front end of the spring contact piece is the most critical, as the displacement accuracy in this area directly affects signal transmission quality. Based on this, the system established an optimization model, focusing on controlling the displacement parameters in this area to maintain them within ±1 micrometer. The real-time monitoring system employs a high-precision sensor array to continuously sample the displacement parameters. Taking the production of a smartphone RF connector spring contact piece as an example, the system collects displacement data 100 times per second and analyzes the data to determine displacement trends. When a continuously increasing displacement value is detected, such as gradually increasing from an initial 1.2 micrometers to 1.8 micrometers, the system will initiate an adjustment process in advance to prevent the displacement from exceeding the allowable range. In actual production, the control of displacement parameters is closely related to factors such as mold life and material properties. For example, in continuous stamping production, mold wear can cause the displacement to gradually deviate from the target value. The system establishes a predictive model by monitoring displacement changes in real time. When the predictive model indicates that the displacement parameter may exceed the range within the next 4 hours, the system will adjust the stamping parameters in advance to ensure continuous and stable product quality. Furthermore, differences in the elastic modulus of different batches of raw materials also affect displacement accuracy; the system will automatically adjust the stamping pressure and speed according to material properties to achieve precise control.

[0040] Step S8: After the stamping is completed, the overall displacement accuracy of the spring sheet is detected by the fiber optic sensor array to determine whether the overall size meets the accuracy requirement of ±0.01mm. If it does not meet the requirement, the multi-stage stamping parameters are re-optimized.

[0041] The displacement of the spring body is obtained through a fiber optic sensor array; the deviation between the displacement and the target value is calculated; if the absolute value of the deviation is greater than a preset first threshold, a parameter optimization process is initiated; the parameter optimization process includes: using a genetic algorithm to iteratively optimize multi-stage stamping parameters to obtain a first optimized parameter set; inputting the first optimized parameter set into the stamping system to execute a new round of stamping; obtaining the displacement of the spring body after the new round of stamping; calculating a new deviation between the displacement after the new round of stamping and the target value; if the absolute value of the new deviation is greater than a preset second threshold, using a particle swarm optimization algorithm to perform a second optimization on the first optimized parameter set to obtain a second optimized parameter set; inputting the second optimized parameter set into the stamping system to complete the final stamping.

[0042] Specifically, the fiber optic sensor array uses grating sensing technology to measure the displacement of the spring sheet. Its principle is to use fiber optic probes to receive reflected light signals and convert changes in light signal intensity into displacement data. The sensor array consists of multiple probes, capable of simultaneously monitoring displacement changes in multiple areas of the spring sheet, providing comprehensive deformation data. For calculating the deviation between the displacement and the target value, the center point of the spring sheet is typically used as a reference, with the target value set within 0.1 mm. For example, during the stamping process, if the displacement at a certain point is 0.15 mm, compared to the target value of 0.08 mm, the deviation is 0.07 mm, exceeding the set threshold, requiring parameter optimization. When optimizing stamping parameters using a genetic algorithm, pressure, speed, and temperature are encoded as genes. For example, the initial population might have twenty parameter combinations, each containing different pressure and speed values. The effectiveness of the parameters is evaluated using a fitness function, and superior individuals are selected for cross-mutation to generate new parameter combinations. Through multiple iterations, such as optimizing from a pressure value of 800 Newtons to 750 Newtons, the displacement gradually approaches the target value. In the secondary optimization, the particle swarm optimization algorithm treats each set of parameters as a particle in space. It sets the initial particle swarm velocity and position, and searches for the optimal solution by updating the particle positions. For example, when the pressure parameter fluctuates around 750 Newtons, the algorithm can quickly find the optimal value of 748 Newtons. The spring sheet stamping system uses a servo motor drive, combined with a high-precision displacement sensor to achieve closed-loop control. When executing the optimized parameters, the system adjusts the stamping pressure and speed through real-time feedback. If a displacement of 0.1-2 millimeters is detected, the pressure is immediately adjusted to ensure displacement control accuracy. A fiber optic sensor array continuously collects data throughout the process, achieving closed-loop monitoring. The sensor sampling frequency can reach 1 kilohertz, ensuring real-time measurement. During the final stamping process, the product quality is comprehensively judged by multi-point displacement data, ensuring uniform deformation of all parts of the spring sheet and meeting radio frequency transmission requirements.

[0043] Step S9: Based on the final test results, update the multi-stage stamping displacement control model and the mold wear compensation model to continuously optimize the forming accuracy and production efficiency of the RF spring sheet.

[0044] Real-time detection values ​​during the RF spring forming process are acquired by sensors; these values ​​are compared with a preset forming degree threshold to obtain a difference value; if the difference value exceeds a preset allowable range, an update mechanism for the displacement control model is triggered; historical production data of the RF spring is acquired, and a mold wear compensation coefficient is determined by analyzing the historical production data; a mapping relationship model between stamping displacement and forming accuracy is established using a machine learning algorithm; the mold wear compensation coefficient and the mapping relationship model are input into the displacement control model to optimize the stamping displacement control parameters, resulting in optimized stamping displacement control parameters; based on the optimized stamping displacement control parameters, the stamping mold displacement is reset, and the RF spring is produced.

[0045] Specifically, real-time detection by displacement sensors is crucial in the RF spring forming process. Modern sensors employ high-precision laser or fiber optic systems capable of detecting micron-level displacement changes. For example, a displacement detector is mounted on the upper end of the spring; when the spring bends, it causes a change in the detection signal, from which the actual displacement value can be calculated. Determining the forming accuracy depends on the appropriate setting of preset thresholds. Typically, the forming degree of RF springs is required to be within ±0.01 mm, a threshold determined based on the product application scenario and process capabilities. When the forming degree of a certain area of ​​the spring exceeds the threshold, the system needs to adjust the control parameters promptly. The update mechanism of the displacement control model involves optimization across multiple dimensions. In actual production, the mold wears down after a certain number of stamping cycles, leading to a decrease in forming accuracy. By analyzing historical data, a mold life curve can be derived, thereby determining the wear compensation coefficient. For instance, statistics from a certain production line show that for every 10,000 stamping cycles, the forming dimension experiences a systematic deviation of 0.005 mm. Machine learning algorithms play a vital role in establishing the mapping relationship between stamping displacement and forming accuracy. Algorithms such as Support Vector Machines can be used, based on historical production data, to train a high-precision predictive model with parameters such as stamping displacement, pressure, and speed as input features and forming accuracy as output features. This model can predict the forming effect under different combinations of process parameters. The introduction of a wear compensation coefficient makes the optimization of control parameters more precise. Specifically, when the mold is used 5,000 times, the system will dynamically adjust the stamping displacement according to the pre-established compensation model. For example, if the original displacement is set to 2 mm, considering the impact of wear, the system will automatically adjust the displacement to 2.01 mm. The optimization of control parameters directly affects production efficiency. The optimized parameters must not only ensure product accuracy but also take into account the production cycle time. In practical applications, the system will automatically calculate the optimal stamping speed and displacement based on product characteristics and process requirements. For example, for a 10 mm long spring sheet, while ensuring accuracy, the stamping speed can be increased to 60 pieces per minute, improving efficiency by 20% compared to before optimization.

[0046] Although the present invention has been described in detail above with general descriptions and specific embodiments, modifications or improvements can be made to it, which will be obvious to those skilled in the art. Therefore, all such modifications or improvements made without departing from the spirit of the present invention fall within the scope of protection claimed by the present invention.

Claims

1. A method for precision stamping and forming of radio frequency springs, characterized in that, The method includes the following steps: S1. A multi-stage stamping division strategy is adopted. For radio frequency springs of different shapes, displacement control models of multiple micro-stamping stages are pre-established according to material properties and forming requirements. Each stage is set with independent displacement rate and pressure parameters to ensure uniform material flow. S2. An array of fiber optic sensors is embedded inside the mold to monitor displacement deviation and stamping force changes in real time during the stamping process and obtain dynamic data of the contact surface between the mold and the material. S3. Based on the displacement deviation and stamping pressure data collected by the fiber optic sensor, calculate the stress distribution of the material flow and determine whether there is a local stress concentration phenomenon. S4. If local stress concentration is detected, dynamically adjust the displacement rate and pressure parameters of the current stamping stage, and use an adaptive control algorithm to optimize the material flow behavior and reduce the impact of stress concentration on forming accuracy. S5. Analyze the minute change trends of mold wear using the displacement and pressure data collected by the fiber optic sensor, establish a data-driven mold wear compensation model, and predict the impact of mold wear on displacement control. S6. Based on the mold wear compensation model, calculate the displacement deviation caused by mold wear, dynamically adjust the mold position and pressure parameters, and compensate for the error caused by wear in real time. S7. Apply the real-time adjusted displacement control parameters to the multi-stage stamping control system to ensure that the displacement accuracy of each stamping stage reaches the micrometer level, meeting the high-frequency transmission performance requirements of the RF spring sheet. S8. After stamping is completed, use a fiber optic sensor array to detect the overall displacement accuracy of the spring sheet and determine whether the overall size meets the accuracy requirement of ±0.01mm. If not, re-optimize the multi-stage stamping parameters. S9. Based on the final test results, update the multi-stage stamping displacement control model and the mold wear compensation model to continuously optimize the forming accuracy and production efficiency of the RF spring sheet.

2. The method for precision stamping of radio frequency springs according to claim 1, characterized in that, S1 includes: The shape characteristics of the radio frequency spring are obtained, and the forming requirements are determined in combination with the material properties. Multiple micro-stamping stages are pre-divided. For each stamping stage, an independent displacement control model is established, and the displacement rate and pressure parameters are set. Based on the material flow state, determine whether the uniform distribution condition is met. If not, adjust the displacement rate and pressure parameters. Obtain the adjusted parameters, update the displacement control model, and recalculate the material flow state; If the material flows uniformly, save the current stage parameters and proceed to the next stamping stage; Repeat the above steps to complete the parameter setting for all stamping stages and obtain the final displacement control model; Machine learning algorithms are used to optimize displacement rate and pressure parameters based on historical data, thereby improving the uniformity of material flow.

3. The method for precision stamping of radio frequency springs according to claim 1, characterized in that, S2 includes: The displacement deviation data and stamping force change data collected by the fiber optic sensor array inside the mold are obtained. The displacement deviation data and stamping force change data are dynamic monitoring data of the contact surface between the mold and the material during the stamping process. If the displacement deviation data exceeds the threshold range, the stamping process adjustment mechanism is triggered to generate optimized parameters for the contact state between the mold and the material. If the stamping force change data exceeds the preset range, the stamping parameter correction module is activated to recalculate the stamping force distribution parameters. Based on the correlation between the displacement deviation data and the stamping force change data, an optimization model for the stamping process is established. The displacement deviation data, stamping force change data, die-material contact state parameters, and stamping force distribution parameters are input into the stamping process optimization model. The stamping process optimization model is trained using a machine learning algorithm to obtain the optimal parameter configuration of the die-material contact surface. Based on the optimal parameter configuration, a parameter adjustment scheme for the stamping process is generated, and adaptive optimization control of the stamping process is performed.

4. The method for precision stamping of radio frequency springs according to any one of claims 1-3, characterized in that, S3 includes: The displacement and impact force data collected by the fiber optic sensor are acquired, and the stress values ​​of the material flow dynamics are generated by combining the preset material parameters with the preset calculation method. Based on the stress values, a stress distribution diagram of the material flow dynamics is plotted; Identify the stress concentration areas in the stress distribution map; The stress value of the concentrated area is obtained. If the stress value of the concentrated area exceeds a preset threshold, it is determined that there is a local stress concentration phenomenon in the material flow dynamics. A preset machine learning algorithm is used to extract features from the stress values ​​in the concentrated area to obtain stress change feature information of the local domain; Based on the stress change characteristics, the degree of stress concentration in the local area is determined; Generate a local stress concentration analysis report, which includes the degree of stress concentration in the local area; Historical stress value data is obtained from a preset data source, and a preset machine learning model is trained using the historical stress value data to obtain the trained machine learning model. The trained machine learning model is used to optimize the accuracy of stress distribution calculation for the material flow dynamics; Based on the optimized stress distribution diagram and local stress concentration analysis report, the material flow stress analysis results are output.

5. The method for precision stamping of radio frequency springs according to any one of claims 1-3, characterized in that, S4 includes: Acquire stress distribution data during the stamping process and determine whether the stress distribution data exceeds a preset threshold; If the stress distribution data exceeds the preset threshold, the current displacement velocity parameters and pressure parameters are obtained, and the adjustment value is calculated. An adaptive control algorithm is used to dynamically adjust the displacement velocity parameter and the pressure parameter according to the adjustment value, so as to obtain the adjusted displacement velocity parameter and pressure parameter; Based on the adjusted displacement velocity parameters and pressure parameters, new stamping parameters are generated, and the operating status of the stamping equipment is updated. Obtain updated stress distribution data and determine whether the updated stress distribution data meets the optimization conditions; If the updated stress distribution data does not meet the optimization conditions, the adjustment value is recalculated, and the dynamic adjustment step is returned until the updated stress distribution data meets the optimization conditions. If the updated stress distribution data meets the optimization conditions, the dynamic adjustment process is terminated, and the final displacement velocity parameters and pressure parameters are recorded. Based on the stress distribution data, the influence of stress concentration on stamping accuracy is determined, and a stamping parameter optimization report is generated.

6. The method for precision stamping of radio frequency springs according to any one of claims 1-3, characterized in that, The S5 includes: Obtain the displacement and pressure values ​​of the mold to obtain basic data on the mold's operating status; For the changes in displacement and pressure, calculate the trend of mold wear and establish a wear trend line. The wear change trend line is input into a pre-established data-driven compensation model to calculate the real-time compensation amount for mold wear. If the real-time compensation amount exceeds the preset threshold, the compensation mechanism is triggered to adjust the displacement control amount and reduce the impact of mold wear on displacement control. Obtain the real-time pressure value of the mold to determine the stress state of the mold; Based on the stress state of the mold, update the parameters of the compensation model and optimize the compensation calculation results; The optimized compensation amount is applied to the displacement control system to achieve real-time compensation for mold wear; Continuously monitor displacement control parameters, evaluate compensation effectiveness, and form a closed-loop control system.

7. The method for precision stamping of radio frequency springs according to any one of claims 1-3, characterized in that, The S7 includes: Obtain the displacement parameters of the multi-stage stamping control system and determine whether the displacement parameters are within the preset micron-level accuracy range; If the displacement parameter exceeds the micrometer-level accuracy range, a real-time adjustment algorithm is used to correct the displacement parameter to obtain the corrected displacement parameter. The control commands of the multi-stage stamping control system are updated based on the corrected displacement parameters. Establish a correlation model between the displacement parameters of the RF spring and the high-frequency transmission performance; The displacement parameters are optimized using the correlation model to obtain optimized displacement parameters that meet the preset high-frequency transmission performance requirements. The control parameters of the multi-stage stamping control system are reconfigured based on the optimized displacement parameters. Obtain the real-time displacement parameters of the multi-stage stamping control system; Determine whether the real-time displacement parameter is stable within the micrometer-level accuracy range; If the real-time displacement parameter exceeds the micrometer-level accuracy range, the process of correcting and optimizing the displacement parameter is repeated.

8. The method for precision stamping of radio frequency springs according to any one of claims 1-3, characterized in that, S9 includes: Real-time detection values ​​are acquired during the radio frequency spring forming process, and the real-time detection values ​​are collected by sensors; The real-time detection value is compared with a preset forming degree threshold to obtain the difference value; If the difference value exceeds the preset allowable range, the displacement control model update mechanism is triggered; Obtain historical production data of radio frequency springs, and determine the mold wear compensation coefficient by analyzing the historical production data; A model for the mapping relationship between stamping displacement and forming accuracy is established using machine learning algorithms; The die wear compensation coefficient and the mapping relationship model are input into the displacement control model to optimize the stamping displacement control parameters, thereby obtaining the optimized stamping displacement control parameters. Based on the optimized stamping displacement control parameters, the stamping die displacement is reset to produce the RF spring sheet.