Control processing method and system based on gold foil processing

Abstract

The invention provides a control processing method and system based on gold foil processing, and the method comprises the steps: collecting real-time processing parameters of a target material in gold foil processing equipment, generating an initial processing parameter set, determining a first control model matched with the target material according to a hot pressing strength gradient sequence in the initial processing parameter set, and carrying out the control processing of the target material according to the first control model. According to an extension rate distribution curve in the initial machining parameter set, a second control model matched with the target material is determined, and based on real-time thickness deviation data and surface uniformity fluctuation data in the gold foil machining process, a target model in the first control model or the second control model is activated, and generating a dynamic adjustment instruction through the target model, inputting the dynamic adjustment instruction to an execution mechanism of the gold foil processing equipment, and correcting the pressure gradient of the hot pressing device or the rate distribution of the extension device. According to the method, the thickness distribution uniformity and the surface morphology stability of the gold foil in a complex processing scene can be improved, and the process quality and consistency of a gold foil finished product are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of data processing and intelligent control, and in particular to a control processing method and system based on gold foil processing. Background Art

[0002] Gold foil processing control aims to achieve precise control of the thickness and surface uniformity of gold foil by adjusting the parameters of the processing equipment. In the existing technology, the dynamic correlation between the hot pressing strength gradient and the extension rate distribution has not been effectively modeled, resulting in the thickness deviation compensation lagging behind the real-time processing state change, and the surface uniformity fluctuation cannot form a synergistic suppression with the extension rate adjustment, resulting in overpressure or extension tearing defects in local areas of the gold foil; at the same time, the static model cannot adapt to the hot pressing deformation differences of materials with different lattice structures, and it is easy to produce parameter control mismatch in complex processing scenarios, resulting in reduced processing accuracy and insufficient process stability. Summary of the invention

[0003] In view of this, the present invention provides a control processing method and system based on gold foil processing.

[0004] The technical solution of the embodiment of the present invention is achieved as follows: On the one hand, the present invention provides a control processing method based on gold foil processing, the method comprising: collecting real-time processing parameters of a target material in a gold foil processing device to generate an initial processing parameter set, the initial processing parameter set comprising a hot pressing intensity gradient sequence and an extension rate distribution curve; determining a first control model adapted to the target material according to the hot pressing intensity gradient sequence in the initial processing parameter set, the first control model comprising a dynamic mapping relationship between hot pressing intensity and thickness change; determining a second control model adapted to the target material according to the extension rate distribution curve in the initial processing parameter set, the second control model comprising a nonlinear association rule between extension rate and surface uniformity; activating a target model in the first control model or the second control model based on real-time thickness deviation data and surface uniformity fluctuation data during the gold foil processing, and generating a dynamic adjustment instruction through the target model; inputting the dynamic adjustment instruction into an actuator of the gold foil processing device to correct the pressure gradient of the hot pressing device or the rate distribution of the extension device.

[0005] On the other hand, the present invention provides a control processing system, including a memory and a processor, wherein the memory stores a computer program executable on the processor, and the processor implements the steps in the above method when executing the program.

[0006] The control processing method based on gold foil processing provided by the present invention generates an initial processing parameter set including a hot pressure strength gradient sequence and an extension rate distribution curve by collecting the real-time processing parameters of the target material in the gold foil processing equipment, constructs a dynamic mapping relationship of the first control model based on the hot pressure strength gradient sequence and a nonlinear association rule of the second control model based on the extension rate distribution curve, activates the target model in real time during the processing process to generate dynamic adjustment instructions and correct equipment parameters, so as to capture the correlation between the thickness of the gold foil and the surface uniformity in combination with the coordinated change law of the hot pressure strength and the extension rate, accurately compensates for the thickness deviation through the dynamic mapping relationship and suppresses the fluctuation of the surface uniformity through the nonlinear rule, and improves the control accuracy of the processing process. In addition, through the multi-dimensional parameter fusion of the hot pressure strength gradient sequence and the extension rate distribution curve, the real-time perception ability of the deformation characteristics of the gold foil material is enhanced, the prediction ability of the thickness change trend is strengthened by using the first control model, and the complex fluctuation pattern of the surface uniformity is analyzed by the second control model, forming a complementary optimization control strategy. Furthermore, based on the dynamic activation mechanism of real-time thickness deviation and surface uniformity fluctuation data, the core control requirements of different processing stages are adaptively matched, key defect areas are corrected first, and parameter interference is eliminated through model collaboration, so as to maintain the uniformity of gold foil thickness distribution and the stability of surface morphology in complex processing scenarios, and ultimately improve the process quality and consistency of gold foil finished products. BRIEF DESCRIPTION OF THE DRAWINGS

[0007] Figure 1 A schematic diagram of the implementation flow of a control processing method based on gold foil processing provided in an embodiment of the present invention.

[0008] Figure 2 A hardware entity schematic diagram of a control processing system provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0009] The embodiment of the present invention provides a control processing method based on gold foil processing, which can be executed by a processor of a control processing system, such as a computer system, a processing system of a gold foil processing device, etc. Figure 1 As shown, the method comprises the following steps: A control processing method based on gold foil processing, the method comprising: Step S100: collecting real-time processing parameters of the target material in the gold foil processing equipment to generate an initial processing parameter set, wherein the initial processing parameter set includes a hot pressing strength gradient sequence and an extension rate distribution curve.

[0010] The hot pressing intensity gradient sequence refers to the sequence of changes in pressure intensity applied by the hot pressing device at different time nodes during the gold foil processing. This sequence is collected through a pressure sensor array and includes the hot pressing intensity values ​​and time interval parameters of multiple continuous processing stages. The extension rate distribution curve represents the distribution law of the movement rate of the extension device transmission system in different processing areas. It is jointly obtained by the displacement sensor and the speed measurement module, including the lateral extension rate baseline value and the longitudinal extension rate change slope. The generation process of the initial processing parameter set is as follows: the dynamic physical property data of the target material in the hot pressing forming stage is captured in real time through a distributed sensor network, in which the hot pressing intensity gradient sequence is collected using multi-band pressure fluctuation detection technology, and the time series characteristics and spatial distribution characteristics of the pressure intensity are simultaneously recorded; the construction of the extension rate distribution curve is based on the real-time rate sampling points obtained by the high-precision laser velocimeter, and a continuous and smooth rate distribution curve is generated through the cubic spline interpolation algorithm. During the implementation process, each data node of the hot pressure intensity gradient sequence contains the pressure intensity value, the action timestamp and the spatial coordinate identifier, forming a multidimensional parameter matrix with time-space correlation; the extension rate distribution curve maps the movement rate of different transmission shafts to a unified time reference coordinate system through normalization processing to ensure the coordination of the extension rate of each area. Optionally, the acquisition frequency of the hot pressure intensity gradient sequence is not less than 1000Hz, and the spatial resolution of the extension rate distribution curve reaches 0.1mm level, so that the initial processing parameter set formed can accurately reflect the real-time operating status of the processing equipment.

[0011] As an implementation mode, in step S100, real-time processing parameters of the target material in the gold foil processing equipment are collected to generate an initial processing parameter set, including: Step S110: synchronously collecting lattice structure characteristic data of the target material in the pre-processing stage through a multi-channel sensor, the lattice structure characteristic data including a grain size distribution map and grain boundary density parameters.

[0012] A multi-channel sensor is a sensor that can collect data from multiple channels at the same time, and can be used to obtain a variety of information about the target material in the pre-processing stage. The lattice structure characteristic data reflects the lattice structure characteristics inside the target material. The grain size distribution map shows the distribution of grain sizes in the target material, and grains of different sizes show different characteristics in the map; the grain boundary density parameter indicates the density of grain boundaries in the target material. By synchronously collecting these data with a multi-channel sensor, we can fully understand the lattice structure characteristics of the target material in the pre-processing stage. For example, for a certain gold foil material, the multi-channel sensor can simultaneously collect its grain size distribution map, showing the distribution range of grain size and the proportion of grains of different sizes. At the same time, it can also obtain the grain boundary density parameter, which reflects the density of the grain boundaries.

[0013] For example, the grain size distribution map refers to a visualization map of the metal crystal microstructure obtained by jointly scanning an X-ray diffractometer and an electron backscatter diffraction system. The map records the diameter distribution data of different grains in the form of a two-dimensional matrix, and each matrix element corresponds to the equivalent diameter value of the grain at a specific coordinate point. The grain boundary density parameter characterizes the ratio of the total length of the grain boundaries per unit area to the surface area of ​​the material, and is quantitatively calculated using the secondary electron imaging technology of a scanning electron microscope. The synchronous acquisition mechanism of the multi-channel sensor is specifically implemented as follows: at the initial moment when the target material enters the preprocessing stage, a data acquisition system based on a time-triggered protocol is started to keep the sampling clock signals of the X-ray diffraction channel, the electron backscatter channel, and the optical microscopy channel strictly synchronized, ensuring that the spatial coordinate alignment accuracy of the grain size data and the grain boundary position data reaches the micron level. In the implementation process, the construction of the grain size distribution map adopts an adaptive grid division algorithm to divide the material surface into several detection units. Local peak detection and neighborhood clustering analysis are performed in each unit to generate a grain size heat map with spatial continuity; the calculation of the grain boundary density parameter is to identify the grain boundary contour line through the Sobel edge detection operator, and the measurement error caused by microscopic pores is eliminated by combining the morphological expansion operation. Optionally, the sampling resolution of the grain size distribution map is set to 0.5μm / pixel, and the update frequency of the grain boundary density parameter matches the material deformation rate in the pretreatment stage.

[0014] Step S120: Divide the target material into at least two sub-material clusters according to the peak interval in the grain size distribution map, and assign a corresponding initial hot-pressing gradient range and an extension rate reference value to each sub-material cluster.

[0015] Exemplarily, the peak interval refers to the numerical range of grain diameters that appear most frequently in the grain size distribution spectrum, and the probability density function fitting algorithm is used to identify the diameter interval boundaries with a cumulative probability of more than 70% in the distribution curve. The sub-material cluster division process adopts an adaptive partitioning technology based on the DBSCAN spatial clustering algorithm to aggregate detection units with similar grain sizes and continuous spatial positions into independent control areas. The initial hot-pressing gradient range is defined as the upper and lower limits of the pressure intensity required to be applied to each sub-material cluster during the hot pressing forming stage, and its value is dynamically calculated based on the grain size-pressure sensitivity curve: for sub-material clusters with smaller grain sizes, due to the higher grain boundary sliding resistance, the initial hot-pressing gradient range is set to a higher pressure range (e.g., 200-300MPa); conversely, sub-material clusters with larger grain sizes use a lower pressure range (e.g., 150-250MPa). The allocation of the extension rate benchmark value follows the grain boundary density-extension rate association rule, which is determined by looking up the pre-stored empirical parameter table. For example, when the grain boundary density of the sub-material cluster exceeds the critical threshold, the extension rate benchmark value is proportionally reduced to prevent grain boundary cracking. For example, the grain boundary density is 15μm / μm 2The corresponding reference speed is 0.8m / s and the density is 10μm / μm 2 During the implementation, the boundary contours of the sub-material clusters were optimized by the Voronoi diagram generation algorithm to ensure that the width of the transition area between adjacent clusters did not exceed two grain diameters to avoid sudden changes in processing parameters.

[0016] Step S130: extracting historical hot-pressing strength compensation parameters and extension rate correction coefficients matching each sub-material cluster from a preset parameter library.

[0017] The preset parameter library is a database that pre-stores various material processing related parameters. The historical hot pressure strength compensation parameters refer to the parameters used to compensate for the hot pressure strength for different materials or material clusters recorded in the previous gold foil processing process; the extension rate correction coefficient is the coefficient used to correct the extension rate. According to the characteristics of each sub-material cluster, the matching historical hot pressure strength compensation parameters and extension rate correction coefficients are extracted from the preset parameter library so as to adjust the hot pressure strength and extension rate in the subsequent processing process. For example, for a specific sub-material cluster, the preset parameter library may record that the hot pressure strength compensation parameters required for the material cluster in previous processing are a specific value range, and the extension rate correction coefficient is a specific coefficient value. For example, the historical hot pressure strength compensation parameters are stored in the process knowledge base module of the relational database, and its data structure includes fields such as the sub-material cluster identifier, the mean grain size, the pressure compensation amount, and the effective timestamp. The matching process uses a multi-dimensional feature similarity retrieval algorithm: the grain size distribution skewness, kurtosis and spatial variation coefficient of the current sub-material cluster are used as query vectors, and the historical records with the smallest Euclidean distance are searched in the preset parameter library to extract the corresponding hot pressing strength compensation parameters. The extraction of the extension rate correction coefficient is based on the fuzzy logic reasoning mechanism: a fuzzy rule library between grain boundary density, extension rate reference value and correction coefficient is established. For example, when the fluctuation range of grain boundary density exceeds 5%, the "high fluctuation-strong correction" rule is triggered to increase the correction coefficient from the reference value of 1.0 to 1.25. Optionally, the timeliness weight of the historical parameters is calculated by an exponential decay function to ensure that the weight of the influence of recent processing data on the correction coefficient accounts for more than 60%, thereby adapting to the time-varying characteristics of material properties.

[0018] Step S140: performing time domain filtering processing on the historical heat pressure strength compensation parameters to generate an optimized heat pressure strength gradient sequence corresponding to each sub-material cluster.

[0019] Exemplarily, the time domain filtering process can adopt a hybrid filtering scheme combining a sliding window weighted average algorithm and a Kalman prediction model. The specific implementation process is: arrange the historical thermal pressure strength compensation parameters in a time series, perform median filtering in a sliding window of length N (N takes 20-30 sampling points) to remove impulse noise, and then use a first-order Kalman filter to predict the pressure compensation trend at the next moment. The generation of the optimized thermal pressure strength gradient sequence follows the pressure superposition principle: the filtered compensation parameters are linearly superimposed with the lower limit of the initial thermal pressure gradient range. For example, when the initial range is 200-300MPa and the compensation parameter is +15MPa, an optimized gradient sequence of 215-315MPa is generated. During the dynamic adjustment process, the time interval of the gradient sequence is adaptively adjusted according to the thermal conductivity of the sub-material cluster: for gold foil materials with higher thermal conductivity, the gradient sequence uses dense time nodes (interval 0.5s); otherwise, it is extended to 1.0s interval to prevent thermal stress accumulation.

[0020] Step S150: performing frequency domain smoothing processing on the extension rate correction coefficient to generate an optimized extension rate distribution curve corresponding to each sub-material cluster.

[0021] Frequency domain smoothing is an operation to smooth the signal in the frequency domain. It can reduce the high-frequency components in the signal and make the signal smoother. Frequency domain smoothing of the extension rate correction coefficient can eliminate the high-frequency fluctuations in the extension rate correction coefficient and make the adjustment of the extension rate more stable. Through this processing, an optimized extension rate distribution curve corresponding to each sub-material cluster is generated, which can more accurately reflect the change of the extension rate of each sub-material cluster during the processing. For example, after frequency domain smoothing, the extension rate correction coefficient that originally had high-frequency fluctuations becomes more stable, and the generated optimized extension rate distribution curve is smoother, which can better guide the extension processing of gold foil. For example, frequency domain smoothing converts the correction coefficient sequence to the frequency domain through fast Fourier transform, uses Butterworth low-pass filter to suppress high-frequency fluctuation components, and the cutoff frequency is set to 1 / 3 of the mechanical resonance frequency of the extension device to avoid exciting vibration modes. The reconstruction process of the optimized extension rate distribution curve adopts the method of combining inverse Fourier transform with cubic spline interpolation: on the basis of retaining the low-frequency trend component, the filtered frequency domain signal is inversely transformed to obtain the time domain reference curve, and then the interpolation points are inserted to make the curve smoothness meet the C² continuity condition. In view of special material characteristics, the smoothing process introduces an adaptive weight mechanism: when the extension rate reference value of the sub-material cluster is detected to exceed 1.5m / s, the cutoff frequency of the filter is automatically reduced to enhance the high-frequency suppression capability and prevent surface ripple defects caused by high-speed extension.

[0022] Step S160: generating an initial processing parameter set by fusing the optimized hot pressing strength gradient sequence and the optimized extension rate distribution curve of each sub-material cluster.

[0023] The optimized hot-pressing strength gradient sequence and optimized extension rate distribution curve obtained after processing of each sub-material cluster are fused, that is, these data are integrated together to form an initial processing parameter set including hot-pressing strength gradient sequence and extension rate distribution curve. This set integrates the processing parameter information of each sub-material cluster and can provide a comprehensive and accurate basis for the subsequent gold foil processing control. For example, the optimized hot-pressing strength gradient sequence and optimized extension rate distribution curve of different sub-material clusters are merged and sorted according to certain rules to finally form a complete initial processing parameter set. For example, the fusion generation process adopts an integrated method combining the spatiotemporal registration algorithm with the conflict resolution strategy. First, the optimized parameters of each sub-material cluster are mapped to a unified processing coordinate system through the coordinate transformation matrix to ensure that the time nodes of the hot-pressing strength gradient sequence are accurately aligned with the spatial position of the extension rate distribution curve. The conflict resolution strategy mainly handles two types of anomalies: when the difference in hot-pressing gradients of adjacent sub-material clusters exceeds the safety threshold, a transition gradient segment is inserted to make it smoothly connected; when the extension rate distribution curve has a rate jump at the cluster boundary, a Sigmoid function is used to generate a gradual transition zone. The final generated initial processing parameter set is stored in the form of a multi-dimensional tensor, including four index axes: time dimension, space dimension, hot pressing strength value, and extension rate value, supporting real-time retrieval and dynamic update. During the implementation process, the verification of the parameter set is performed through the digital twin simulation system: the initial parameters are input into the virtual processing model, the thickness distribution and surface uniformity indicators are simulated and calculated, and if the results exceed the expected range, the parameter rollback mechanism is triggered to re-execute the optimization process of steps S140-S150.

[0024] As an implementation manner, when determining the first control model and the second control model, the following steps are included: Step S10: input the optimized heat pressure intensity gradient sequence into the preset recursive neural network model, update the hidden layer weights through multi-cycle training iterations, and generate a heat pressure intensity compensation function in the first control model. The heat pressure intensity compensation function is the target model in the first control model.

[0025] The preset recursive neural network model is a neural network with feedback connections, which can process sequence data and capture the timing information in the data. The optimized hot pressure intensity gradient sequence is obtained after time domain filtering, which more accurately reflects the changing law of hot pressure intensity of each sub-material cluster during the processing. The optimized hot pressure intensity gradient sequence is input into the recursive neural network model, and the model processes and learns the input data. During the multi-cycle training process, the weights of the hidden layer are continuously updated iteratively, so that the model can better fit the relationship between hot pressure intensity and thickness change. Finally, the hot pressure intensity compensation function in the first control model is generated, which can predict the intensity value that needs to be compensated according to the current hot pressure intensity situation to ensure that the thickness of the gold foil meets expectations. For example, in a gold foil processing scenario, the optimized hot pressure intensity gradient sequence is input into the recursive neural network model. After multiple training cycles, the model generates a hot pressure intensity compensation function, which can accurately calculate the hot pressure intensity compensation value required to achieve the target thickness according to the real-time change of hot pressure intensity.

[0026] Step S20: input the optimized extension rate distribution curve into the convolutional neural network model, adjust the convolution kernel parameters through the back propagation algorithm, and generate an extension rate correction mapping table in the second control model; the extension rate correction mapping table is the target model in the second control model.

[0027] The convolutional neural network model is good at processing data with spatial structure and can automatically extract features from the data. The optimized extension rate distribution curve is obtained after frequency domain smoothing, which more accurately reflects the change of the extension rate of each sub-material cluster during the processing process. The optimized extension rate distribution curve is input into the convolutional neural network model, and the model uses the convolution kernel to perform convolution operation on the data to extract the characteristic relationship between the extension rate and the surface uniformity. Through the back propagation algorithm, the parameters of the convolution kernel are adjusted according to the output error of the model, so that the model can better learn the nonlinear association rules between the extension rate and the surface uniformity. Finally, the extension rate correction mapping table in the second control model is generated, which records the corresponding surface uniformity correction information under different extension rates and can be used to guide the adjustment of the extension rate. For example, in actual processing, the optimized extension rate distribution curve is input into the convolutional neural network model, and after the convolution kernel parameters are adjusted by the back propagation algorithm, the extension rate correction mapping table is generated. According to this table, the extension rate can be adjusted according to different surface uniformity conditions.

[0028] Step S30: After the recursive neural network model and the convolutional neural network model converge, the key feature vectors in the hot pressure strength compensation function and the extension rate correction mapping table are extracted respectively.

[0029] After multiple rounds of training, the output error of the recursive neural network model and the convolutional neural network model gradually decreases and tends to be stable, that is, it reaches a convergence state. After convergence, it is necessary to extract key feature vectors from the hot pressure strength compensation function and the extension rate correction mapping table. Key feature vectors are vectors that can represent the core characteristics and performance of the model. By extracting these vectors, the adaptability of the model can be better evaluated and used for subsequent similarity matching and other operations. For example, after the model converges, through specific algorithms and methods, key feature vectors that can reflect the relationship between hot pressure strength and thickness change and the relationship between extension rate and surface uniformity are extracted from the hot pressure strength compensation function and the extension rate correction mapping table.

[0030] As an implementation mode, step S30, respectively extracting key feature vectors in the hot pressing strength compensation function and the extension rate correction mapping table, includes: Step S31: traverse the weight distribution of each hidden layer node in the recursive neural network model, extract the weight change trajectory that is temporally associated with the hot pressure intensity gradient sequence, and generate a first weight matrix.

[0031] The weight distribution of the hidden layer nodes in the recursive neural network model reflects the importance the model attaches to different input data during the learning process. Traverse the weight distribution of each hidden layer node to find the weights that are temporally associated with the thermal pressure intensity gradient sequence. The change trajectory of these weights can reflect the impact of thermal pressure intensity on the model output at different time points. These weight change trajectories are sorted to generate the first weight matrix. This matrix records the weight information related to the thermal pressure intensity gradient sequence, providing a basis for subsequent analysis of the characteristics of the thermal pressure intensity compensation function. For example, in the recursive neural network model, by traversing the weights of the hidden layer nodes one by one, those weights related to the temporal changes of the thermal pressure intensity gradient sequence are screened out, and their change trajectories are combined into the first weight matrix.

[0032] Step S32: Perform sliding correlation analysis of adjacent time windows on the first weight matrix to identify a target weight subset that maintains a stable fluctuation pattern over multiple training cycles.

[0033] Sliding correlation analysis of adjacent time windows is a method for analyzing the correlation of data in adjacent time intervals. This analysis is performed on the first weight matrix, and by calculating the correlation between weights in different time windows, those weights that maintain a stable fluctuation pattern over multiple training cycles are found. These weights constitute the target weight subset, which has an important stabilizing effect on the output of the heat pressure strength compensation function. For example, a sliding correlation analysis is performed on the first weight matrix, and the correlation coefficients of the weights are calculated in different time windows. The weights with stable correlation coefficients and consistent fluctuation patterns over multiple training cycles are screened out to form the target weight subset.

[0034] Step S33: Filter the activation weight nodes directly connected to the output end of the heat pressure strength compensation function from the target weight subset, extract their activation direction and amplitude data, and generate a first feature vector set.

[0035] The target weight subset contains multiple weight nodes, but not all nodes directly affect the output of the heat pressure intensity compensation function. The activation weight nodes directly connected to the output of the heat pressure intensity compensation function are screened out from the target weight subset. The activation of these nodes directly determines the output value of the heat pressure intensity compensation function. The activation direction (i.e., whether the weight is positive or negative, reflecting the enhancement or inhibition of the output) and amplitude data (i.e., the size of the weight, reflecting the degree of influence on the output) of these activation weight nodes are extracted, and these data are combined into the first eigenvector set. This set can accurately describe the key features of the heat pressure intensity compensation function. For example, in the target weight subset, find the activation weight nodes that are directly connected to the output of the heat pressure intensity compensation function, record their activation direction and amplitude, and generate the first eigenvector set.

[0036] Step S34: synchronously traverse the activation area distribution of each convolution kernel in the convolutional neural network model, extract the activation response pattern that is spatially correlated with the extension rate distribution curve, and generate a second weight matrix.

[0037] The convolution kernels in the convolutional neural network model will generate activation responses in different areas when processing input data. Synchronously traverse the activation area distribution of each convolution kernel to find activation response patterns that are spatially associated with the extension rate distribution curve. These patterns reflect the feature extraction of the convolution kernel when processing the extension rate distribution curve. These activation response patterns are sorted to generate a second weight matrix. This matrix records the convolution kernel activation information related to the extension rate distribution curve, and provides a basis for subsequent analysis of the characteristics of the extension rate correction mapping table. For example, in the convolutional neural network model, traverse the activation areas of each convolution kernel at the same time, screen out those activation response patterns related to the spatial characteristics of the extension rate distribution curve, and form them into a second weight matrix.

[0038] Step S35: Perform cross-channel activation peak detection on the second weight matrix to locate the target convolution kernel group that is repeatedly triggered in the key decision path of the extension rate correction mapping table.

[0039] Cross-channel activation peak detection is a method for detecting activation peaks of data on different channels. This detection is performed on the second weight matrix, and by analyzing the activation peaks on different channels, the convolution kernel groups that are repeatedly triggered in the key decision path of the stretch rate correction map are located. These target convolution kernel groups play an important role in determining the output results of the stretch rate correction map, and their repeated triggering indicates their key influence on the relationship between stretch rate and surface uniformity. For example, cross-channel activation peak detection is performed on the second weight matrix to find the convolution kernel groups with multiple activation peaks in the key decision steps of the stretch rate correction map, and they are located as target convolution kernel groups.

[0040] Step S36: Separate the activation response sequence synchronized with the change of the surface uniformity index from the target convolution kernel group, extract its spatial frequency and intensity distribution data, and generate a second feature vector set.

[0041] The target convolution kernel group contains activation responses of multiple convolution kernels, but not all responses are synchronized with the changes in the surface uniformity index. The activation response sequences that are synchronized with the changes in the surface uniformity index are separated from the target convolution kernel group. These sequences reflect the characteristics of the convolution kernel when processing information related to surface uniformity. The spatial frequency (i.e., the frequency of change of the activation response in space) and intensity distribution data (i.e., the distribution of the intensity of the activation response in space) of these activation response sequences are extracted, and these data are combined into a second feature vector set. This set can accurately describe the key features of the relationship between the extension rate correction mapping table and surface uniformity. For example, in the target convolution kernel group, those activation response sequences that are synchronized with the changes in the surface uniformity index are screened out, their spatial frequency and intensity distribution are recorded, and the second feature vector set is generated.

[0042] Step S37: Align the first feature vector set and the second feature vector set across model feature dimensions to eliminate feature scale differences caused by different neural network structures.

[0043] Since the first feature vector set and the second feature vector set come from the recursive neural network model and the convolutional neural network model respectively, these two models have different structures and feature extraction methods, resulting in differences in the feature dimensions and scales of the two feature vector sets. In order to be able to effectively compare and analyze them, they need to be aligned across model feature dimensions. Through specific algorithms and methods, the dimensions and scales of the two feature vector sets are adjusted to eliminate the feature scale differences caused by different neural network structures. For example, the first feature vector set and the second feature vector set are processed using normalization and standardization methods to make their feature dimensions and scales consistent.

[0044] Step S38: extracting key feature vectors that dynamically match the real-time processing parameters from the hot pressing strength compensation function and the extension rate correction mapping table according to the aligned feature vector set.

[0045] After completing the cross-model feature dimension alignment of the first feature vector set and the second feature vector set, key feature vectors that dynamically match the real-time processing parameters are further extracted from the heat pressure strength compensation function and the extension rate correction mapping table based on the aligned feature vector set. The real-time processing parameters will continue to change with the gold foil processing process. These key feature vectors can dynamically adjust the output of the model according to the real-time parameters to better adapt to the needs of the processing process. For example, combined with the real-time heat pressure strength and extension rate data, the key feature vectors that can accurately reflect the current processing status are selected from the aligned feature vector set, and are used for the dynamic adjustment of the heat pressure strength compensation function and the extension rate correction mapping table respectively.

[0046] Step S40: performing similarity matching between the key feature vector and the lattice structure feature data collected in real time, and verifying the compatibility of the first control model and the second control model.

[0047] The key feature vector represents the core features of the first control model and the second control model, while the lattice structure feature data collected in real time reflects the current lattice structure characteristics of the target material. The key feature vector is matched with the lattice structure feature data collected in real time for similarity, and the similarity index between the two, such as cosine similarity, is calculated to determine whether the first control model and the second control model are compatible with the current state of the target material. If the similarity is high, it means that the model can better reflect the processing characteristics of the target material and has good adaptability; otherwise, it means that the model may need further adjustment. For example, the cosine similarity between the key feature vector and the lattice structure feature data collected in real time is calculated. If the similarity reaches a certain threshold, the model is considered to be compatible; if the similarity is low, the model needs to be re-evaluated.

[0048] Step S50: When the similarity matching result is lower than a preset threshold, the training cycle of the recursive neural network model and the convolutional neural network model is readjusted until the key feature vector meets the adaptability condition.

[0049] The preset threshold is a pre-set similarity critical value used to determine whether the model is compatible with the target material. When the similarity matching result is lower than the preset threshold, it means that the first control model and the second control model are poorly compatible with the target material. At this time, it is necessary to readjust the training cycle of the recursive neural network model and the convolutional neural network model. By increasing or decreasing the training cycle, changing the training intensity and parameter update times of the model, the model can better learn the characteristics of the target material. The similarity matching process is repeated until the similarity between the key feature vector and the lattice structure feature data collected in real time reaches the preset threshold, that is, the adaptability condition is met. For example, when the similarity matching result is lower than the preset threshold, the training cycle of the recursive neural network model and the convolutional neural network model is increased, and training and similarity matching are performed again until the adaptability requirements are met.

[0050] Step S200: determining a first control model adapted to the target material according to the hot pressing intensity gradient sequence in the initial processing parameter set, wherein the first control model includes a dynamic mapping relationship between hot pressing intensity and thickness change.

[0051] The hot pressing intensity gradient sequence reflects the change of hot pressing intensity during the processing. By analyzing and processing the hot pressing intensity gradient sequence in the initial processing parameter set, the first control model adapted to the target material is determined. This model describes the dynamic mapping relationship between hot pressing intensity and the change in gold foil thickness, that is, how the change in hot pressing intensity affects the change in gold foil thickness. For example, when the hot pressing intensity increases, the thickness of the gold foil may decrease accordingly. The first control model can accurately reflect this dynamic change relationship, thereby providing a basis for adjusting the hot pressing intensity according to the change in gold foil thickness in the future.

[0052] Step S300: Determine a second control model adapted to the target material according to the extension rate distribution curve in the initial processing parameter set, wherein the second control model includes a nonlinear association rule between the extension rate and the surface uniformity.

[0053] The extension rate distribution curve shows the change of the extension rate of the gold foil during the extension process. By studying and analyzing the extension rate distribution curve in the initial processing parameter set, the second control model adapted to the target material is determined. The model contains the nonlinear association rules between the extension rate and the surface uniformity of the gold foil, that is, how the change of the extension rate affects the surface uniformity of the gold foil in a nonlinear way. For example, when the extension rate is too fast, the surface of the gold foil may become uneven. The second control model can describe this complex nonlinear relationship so that the extension rate can be adjusted according to the surface uniformity requirements during the processing.

[0054] Step S400: based on the real-time thickness deviation data and surface uniformity fluctuation data during the gold foil processing, activate the target model in the first control model or the second control model, and generate dynamic adjustment instructions through the target model.

[0055] Real-time thickness deviation data refers to the deviation data between the actual thickness of the gold foil and the expected thickness during the gold foil processing; surface uniformity fluctuation data refers to the fluctuation of the surface uniformity of the gold foil during the processing. Based on these real-time data, it is determined whether the first control model or the target model in the second control model needs to be activated. If the real-time thickness deviation data exceeds a certain threshold, it means that there is a large deviation in the thickness of the gold foil, and the target model in the first control model is activated at this time; if the surface uniformity fluctuation data exceeds a certain threshold, it means that there is a large fluctuation in the surface uniformity of the gold foil, and the target model in the second control model is activated at this time. After the target model is activated, dynamic adjustment instructions are generated according to the rules and algorithms in the model to adjust the relevant parameters of the gold foil processing equipment. For example, when the real-time thickness deviation data exceeds the threshold, the hot pressure strength compensation function in the first control model is activated, and a dynamic adjustment instruction for adjusting the hot pressure strength is generated according to the function.

[0056] As an implementation method, step S400, based on the real-time thickness deviation data and surface uniformity fluctuation data during the gold foil processing, activates the target model in the first control model or the second control model, and generates dynamic adjustment instructions through the target model, including: Step S410: Real-time monitoring of the thickness deviation trend line and surface uniformity oscillation amplitude during the gold foil processing.

[0057] The thickness deviation trend line reflects the trend of the thickness deviation of the gold foil over time. By real-time monitoring, the direction and degree of the thickness deviation of the gold foil can be understood. The surface uniformity oscillation amplitude indicates the fluctuation range of the surface uniformity of the gold foil within a certain period of time. Real-time monitoring of the amplitude can timely detect abnormal changes in the surface uniformity of the gold foil. For example, the thickness and surface uniformity data of the gold foil are collected in real time by a thickness sensor and a surface detection device, and a thickness deviation trend line is drawn and the surface uniformity oscillation amplitude is calculated. In specific implementation, for example, the following method can be referred to, the longitudinal thickness data of the gold foil is continuously collected at a sampling frequency of not less than 100Hz, and a thickness deviation time series curve is formed after sliding average filtering. The curve takes the process standard thickness value as the baseline and displays the thickness deviation of each detection point in real time. The surface uniformity oscillation amplitude is quantitatively calculated by a multi-spectral imaging system. A high-resolution linear array CCD camera is installed at the exit of the extension device to capture the reflected light intensity distribution of the gold foil surface at a rate of 50 frames per second. The surface texture consistency parameters are extracted by the grayscale co-occurrence matrix algorithm, and the oscillation frequency components obtained by wavelet transform decomposition are combined to generate a uniformity oscillation waveform containing amplitude values ​​and phase angles. During the monitoring process, the time window length of the thickness deviation trend line is set to the most recent 30-second processing period, and the calculation window of the surface uniformity oscillation amplitude is a 10-second moving average period, ensuring that the detection data can reflect the immediate working conditions while avoiding noise interference.

[0058] Step S420: When the thickness deviation trend line exceeds the first warning threshold and the surface uniformity oscillation amplitude is in a preset stable range, activating the hot pressing strength compensation function in the first control model.

[0059] The first warning threshold is a preset critical value of thickness deviation. When the thickness deviation trend line exceeds the threshold, it means that the thickness deviation of the gold foil has reached a level that requires adjustment. The surface uniformity oscillation amplitude is in the preset stable range, indicating that the surface uniformity of the gold foil is currently in a normal state. In this case, the hot pressure intensity compensation function in the first control model is activated, which is used to compensate and adjust the hot pressure intensity according to the thickness deviation. For example, when the thickness deviation trend line exceeds the first warning threshold, and the surface uniformity oscillation amplitude is within the preset stable range, the hot pressure intensity compensation function in the first control model is activated to prepare for adjusting the hot pressure intensity.

[0060] Step S430: Generate a dynamic pressure adjustment instruction for the hot pressing device according to the gradient compensation rule in the hot pressing intensity compensation function.

[0061] The gradient compensation rule in the hot pressing strength compensation function specifies the specific way to adjust the hot pressing strength according to the thickness deviation. According to the rule, combined with the current thickness deviation data, a dynamic pressure adjustment instruction for the hot pressing device is generated. The instruction contains the pressure value and adjustment method that the hot pressing device needs to adjust to ensure that the thickness of the gold foil can be restored to the expected range. For example, the hot pressing strength compensation function calculates the hot pressing strength value that needs to be increased or decreased according to the current thickness deviation, converts this information into a dynamic pressure adjustment instruction, and sends it to the hot pressing device. For example, the gradient compensation rule is encoded as a three-dimensional lookup table structure, whose input dimensions include the current pressure gradient value, the slope of the thickness deviation trend line, and the cumulative deviation, and the output is the compensation coefficient matrix of each pressure zone. The generation process of the dynamic pressure adjustment instruction is specifically as follows: the real-time thickness deviation data is normalized and input into the compensation function, the compensation pressure value of each pressure zone is obtained by interpolation calculation, and it is converted into the opening adjustment instruction of the hydraulic servo valve in combination with the response characteristic curve of the hot pressing device. For example, when the thickness deviation of the central area is detected to reach +4μm, the compensation function generates an instantaneous pressure compensation of -8MPa in the corresponding partition, and the adjacent partitions are weighted by distance to allocate a gradient compensation value of -4MPa to -2MPa. The command transmission adopts a timestamp synchronization protocol to ensure that the adjustment actions of each pressure partition are executed synchronously within ±5ms to avoid uneven stress caused by pressure wave propagation.

[0062] Step S440: When the surface uniformity oscillation amplitude exceeds the second warning threshold and the thickness deviation trend line is in a preset stable interval, activating the extension rate correction mapping table in the second control model.

[0063] The second warning threshold is a pre-set critical value of the surface uniformity oscillation amplitude. When the surface uniformity oscillation amplitude exceeds the threshold, it indicates that the surface uniformity of the gold foil has fluctuated greatly. The thickness deviation trend line is in the preset stable interval, indicating that the thickness of the gold foil is currently in a normal state. In this case, the extension rate correction mapping table in the second control model is activated, which is used to correct and adjust the extension rate according to the surface uniformity fluctuation. For example, when the surface uniformity oscillation amplitude exceeds the second warning threshold, and the thickness deviation trend line is in the preset stable interval, the extension rate correction mapping table in the second control model is activated to prepare for adjusting the extension rate.

[0064] Step S450: Generate a dynamic rate adjustment instruction for the extension device according to the rate distribution rule in the extension rate correction mapping table.

[0065] The rate distribution rules in the extension rate correction mapping table specify the specific way to adjust the extension rate according to the surface uniformity fluctuation. According to the rules, combined with the current surface uniformity fluctuation data, a dynamic rate adjustment instruction for the extension device is generated. The instruction contains the rate value and adjustment method that the extension device needs to adjust to ensure that the surface uniformity of the gold foil can be restored to the expected range. For example, the extension rate correction mapping table calculates the extension rate value that needs to be increased or decreased according to the current surface uniformity fluctuation, converts this information into a dynamic rate adjustment instruction, and sends it to the extension device. Exemplarily, the rate distribution rules are stored as two-dimensional polar coordinate grid data, and each grid unit is associated with a rate correction coefficient for a specific azimuth interval and a transmission chain position. The generation process of the dynamic rate adjustment instruction is to input the real-time detected surface uniformity oscillation amplitude and phase information into the correction mapping table, calculate the target rate value of each transmission unit by bilinear interpolation, and then convert it into the speed-torque curve of the servo motor according to the dynamic model of the transmission system. For example, when the uniformity amplitude exceeds the standard in the 90-degree azimuth area, the correction mapping table generates a +0.2m / s rate increase instruction in this area, and the adjacent 120-degree area is synchronously adjusted to -0.1m / s to form a rate gradient balance. The instructions are issued using a hierarchical verification mechanism, and the motion controller first verifies the mechanical feasibility of the instructions to exclude dangerous instructions that exceed the motor torque limit or the transmission belt speed ratio limit.

[0066] Step S460: If the thickness deviation trend line and the surface uniformity oscillation amplitude both exceed the warning threshold, the dynamic pressure adjustment instruction generated by the first control model is executed first, and the adjustment instruction of the second control model is delayed until the hot pressing device completes the correction operation.

[0067] When the thickness deviation trend line and the surface uniformity oscillation amplitude both exceed their respective warning thresholds, it indicates that there are major problems with the thickness and surface uniformity of the gold foil. Since the thickness problem may have a more critical impact on the quality of the gold foil, the dynamic pressure adjustment instruction generated by the first control model is executed first to adjust the hot pressing intensity to restore the thickness of the gold foil as soon as possible. After the hot pressing device completes the correction operation, the adjustment instruction of the second control model is executed to adjust the extension rate to improve the surface uniformity of the gold foil. For example, when the thickness deviation trend line and the surface uniformity oscillation amplitude both exceed the warning threshold, the pressure of the hot pressing device is first adjusted according to the dynamic pressure adjustment instruction generated by the first control model. After the hot pressing device completes the pressure adjustment and stabilizes the thickness of the gold foil, the rate of the extension device is adjusted according to the adjustment instruction generated by the second control model.

[0068] Step S500: inputting a dynamic adjustment instruction to an actuator of the gold foil processing equipment to correct a pressure gradient of a hot pressing device or a rate distribution of an extension device.

[0069] Dynamic adjustment instructions are instructions for adjusting processing parameters generated according to real-time data and control models during the gold foil processing process. These instructions are input into the actuator of the gold foil processing equipment, and the actuator corrects the pressure gradient of the hot pressing device or the rate distribution of the extension device according to the instructions. For example, the dynamic pressure adjustment instruction of the hot pressing device is input into the actuator of the hot pressing device, and the actuator adjusts the pressure of each area in the hot pressing device according to the instruction so that the pressure gradient meets the adjustment requirements; the dynamic rate adjustment instruction of the extension device is input into the actuator of the extension device, and the actuator adjusts the rate of the extension device at different positions according to the instruction so that the rate distribution is corrected. Exemplarily, the dual over-limit working condition processing adopts a priority arbitration strategy to establish a quantitative influence coefficient matrix of thickness deviation and surface uniformity. When the influence coefficient of thickness deviation exceeds 1.5 times of uniformity, the hot pressing correction is determined to be a priority task. The specific implementation of the delayed execution mechanism includes: temporarily storing the rate adjustment instruction generated by the second control model in the ring buffer, and starting the hot pressing correction completion monitoring thread at the same time. When the feedback pressure gradient of the hot pressing device reaches 95% of the target value and remains stable for more than 3 seconds, the extension adjustment instruction in the buffer is released. During the waiting period, the system continuously monitors the surface uniformity status. If the deterioration rate exceeds the preset safety threshold (such as an increase of more than 10% per minute), the hot pressure correction process is interrupted and the emergency collaborative adjustment mode is activated. In the collaborative mode, the hot pressure compensation amount is reduced by 50%, freeing up system resources to synchronously perform some extension rate adjustments to prevent overall processing failure caused by a single parameter correction.

[0070] As an implementation method, step S500, inputting a dynamic adjustment instruction to an actuator of the gold foil processing equipment to correct a pressure gradient of a hot pressing device or a rate distribution of an extension device, includes: Step S510: According to the gradient compensation value in the dynamic pressure adjustment instruction, the instantaneous pressure value of each pressure zone in the hot pressing device is adjusted in stages.

[0071] The gradient compensation value in the dynamic pressure adjustment instruction is used to compensate for the numerical value of the hot pressing intensity deviation. According to this value, the instantaneous pressure value of each pressure partition in the hot pressing device is adjusted in stages. Staged adjustment can avoid the adverse effects of excessive pressure adjustment on the gold foil, making the pressure adjustment more stable and precise. For example, the dynamic pressure adjustment instruction requires the pressure of a certain pressure partition to be increased. According to the gradient compensation value, the pressure increase process is divided into several stages, and a certain pressure value is increased in each stage to gradually achieve the desired pressure adjustment effect. Exemplarily, the gradient compensation value is stored in the form of a three-dimensional pressure tensor, including an axial pressure component, a radial pressure component and a time phase parameter. The staged adjustment process adopts a pressure wavefront propagation control strategy to divide the working surface of the hot pressing device into a 5×5 grid pressure partition. The instantaneous pressure value of each partition is incrementally adjusted according to the corresponding element in the compensation value matrix. The adjustment phase is divided into a warm-up period, a main adjustment period, and a stable period: During the warm-up period, a 10% trial load of the compensation value is performed, and the duration is set to 1.2 times the response delay time of the hydraulic system (usually 200-300ms); During the main adjustment period, the remaining 90% of the compensation value is applied three times according to the exponential decay curve, and the interval between each loading is 1 / 4 of the system mechanical resonance period (about 50ms) to avoid exciting pressure oscillations; During the stable period, the final pressure value is maintained until the thickness feedback data meets the standard. For example, when the dynamic pressure adjustment instruction requires the pressure value of partition A3 to increase by 15MPa, the system applies 1.5MPa during the warm-up period, and loads 4.5MPa, 6.0MPa and 3.0MPa three times during the main adjustment period to ensure the continuity of the second-order derivative of the pressure climb curve.

[0072] Step S520: After each pressure adjustment, collect the real-time feedback data of the gold foil thickness and perform deviation calculation with the predicted value of the hot pressure strength compensation function.

[0073] After each adjustment of the pressure of the hot pressing device, the real-time thickness data of the gold foil is collected through the thickness sensor to obtain the real-time feedback data of the gold foil thickness. This data is compared with the predicted value of the hot pressing strength compensation function, and the deviation between the two is calculated. Through the deviation calculation, it is possible to understand whether the pressure adjustment has achieved the expected effect and whether further pressure adjustment is required. For example, after a pressure adjustment, the real-time thickness data of the gold foil is collected, compared with the thickness value predicted by the hot pressing strength compensation function, and the thickness deviation value is calculated.

[0074] Step S530: If the deviation value continues to decrease, the current pressure gradient adjustment direction is maintained until the target thickness range is reached.

[0075] If the deviation value continues to decrease, it means that the current pressure gradient adjustment direction is correct and is moving towards restoring the thickness of the gold foil to the target thickness range. In this case, maintain the current pressure gradient adjustment direction and continue to adjust the pressure until the thickness of the gold foil reaches the target thickness range. For example, after several consecutive pressure adjustments, the calculated thickness deviation value gradually decreases, indicating that the pressure adjustment is effective. Continue to adjust the pressure in the current adjustment direction until the thickness of the gold foil meets the requirements. Exemplarily, the criterion for the continuous reduction of the deviation is: the absolute value decrease rate of the deviation for three consecutive sampling cycles exceeds 5%, and the directional consistency coefficient is greater than 0.85. The maintenance of the pressure gradient adjustment direction is achieved by locking the servo valve control current, and the update cycle of the control signal is extended to three times the normal cycle (about 300ms). At the same time, the pressure micro-oscillation suppression algorithm is started, and a high-frequency disturbance signal of ±0.5MPa (frequency 50Hz) is superimposed near the target pressure value to eliminate the hysteresis effect caused by static friction. The boundary condition of the target thickness range is set to ±1.5 times the standard deviation of the process standard value. When the actual thickness value enters this range, the system automatically switches to the pressure maintenance mode and adjusts the working state of the hydraulic system to constant pressure output.

[0076] Step S540: If the deviation value shows an expanding trend, the pressure gradient is adjusted in the opposite direction and the parameter iteration of the heat pressure strength compensation function is re-triggered.

[0077] If the deviation value shows an expanding trend, it means that the current pressure gradient adjustment direction may be incorrect, and the pressure gradient needs to be adjusted in the opposite direction. At the same time, the parameter iteration of the hot pressure strength compensation function is retriggered, that is, the parameters of the hot pressure strength compensation function are recalculated to find a more suitable pressure adjustment method. For example, during the pressure adjustment process, it is found that the thickness deviation value gradually increases. At this time, the pressure adjustment direction is changed, and the parameters of the hot pressure strength compensation function are recalculated, and the pressure is adjusted again. Exemplarily, the warning trigger condition for the deviation expansion is: the deviation growth rate of two adjacent sampling cycles exceeds 10%, or the direction consistency coefficient is less than -0.7. The reverse adjustment strategy adopts a hybrid algorithm combining Bang-Bang control and fuzzy logic: first, emergency reverse compensation is performed, and the current pressure gradient value is reversely applied with a compensation amount of 200% within 50ms; then the fuzzy reasoning engine is started to generate an accurate reverse adjustment coefficient based on the deviation expansion rate, mold temperature change rate and material yield strength parameters. The parameter iteration process resets the training cycle counter of the hot pressure strength compensation function, adopts an incremental learning mechanism, inserts the current working condition data as a new sample into the first place of the training set, and recalculates the hidden layer weight matrix. During the iteration process, the learning rate is increased to five times the initial value to accelerate the convergence of the model.

[0078] Step S550: dynamically adjusting the transmission speed of the extension device in different sections according to the speed distribution parameters in the dynamic speed adjustment instruction.

[0079] The rate distribution parameters in the dynamic rate adjustment instruction specify the transmission rate that the extension device needs to adjust in different sections. According to these parameters, the transmission rate of the extension device in different sections is dynamically adjusted so that the rate distribution of the extension device meets the adjustment requirements. For example, the dynamic rate adjustment instruction requires an increase in the extension rate in a certain section and a decrease in the extension rate in another section. According to these requirements, the transmission rate of the extension device in the corresponding section is adjusted. Exemplarily, the rate distribution parameters are encoded as an angular velocity-linear velocity mapping table in a polar coordinate system, and each section corresponds to a target linear velocity value within a specific azimuth range. The dynamic adjustment process uses a multi-axis synchronous interpolation algorithm to decompose the transmission system into a main drive shaft and six driven shafts. The main shaft runs at a reference rate, and the driven shaft calculates the phase difference compensation amount according to the mapping table parameters. During the adjustment process, the acceleration of each drive shaft is limited to within 5m / s² to prevent gold foil wrinkles caused by inertial overshoot. For example, when the dynamic rate adjustment instruction requires the linear speed of the 120-150 degree section to be increased by 0.3 m / s, the system first calculates the angular acceleration requirement of the driven shaft corresponding to the section, and then implements the speed ramp-up through real-time adjustment of the electronic gear ratio to ensure that the speed difference between adjacent sections does not exceed 0.1 m / s.

[0080] Step S560: During the rate adjustment process, the optical sensor is used to detect the change in the uniformity of the gold foil surface in real time, and the correction coefficient in the extension rate correction mapping table is dynamically updated according to the detection result.

[0081] In the process of adjusting the rate of the extension device, an optical sensor is used to detect the change in the uniformity of the gold foil surface in real time. The optical sensor can obtain the uniformity data of the gold foil surface by detecting information such as reflected light or transmitted light on the gold foil surface. According to the detection results, the correction coefficient in the extension rate correction mapping table is dynamically updated so that the correction coefficient can more accurately reflect the relationship between the current uniformity of the gold foil surface and the extension rate, thereby better guiding the adjustment of the extension rate. For example, the optical sensor detects that the uniformity of a certain area on the gold foil surface has deteriorated. According to the detection result, the correction coefficient corresponding to the area in the extension rate correction mapping table is adjusted to adjust the extension rate of the area.

[0082] As an implementation mode, step S560, using an optical sensor to detect the change in the uniformity of the gold foil surface in real time, and dynamically updating the correction coefficient in the extension rate correction mapping table according to the detection result, may specifically include: Step S561: Continuously collect the reflected light intensity distribution of the gold foil surface in different sections of the extension device through the multi-spectral imaging unit to generate a real-time surface image data set.

[0083] The multispectral imaging unit can collect the reflected light intensity distribution information of the gold foil surface in multiple spectral bands. The unit continuously collects the reflected light intensity distribution of the gold foil surface in different sections of the extension device, organizes and stores the collected data, and generates a real-time surface image data set. This data set contains the reflected light intensity information of the gold foil surface at different positions and different spectral bands, which can reflect the microscopic characteristics and uniformity of the gold foil surface. For example, the multispectral imaging unit continuously collects the reflected light intensity of the gold foil surface in each section of the extension device over a period of time, and combines these data into a real-time surface image data set.

[0084] Step S562: grayscale gradient analysis is performed on the real-time surface image data set to extract the surface texture direction consistency parameter and the reflection intensity dispersion index corresponding to each extended segment, and generate a uniformity distribution index set.

[0085] Grayscale gradient analysis (such as directional gradient histogram algorithm) is used to analyze the changes in grayscale values ​​in an image. Grayscale gradient analysis of the real-time surface image data set can extract the surface texture directional consistency parameters and reflective intensity dispersion indexes corresponding to each extended section. The surface texture directional consistency parameter reflects the consistency of the texture direction of the gold foil surface. The higher the consistency, the more uniform the surface; the reflective intensity dispersion index indicates the dispersion of the reflective intensity of the gold foil surface. The smaller the dispersion, the more uniform the surface. These parameters and indicators are sorted out to generate a uniformity distribution index set, which can comprehensively reflect the uniformity of the gold foil surface in each extended section. For example, through grayscale gradient analysis, the surface texture directional consistency parameters and reflective intensity dispersion indexes of the gold foil surface in each extended section are calculated, and these data are combined into a uniformity distribution index set.

[0086] Step S563: compare the uniformity distribution index set with the reference uniformity template stored in the extension rate correction mapping table section by section to identify abnormal extension sections whose surface uniformity deviation exceeds a threshold under the current rate distribution.

[0087] The reference uniformity template stored in the extension rate correction mapping table is a pre-set reference standard for the uniformity of the gold foil surface. The uniformity distribution index set is compared with the reference uniformity template section by section, that is, the uniformity index of each extension section of the gold foil surface is compared with the reference template, and the surface uniformity deviation of each section is calculated. When the surface uniformity deviation of a section exceeds a preset threshold, the section is identified as an abnormal extension section. For example, the surface texture directional consistency parameter and the reflective intensity dispersion index of each extension section in the uniformity distribution index set are compared with the reference uniformity template, and the deviation is calculated. If the deviation of a section exceeds the threshold, it is marked as an abnormal extension section. As an exemplary implementation method, the reference uniformity template can be derived from a statistical model under the historical optimal process parameters, and stored as a two-dimensional lookup table containing azimuth intervals, directional consistency reference values, and reflective intensity dispersion reference values. The comparison process uses a double difference analysis method: first, the absolute deviation between the real-time directional consistency parameter and the reference value is calculated. When the deviation exceeds 15%, it is marked as a directional anomaly; the relative change rate of the dispersion of the reflective intensity is calculated simultaneously, and when it exceeds 20%, it is marked as a reflection anomaly. The comprehensive judgment logic of the abnormal extension section is: when a section meets the conditions of both directional anomaly and reflection anomaly, or a single abnormal indicator exceeds 2 times the threshold, it is judged as a severely deviated section. For example, in the 60-120° section, a directional consistency deviation of 32% and a reflection dispersion change rate of 18% were detected. Although they did not meet the double abnormality standard, they were still identified as abnormal sections because the directional deviation exceeded twice the threshold (15%×2=30%).

[0088] Step S564: According to the deviation direction and amplitude of the abnormal extension segment, the corresponding rate compensation mode is matched from the historical correction record, and the correction coefficient weight associated with the abnormal extension segment in the extension rate correction mapping table is adjusted based on the rate compensation mode.

[0089] The deviation direction and amplitude of the abnormal extension section reflect the degree and direction of the surface uniformity of the section deviating from the benchmark standard. Based on this information, the matching rate compensation mode is searched from the historical correction record, and the rate adjustment mode used in the past to deal with similar surface uniformity problems is stored in the historical correction record. Based on the matched rate compensation mode, the correction coefficient weight associated with the abnormal extension section in the extension rate correction mapping table is adjusted to change the extension rate adjustment method of the extension section. For example, if the surface uniformity deviation of the abnormal extension section is caused by the extension rate being too fast, the corresponding rate reduction compensation mode is found from the historical correction record, and the correction coefficient weight associated with the abnormal extension section in the extension rate correction mapping table is adjusted according to the mode to reduce the extension rate of the extension section. Exemplarily, the historical correction record is stored in the process knowledge graph of the relational database, and each record contains the abnormal type code (direction deviation / reflection deviation / compound deviation), the deviation amplitude range (0-15% / 15-30% / 30%+) and the effective compensation mode hash value. The matching process uses a similarity retrieval algorithm based on random forests: the directional consistency deviation rate, reflective discreteness change rate and azimuth position of the current abnormal section are input into the pre-trained matching model, and the three closest historical cases are output. The specific operation of the rate compensation mode adjustment is: select the historical case with the highest matching degree, scale its compensation coefficient weight according to the current deviation amplitude ratio, and perform weighted fusion with the original weight in the extended rate correction mapping table. For example, when the historical case shows a directional deviation of 25%, a correction coefficient of +0.12 is used. When a 30% deviation is currently detected, the correction coefficient is increased to +0.15 (0.12×30 / 25=0.144≈0.15), and the weight value of the section in the mapping table is adjusted from 1.0 to 1.15.

[0090] Step S565: After the correction coefficient weight is adjusted, the local rate fine-tuning operation of the extension device is triggered, and the acquisition period of the real-time surface image data set is synchronously updated to verify whether the corrected surface uniformity distribution reaches the target convergence range.

[0091] After adjusting the correction coefficient weight associated with the abnormal extension section in the extension rate correction mapping table, the extension device is triggered to perform a local rate fine-tuning operation on the abnormal extension section, that is, to adjust the extension rate of the extension section by a small margin. At the same time, the acquisition cycle of the real-time surface image data set is synchronously updated to obtain the corrected gold foil surface uniformity data in a more timely manner. By analyzing the updated real-time surface image data set, it is verified whether the corrected surface uniformity distribution reaches the target convergence range, that is, whether it is close to or reaches the standard of the benchmark uniformity template. For example, after adjusting the correction coefficient weight, the extension device fine-tunes the extension rate of the abnormal extension section, and at the same time shortens the acquisition cycle of the real-time surface image data set, collects the corrected surface image data, and analyzes whether the surface uniformity is improved. Exemplarily, the local rate fine-tuning operation is implemented by a distributed motion control system, converting the corrected extension rate correction mapping table into independent rate setting values ​​for six process sections, and generating a trapezoidal speed curve instruction including a target rate, acceleration limit and transition time. The trigger mechanism uses hardware-in-the-loop verification. Before the command is issued, the execution effect is simulated through the digital twin system. When the predicted uniformity improvement rate exceeds 50%, the actual equipment adjustment is triggered. The acquisition cycle update strategy is: within three feedback cycles (25ms) after the fine-tuning is executed, the sampling frequency is increased to 240fps, and then dynamically adjusted according to the uniformity change slope. If the improvement rate meets expectations, it is gradually restored to the baseline sampling rate. The target convergence range is defined as a composite condition of directional consistency parameter deviation ≤5% and reflective intensity discreteness change rate ≤8%. During verification, three consecutive sampling cycles must meet the dual index requirements at the same time.

[0092] Step S566: If the target convergence range is not reached, the updated uniformity distribution index set is re-extracted and deviation identification and correction coefficient weight adjustment are iteratively performed until the surface uniformity index of all extended sections meets the preset process standards.

[0093] If the verification results show that the corrected surface uniformity distribution does not reach the target convergence range, it means that the current adjustment is not sufficient and needs further adjustment. Re-extract the updated uniformity distribution index set, perform deviation identification again, find out the abnormal extension sections that still have problems, and adjust the correction coefficient weight according to the new deviation. Repeat this process, iterate the deviation identification and correction coefficient weight adjustment continuously until the surface uniformity indicators of all extension sections meet the preset process standards. For example, after one adjustment, it is found that the surface uniformity of some extension sections still does not reach the target convergence range. Re-extract the uniformity distribution index set, perform deviation analysis again, adjust the correction coefficient weight, and continue to adjust until the surface uniformity of all extension sections meets the requirements.

[0094] As an implementation mode, the method provided by the present invention further includes, in a feedback calibration phase after the dynamic adjustment instruction is executed, specifically performing the following operations: Step S600: collecting final thickness distribution data and surface uniformity detection graph after gold foil processing is completed.

[0095] After the gold foil processing is completed, the final thickness distribution data of the gold foil is collected by the thickness detection equipment. The data reflects the thickness change of the gold foil after the entire processing process. At the same time, the surface detection equipment is used to obtain the surface uniformity detection map of the gold foil, which shows the distribution of the surface uniformity of the gold foil. For example, a high-precision thickness sensor is used to measure the thickness of each position of the gold foil to obtain the final thickness distribution data; the optical detection equipment is used to scan the surface of the gold foil to generate a surface uniformity detection map. Exemplarily, the final thickness distribution data is obtained by a high-precision laser scanning array, which consists of 48 groups of confocal displacement sensors distributed at intervals of 5mm along the transmission direction of the gold foil. Each sensor captures the axial thickness value at a sampling frequency of 1kHz, and generates a three-dimensional thickness distribution cloud map after Kalman filtering and noise reduction. The generation of the surface uniformity detection map adopts a multi-spectral confocal microscopy system to scan the gold foil surface under dual wavelengths of visible light (532nm) and near-infrared (785nm), and combines the moiré fringe interference technology to quantify the surface roughness parameters to form a three-dimensional topological map containing azimuth coordinates, roughness values ​​and texture direction angles.

[0096] Step S700: Compare the final thickness distribution data with the predicted thickness range of the first control model to generate a hot pressing model error signal.

[0097] The predicted thickness range of the first control model is the thickness range of the gold foil predicted based on the dynamic mapping relationship between hot pressing intensity and thickness change. The final thickness distribution data collected is compared with the predicted thickness range, the difference between the two is calculated, and a hot pressing model error signal is generated. This signal reflects the accuracy of the first control model in predicting the thickness of the gold foil, as well as the difference between the effect of hot pressing intensity on the thickness of the gold foil during actual processing and the model prediction. For example, the thickness value of each position in the final thickness distribution data is compared with the thickness range predicted by the first control model, the deviation value is calculated, and these deviation values ​​are combined into a hot pressing model error signal. Exemplarily, the comparison process uses a three-dimensional space registration algorithm to align the coordinates of the measured thickness distribution cloud map with the predicted thickness grid output by the first control model, and the absolute deviation is calculated at each registration point.

[0098] Step S800: performing overlap analysis on the surface uniformity detection map and the expected uniformity template of the second control model to generate an extended model error signal.

[0099] The expected uniformity template of the second control model is a reference standard for the surface uniformity of the gold foil set according to the nonlinear association rule between the extension rate and the surface uniformity. The surface uniformity detection map and the expected uniformity template are overlapped, that is, the similarity between the two is compared and the overlap is calculated. According to the difference in overlap, an extension model error signal is generated. This signal reflects the accuracy of the second control model in predicting the surface uniformity of the gold foil, as well as the difference between the influence of the extension rate on the surface uniformity of the gold foil during the actual processing and the model expectation. For example, the surface uniformity detection map is compared with the expected uniformity template through image analysis technology, the overlap is calculated, and the extension model error signal is generated according to the deviation of the overlap. Exemplarily, the overlap analysis adopts a hybrid algorithm combining phase correlation method and feature matching: first, Gabor filtering is performed on the surface uniformity detection map to extract the texture direction feature map; then, multi-scale pyramid matching is performed with the expected uniformity template to calculate the similarity integral value at each scale. The extended model error signal is generated, for example, by extracting low-frequency difference components through discrete cosine transform for the main error component and capturing high-frequency texture deviations using local binary pattern analysis for the secondary error component. The error signal is stored in polar coordinates, and each azimuth interval is associated with a radial error value, a tangential error value, and a comprehensive error intensity.

[0100] Step S900: adjusting the gradient compensation rule in the first control model according to the system error component in the thermal pressure model error signal.

[0101] The error signal of the hot pressing model contains systematic error components and random error components. The systematic error component is the error caused by defects in the model itself or some fixed factors in the processing process. According to the systematic error component in the error signal of the hot pressing model, the gradient compensation rule in the first control model is adjusted to improve the accuracy and adaptability of the model.

[0102] Step S1000: Optimizing the extension rate correction coefficient in the second control model according to the random error component in the extension model error signal.

[0103] The random error component in the error signal of the extension model is the error caused by some random factors in the processing process, such as slight differences in materials, slight changes in the environment, etc. According to the random error component in the error signal of the extension model, the extension rate correction coefficient in the second control model is optimized to improve the adaptability of the model to random factors and the prediction accuracy of the surface uniformity of the gold foil. For example, the random error component in the error signal of the extension model is analyzed, and the extension rate correction coefficient in the second control model is adjusted according to the error situation, so that the model can better cope with random changes in the processing process and more accurately adjust the extension rate to ensure the surface uniformity of the gold foil.

[0104] Step S1100: synchronously updating the adjusted gradient compensation rule and extension rate correction coefficient to the historical parameter library of the initial processing parameter set.

[0105] The adjusted and optimized gradient compensation rules in the first control model and the extension rate correction coefficient in the second control model are synchronously updated to the historical parameter library of the initial processing parameter set. The historical parameter library is used to store various parameters and model information in previous processing processes. The updated parameters and coefficients can provide a more accurate reference for subsequent gold foil processing and improve the quality and efficiency of processing. For example, the adjusted gradient compensation rules and extension rate correction coefficients are written into the historical parameter library. In subsequent processing, when encountering similar materials or processing conditions, these updated parameters and coefficients can be directly obtained from the historical parameter library and applied to the new processing process.

[0106] As an implementation method, the process of adjusting the gradient compensation rule and the extension optimization rate correction coefficient includes the following steps: Step S910: Identify the steady-state deviation component in the thermal pressure model error signal, and calculate the steady-state compensation amount by sliding window averaging method.

[0107] The steady-state deviation component in the error signal of the hot pressing model refers to the deviation part of the error signal that is relatively stable and does not change rapidly with time. The steady-state deviation component is identified by a specific signal processing method, and then processed by the sliding window averaging method. The sliding window averaging method is a method for smoothing the signal. By averaging the signal within a certain time window, a relatively stable value, i.e., the steady-state compensation amount, is obtained. The steady-state compensation amount is used to compensate for the steady-state deviation in the hot pressing model and improve the accuracy of the model. For example, the error signal of the hot pressing model is analyzed to find the steady-state deviation component therein, and the sliding window averaging method is used to average the component within a set time window to obtain the steady-state compensation amount. Exemplarily, the identification of the steady-state deviation component is based on the time series stationarity test, and the ADF unit root test method is used to screen out the error components with significant non-periodicity. The window width of the sliding window averaging method is set to an integer multiple of the processing cycle (such as 30-50 sampling points), and the window step is equal to the mechanical response cycle of the hot pressing device (such as 200ms). The calculation of the steady-state compensation is implemented as the weighted average of the error values ​​in the window, and the weight coefficient is set according to the importance of the spatial position: the weight of the central area is 1.2, and the weight of the edge area is 0.8. For example, when the steady-state deviation of the central area is detected in five consecutive windows +2.8μm, the steady-state compensation of the area is calculated to be +3.1MPa (such as 2.8μm×1.1μm / MPa conversion coefficient).

[0108] Step S920: adjusting the reference pressure gradient in the thermal pressure intensity compensation function according to the steady-state compensation amount.

[0109] According to the calculated steady-state compensation amount, the reference pressure gradient in the hot pressing strength compensation function is adjusted. The reference pressure gradient is an important parameter in the hot pressing strength compensation function, which determines the basic change trend of the hot pressing strength during the processing. By adjusting the reference pressure gradient, the steady-state deviation in the hot pressing model can be compensated, so that the adjustment of the hot pressing strength can more accurately meet the actual change requirements of the gold foil thickness. For example, according to the steady-state compensation amount, the reference pressure gradient in the hot pressing strength compensation function is increased or decreased, so that the adjustment of the hot pressing strength can better adapt to the processing conditions of the gold foil.

[0110] Step S1010: Identify the high-frequency oscillation component in the extended model error signal, and extract the effective correction component through a low-pass filter.

[0111] The high-frequency oscillation component in the extended model error signal refers to the part of the error signal with a higher frequency of change. These high-frequency oscillations may be caused by some random interference or rapidly changing factors in the processing process. These high-frequency oscillation components are identified by signal processing methods, and then processed by a low-pass filter. A low-pass filter is a filter that allows low-frequency signals to pass through and suppresses high-frequency signals. The low-pass filter can be used to extract the effective correction components in the extended model error signal, that is, those low-frequency signal components that are of practical significance for the correction of the extension rate. For example, the extended model error signal is subjected to spectral analysis to find the high-frequency oscillation components, and the signal is filtered using a low-pass filter to extract the effective correction components.

[0112] Step S1020: adjusting the rate distribution weight in the extended rate correction mapping table according to the effective correction component.

[0113] According to the extracted effective correction component, the rate distribution weight in the extension rate correction mapping table is adjusted. The rate distribution weight determines the extension rate adjustment range and method of the extension device in different sections. By adjusting the rate distribution weight, the extension rate can be adjusted more accurately according to the effective correction component in the extension model error signal to improve the surface uniformity of the gold foil. For example, the rate distribution weight of a certain section in the extension rate correction mapping table is increased or decreased according to the effective correction component, so that the extension rate of the section is adjusted accordingly.

[0114] Step S1110: replacing the original model parameters with the updated hot pressing strength compensation function and the extension rate correction mapping table.

[0115] The adjusted hot pressing strength compensation function and extension rate correction mapping table replace the original model parameters, so that the parameters in the first control model and the second control model are updated. The updated model parameters can more accurately reflect the relationship between the hot pressing strength and thickness change, the extension rate and the surface uniformity during the gold foil processing, thereby improving the control accuracy of the model and the quality of gold foil processing. For example, the updated hot pressing strength compensation function and extension rate correction mapping table are stored in the model to replace the original parameters, and the updated model is used for control and adjustment in the subsequent processing process.

[0116] As an implementation mode, the method provided by the present invention further includes performing the following pre-processing operations before starting the device, which may specifically include: Step S1: parse the lattice symmetry identifier and the atomic spacing coding segment in the lattice type code corresponding to the target material to generate a lattice feature description vector.

[0117] The lattice type code corresponding to the target material is a code used to represent the lattice structure characteristics of the target material. Among them, the lattice symmetry identifier represents the symmetry property of the lattice, and the atomic spacing coding segment represents the spacing information between atoms. By parsing the lattice type code, the lattice symmetry identifier and the atomic spacing coding segment are extracted, and these information are combined into a vector, namely the lattice feature description vector. This vector can comprehensively describe the lattice structure characteristics of the target material and provide a basis for subsequent material matching and parameter selection. For example, the lattice type code of the target material is parsed, the lattice symmetry identifier and the atomic spacing coding segment are extracted, and these information are combined into a vector according to certain rules to obtain the lattice feature description vector.

[0118] Step S2: Input the lattice feature description vector into the preloaded material property database for multi-dimensional similarity search, and screen out a set of candidate reference templates whose similarity with the current lattice type code exceeds a matching threshold.

[0119] The preloaded material property database stores the property information of various materials and related benchmark templates. The generated lattice feature description vector is input into the database for multi-dimensional similarity retrieval, that is, the similarity between the lattice feature description vector and the feature vectors of each material in the database is compared from multiple angles. The material benchmark templates whose similarity with the current lattice type code exceeds the matching threshold are screened out to form a candidate benchmark template set. The matching threshold is a pre-set similarity critical value, and only benchmark templates whose similarity exceeds the threshold will be selected. For example, the lattice feature description vector is compared with each material feature vector in the material property database, the similarity is calculated, and the benchmark templates whose similarity exceeds the matching threshold are screened out to form a candidate benchmark template set.

[0120] Step S3: Perform the following operations on each candidate reference template in the candidate reference template set: extract the average thickness deviation value and surface uniformity compliance rate in its historical application records, and calculate the comprehensive process efficiency score.

[0121] Each candidate benchmark template in the candidate benchmark template set has its historical application records, which contain some process indicators of the template in the previous gold foil processing process, such as the average thickness deviation value and the surface uniformity compliance rate. The average thickness deviation value reflects the average degree of deviation between the actual thickness of the gold foil and the expected thickness when the template is used to process the gold foil; the surface uniformity compliance rate indicates the proportion of the surface uniformity of the gold foil that meets the process standard when the template is used to process the gold foil. Based on these indicators, the comprehensive process efficiency score of each candidate benchmark template is calculated, which comprehensively considers the performance of the template in controlling the thickness and surface uniformity of the gold foil. For example, for each template in the candidate benchmark template set, the average thickness deviation value and the surface uniformity compliance rate are extracted from its historical application records, and the comprehensive process efficiency score is calculated according to a certain calculation method.

[0122] Step S4: Prioritize the candidate benchmark template set according to the comprehensive process efficiency score, and select the candidate benchmark template with the highest score as the optimal benchmark hot pressing gradient template.

[0123] According to the calculated comprehensive process efficiency score, the candidate benchmark template set is prioritized, and the template with the higher score is ranked first. The candidate benchmark template with the highest score is selected as the optimal benchmark hot pressing gradient template. This template has the best performance in controlling the thickness and surface uniformity of the gold foil, and can provide the most suitable hot pressing gradient reference for subsequent gold foil processing. For example, the candidate benchmark template set is sorted from high to low according to the comprehensive process efficiency score, and the template ranked first is selected as the optimal benchmark hot pressing gradient template.

[0124] Step S5: extracting a set of extension rate reference curves that are process-related to the optimal benchmark hot-pressing gradient template from the material property database.

[0125] The material property database stores the process information of various materials, including the correlation between hot pressing gradient and extension rate. From the database, a set of extension rate reference curves that are process-related to the optimal benchmark hot pressing gradient template are extracted. These curves reflect the influence of different extension rates on the processing effect when the optimal benchmark hot pressing gradient template is used for gold foil processing. For example, the extension rate reference curves associated with the optimal benchmark hot pressing gradient template are searched in the material property database, and these curves are combined into an extension rate reference curve set.

[0126] Step S6: Verify the synergy between rate distribution and thermal pressure gradient for each curve in the extension rate reference curve set, remove abnormal curves with rate mutation or thermal pressure conflict, and generate an optimized extension rate reference curve set.

[0127] For each curve in the extension rate reference curve set, the synergy between the rate distribution and the hot-pressing gradient is verified, that is, whether the rate distribution in the curve and the hot-pressing gradient in the optimal benchmark hot-pressing gradient template can work together, and whether there is a rate mutation or hot-pressing conflict. Rate mutation refers to a large change in the extension rate in a short period of time, and hot-pressing conflict refers to the interaction between the hot-pressing gradient and the extension rate that causes problems in the processing process. Eliminate the curves with these abnormalities, and form the remaining curves into an optimized extension rate reference curve set. For example, analyze each curve in the extension rate reference curve set to check whether there is a rate mutation or hot-pressing conflict, eliminate the abnormal curves, and obtain the optimized extension rate reference curve set.

[0128] As an implementation mode, step S6, verifying the synergy between the rate distribution and the thermal pressure gradient for each curve in the extension rate reference curve set, may specifically include: Step S61: extracting the rate distribution data of a single curve to be verified from the extension rate reference curve set, and synchronously obtaining a thermal pressure gradient sequence in the optimal reference thermal pressure gradient template that is time-aligned with the curve to be verified.

[0129] Select a curve to be verified from the extension rate reference curve set, and extract the rate distribution data of the curve, which reflects the change of the extension rate with time or position. At the same time, obtain the thermostatic gradient sequence in the optimal benchmark thermostatic gradient template that is aligned with the time of the curve to be verified, and ensure that the rate distribution data and the thermostatic gradient sequence are corresponding in time, so as to perform collaborative verification. For example, select a curve from the extension rate reference curve set, extract its rate distribution data, and find the thermostatic gradient sequence corresponding to the time of the curve from the optimal benchmark thermostatic gradient template.

[0130] Step S62: According to the phase nodes of the thermal pressure gradient sequence, a corresponding coordinated timestamp identifier is marked on the curve to be verified, and a rate distribution curve with a timing mark is generated.

[0131] The phase node of the thermal pressure gradient sequence refers to the time point at which the thermal pressure gradient changes significantly. According to these phase nodes, the corresponding collaborative timestamp identifier is marked on the curve to be verified, so that the curve to be verified has time sequence information and a rate distribution curve with time sequence marks is generated. This curve can clearly show the temporal correspondence between the extension rate and the thermal pressure gradient, which is convenient for subsequent synergy analysis. For example, according to the phase node of the thermal pressure gradient sequence, the corresponding timestamp is marked on the curve to be verified to form a rate distribution curve with time sequence marks.

[0132] Step S63: traverse each coordinated timestamp identifier in the rate distribution curve with timing mark, detect whether the rate change slope in its adjacent time interval exceeds the preset mutation threshold, and identify the abnormal time interval with rate mutation.

[0133] Traverse each collaborative timestamp identifier in the rate distribution curve with timing marks, and calculate the rate change slope in its adjacent time interval. The rate change slope represents the rate of change of the extended rate in the adjacent time. Compare the calculated rate change slope with the preset mutation threshold. When the slope exceeds the threshold, it indicates that there is a rate mutation in the adjacent time interval, and the time interval is identified as an abnormal time interval. For example, for each collaborative timestamp identifier in the rate distribution curve with timing marks, calculate the rate change slope of the adjacent time points before and after it. If the slope exceeds the preset mutation threshold, the adjacent time interval is marked as an abnormal time interval.

[0134] Step S64: when an abnormal time interval is identified, the thermal pressure gradient value of the corresponding time node in the optimal reference thermal pressure gradient template is traced back to determine whether the pressure fluctuation direction of the thermal pressure gradient value interferes with the rate mutation direction in reverse.

[0135] When an abnormal time interval with a sudden rate change is identified, the hot pressure gradient value of the corresponding time node in the optimal benchmark hot pressure gradient template is traced back. Analyze the pressure fluctuation direction of the hot pressure gradient value, that is, whether the hot pressure intensity increases or decreases, and the direction of the rate mutation, that is, whether the extension rate increases or decreases, to see if there is reverse interference. Reverse interference means that the direction of change of the hot pressure gradient and the direction of change of the extension rate are contradictory, which may cause problems in the processing process. For example, when an abnormal time interval is identified, check the hot pressure gradient value of the corresponding time in the optimal benchmark hot pressure gradient template to determine whether the hot pressure intensity increases while the extension rate suddenly decreases, or the hot pressure intensity decreases while the extension rate suddenly increases. If such a contradiction exists, it is considered that reverse interference has occurred.

[0136] Step S65: If the reverse interference intensity exceeds the conflict threshold, the abnormal time interval is marked as a pressure conflict area, and the starting position and duration of the pressure conflict area in the velocity distribution curve are calculated.

[0137] When it is determined that the pressure fluctuation direction of the thermal pressure gradient value and the rate mutation direction produce reverse interference, the reverse interference intensity is calculated. The reverse interference intensity is compared with the conflict threshold. If it exceeds the threshold, the abnormal time interval is marked as a pressure conflict area. At the same time, the starting position and duration of the pressure conflict area in the rate distribution curve are calculated for subsequent processing. For example, the reverse interference intensity is calculated by a certain calculation method. If it exceeds the conflict threshold, the abnormal time interval is marked as a pressure conflict area, and its start time and end time in the rate distribution curve are recorded, and the duration is calculated.

[0138] Step S66: According to the starting position and duration of the pressure conflict area, a rate correction interval is defined on the curve to be verified, and an alternative rate smoothing segment is generated based on the correction records of similar conflicts in the historical parameter library.

[0139] According to the starting position and duration of the pressure conflict area, a rate correction interval is defined on the curve to be verified, which includes the pressure conflict area and a certain range around it. Based on the correction records of similar conflicts in the historical parameter library, an alternative rate smooth segment is generated. This smooth segment is used to replace the part of the rate mutation and pressure conflict in the curve to be verified, making the rate distribution smoother and more reasonable. For example, according to the start and end time of the pressure conflict area, a rate correction interval is determined on the curve to be verified, and correction records of similar conflicts are found from the historical parameter library, and an alternative rate smooth segment is generated based on these records.

[0140] Step S67: inserting the alternative rate smoothing segment into the rate correction interval to generate an optimized extended rate curve, and deleting abnormal data points in the original rate distribution curve that overlap with the pressure conflict area.

[0141] The generated alternative rate smooth segment is inserted into the rate correction interval of the curve to be verified, replacing the problematic part of the original curve to generate an optimized extended rate curve. At the same time, the abnormal data points in the original rate distribution curve that overlap with the pressure conflict area are deleted to make the optimized extended rate curve more accurate and reliable. For example, the alternative rate smooth segment is inserted into the specified position of the curve to be verified, and the abnormal data points corresponding to the pressure conflict area in the original curve are deleted to form an optimized extended rate curve.

[0142] Step S68: performing integrity check on all optimized extension rate curves that have completed the replacement insertion operation in the extension rate reference curve set, and removing residual curves that still contain uncorrected conflict areas.

[0143] Perform integrity check on all optimized extension rate curves obtained through substitution insertion in the extension rate reference curve set to check whether there are any uncorrected conflicting areas in the curves. If there are any uncorrected conflicting areas, it means that the curve still has problems and it should be removed. Through integrity check, it is ensured that the optimized extension rate curve set only contains curves without conflicts and anomalies. For example, perform a detailed check on each optimized extension rate curve to find out whether there are any areas with sudden rate changes or pressure conflicts. If so, remove the curve from the set.

[0144] Step S69: merging the optimized extension rate curves that have passed the verification into an optimized extension rate reference curve set.

[0145] The optimized extension rate curves that have passed the integrity check are merged together to form an optimized extension rate reference curve set. The curves in this set have good synergy in rate distribution and hot pressing gradient, and can provide a more reasonable extension rate reference for gold foil processing. For example, all the optimized extension rate curves that have passed the check are integrated into one set to obtain the optimized extension rate reference curve set.

[0146] Step S7: According to the historical processing stability index of each optimized extension rate reference curve in the optimized extension rate reference curve set, the optimized extension rate reference curve with the smallest fluctuation coefficient is selected as the optimal extension rate reference curve.

[0147] Each curve in the optimized extension rate reference curve set has its historical processing stability index. The fluctuation coefficient is an important index to measure the stability of the curve. The smaller the fluctuation coefficient, the smaller the fluctuation of the curve in the historical processing process, and the better the stability. According to the historical processing stability index of each optimized extension rate reference curve, the curve with the smallest fluctuation coefficient is selected as the optimal extension rate reference curve, which can provide the most stable extension rate reference for gold foil processing. For example, the fluctuation coefficient is calculated for each curve in the optimized extension rate reference curve set, and the curve with the smallest fluctuation coefficient is selected as the optimal extension rate reference curve.

[0148] Step S8: Align the optimal reference hot pressing gradient template and the optimal extension rate reference curve in process timing to generate a reference configuration of the initial processing parameter set.

[0149] The optimal benchmark hot pressing gradient template and the optimal extension rate benchmark curve are aligned in process timing, that is, the hot pressing gradient and the extension rate are matched in time, so that the two can work together to complete the processing of gold foil. Through process timing alignment, a benchmark configuration of the initial processing parameter set is generated, which provides an initial and reasonable reference for hot pressing strength and extension rate parameters for subsequent gold foil processing. For example, the optimal benchmark hot pressing gradient template and the optimal extension rate benchmark curve are aligned in time order, and the aligned parameters are combined into the benchmark configuration of the initial processing parameter set.

[0150] As an embodiment, the method provided by the present invention may further include: Step S1200: synchronously collect the reflected light intensity distribution and the transmitted light attenuation spectrum of the gold foil surface in the transmission direction of the stretching device through a multi-angle optical scanning unit to generate a multimodal surface morphology data set.

[0151] The multi-angle optical scanning unit can scan the surface of the gold foil from multiple angles, and simultaneously collect the reflected light intensity distribution and the transmitted light attenuation spectrum of the gold foil surface in the transmission direction of the extension device. The reflected light intensity distribution reflects the reflection characteristics of the gold foil surface at different angles, and the transmitted light attenuation spectrum shows the attenuation of light when it passes through the gold foil. The collected data are integrated to generate a multimodal surface morphology data set, which contains rich microstructure and morphology information on the gold foil surface. For example, the multi-angle optical scanning unit scans the gold foil surface from different angles, and simultaneously obtains the reflected light intensity distribution and the transmitted light attenuation spectrum, which are combined into a multimodal surface morphology data set.

[0152] Step S1300: performing spatial frequency decomposition on the multimodal surface morphology data set, separating the low-frequency component representing the thickness variation and the high-frequency component representing the micro cracks, and generating a thickness distribution map and a crack characteristic map.

[0153] Spatial frequency decomposition is a method for processing images or data to decompose data into components of different frequencies. The multimodal surface morphology data set is subjected to spatial frequency decomposition to separate the low-frequency component representing the thickness change and the high-frequency component representing the microcracks. The low-frequency component reflects the overall change in the thickness of the gold foil surface, while the high-frequency component reflects the characteristics of the microcracks on the gold foil surface. Based on the separated low-frequency and high-frequency components, a thickness distribution map and a crack feature map are generated respectively. These two maps can intuitively display the thickness change and microcracks on the gold foil surface. For example, the multimodal surface morphology data set is processed using a spatial frequency decomposition algorithm to separate the low-frequency and high-frequency components, and a thickness distribution map is drawn based on the low-frequency component, and a crack feature map is drawn based on the high-frequency component.

[0154] Step S1400: input the thickness distribution map into the pre-trained surface defect recognition model, detect the gradient direction consistency of the thickness mutation area by sliding the convolution kernel, and identify the local thickness abnormality band perpendicular to the transmission direction of the stretching device.

[0155] The pre-trained surface defect recognition model is obtained through a large amount of data training and can identify various defects on the surface of gold foil. The thickness distribution map is input into the model, and the model slides on the map through the convolution kernel to detect the consistency of the gradient direction in the thickness mutation area. The gradient direction consistency reflects the regularity and continuity of the thickness change. When the gradient direction is inconsistent, there may be thickness anomalies. Through detection, local thickness anomaly bands perpendicular to the transmission direction of the stretching device are identified, which may affect the quality and performance of the gold foil. For example, the thickness distribution map is input into the surface defect recognition model, and the model uses the convolution kernel to slide on the map, calculate the gradient direction of each area, find the area with inconsistent gradient direction and perpendicular to the transmission direction of the stretching device, and identify it as a local thickness anomaly band.

[0156] Step S1500: input the crack feature map into the surface defect recognition model, extract the angle data between the expansion direction of the crack branch and the main stress direction, and screen out the accelerated expansion cracks with an angle less than a preset angle.

[0157] The crack feature map is input into the surface defect recognition model. The model analyzes the crack feature map and extracts the angle data between the expansion direction of the crack branch and the principal stress direction. The principal stress direction refers to the direction of the main stress to which the gold foil is subjected during processing. The angle between the expansion direction of the crack branch and the principal stress direction will affect the crack expansion speed. Cracks with angles less than the preset angle are screened out. These cracks will accelerate expansion under the action of the principal stress and are called accelerated expansion cracks. For example, the crack feature map is input into the surface defect recognition model. The model calculates the angle between the expansion direction of each crack branch and the principal stress direction, and screens out cracks with angles less than the preset angle as accelerated expansion cracks.

[0158] Step S1600: Generate a thermal pressure compensation area coordinate set associated with the first control model according to the width and gradient direction of the local thickness anomaly zone, and calculate the pressure compensation priority weight of the corresponding area.

[0159] According to the width and gradient direction of the identified local thickness anomaly band, the area that needs to be hot-pressed is determined. A hot-pressed compensation area coordinate set associated with the first control model is generated, and the coordinate set clarifies the specific location where hot-pressed compensation is required. At the same time, according to the severity of the local thickness anomaly band and the degree of influence on the quality of the gold foil, the pressure compensation priority weight of the corresponding area is calculated. The higher the priority weight, the more priority the area needs to be hot-pressed. For example, according to the width and gradient direction of the local thickness anomaly band, its specific position on the gold foil surface is determined, and a hot-pressed compensation area coordinate set is generated, and the pressure compensation priority weight is calculated for each area according to the severity of the anomaly band.

[0160] Step S1700: Generate an extension inhibition region coordinate set associated with the second control model based on the angle data and branch length of the accelerated extension crack, and calculate the rate inhibition urgency level of the corresponding region.

[0161] According to the angle data and branch length of the screened accelerated expansion cracks, the area that needs to be suppressed is determined. A coordinate set of the extension suppression area associated with the second control model is generated, and the coordinate set clarifies the specific location where the extension suppression is required. At the same time, according to the expansion speed of the accelerated expansion cracks and the degree of harm to the quality of the gold foil, the rate suppression urgency level of the corresponding area is calculated. The higher the urgency level, the more the area needs to suppress the extension rate as soon as possible. For example, according to the angle data and branch length of the accelerated expansion cracks, its specific position on the gold foil surface is determined, and the extension suppression area coordinate set is generated. The rate suppression urgency level is calculated for each area according to the expansion of the cracks.

[0162] Step S1800: inputting the thermal pressure compensation area coordinate set and the pressure compensation priority weight into the first control model to generate a dynamic pressure compensation instruction set covering the local thickness anomaly zone.

[0163] The generated hot pressure compensation area coordinate set and pressure compensation priority weight are input into the first control model. Based on this information and the dynamic mapping relationship between hot pressure intensity and thickness change, the first control model generates a dynamic pressure compensation instruction set covering the local thickness abnormality band. The instruction set includes the pressure value and adjustment method that need to be adjusted for each hot pressure compensation area to eliminate the local thickness abnormality band and make the thickness of the gold foil more uniform. For example, the hot pressure compensation area coordinate set and pressure compensation priority weight are input into the first control model. The model calculates the hot pressure intensity that needs to be increased or decreased in each area based on these data and generates a dynamic pressure compensation instruction set.

[0164] Step S1900: Inputting the extension inhibition region coordinate set and the rate inhibition urgency level into the second control model to generate a gradient rate attenuation instruction set for inhibiting accelerated extension cracks.

[0165] The generated extension inhibition area coordinate set and rate inhibition urgency level are input into the second control model. Based on this information and the nonlinear association rule between extension rate and surface uniformity, the second control model generates a gradient rate attenuation instruction set for inhibiting accelerated expansion cracks. The instruction set includes the extension rate value and adjustment method that need to be adjusted for each extension inhibition area to inhibit the expansion of accelerated expansion cracks and ensure the quality of the gold foil surface. For example, the extension inhibition area coordinate set and rate inhibition urgency level are input into the second control model. The model calculates the extension rate that needs to be reduced in each area based on these data and generates a gradient rate attenuation instruction set.

[0166] Step S2000: adjusting the pressure distribution of the hot pressing device in the compensation area according to the dynamic pressure compensation instruction set, and executing the gradient rate attenuation instruction set to reduce the transmission rate of the extension device in the suppression area.

[0167] According to the generated dynamic pressure compensation instruction set, the pressure distribution of the hot pressing device in the hot pressing compensation area is adjusted so that the hot pressing strength meets the requirements of the instruction set to eliminate the local thickness abnormal zone. At the same time, the gradient rate attenuation instruction set is executed to reduce the transmission rate of the extension device in the extension inhibition area to inhibit the expansion of accelerated expansion cracks. For example, the dynamic pressure compensation instruction set is sent to the actuator of the hot pressing device, and the actuator adjusts the pressure of the hot pressing device in the compensation area according to the instruction; the gradient rate attenuation instruction set is sent to the actuator of the extension device, and the actuator reduces the transmission rate of the extension device in the inhibition area according to the instruction.

[0168] Step S2100: After the adjustment operation is performed, the multimodal surface morphology data set of the compensation area and the suppression area is re-collected to verify whether the gradient direction consistency of the thickness anomaly band is restored to within the allowable threshold and whether the angle of the accelerated expansion crack is increased to a safe range.

[0169] After adjusting the pressure of the hot pressing device and the rate of the stretching device, the multi-modal surface morphology data set of the compensation area and the suppression area is collected again using the multi-angle optical scanning unit. By analyzing the newly collected data set, it is verified whether the gradient direction consistency of the local thickness anomaly band is restored to the allowable threshold, that is, whether the thickness change becomes more uniform; at the same time, it is verified whether the angle of the accelerated expansion crack increases to a safe range, that is, whether the crack expansion is effectively suppressed. For example, the multi-modal surface morphology data set is collected again, the thickness distribution map and the crack characteristic map are analyzed, and it is checked whether the gradient direction consistency of the local thickness anomaly band and the angle of the accelerated expansion crack meet the requirements.

[0170] Step S2200: If the verification condition is not met, the pressure compensation priority weight and the rate suppression urgency level are iteratively updated, and the dynamic pressure compensation instruction set and the gradient rate attenuation instruction set are regenerated until the abnormality is eliminated.

[0171] If the verification results show that the gradient direction consistency of the local thickness anomaly band has not recovered to the allowable threshold, or the angle of the accelerated expansion crack has not increased to the safe range, it means that the adjustment operation is not sufficient and further adjustment is required. Iteratively update the pressure compensation priority weight and rate suppression urgency level, and regenerate the dynamic pressure compensation instruction set and gradient rate attenuation instruction set according to the updated weights and levels. Repeat this process, continuously adjust and verify until the abnormal conditions of the local thickness anomaly band and accelerated expansion crack are eliminated. For example, adjust the pressure compensation priority weight and rate suppression urgency level according to the verification results, regenerate the instruction set, perform the adjustment operation again, and then verify again until the abnormality is eliminated.

[0172] Step S2300: After the anomaly is eliminated, the final adjustment parameters and verification results are recorded in the historical parameter library, and the local feature detection rules of the surface defect recognition model are updated.

[0173] When the abnormal conditions of the local thickness abnormal band and the accelerated expansion crack are eliminated, the final adjustment parameters, such as the pressure adjustment value of the hot pressing device and the rate adjustment value of the extension device, and the verification results, such as the elimination of the thickness abnormal band and the suppression of crack expansion, are recorded in the historical parameter library. These records can provide a reference for subsequent gold foil processing and help deal with similar problems. At the same time, based on the experience and data in this processing process, the local feature detection rules of the surface defect recognition model are updated so that the model can more accurately identify and process defects on the gold foil surface. For example, the final adjustment parameters and verification results are stored in the historical parameter library, and the local feature detection rules of the surface defect recognition model are modified and optimized based on the data in the processing process.

[0174] Figure 2 A hardware entity diagram of a control processing system provided by an embodiment of the present invention, such as Figure 2 As shown, the hardware entity of the control processing system 1000 includes: a processor 1001 and a memory 1002, wherein the memory 1002 stores a computer program that can be run on the processor 1001, and the processor 1001 implements the steps in the method of any of the above embodiments when executing the program.

Claims

1. A control processing method based on gold foil processing, characterized in that: The method comprises: Collecting real-time processing parameters of the target material in the gold foil processing equipment to generate an initial processing parameter set, wherein the initial processing parameter set includes a hot pressing strength gradient sequence and an extension rate distribution curve; Determining a first control model adapted to the target material according to the hot pressing intensity gradient sequence in the initial processing parameter set, wherein the first control model includes a dynamic mapping relationship between hot pressing intensity and thickness change; Determining a second control model adapted to the target material according to the extension rate distribution curve in the initial processing parameter set, wherein the second control model includes a nonlinear association rule between the extension rate and the surface uniformity; Based on the real-time thickness deviation data and surface uniformity fluctuation data during the gold foil processing, activating the target model in the first control model or the second control model, and generating dynamic adjustment instructions through the target model; The dynamic adjustment instruction is input into the actuator of the gold foil processing equipment to correct the pressure gradient of the hot pressing device or the rate distribution of the stretching device.

2. The method according to claim 1, characterized in that The collecting of real-time processing parameters of the target material in the gold foil processing equipment to generate an initial processing parameter set includes: Synchronously collecting lattice structure characteristic data of the target material in the pre-processing stage through a multi-channel sensor, wherein the lattice structure characteristic data includes a grain size distribution map and a grain boundary density parameter; According to the peak interval in the grain size distribution map, the target material is divided into at least two sub-material clusters, and a corresponding initial hot-pressing gradient range and an extension rate reference value are assigned to each sub-material cluster; Extract historical hot-pressing strength compensation parameters and extension rate correction coefficients matching each sub-material cluster from a preset parameter library; Performing time-domain filtering on the historical heat-pressure intensity compensation parameters to generate an optimized heat-pressure intensity gradient sequence corresponding to each sub-material cluster; Performing frequency domain smoothing processing on the extension rate correction coefficient to generate an optimized extension rate distribution curve corresponding to each sub-material cluster; The initial processing parameter set is generated by fusing the optimized hot pressing strength gradient sequence and the optimized extension rate distribution curve of each sub-material cluster.

3. The method according to claim 2, characterized in that The determining of the first control model and the second control model includes: Inputting the optimized heat pressure intensity gradient sequence into a preset recursive neural network model, iteratively updating the hidden layer weights through multi-cycle training, and generating a heat pressure intensity compensation function in the first control model, wherein the heat pressure intensity compensation function is a target model in the first control model; Inputting the optimized extension rate distribution curve into a convolutional neural network model, adjusting the convolution kernel parameters through a back-propagation algorithm, and generating an extension rate correction mapping table in the second control model; the extension rate correction mapping table is a target model in the second control model; After the recursive neural network model and the convolutional neural network model converge, respectively extracting key feature vectors in the hot pressure strength compensation function and the extension rate correction mapping table; Performing similarity matching between the key feature vector and the lattice structure feature data collected in real time to verify the compatibility of the first control model and the second control model; When the similarity matching result is lower than a preset threshold, the training cycles of the recursive neural network model and the convolutional neural network model are readjusted until the key feature vector meets the adaptability condition.

4. The method according to claim 3, characterized in that: Based on the real-time thickness deviation data and surface uniformity fluctuation data in the gold foil processing process, activating the target model in the first control model or the second control model, and generating dynamic adjustment instructions through the target model, including: Real-time monitoring of the thickness deviation trend line and surface uniformity oscillation amplitude during gold foil processing; When the thickness deviation trend line exceeds a first warning threshold and the surface uniformity oscillation amplitude is within a preset stable range, activating a hot pressing intensity compensation function in the first control model; Generating a dynamic pressure adjustment instruction of the hot pressing device according to the gradient compensation rule in the hot pressing intensity compensation function; When the surface uniformity oscillation amplitude exceeds a second warning threshold and the thickness deviation trend line is in a preset stable interval, activating an extension rate correction mapping table in the second control model; Generating a dynamic rate adjustment instruction for the extension device according to the rate distribution rule in the extension rate correction mapping table; If the thickness deviation trend line and the surface uniformity oscillation amplitude both exceed the warning threshold, the dynamic pressure adjustment instruction generated by the first control model is executed first, and the adjustment instruction of the second control model is delayed until the hot pressing device completes the correction operation.

5. The method according to claim 4, characterized in that The step of inputting the dynamic adjustment instruction to the actuator of the gold foil processing equipment to correct the pressure gradient of the hot pressing device or the rate distribution of the stretching device in real time includes: According to the gradient compensation value in the dynamic pressure adjustment instruction, the instantaneous pressure value of each pressure zone in the hot pressing device is adjusted in stages; After each pressure adjustment, real-time feedback data of the thickness of the gold foil is collected and deviation calculation is performed with the predicted value of the hot pressure strength compensation function; If the deviation value continues to decrease, the current pressure gradient adjustment direction is maintained until the target thickness range is reached; If the deviation value shows an expanding trend, the pressure gradient is adjusted in the opposite direction and the parameter iteration of the thermal pressure intensity compensation function is re-triggered; Dynamically adjust the transmission speed of the extension device in different sections according to the speed distribution parameters in the dynamic speed adjustment instruction; During the rate adjustment process, the optical sensor is used to detect the change in the uniformity of the gold foil surface in real time, and the correction coefficient in the extension rate correction mapping table is dynamically updated according to the detection result.

6. The method according to claim 1, characterized in that The method also includes a feedback calibration phase after the dynamic adjustment instruction is executed, including the following operations: Collect the final thickness distribution data and surface uniformity detection map after gold foil processing; Comparing the final thickness distribution data with the predicted thickness range of the first control model to generate a hot pressing model error signal; Performing overlap analysis on the surface uniformity detection map and the expected uniformity template of the second control model to generate an extended model error signal; adjusting a gradient compensation rule in the first control model according to a system error component in the thermal pressure model error signal; Optimizing the extension rate correction coefficient in the second control model according to the random error component in the extension model error signal; The adjusted gradient compensation rule and extension rate correction coefficient are synchronously updated to the historical parameter library of the initial processing parameter set.

7. The method according to claim 6, characterized in that The process of adjusting the gradient compensation rule and the extension optimization rate correction coefficient includes: Identifying a steady-state deviation component in the thermal pressure model error signal, and calculating a steady-state compensation amount by a sliding window averaging method; adjusting the reference pressure gradient in the thermal pressure intensity compensation function according to the steady-state compensation amount; Identifying high frequency oscillation components in the extended model error signal and extracting effective correction components through a low pass filter; adjusting the rate distribution weight in the extended rate correction mapping table according to the effective correction component; The updated hot pressure strength compensation function and elongation rate correction mapping table replace the original model parameters.

8. The method according to claim 1, characterized in that The method also includes performing the following pre-processing operations before the device is started, including: Parse the lattice symmetry identifier and the atomic spacing encoding segment in the lattice type code corresponding to the target material to generate a lattice feature description vector; Inputting the lattice feature description vector into a preloaded material property database for multi-dimensional similarity search, and screening out a candidate reference template set whose similarity with the current lattice type code exceeds a matching threshold; Performing the following operations on each candidate reference template in the candidate reference template set: extracting the average thickness deviation value and surface uniformity compliance rate in its historical application records, and calculating a comprehensive process efficiency score; Prioritizing the candidate reference template set according to the comprehensive process efficiency score, and selecting the candidate reference template with the highest score as the optimal reference hot pressing gradient template; Extracting a set of extension rate reference curves that are process-related to the optimal benchmark hot-pressing gradient template from a material property database; Verify the synergy between rate distribution and thermal pressure gradient for each curve in the extension rate reference curve set, remove abnormal curves with rate mutation or thermal pressure conflict, and generate an optimized extension rate reference curve set; According to the historical processing stability index of each optimized extension rate reference curve in the optimized extension rate reference curve set, the optimized extension rate reference curve with the smallest fluctuation coefficient is selected as the optimal extension rate reference curve; The optimal reference hot pressing gradient template and the optimal extension rate reference curve are aligned in process timing to generate a reference configuration of an initial processing parameter set.

9. The method according to claim 8, characterized in that The verifying the synergy between the rate distribution and the thermal pressure gradient for each curve in the extension rate reference curve set includes: Extracting rate distribution data of a single curve to be verified from the extension rate reference curve set, and synchronously acquiring a thermal pressure gradient sequence in the optimal reference thermal pressure gradient template that is time-aligned with the curve to be verified; According to the phase nodes of the thermal pressure gradient sequence, a corresponding coordinated timestamp identifier is marked on the curve to be verified, and a rate distribution curve with a timing mark is generated; Traversing each coordinated timestamp identifier in the rate distribution curve with timing marks, detecting whether the rate change slope in the adjacent time interval exceeds a preset mutation threshold, and identifying abnormal time intervals with rate mutations; When an abnormal time interval is identified, the thermal pressure gradient value of the corresponding time node in the optimal reference thermal pressure gradient template is traced back to determine whether the pressure fluctuation direction of the thermal pressure gradient value is reversely interfering with the direction of the rate mutation; If the reverse interference intensity exceeds the conflict threshold, the abnormal time interval is marked as a pressure conflict area, and the starting position and duration of the pressure conflict area in the velocity distribution curve are calculated; According to the starting position and duration of the pressure conflict area, a rate correction interval is defined on the curve to be verified, and an alternative rate smoothing segment is generated based on the correction records of similar conflicts in the historical parameter library; Inserting the alternative rate smoothing segment into the rate correction interval to generate an optimized extended rate curve, and deleting abnormal data points in the original rate distribution curve that overlap with the pressure conflict area; Performing integrity check on all optimized extension rate curves that have completed the replacement insertion operation in the extension rate reference curve set, and removing residual curves that still contain uncorrected conflict areas; The optimized extension rate curves that have passed the verification are merged into the optimized extension rate reference curve set.

10. A control processing system, comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor executes the program, the steps in the method according to any one of claims 1 to 9 are implemented.

Images