Method and system for real-time precise regulation of sheet impedance line

Abstract

This invention relates to the field of impedance control technology, and in particular to a method and system for real-time and precise control of the impedance line of a thin plate. The method involves acquiring the surface texture image and impedance distribution curve along the impedance line of the thin plate to be controlled, performing multi-scale feature extraction, and establishing a correlation mapping between surface quality and impedance stability. Based on this, the impedance line is divided into multiple processing units and assigned a control priority. A partitioned control strategy is constructed, incorporating impedance target values, surface quality constraints, and boundary continuity constraints. A laser energy modulation sequence and scanning speed coordination scheme for each unit are generated. Impedance change trajectory data are processed sequentially according to priority and collected. The trajectory data and energy modulation sequence are time-series aligned and analyzed to extract time delay and cumulative effect characteristics, establishing a temporal coupling relationship. Based on this coupling relationship, the expected impedance value of the unprocessed unit is calculated, and the partitioned strategy and parameters are corrected. This process is iterated until all units are completed, achieving real-time and precise control of the thin plate impedance line, thus improving control accuracy and efficiency.

Description

Technical Field

[0001] This invention relates to the field of impedance control technology, and in particular to a method and system for real-time and precise control of impedance lines in thin plates. Background Technology

[0002] In the field of laser impedance tuning technology for thin-plate impedance lines, the conventional approach is to set fixed laser parameters based on the target resistance value, and remove part of the material through one or more scans to adjust the impedance value. During the processing, operators often rely on the final resistance measurement value to determine whether it meets the standard, and if the deviation is large, repeated adjustments are made. This method generally focuses on the convergence of impedance values, but ignores the influence of the surface micromorphology of the impedance line on long-term stability. For example, surface defects such as scratches, cracks, or material accumulation can cause impedance drift, leading to a decrease in product reliability.

[0003] Current conventional methods suffer from two main drawbacks. First, they lack quantitative analysis and quality assessment of the impedance line surface texture, making it impossible to establish a correlation model between surface defects and impedance stability. Consequently, the processing strategy cannot be dynamically adjusted based on surface quality during impedance adjustment, making it prone to significant fluctuations in impedance values ​​induced by local defects. Second, traditional methods typically treat the entire impedance line as a whole for global parameter setting, without differentiated zoning based on surface conditions and impedance distribution in different regions. This "one-size-fits-all" processing approach cannot accommodate the quality constraints of different sections and makes it difficult to guarantee impedance continuity at the boundaries of adjacent processing areas. Ultimately, this results in insufficient impedance adjustment accuracy, frequent over- or under-adjustment, requiring repeated readjustments, reducing efficiency and increasing costs. Therefore, existing technologies struggle to achieve high-precision, high-efficiency, and high-stability control of thin-plate impedance lines, necessitating a refined control scheme that comprehensively considers surface quality and impedance distribution. Summary of the Invention

[0004] The present invention provides a method and system for real-time and precise control of impedance lines in thin plates, which can solve the problems in the prior art.

[0005] A first aspect of the present invention provides a method for real-time and precise control of impedance lines in thin plates, comprising: Obtain the surface texture image of the impedance line of the thin plate to be tunable and the impedance distribution curve along the line; Multi-scale feature extraction is performed on the surface texture image to identify surface defect regions and uniform regions of the impedance line, and a correlation mapping between surface quality and impedance stability is established. Based on the correlation mapping, the impedance line is divided into multiple processing units and assigned a resistance adjustment priority. For each processing unit, a partitioned resistance adjustment strategy is constructed, which includes the impedance target value, surface quality constraints and the continuity constraints of the adjacent unit boundary. Based on the partitioned impedance tuning strategy, a laser energy modulation sequence and scanning speed coordination scheme for each processing unit are generated. The processing units are selected sequentially according to the impedance tuning priority and the laser beam is driven to perform impedance tuning processing. The impedance change trajectory data after processing is collected. The impedance change trajectory data is time-aligned with the laser energy modulation sequence to extract the time delay characteristics and cumulative effect characteristics of the impedance response and establish a time-series coupling relationship. The expected impedance response of the unprocessed unit is calculated based on the time-series coupling relationship. The impedance target value and constraint parameters in the partitioned impedance tuning strategy are corrected. The laser energy modulation sequence and scanning speed coordination scheme of the unprocessed unit are updated. The process is iteratively executed until all processing units are completed.

[0006] Multi-scale feature extraction is performed on the surface texture image to identify surface defect regions and uniform regions of the impedance line, and a correlation mapping between surface quality and impedance stability is established, including: The surface texture image is slid-scanned along the impedance line direction with multiple preset window sizes. Local variance is calculated in each window and arranged along the impedance line direction to form a variance distribution curve. After smoothing and filtering each variance distribution curve, the second derivative is calculated. The position of the positive and negative sign of the second derivative at each scale is identified. The position of the sign change that occurs in multiple scales is determined as the texture segmentation boundary. The impedance line is divided into multiple texture continuous segments by the texture segmentation boundary. The fluctuation amplitude of the local variance in each texture continuous segment is calculated. Segments with fluctuation amplitude exceeding the continuity threshold are marked as surface defect regions, and segments with fluctuation amplitude below the continuity threshold are marked as uniform regions. Extract the impedance sub-curves corresponding to each continuous segment of the texture on the impedance distribution curve, perform linear fitting on each impedance sub-curve and calculate the fitting residual; use the fluctuation amplitude as the surface quality index and the fitting residual as the impedance stability index, extract the influence coefficient of the fluctuation amplitude on the fitting residual through linear regression, and establish the mapping relationship from the surface quality index to the impedance stability index.

[0007] Based on the aforementioned correlation mapping, the impedance line is divided into multiple processing units and assigned impedance adjustment priorities. For each processing unit, a partitioned impedance adjustment strategy is constructed, including impedance target value, surface quality constraints, and continuity constraints of adjacent unit boundaries, comprising: The influence coefficient is extracted based on the correlation mapping. The rate of change of the influence coefficient is calculated along the impedance line direction. The second derivative of the rate of change curve is calculated. The position where the second derivative turns from negative to positive is identified as the quality-sensitive boundary. The quality-sensitive boundary is used as the processing unit boundary. For each processing unit, the surface quality adjustment cost coefficient is calculated based on the correlation mapping. The root mean square deviation between the impedance distribution curve within the processing unit and the target impedance specification is extracted as the impedance adjustment requirement. This is multiplied by the surface quality adjustment cost coefficient to obtain a processing difficulty score. Based on the processing difficulty score, impedance adjustment priorities are assigned from high to low. A target impedance segment corresponding to the spatial position of the processing unit is extracted from the target impedance specification. The center value of the target impedance segment is calculated as the target impedance value. The maximum allowable change of the surface quality index when the target impedance value is reached is calculated based on the surface quality adjustment cost coefficient as the surface quality constraint. Extract the impedance gradient between the boundary position of the processing unit and the boundary position of the adjacent processing unit, and set the upper limit of the difference between the two impedance gradients as the continuity constraint of the adjacent unit boundary. By using the target impedance value as the optimization objective and the surface quality constraint and the continuity constraint of the adjacent element boundary as the boundary conditions for resistance adjustment, a partitioned resistance adjustment strategy is obtained.

[0008] Based on the aforementioned partitioned impedance tuning strategy, a laser energy modulation sequence and scanning speed coordination scheme for each processing unit are generated. Processing units are selected sequentially according to tuning priority, and the laser beam is driven to perform impedance tuning. Impedance change trajectory data after processing is collected, including: For each processing unit, the impedance target value and surface quality constraint in the partitioned resistance adjustment strategy are extracted, the impedance distribution curve of the processing unit and the deviation curve of the impedance target value are obtained, gradient analysis is performed on the deviation curve to identify the impedance steep change area and the impedance smooth area, and the resistance adjustment point density is set differently. The upper limit of energy deposition at each trimming point is determined based on surface quality constraints. The amount of energy deposition at each trimming point is determined based on the deviation curve. The energy deposition amount is compared with the upper limit of energy deposition to generate a laser energy modulation sequence. The boundary continuity constraint in the partitioned trimming strategy is transformed into a laser action time constraint at the boundary and the boundary scanning speed is determined. The boundary scanning speed and the energy deposition amount at each trimming point are collaboratively mapped using an interpolation algorithm to generate a scanning speed distribution curve, thus forming a collaborative scanning speed scheme. Select a portion of the processing units according to the impedance adjustment priority, convert the scanning speed coordination scheme into the laser beam motion trajectory, and drive the laser beam to perform impedance adjustment processing on the processing units. During the impedance adjustment processing, the impedance response signal of each impedance adjustment point is collected and spatiotemporally registered with the spatial coordinates to form impedance change trajectory data.

[0009] The impedance change trajectory data is time-aligned with the laser energy modulation sequence to extract the time delay characteristics and cumulative effect characteristics of the impedance response, and to establish a time-series coupling relationship, including: The impedance response signal of each tuning point in the impedance change trajectory data is extracted, the laser application time of each tuning point in the laser energy modulation sequence is extracted, the impedance response signal is segmented based on the laser application time, the response start position of each segment is identified by rising edge detection, and the time interval from the trigger reference to the response start position is calculated as the time delay characteristic. The time delay characteristics are arranged along the impedance line to form a delay sequence. The extreme value search of the delay sequence is performed to identify the peak and valley positions. The spacing between adjacent peak positions is extracted and correlated with the spatial dimensions of the processing unit. For each impedance adjustment point, the laser energy values ​​of the current point and previous historical points are obtained. The energy values ​​of the historical points are exponentially decayed and weighted according to time sequence and accumulated to obtain the equivalent cumulative energy. The impedance response amplitude of the current point is extracted. The equivalent cumulative energy and the impedance response amplitude are used to construct a bivariate sequence and the autocorrelation coefficient is calculated as the cumulative effect characteristic. The time delay characteristics and cumulative effect characteristics are combined according to the spatial location of the damping point to form a time-series coupling relationship.

[0010] Based on the aforementioned temporal coupling relationship, the expected impedance response of the unprocessed unit is calculated. The impedance target value and constraint parameters in the partitioned impedance tuning strategy are corrected. The laser energy modulation sequence and scanning speed coordination scheme of the unprocessed unit are updated. This process is iteratively executed until all processing units are completed, including: The temporal coupling relationship is decomposed, and the instantaneous and delayed components of the impedance response are extracted. A two-component superposition model of the impedance response under the action of the laser energy modulation sequence is established, and the model parameters are calibrated based on the impedance change trajectory data of the processed unit. The initial laser energy modulation sequence of the unprocessed unit is input into the dual-component superposition model. The instantaneous influence of the instantaneous component on the impedance and the cumulative influence of the delayed component on the impedance are calculated. The expected impedance response value of each unprocessed unit is synthesized, and it is determined whether the impedance target value and boundary continuity constraint set by the partitioned impedance tuning strategy are met. For unprocessed units that do not meet the conditions, the joint correction vector of the impedance target value and surface quality constraint is calculated. Based on the joint correction vector, the temporal distribution characteristics of the unprocessed unit in the laser energy modulation sequence and the spatial distribution characteristics in the scanning speed coordination scheme are adjusted, and the processing parameters are updated. Process each unit sequentially according to its resistance adjustment priority, update the parameters of the impedance response dual-component superposition model synchronously, and iteratively execute the prediction, correction and update process until all processed units have completed resistance adjustment.

[0011] The temporal coupling relationship is decomposed, and the instantaneous and delayed components of the impedance response are extracted. A two-component superposition model of the impedance response under the action of the laser energy modulation sequence is established. The model parameters are calibrated based on the impedance change trajectory data of the processed unit, including: Identify the time evolution pattern of impedance response in the time-series coupling relationship, divide the impedance response that occurs synchronously with the current laser energy modulation into an instantaneous component, and divide the impedance response that continues to evolve due to the influence of previous processing into a delayed component. For the instantaneous component, the instantaneous component is directly calculated from the laser energy modulation sequence value at the current moment. For the delayed component, the delayed component is recursively updated based on the delayed component value at the previous moment and the current laser energy modulation sequence value. The two calculation methods are combined in parallel to form a two-component superposition model of impedance response. From the impedance change trajectory data of the processed unit, the component that changes synchronously with the laser energy modulation sequence is extracted as the instantaneous component training set, and the component that lags behind the laser energy modulation sequence is extracted as the delayed component training set. The required mapping coefficients are directly calculated using the instantaneous component training set calibration, and the required recursive coefficients and initial states are recursively updated using the delayed component training set calibration, thus completing the parameter calibration of the impedance response dual-component superposition model.

[0012] A second aspect of the present invention provides a real-time precise control system for the impedance line of a thin plate, comprising: The image acquisition unit is used to acquire the surface texture image of the impedance line of the thin plate to be adjusted and the impedance distribution curve along the line; The correlation mapping unit is used to perform multi-scale feature extraction on the surface texture image, identify the surface defect region and uniform region of the impedance line, and establish a correlation mapping between surface quality and impedance stability. The partitioning strategy unit is used to divide the impedance line into multiple processing units according to the association mapping and assign impedance adjustment priority. For each processing unit, a partitioned impedance adjustment strategy including impedance target value, surface quality constraint and adjacent unit boundary continuity constraint is constructed. The processing implementation unit is used to generate a laser energy modulation sequence and scanning speed coordination scheme for each processing unit based on the partitioned impedance tuning strategy, select processing units in sequence according to impedance tuning priority and drive the laser beam to perform impedance tuning processing, and collect impedance change trajectory data after processing. The timing analysis unit is used to perform timing alignment analysis on the impedance change trajectory data and the laser energy modulation sequence, extract the time delay characteristics and cumulative effect characteristics of the impedance response, and establish a timing coupling relationship. The iterative correction unit is used to calculate the expected impedance response value of the unprocessed unit according to the time-series coupling relationship, correct the impedance target value and constraint parameters in the partitioned impedance tuning strategy, update the laser energy modulation sequence and scanning speed coordination scheme of the unprocessed unit, and iteratively execute until all processing units are completed.

[0013] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0014] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0015] This invention divides the process into multiple processing units. Each processing unit is assigned a differentiated priority based on the correlation mapping between surface texture and impedance stability. The coordinated scheme of laser energy and scanning speed is executed precisely according to priority. The defect area and uniform area on the impedance line surface obtain appropriate impedance adjustment parameters, effectively avoiding local overburning or insufficient impedance adjustment, so that the impedance value of the entire line approaches the target and is evenly distributed.

[0016] The partitioned impedance adjustment strategy integrates surface quality constraints and adjacent unit boundary continuity constraints. During the impedance adjustment process, each processing unit not only meets its own impedance target value requirements, but also ensures a smooth impedance transition between units, eliminating step-like abrupt changes. The overall impedance curve is continuous without abnormal fluctuations, and the consistency and reliability of the thin plate impedance line are significantly improved.

[0017] Timing alignment analysis of impedance change trajectory data and laser energy modulation sequence reveals the delay characteristics and cumulative effect of impedance response. The expected impedance response value of the unprocessed unit is dynamically calculated and fed back to correct the strategy parameters of the remaining units. During the iterative execution, the processing deviation of the previous unit is canceled in real time, and the impedance target value and constraint parameters are continuously adaptively adjusted. Finally, the impedance value of all processed units is accurately locked within the target range. No manual intervention is required throughout the process, and the finished product qualification rate of thin plate impedance lines is greatly improved. Attached Figure Description

[0018] Figure 1 This is a flowchart illustrating the real-time precise control method for thin-plate impedance lines according to an embodiment of the present invention. Figure 2 A schematic diagram of the process for establishing the correlation mapping between surface quality and impedance stability in an embodiment of the present invention. Detailed Implementation

[0019] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0020] The technical solution of the present invention will be described in detail below with reference to specific embodiments. These specific embodiments can be combined with each other, and the same or similar concepts or processes may not be described again in some embodiments.

[0021] Figure 1 This is a schematic flowchart of the real-time precise control method for thin-plate impedance lines according to an embodiment of the present invention. The present invention provides a real-time precise control method for thin-plate impedance lines, comprising: Obtain the surface texture image of the impedance line of the thin plate to be tunable and the impedance distribution curve along the line; Multi-scale feature extraction is performed on the surface texture image to identify surface defect regions and uniform regions of the impedance line, and a correlation mapping between surface quality and impedance stability is established. Based on the correlation mapping, the impedance line is divided into multiple processing units and assigned a resistance adjustment priority. For each processing unit, a partitioned resistance adjustment strategy is constructed, which includes the impedance target value, surface quality constraints and the continuity constraints of the adjacent unit boundary. Based on the partitioned impedance tuning strategy, a laser energy modulation sequence and scanning speed coordination scheme for each processing unit are generated. The processing units are selected sequentially according to the impedance tuning priority and the laser beam is driven to perform impedance tuning processing. The impedance change trajectory data after processing is collected. The impedance change trajectory data is time-aligned with the laser energy modulation sequence to extract the time delay characteristics and cumulative effect characteristics of the impedance response and establish a time-series coupling relationship. The expected impedance response of the unprocessed unit is calculated based on the time-series coupling relationship. The impedance target value and constraint parameters in the partitioned impedance tuning strategy are corrected. The laser energy modulation sequence and scanning speed coordination scheme of the unprocessed unit are updated. The process is iteratively executed until all processing units are completed.

[0022] Figure 2 This is a schematic diagram illustrating the process of establishing a correlation mapping between surface quality and impedance stability according to an embodiment of the present invention. Multi-scale feature extraction is performed on the surface texture image to identify surface defect regions and uniform regions of the impedance line, establishing a correlation mapping between surface quality and impedance stability, including: The surface texture image is slid-scanned along the impedance line direction with multiple preset window sizes. Local variance is calculated in each window and arranged along the impedance line direction to form a variance distribution curve. After smoothing and filtering each variance distribution curve, the second derivative is calculated. The position of the positive and negative sign of the second derivative at each scale is identified. The position of the sign change that occurs in multiple scales is determined as the texture segmentation boundary. The impedance line is divided into multiple texture continuous segments by the texture segmentation boundary. The fluctuation amplitude of the local variance in each texture continuous segment is calculated. Segments with fluctuation amplitude exceeding the continuity threshold are marked as surface defect regions, and segments with fluctuation amplitude below the continuity threshold are marked as uniform regions. Extract the impedance sub-curves corresponding to each texture continuous segment on the impedance distribution curve, perform linear fitting on each impedance sub-curve and calculate the fitting residual; By using fluctuation amplitude as a surface quality indicator and fitting residual as an impedance stability indicator, the influence coefficient of fluctuation amplitude on fitting residual is extracted through linear regression, and a mapping relationship between surface quality indicators and impedance stability indicators is established.

[0023] After acquiring the surface texture image of the thin plate to be adjusted, multi-scale sliding scanning of the image with multiple preset window sizes is the foundation for achieving fine texture segmentation. The preset window sizes are typically determined based on the linewidth of the impedance line and the scale range of the defects. For example, three levels can be set: fine-scale window, medium-scale window, and coarse-scale window, corresponding to the detection needs of local micro-scratches along the impedance line, medium-range material inhomogeneity areas, and large-scale macro-defects, respectively. When progressively sliding the surface texture image along the impedance line, the local variance of the pixel grayscale values ​​within each window is calculated. The local variance values ​​at each location are arranged sequentially to form a variance distribution curve along the impedance line. Since different window sizes have different sensitivities to texture changes, fine-scale windows can capture local abrupt changes, while coarse-scale windows reflect the overall texture change trend over a larger area. Therefore, each preset window size corresponds to an independent variance distribution curve.

[0024] Smoothing filters are applied to the variance distribution curves obtained at each scale to suppress pseudo-fluctuations introduced by image acquisition noise or local pixel jitter, ensuring the stability of subsequent derivative calculations. Gaussian kernel filtering can be used for smoothing, with the standard deviation parameter of the filter kernel adaptively adjusted according to the corresponding window size; a larger window results in a larger standard deviation, maintaining consistency in smoothness across scales. After smoothing, the second derivative is calculated for each variance distribution curve. The positions where the sign of the second derivative changes correspond to the inflection points of the variance distribution curve. These inflection points physically represent the turning points in the rate of texture change, signals of fundamental changes in texture structure. The positions where the sign of the second derivative changes in the variance distribution curves at each scale are identified and recorded as candidate texture segmentation boundaries for each scale. Since candidate boundaries detected at a single scale may contain pseudo-boundaries caused by noise, the intersection of candidate boundaries at multiple scales is further calculated. Only positions where the sign changes at multiple scales are ultimately determined as valid texture segmentation boundaries, effectively eliminating misjudgments caused by single-scale detection.

[0025] Using defined texture segmentation boundaries as demarcation points, the impedance line is divided into several texture continuous segments. The texture structure within each continuous segment should theoretically remain relatively consistent; therefore, the fluctuation amplitude of pixel variance within that segment can reflect the uniformity of the surface quality. Specifically, the range or standard deviation of the local variance sequence within each continuous segment is calculated as a quantitative indicator of fluctuation amplitude, and a continuity threshold is set to distinguish between surface defect areas and uniform areas. Texture continuous segments with fluctuation amplitudes exceeding the continuity threshold indicate significant irregular changes in the texture within that segment, suggesting surface defects such as scratches, pits, material peeling, or impurity adhesion, and are thus marked as surface defect areas. Texture continuous segments with fluctuation amplitudes below the continuity threshold indicate uniform surface texture, and are marked as uniform areas. The continuity threshold can be determined by combining the process specifications and experimental calibration data of the thin plate material, or it can be adaptively calculated based on the statistical distribution of the fluctuation amplitudes of all continuous texture segments, for example, by taking the mean of all fluctuation amplitude values ​​plus a certain multiple of the standard deviation as the threshold.

[0026] After classifying and labeling the texture continuum segments, impedance sub-curves corresponding to the spatial positions of each texture continuum segment are extracted from the impedance distribution curves along the line. The extraction of impedance sub-curves depends on the spatial alignment between the surface texture image coordinate system and the impedance distribution curve coordinate system; therefore, prior calibration is required to ensure accurate positional correspondence between the two along the impedance line direction. A linear regression model is then performed on each impedance sub-curve, using the position coordinates on the impedance line as the independent variable and the corresponding impedance measurement value as the dependent variable. The residual sequence between the fitted line and the actual impedance sub-curve is calculated, and the root mean square value of the residuals is used as the impedance stability index for that texture continuum segment. A larger fitting residual indicates a greater deviation of the impedance value from the linear trend, resulting in poorer impedance stability; a smaller fitting residual indicates a more stable impedance distribution and better impedance stability.

[0027] The fluctuation amplitude of each continuous segment of the texture is used as a surface quality index, denoted as . ,in Let be the number of the continuous texture segment; the root mean square value of the fitting residual of the corresponding segment is used as the impedance stability index, denoted as . Collect all texture contiguous segments. For the data pairs, the influence coefficient of fluctuation amplitude on the fitted residuals is extracted using linear regression. The linear regression model can be expressed as follows: ,in The influence coefficient reflects the intensity of the effect of surface quality fluctuations on impedance stability. The intercept term reflects the baseline level of impedance stability when the surface quality is ideal. The solution is obtained by fitting the data of all texture continuity segments using the least squares method. and The estimated values ​​are used to establish a quantitative mapping relationship from surface quality indicators to impedance stability indicators.

[0028] This mapping relationship physically reveals the intrinsic correlation between the uniformity of surface texture and the stability of impedance distribution in thin plates: surface defect areas, due to localized material inhomogeneity and damaged conductive paths, exhibit significant impedance fluctuations along the line, resulting in large fitting residuals; while uniform areas, due to good material consistency, show gentle impedance changes along the line, leading to smaller fitting residuals. By establishing this correlation, subsequent zoned impedance adjustment strategies can apply stronger impedance adjustment interventions to surface defect areas while maintaining relatively mild processing parameters for uniform areas. This improves overall impedance accuracy while avoiding unnecessary thermal damage or material modification to already compliant areas. This mapping relationship also provides a priori basis for predicting the impedance response characteristics of unprocessed units, making the adjustment of impedance target values ​​and constraint parameters during iterative correction more physically interpretable and reliable.

[0029] Based on the aforementioned correlation mapping, the impedance line is divided into multiple processing units and assigned impedance adjustment priorities. For each processing unit, a partitioned impedance adjustment strategy is constructed, including impedance target value, surface quality constraints, and continuity constraints of adjacent unit boundaries, comprising: The influence coefficient is extracted based on the correlation mapping. The rate of change of the influence coefficient is calculated along the impedance line direction. The second derivative of the rate of change curve is calculated. The position where the second derivative turns from negative to positive is identified as the quality-sensitive boundary. The quality-sensitive boundary is used as the processing unit boundary. For each processing unit, the surface quality adjustment cost coefficient is calculated based on the correlation mapping. The root mean square deviation between the impedance distribution curve within the processing unit and the target impedance specification is extracted as the impedance adjustment requirement. This is multiplied by the surface quality adjustment cost coefficient to obtain a processing difficulty score. Based on the processing difficulty score, impedance adjustment priorities are assigned from high to low. A target impedance segment corresponding to the spatial position of the processing unit is extracted from the target impedance specification. The center value of the target impedance segment is calculated as the target impedance value. The maximum allowable change of the surface quality index when the target impedance value is reached is calculated based on the surface quality adjustment cost coefficient as the surface quality constraint. Extract the impedance gradient between the boundary position of the processing unit and the boundary position of the adjacent processing unit, and set the upper limit of the difference between the two impedance gradients as the continuity constraint of the adjacent unit boundary. By using the target impedance value as the optimization objective and the surface quality constraint and the continuity constraint of the adjacent element boundary as the boundary conditions for resistance adjustment, a partitioned resistance adjustment strategy is obtained.

[0030] After obtaining the correlation mapping between surface quality and impedance stability, the impedance line needs to be divided into several physically meaningful processing units along its extension direction, and each unit needs to be assigned a reasonable impedance adjustment priority, thus providing a structured operational framework for subsequent laser processing. The core basis for dividing the processing units is the distribution law of the influence coefficients in the correlation mapping along the impedance line direction. The influence coefficients reflect the sensitivity of surface quality changes to impedance stability, and their values ​​differ at different locations. Therefore, the influence coefficients are calculated point by point along the impedance line direction to obtain the influence coefficient distribution sequence.

[0031] The first-order rate of change of the influence coefficient distribution sequence along its spatial location is calculated to obtain a rate of change curve, which describes how quickly the influence coefficients change spatially. Based on this, the second derivative of the rate of change curve is taken to obtain the second derivative sequence. The sign change of the second derivative has important physical implications: when the second derivative changes from negative to positive, the rate of change of the influence coefficient at the corresponding location changes from deceleration to acceleration. This means that the influence of surface quality on impedance stability near that location begins to significantly increase, representing a boundary point where the quality sensitivity undergoes a structural transformation; this is defined as the quality-sensitive boundary. Using all quality-sensitive boundaries as the dividing boundaries for processing units, the entire impedance line is divided into several processing units. The quality sensitivity within each processing unit is relatively uniform, facilitating the subsequent development of targeted impedance adjustment strategies.

[0032] After completing the processing unit division, the processing difficulty of each unit needs to be quantitatively assessed, and resistance adjustment priorities assigned accordingly. Surface quality adjustment cost coefficient. Characterization of the first The processing cost required for surface quality improvement in a processing unit is derived from the local parameters of the associated mapping within that unit. The larger the value, the more sensitive the surface quality of the unit is to laser processing response, and the more difficult it is to adjust. Impedance adjustment requirement. Defined as the first The root mean square deviation between the measured impedance distribution curve and the target impedance specification within each processing unit is calculated using the following formula: ,in For the first The total number of sampling points within each processing unit For the first The first processing unit The measured impedance values ​​at each sampling point This represents the target impedance specification value for the corresponding location. Manufacturing difficulty score. It is obtained by multiplying the surface quality adjustment cost coefficient and the impedance adjustment requirement, i.e. .according to The processing units are sorted from high to low. The higher the score, the greater the impedance deviation and the more difficult the adjustment. These units should be processed first to avoid the impedance of high-difficulty units shifting further due to the cumulative effect after the low-priority units are processed, thereby improving the overall control accuracy.

[0033] For each processing unit, a target impedance segment corresponding to the spatial location of the unit is extracted from the target impedance specification. The arithmetic mean of the impedance values ​​at all sampling points within this target impedance segment is calculated and used as the target impedance value for that processing unit. The target impedance value represents the overall impedance level that the processing unit is expected to achieve after impedance tuning, serving as the core optimization objective for subsequent laser energy modulation. Simultaneously, the cost coefficient is adjusted based on surface quality. Calculate the maximum allowable change in surface quality parameters while achieving the target impedance value. Specifically, The derivation is made through the inverse mapping relationship of the correlation mapping, that is, given a change in impedance... (That Under the condition of (the average measured impedance of this unit), combined with Determine the maximum permissible fluctuation range of surface quality indicators during processing, ensuring they do not exceed the material damage threshold. The surface quality constraints constrain the processing unit to ensure that the laser trimming process does not cause irreversible damage such as surface microcracks or localized ablation due to excessive energy concentration.

[0034] Impedance continuity between adjacent processing units is crucial for ensuring the overall electrical performance uniformity of the thin plate. At the boundary of each processing unit, the impedance gradient value of the measured impedance distribution curve at that boundary is extracted. The impedance gradient is defined as the first derivative of the impedance value along its spatial position, reflecting the steepness of the impedance change near the boundary. For adjacent units... The and the first Each processing unit extracts its impedance gradient at its common boundary. and The absolute value of the difference between the two Set upper limit That is, requirements This serves as a continuity constraint for the boundaries of adjacent units. The value of is determined based on the design specifications of the thin plate impedance line and the requirements for impedance abrupt change tolerance in practical applications. It is usually calibrated according to the maximum allowable gradient difference between adjacent sections in the target impedance specification. The introduction of this constraint can effectively prevent impedance steps or abrupt changes at the boundary after adjacent processing units have completed their respective impedance adjustments, ensuring a smooth transition along the impedance line.

[0035] Target impedance value As the target for resistance adjustment optimization in each processing unit, surface quality is constrained. Continuity constraints with adjacent element boundaries Together, these serve as boundary conditions for impedance tuning, forming a comprehensive zoned impedance tuning strategy. This strategy stores the target parameters and constraints of each processing unit in a structured format, providing a complete input specification for the subsequent generation of laser energy modulation sequences and scanning speed coordination schemes. While the strategies of each processing unit are independent, they are globally interconnected through boundary continuity constraints. This allows for precise control of the entire impedance line, enabling both refined processing of local characteristics and maintaining overall impedance distribution consistency at the macroscopic level. This achieves high-precision zoned collaborative control of the thin-plate impedance line.

[0036] Based on the aforementioned partitioned impedance tuning strategy, a laser energy modulation sequence and scanning speed coordination scheme for each processing unit are generated. Processing units are selected sequentially according to tuning priority, and the laser beam is driven to perform impedance tuning. Impedance change trajectory data after processing is collected, including: For each processing unit, the impedance target value and surface quality constraint in the partitioned resistance adjustment strategy are extracted, the impedance distribution curve of the processing unit and the deviation curve of the impedance target value are obtained, gradient analysis is performed on the deviation curve to identify the impedance steep change area and the impedance smooth area, and the resistance adjustment point density is set differently. The upper limit of energy deposition at each trimming point is determined based on surface quality constraints. The amount of energy deposition at each trimming point is determined based on the deviation curve. The energy deposition amount is compared with the upper limit of energy deposition to generate a laser energy modulation sequence. The boundary continuity constraint in the partitioned trimming strategy is transformed into a laser action time constraint at the boundary and the boundary scanning speed is determined. The boundary scanning speed and the energy deposition amount at each trimming point are collaboratively mapped using an interpolation algorithm to generate a scanning speed distribution curve, thus forming a collaborative scanning speed scheme. Select a portion of the processing units according to the impedance adjustment priority, convert the scanning speed coordination scheme into the laser beam motion trajectory, and drive the laser beam to perform impedance adjustment processing on the processing units. During the impedance adjustment processing, the impedance response signal of each impedance adjustment point is collected and spatiotemporally registered with the spatial coordinates to form impedance change trajectory data.

[0037] For each processing unit, the corresponding impedance target value and surface quality constraint are extracted from the zoned impedance adjustment strategy. The impedance distribution curve of the processing unit is then subtracted from the impedance target value point by point to obtain the deviation curve along the line. The deviation curve intuitively reflects the magnitude and spatial distribution of the difference between the impedance and the target specification at each location. A first-order gradient is calculated along the spatial coordinate direction of the deviation curve. Sections with large absolute gradient values ​​indicate a sharp change in impedance deviation, defined as the impedance steep change region; sections with small absolute gradient values ​​and gradual changes are defined as the impedance smooth change region. The impedance steep change region is sensitive to laser impedance adjustment. If the density of adjustment points is insufficient, it can easily lead to local over-adjustment or under-adjustment, resulting in new non-uniformities in the impedance distribution after adjustment. Therefore, in the impedance steep change region, it is necessary to differentially increase the density of adjustment points to make the laser action points more concentrated, thereby achieving fine control of the impedance steep change section. In the impedance smooth change region, the impedance deviation changes slowly, and the density of adjustment points can be appropriately reduced to decrease unnecessary laser irradiation times, thus protecting the surface structure of the thin plate and improving overall processing efficiency.

[0038] When determining the energy deposition amount at each trimming point, both the impedance deviation curve and surface quality constraints must be considered simultaneously. According to the impedance deviation curve, the amount of laser energy required at each trimming point determines the required energy deposition amount: the larger the deviation, the higher the required energy deposition amount; the smaller the deviation, the lower the required energy deposition amount. Specifically, let the first trimming point... The impedance deviation at each adjustment point is The corresponding initial energy deposition requirement is The two are converted using a pre-calibrated impedance-energy response relationship, which is obtained through multiple sets of experimental fitting during the material calibration stage. Simultaneously, surface quality constraints limit the upper limit of laser energy deposition at each adjustable resistance point, denoted as the... The upper limit of energy deposition at each resistance adjustment point is: This upper limit is derived from the surface quality constraint parameters of the processing unit, ensuring that laser processing will not cause damage to the thin plate surface beyond the allowable range. The initial energy deposition requirement... With upper limit of energy deposition Perform point-by-point comparison: If Then directly adopt As the actual energy deposition at that point; if The actual energy deposition amount is then cut off to Simultaneously, supplementary processing planning is carried out for this point during subsequent iterative correction phases. All trimming points are processed one by one according to the above rules, and arranged sequentially to form a laser energy modulation sequence. This sequence completely records the laser energy output command corresponding to each trimming point.

[0039] The scanning speed coordination scheme is generated based on boundary continuity constraints. The partitioned impedance tuning strategy specifies the impedance gradient difference constraint between adjacent processing units at their common boundary. To implement this constraint at the laser processing execution level, it needs to be transformed into a laser interaction time constraint at the boundary. The laser interaction time at the boundary directly determines the local energy deposition density, thus affecting the smoothness of the impedance transition on both sides of the boundary. Based on the required energy deposition amount and laser beam area at the boundary, the upper limit of the laser interaction time at the boundary is calculated. Then, combined with the path length of the laser beam at the boundary, the scanning speed constraint value at the boundary is calculated, denoted as... . The determination of the speed ensures that when the laser beam passes through the common boundary of adjacent processing units, it can complete the necessary energy deposition without causing local overheating due to excessively low speed or insufficient energy due to excessively high speed, thereby ensuring the continuity of impedance transition on both sides of the boundary.

[0040] Determine the boundary scan speed Subsequently, an interpolation algorithm was used to co-map the boundary scan velocity with the energy deposition at each adjustable resistivity point, generating a complete scan velocity distribution curve. The scan velocity at each adjustable resistivity point... The actual energy deposition at this point The scan speed (or the truncated value) is determined by both the laser power parameter and the energy deposition rate: Under a fixed laser power, a higher energy deposition rate results in a lower scan speed; conversely, a lower energy deposition rate results in a higher scan speed. Scan speed at boundary positions. As constraint anchor points, cubic spline interpolation or piecewise linear interpolation methods are used to smoothly connect the speed requirements of each trimming point with the boundary speed constraints, avoiding abrupt changes in scanning speed in space and preventing local anomalies in laser energy distribution caused by speed abrupt changes. After interpolation, the output scanning speed distribution curve covers all trimming points and boundary areas within the processing unit, forming a complete scanning speed coordination scheme. This scheme, together with the laser energy modulation sequence, constitutes a complete control command set for driving the laser beam to perform trimming processing.

[0041] Processing units are selected sequentially from high to low priority based on impedance adjustment. The velocity distribution curve in the scanning speed coordination scheme is converted into the motion trajectory of the laser beam in the processing plane. Specifically, this conversion requires combining the spatial coordinate system of the processing unit with the kinematic parameters of the laser scanning head. The coordinates, corresponding velocity values, and energy modulation commands of each impedance adjustment point are integrated into a time-sequential motion control command, driving the laser beam to perform impedance adjustment processing on the selected processing units according to the planned trajectory. During impedance adjustment processing, impedance response signals are acquired in real time at each impedance adjustment point. The acquisition method employs a four-terminal impedance measurement method triggered synchronously with the laser scanning. The measurement time is strictly synchronized with the arrival time of the laser beam at each point, ensuring that the acquired impedance response signals accurately correspond to the laser action event in the time dimension. The acquired impedance response signals carry the spatial coordinate information of each impedance adjustment point. A spatiotemporal registration algorithm binds the impedance response signals with the corresponding spatial coordinates, forming impedance change trajectory data indexed by the spatial coordinates. During the spatiotemporal registration process, the timestamp of the laser beam's motion trajectory is used as a reference to map the impedance acquisition time to the corresponding spatial coordinate position. This eliminates the spatiotemporal misalignment error caused by acquisition delay or signal transmission jitter, ensuring that the impedance change trajectory data can accurately reflect the true impedance response state after laser trimming at each spatial position. This provides a reliable data foundation for subsequent time-series coupling analysis and strategy correction.

[0042] The impedance change trajectory data is time-aligned with the laser energy modulation sequence to extract the time delay characteristics and cumulative effect characteristics of the impedance response, and to establish a time-series coupling relationship, including: The impedance response signal of each tuning point in the impedance change trajectory data is extracted, the laser application time of each tuning point in the laser energy modulation sequence is extracted, the impedance response signal is segmented based on the laser application time, the response start position of each segment is identified by rising edge detection, and the time interval from the trigger reference to the response start position is calculated as the time delay characteristic. The time delay characteristics are arranged along the impedance line to form a delay sequence. The extreme value search of the delay sequence is performed to identify the peak and valley positions. The spacing between adjacent peak positions is extracted and correlated with the spatial dimensions of the processing unit. For each impedance adjustment point, the laser energy values ​​of the current point and previous historical points are obtained. The energy values ​​of the historical points are exponentially decayed and weighted according to time sequence and accumulated to obtain the equivalent cumulative energy. The impedance response amplitude of the current point is extracted. The equivalent cumulative energy and the impedance response amplitude are used to construct a bivariate sequence and the autocorrelation coefficient is calculated as the cumulative effect characteristic. The time delay characteristics and cumulative effect characteristics are combined according to the spatial location of the damping point to form a time-series coupling relationship.

[0043] After laser trimming of each processing unit, the acquired impedance change trajectory data records the dynamic evolution of impedance at each trimming point before and after laser application. These trajectory data have an inherent temporal correlation with the laser energy modulation sequence, requiring precise timing alignment analysis to reveal the physical laws governing the impedance response. The core of timing alignment lies in calibrating the impedance acquisition clock and the laser control clock using a unified reference system, eliminating the inherent deviation between the two signals at the start of acquisition, and ensuring the consistency of the time base for subsequent feature extraction.

[0044] For the impedance change trajectory data, impedance response signal segments corresponding to each tuning point are extracted one by one. The impedance response signal of each tuning point is a continuous sequence that changes over time, reflecting the evolution of the impedance value at that point before and after the application of laser energy. Simultaneously, the laser application time for each tuning point is extracted from the laser energy modulation sequence; this time is defined as the precise timestamp at which the laser pulse or continuous waveform begins to act at that point. Using the laser application time as the trigger reference, the impedance response signal is segmented. The segmentation window covers several sampling points before the laser application time to confirm the baseline level, and a sufficiently long time range after the laser application time to fully capture the rise, stabilization, and possible attenuation processes of the impedance response.

[0045] After segmentation, a rising edge detection algorithm is executed on each signal segment to identify the starting position of the impedance response. The specific implementation of rising edge detection is as follows: using the average signal value before the laser application moment within the segmented window as a baseline reference value, a detection threshold higher than the baseline reference value is set, for example, the baseline reference value plus a multiple of the baseline standard deviation. After the laser application moment, sampling points are scanned sequentially. When the signal value first exceeds the detection threshold and remains above the threshold for several consecutive subsequent sampling points, the sampling point at which the threshold is first exceeded is marked as the response start position. The time interval between the trigger reference (laser application moment) and the response start position is calculated; this time interval is the time delay characteristic of the impedance adjustment point, denoted as . ,in This is the index for the resistance adjustment point. The physical meaning is the thermal conduction and material response hysteresis time between the laser energy deposited onto the thin plate material and the detectable impedance change.

[0046] Time delay characteristics of each resistance adjustment point Arranged spatially along the impedance line direction, a delay sequence is formed. This delay sequence is a one-dimensional sequence with the spatial coordinates of the adjustment points on the horizontal axis and the time delay on the vertical axis. An extreme value search algorithm is applied to this delay sequence to identify peak positions (indices of local maxima) and valley positions (indices of local minima). Peak positions typically correspond to regions on the impedance line where the material response is slow, such as areas with dense surface defects or abnormal thermal conductivity; valley positions correspond to uniform regions where the material response is fast. The distance between adjacent peak positions is extracted, converted into spatial length, and correlated with the spatial dimensions of the processing unit. This verifies whether the periodic structure of the delay sequence spatially matches the boundary of the processing unit, providing a spatial positioning basis for subsequent correction of the partitioned impedance adjustment strategy.

[0047] The extraction of cumulative effect characteristics was performed independently for each adjustable resistance point. For the point numbered... The system identifies the impedance adjustment points and obtains the laser energy value at the current point, as well as the laser energy values ​​of all historical points preceding the current point in time sequence. The contribution of the energy values ​​of historical points to the impedance response of the current point decreases with increasing time interval. An exponential decay weighting method is used to process the historical energy: historical points closer to the current point in time are assigned higher weights, and historical points farther away are assigned lower weights. The decay rate is determined by an exponential decay coefficient. control, The value of reflects the combined characteristics of the thermal diffusion rate and impedance recovery rate of the thin plate material. The equivalent cumulative energy at the current point is obtained by summing the historical energy values ​​after exponential decay weighting. Its calculation form is ,in For the first Historical points (including the current point, i.e.) The laser energy value at (time). For the historical location index, This is the exponential decay coefficient.

[0048] Extract current location impedance response amplitude , defined as the absolute offset of the impedance change of the impedance response signal at that point relative to the baseline after the response stabilizes. The equivalent cumulative energy at each impedance adjustment point... With impedance response amplitude Constructing a bivariate sequence, i.e., using For indexing, with This is an ordered set of element pairs. The autocorrelation coefficient of this bivariate sequence is calculated using the standard Pearson correlation coefficient method. The correlation between the equivalent cumulative energy sequence and the impedance response amplitude sequence is calculated at different time lags, resulting in an autocorrelation function curve with the time lag as the horizontal axis and the correlation coefficient as the vertical axis. This autocorrelation function curve serves as a quantitative representation of the cumulative effect. The time lag corresponding to its peak position reflects the time span during which the historical laser energy has the most significant impact on the current impedance response, while the magnitude of the peak correlation coefficient reflects the strength of the cumulative effect.

[0049] After extracting the time delay characteristics and cumulative effect characteristics separately, the two types of characteristics are combined according to the spatial coordinates of each adjustment point. For each adjustment point... Its time delay characteristics Extreme value structure information and equivalent cumulative energy of delayed sequences Impedance response amplitude The key parameters of the autocorrelation function curve (peak time lag and peak correlation coefficient) are integrated into a multidimensional feature vector and bound to the spatial coordinates of that point for storage. The multidimensional feature vectors of all trimmed points are arranged in spatial order to form a time-series coupling relationship data structure covering the entire impedance line. This time-series coupling relationship data structure fully describes the dynamic temporal relationship between laser energy application and impedance response at each location on the impedance line, providing an accurate physical model basis for subsequent calculations of the expected impedance response of the unprocessed unit and correction of the partitioned trimming strategy. The spatial resolution of the time-series coupling relationship is consistent with the spacing of the trimmed points, supporting point-by-point impedance prediction and strategy correction, ensuring the uniformity and accuracy of the trimming results across the entire impedance line.

[0050] Based on the aforementioned temporal coupling relationship, the expected impedance response of the unprocessed unit is calculated. The impedance target value and constraint parameters in the partitioned impedance tuning strategy are corrected. The laser energy modulation sequence and scanning speed coordination scheme of the unprocessed unit are updated. This process is iteratively executed until all processing units are completed, including: The temporal coupling relationship is decomposed, and the instantaneous and delayed components of the impedance response are extracted. A two-component superposition model of the impedance response under the action of the laser energy modulation sequence is established, and the model parameters are calibrated based on the impedance change trajectory data of the processed unit. The initial laser energy modulation sequence of the unprocessed unit is input into the dual-component superposition model. The instantaneous influence of the instantaneous component on the impedance and the cumulative influence of the delayed component on the impedance are calculated. The expected impedance response value of each unprocessed unit is synthesized, and it is determined whether the impedance target value and boundary continuity constraint set by the partitioned impedance tuning strategy are met. For unprocessed units that do not meet the conditions, the joint correction vector of the impedance target value and surface quality constraint is calculated. Based on the joint correction vector, the temporal distribution characteristics of the unprocessed unit in the laser energy modulation sequence and the spatial distribution characteristics in the scanning speed coordination scheme are adjusted, and the processing parameters are updated. Process each unit sequentially according to its resistance adjustment priority, update the parameters of the impedance response dual-component superposition model synchronously, and iteratively execute the prediction, correction and update process until all processed units have completed resistance adjustment.

[0051] After acquiring the impedance change trajectories of all processed units and establishing the temporal coupling relationship, the temporal coupling relationship needs to be further decomposed into two independent components: an instantaneous component and a delayed component. The instantaneous component reflects the impedance change of the material in a very short time after the laser energy is applied, mainly contributed by the instantaneous change in resistivity caused by local thermal effects. The delayed component reflects the continuous evolution of impedance caused by slow processes such as thermal diffusion, stress relaxation, and microstructure rearrangement, and its time span is much longer than that of the instantaneous component. After modeling the two components separately, a dual-component superposition model of the impedance response is constructed. The model is a linear superposition of the instantaneous and delayed components. The instantaneous component takes the laser energy and local state of the material at the current trimming point as input, while the delayed component takes the equivalent cumulative energy of the historical processing sequence as input. Using the impedance change trajectory data of the processed units, the parameters in the dual-component superposition model are calibrated by least-squares fitting to ensure that the model has sufficient fitting accuracy for the impedance response of the processed units. The calibration error must be lower than the preset impedance accuracy threshold; otherwise, the model structure needs to be corrected or a nonlinear correction term needs to be introduced before recalibration.

[0052] After model calibration, the initial laser energy modulation sequence of the unprocessed unit is used as input to calculate the expected impedance response of each unprocessed unit under the two-component superposition model. For the first... Let the instantaneous component of an unprocessed unit have an instantaneous effect on the impedance as value. The cumulative effect of the delay component on the impedance is... Then the expected value of the impedance response of the unit. satisfy .in, Determined by the laser power density and interaction time of the current unit. The energy is obtained by superimposing the equivalent cumulative energy of historically processed units after being converted using an exponential decay model. The time constant of the decay process is determined during the calibration phase. The calculation yields... Then, it is compared with the impedance target value of that unit in the partitioned impedance adjustment strategy, and at the same time, it is checked whether the impedance gradient difference at the common boundary of adjacent units satisfies the boundary continuity constraint. If If the deviation from the target impedance value exceeds the allowable range, or the boundary impedance gradient difference exceeds the upper limit, the unit is determined to not meet the conditions and needs to enter the joint correction process.

[0053] For unprocessed units that do not meet the conditions, a joint correction vector for the impedance target value and surface quality constraints is calculated. The joint correction vector consists of two components: an impedance target correction component to compensate for the deviation between the expected and target values, and a surface quality constraint correction component to prevent the laser energy adjustment amplitude from exceeding the allowable variation range of the surface quality indicators. Let the... The impedance target correction for each unprocessed unit is The surface quality constraint correction amount is Then the joint correction vector . The impedance response is determined by the difference between the expected value and the target value, as well as the propagation effect of the delay component in subsequent elements. It is necessary to comprehensively consider the impact of the current element's correction on the expected value of the downstream element's impedance to avoid local overcorrection that could lead to an imbalance in the global impedance distribution. The laser energy adjustment range is constrained based on the maximum allowable change in surface quality indicators to ensure that the corrected processing parameters do not cause surface quality deterioration beyond specifications.

[0054] Based on the joint correction vector, the laser energy modulation sequence and scanning speed coordination scheme of the unprocessed unit are adjusted respectively. Regarding the adjustment of the temporal distribution characteristics of the laser energy modulation sequence, based on... The sign and amplitude of the energy deposition at each impedance adjustment point in the sequence are used to redistribute the energy, prioritizing the adjustment of time-domain segments that contribute significantly to the impedance response, while keeping the overall energy of the sequence within the material's tolerance range. Regarding the spatial distribution characteristics adjustment of the scan rate coordination scheme, based on... The constraints are used to smooth the distribution of scanning speed at various spatial locations, avoiding abrupt changes in speed that could lead to abnormal local energy density. After adjustment, the updated laser energy modulation sequence and scanning speed coordination scheme are re-input into the two-component superposition model to verify whether the corrected impedance response expectation value meets the impedance target value and boundary continuity constraints. If it still does not meet the constraints, iterative correction continues until the conditions are met or the maximum number of iterations is reached.

[0055] During the actual processing phase, laser trimming is performed on each unprocessed unit sequentially according to trimming priority. After each unit is processed, the measured impedance change trajectory data of that unit is collected and incorporated into the parameter update process of the dual-component superposition model. Using the newly acquired data, the parameters of the instantaneous and delayed components in the model are corrected online, so that the model's accuracy in predicting the impedance response of subsequent unprocessed units continuously improves as the processing progresses. The parameter update adopts the recursive least squares method, with newly acquired data as the main weight and historical data weighted by a forgetting factor, ensuring that the model parameters can reflect the dynamic changes in the material processing state in a timely manner, while avoiding excessive interference from historical data on the current state.

[0056] After parameter updates are completed, the entire process of prediction, correction, and update is re-executed for the remaining unprocessed units: the updated model parameters and the current laser energy modulation sequence of each unprocessed unit are input into the dual-component superposition model to calculate the new expected impedance response value. It is then determined whether the constraints are met. For units that do not meet the conditions, the joint correction vector is recalculated and the processing parameters are updated. Processing is then performed according to priority. This iterative process continues until all processed units have completed impedance adjustment. In the final verification stage, the measured impedance distribution of all processed units is comprehensively checked to confirm that the impedance values ​​of each unit fall within the target specification range, and that the impedance gradient difference at the boundary of adjacent units meets the continuity constraint. This achieves real-time, precise control of the impedance along the thin plate impedance line and ensures global consistency.

[0057] The temporal coupling relationship is decomposed, and the instantaneous and delayed components of the impedance response are extracted. A two-component superposition model of the impedance response under the action of the laser energy modulation sequence is established. The model parameters are calibrated based on the impedance change trajectory data of the processed unit, including: Identify the time evolution pattern of impedance response in the time-series coupling relationship, divide the impedance response that occurs synchronously with the current laser energy modulation into an instantaneous component, and divide the impedance response that continues to evolve due to the influence of previous processing into a delayed component. For the instantaneous component, the instantaneous component is directly calculated from the laser energy modulation sequence value at the current moment. For the delayed component, the delayed component is recursively updated based on the delayed component value at the previous moment and the current laser energy modulation sequence value. The two calculation methods are combined in parallel to form a two-component superposition model of impedance response. From the impedance change trajectory data of the processed unit, the component that changes synchronously with the laser energy modulation sequence is extracted as the instantaneous component training set, and the component that lags behind the laser energy modulation sequence is extracted as the delayed component training set. The required mapping coefficients are directly calculated using the instantaneous component training set calibration, and the required recursive coefficients and initial states are recursively updated using the delayed component training set calibration, thus completing the parameter calibration of the impedance response dual-component superposition model.

[0058] After obtaining the temporal coupling relationship, further structural decomposition is needed to distinguish the contributions of different physical mechanisms to the impedance response. The impedance response is not a single source on the time axis: when a laser beam acts on the impedance lines of a thin plate, part of the impedance change occurs almost synchronously with the laser energy modulation, reflecting the instantaneous thermal effect of the material under the current laser energy input; another part of the impedance change lags behind the laser energy modulation, originating from the heat accumulation in the previously processed area, the release of residual stress, and the continuous evolution of the microstructure, which is a delayed effect. Distinguishing between these two types of responses and establishing independent computational paths is the foundation for building accurate prediction models.

[0059] When identifying the temporal evolution pattern of impedance response in a temporally coupled relationship, the timestamp of the laser energy modulation sequence is used as a reference to perform point-by-point time delay analysis on the acquired impedance change trajectory data. The portion of the impedance response signal that is highly synchronized with the current laser energy modulation sequence in time and has a correlation coefficient exceeding a set threshold is classified as the instantaneous component, characterizing the impedance change behavior directly induced by the laser energy. The portion of the impedance response signal with significant time delay that continues to evolve after the laser energy modulation ends is classified as the delayed component, characterizing the continued impact of previous processing on the current region's impedance. This classification process relies on cross-correlation analysis of the impedance response signal: the cross-correlation function values ​​of the impedance change trajectory data and the laser energy modulation sequence at different time delays are calculated. Responses with a peak value corresponding to a time delay close to zero belong to the instantaneous component, while responses with a peak value corresponding to a time delay significantly greater than zero belong to the delayed component. Through statistical analysis of the time delay distribution, the time delay threshold for classifying the instantaneous and delayed components can be determined, ensuring that the physical meanings of the two types of components are clear and do not overlap.

[0060] For the calculation of instantaneous components, a direct mapping method based on the current laser energy modulation sequence value is adopted. Let the current time... The laser energy modulation sequence value is The instantaneous component impedance response value is Then the two are connected through the mapping coefficient. Establish a linear relationship, that is Mapping coefficients here The instantaneous impedance change caused by a unit laser energy input is closely related to the thermal resistivity of the thin plate material, the laser spot size, and the duration of the interaction. For the calculation of the delay component, a recursive update strategy is adopted: assuming the previous time step... The delay component value The current laser energy modulation sequence value is Then the current time delay component From recursion coefficients Coupling coefficient with energy Jointly decided, the recursive relationship is .in, The self-decay rate characterizes the delayed component, reflecting the gradual decay of the effects of previous processing over time; This characterizes the new contribution of the current laser energy to the delay component, reflecting the continuous superposition of thermal accumulation effects. By superimposing the instantaneous and delayed components in parallel, a two-component superposition model of the impedance response is obtained, which is the predicted value of the total impedance response at the current moment. satisfy This allows the contributions of both physical mechanisms to be incorporated into the same prediction framework.

[0061] When constructing the training set, starting from the impedance change trajectory data of the processed units, the impedance response signal is separated and extracted according to the aforementioned time delay threshold. Impedance response data points that change synchronously with the laser energy modulation sequence and whose time delay is within the threshold range constitute the instantaneous component training set. Each record in this training set contains the laser energy modulation sequence value at the current moment. and the corresponding instantaneous impedance response observations Impedance response data points that lag behind the laser energy modulation sequence, have a time delay exceeding a threshold, and continue to change after the laser energy modulation ends constitute the delay component training set. Each record in this training set contains the delay component observation value from the previous time step. Current laser energy modulation sequence value and the current time delay component observation value The quality of the training set directly affects the accuracy of model parameter calibration. Therefore, outlier data points need to be removed during the extraction process. An outlier detection method based on residuals is used to exclude data points with residuals exceeding three times the standard deviation from the training set.

[0062] Calibrate mapping coefficients using instantaneous component training set At that time, with the objective of minimizing the mean square error between the predicted and observed instantaneous component values, the solution is obtained through least squares regression. The optimal estimate is obtained. Specifically, a system of regression equations is constructed for all data points in the training set, and solved using matrix inversion or gradient descent methods to obtain the optimal estimate. The calibration results. The recursive coefficients are calibrated using the delayed component training set. Energy coupling coefficient and initial state Then, the recursive relation is expanded into a multi-step prediction form, and the delayed component observations at all times in the training set are used as supervision signals to construct a multi-parameter joint optimization problem. An alternating iterative optimization strategy is adopted: the initial state is fixed. ,optimization and ; then fix and ,optimization Repeat the iterations until the parameters converge, thus completing the calibration of the parameters related to the delay component. Initial state The physical meaning of impedance is the initial effect of the thermal effect accumulated inside the material at the beginning of the processing sequence on the impedance. It can usually be set to zero before the first processing unit, but in subsequent processing units, it needs to be initialized according to the cumulative effect of the previous processing.

[0063] After parameter calibration, the two-component superposition model is validated. A portion of impedance change trajectory data from the processed units is reserved as a validation set. The calibrated model parameters are substituted into the two-component superposition model to calculate the predicted impedance response values ​​at each time step. The predicted error is calculated by comparing it with the measured impedance response in the validation set. If the prediction error exceeds the preset accuracy requirement, the training set needs to be expanded or the time delay threshold adjusted before recalibration until the model's prediction accuracy meets the requirements. The validated dual-component superposition model will serve as the core tool for calculating the expected impedance response of subsequent unprocessed units, providing a reliable prediction basis for the dynamic correction of the partitioned impedance adjustment strategy and ensuring that the impedance adjustment accuracy of the entire impedance line continues to improve during iterative processing.

[0064] A second aspect of the present invention provides a real-time precise control system for the impedance line of a thin plate, comprising: The image acquisition unit is used to acquire the surface texture image of the impedance line of the thin plate to be adjusted and the impedance distribution curve along the line; The correlation mapping unit is used to perform multi-scale feature extraction on the surface texture image, identify the surface defect region and uniform region of the impedance line, and establish a correlation mapping between surface quality and impedance stability. The partitioning strategy unit is used to divide the impedance line into multiple processing units according to the association mapping and assign impedance adjustment priority. For each processing unit, a partitioned impedance adjustment strategy including impedance target value, surface quality constraint and adjacent unit boundary continuity constraint is constructed. The processing implementation unit is used to generate a laser energy modulation sequence and scanning speed coordination scheme for each processing unit based on the partitioned impedance tuning strategy, select processing units in sequence according to impedance tuning priority and drive the laser beam to perform impedance tuning processing, and collect impedance change trajectory data after processing. The timing analysis unit is used to perform timing alignment analysis on the impedance change trajectory data and the laser energy modulation sequence, extract the time delay characteristics and cumulative effect characteristics of the impedance response, and establish a timing coupling relationship. The iterative correction unit is used to calculate the expected impedance response value of the unprocessed unit according to the time-series coupling relationship, correct the impedance target value and constraint parameters in the partitioned impedance tuning strategy, update the laser energy modulation sequence and scanning speed coordination scheme of the unprocessed unit, and iteratively execute until all processing units are completed.

[0065] A third aspect of the present invention provides an electronic device, comprising: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the aforementioned method.

[0066] A fourth aspect of the present invention provides a computer-readable storage medium having stored thereon computer program instructions that, when executed by a processor, implement the aforementioned method.

[0067] This invention can be a method, apparatus, system, and / or computer program product. The computer program product may include a computer-readable storage medium having computer-readable program instructions loaded thereon for performing various aspects of the invention.

[0068] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for real-time and precise control of impedance lines in thin plates, characterized in that, include: Obtain the surface texture image of the impedance line of the thin plate to be tunable and the impedance distribution curve along the line; Multi-scale feature extraction is performed on the surface texture image to identify surface defect regions and uniform regions of the impedance line, and a correlation mapping between surface quality and impedance stability is established. Based on the correlation mapping, the impedance line is divided into multiple processing units and assigned a resistance adjustment priority. For each processing unit, a partitioned resistance adjustment strategy is constructed, which includes the impedance target value, surface quality constraints and the continuity constraints of the adjacent unit boundary. Based on the partitioned impedance tuning strategy, a laser energy modulation sequence and scanning speed coordination scheme for each processing unit are generated. The processing units are selected sequentially according to the impedance tuning priority and the laser beam is driven to perform impedance tuning processing. The impedance change trajectory data after processing is collected. The impedance change trajectory data is time-aligned with the laser energy modulation sequence to extract the time delay characteristics and cumulative effect characteristics of the impedance response and establish a time-series coupling relationship. The expected impedance response of the unprocessed unit is calculated based on the time-series coupling relationship. The impedance target value and constraint parameters in the partitioned impedance tuning strategy are corrected. The laser energy modulation sequence and scanning speed coordination scheme of the unprocessed unit are updated. The process is iteratively executed until all processing units are completed.

2. The method according to claim 1, characterized in that, Multi-scale feature extraction is performed on the surface texture image to identify surface defect regions and uniform regions of the impedance line, and a correlation mapping between surface quality and impedance stability is established, including: The surface texture image is slid-scanned along the impedance line direction with multiple preset window sizes. Local variance is calculated in each window and arranged along the impedance line direction to form a variance distribution curve. After smoothing and filtering each variance distribution curve, the second derivative is calculated. The position of the positive and negative sign of the second derivative at each scale is identified. The position of the sign change that occurs in multiple scales is determined as the texture segmentation boundary. The impedance line is divided into multiple texture continuous segments by the texture segmentation boundary. The fluctuation amplitude of the local variance in each texture continuous segment is calculated. Segments with fluctuation amplitude exceeding the continuity threshold are marked as surface defect regions, and segments with fluctuation amplitude below the continuity threshold are marked as uniform regions. Extract the impedance sub-curves corresponding to each continuous segment of the texture on the impedance distribution curve, perform linear fitting on each impedance sub-curve and calculate the fitting residual; use the fluctuation amplitude as the surface quality index and the fitting residual as the impedance stability index, extract the influence coefficient of the fluctuation amplitude on the fitting residual through linear regression, and establish the mapping relationship from the surface quality index to the impedance stability index.

3. The method according to claim 1, characterized in that, Based on the aforementioned correlation mapping, the impedance line is divided into multiple processing units and assigned impedance adjustment priorities. For each processing unit, a partitioned impedance adjustment strategy is constructed, including impedance target value, surface quality constraints, and continuity constraints of adjacent unit boundaries, comprising: The influence coefficient is extracted based on the correlation mapping. The rate of change of the influence coefficient is calculated along the impedance line direction. The second derivative of the rate of change curve is calculated. The position where the second derivative turns from negative to positive is identified as the quality-sensitive boundary. The quality-sensitive boundary is used as the processing unit boundary. For each processing unit, the surface quality adjustment cost coefficient is calculated based on the correlation mapping. The root mean square deviation between the impedance distribution curve within the processing unit and the target impedance specification is extracted as the impedance adjustment requirement. This is multiplied by the surface quality adjustment cost coefficient to obtain a processing difficulty score. Based on the processing difficulty score, impedance adjustment priorities are assigned from high to low. A target impedance segment corresponding to the spatial position of the processing unit is extracted from the target impedance specification. The center value of the target impedance segment is calculated as the target impedance value. The maximum allowable change of the surface quality index when the target impedance value is reached is calculated based on the surface quality adjustment cost coefficient as the surface quality constraint. Extract the impedance gradient between the boundary position of the processing unit and the boundary position of the adjacent processing unit, and set the upper limit of the difference between the two impedance gradients as the continuity constraint of the adjacent unit boundary. By using the target impedance value as the optimization objective and the surface quality constraint and the continuity constraint of the adjacent element boundary as the boundary conditions for resistance adjustment, a partitioned resistance adjustment strategy is obtained.

4. The method according to claim 1, characterized in that, Based on the aforementioned partitioned impedance tuning strategy, a laser energy modulation sequence and scanning speed coordination scheme for each processing unit are generated. Processing units are selected sequentially according to tuning priority, and the laser beam is driven to perform impedance tuning. Impedance change trajectory data after processing is collected, including: For each processing unit, the impedance target value and surface quality constraint in the partitioned resistance adjustment strategy are extracted, the impedance distribution curve of the processing unit and the deviation curve of the impedance target value are obtained, gradient analysis is performed on the deviation curve to identify the impedance steep change area and the impedance smooth area, and the resistance adjustment point density is set differently. The upper limit of energy deposition at each trimming point is determined based on surface quality constraints. The amount of energy deposition at each trimming point is determined based on the deviation curve. The energy deposition amount is compared with the upper limit of energy deposition to generate a laser energy modulation sequence. The boundary continuity constraint in the partitioned trimming strategy is transformed into a laser action time constraint at the boundary and the boundary scanning speed is determined. The boundary scanning speed and the energy deposition amount at each trimming point are collaboratively mapped using an interpolation algorithm to generate a scanning speed distribution curve, thus forming a collaborative scanning speed scheme. Select a portion of the processing units according to the impedance adjustment priority, convert the scanning speed coordination scheme into the laser beam motion trajectory, and drive the laser beam to perform impedance adjustment processing on the processing units. During the impedance adjustment processing, the impedance response signal of each impedance adjustment point is collected and spatiotemporally registered with the spatial coordinates to form impedance change trajectory data.

5. The method according to claim 1, characterized in that, The impedance change trajectory data is time-aligned with the laser energy modulation sequence to extract the time delay characteristics and cumulative effect characteristics of the impedance response, and to establish a time-series coupling relationship, including: The impedance response signal of each tuning point in the impedance change trajectory data is extracted, the laser application time of each tuning point in the laser energy modulation sequence is extracted, the impedance response signal is segmented based on the laser application time, the response start position of each segment is identified by rising edge detection, and the time interval from the trigger reference to the response start position is calculated as the time delay characteristic. The time delay characteristics are arranged along the impedance line to form a delay sequence. The extreme value search of the delay sequence is performed to identify the peak and valley positions. The spacing between adjacent peak positions is extracted and correlated with the spatial dimensions of the processing unit. For each impedance adjustment point, the laser energy values ​​of the current point and previous historical points are obtained. The energy values ​​of the historical points are exponentially decayed and weighted according to time sequence and accumulated to obtain the equivalent cumulative energy. The impedance response amplitude of the current point is extracted. The equivalent cumulative energy and the impedance response amplitude are used to construct a bivariate sequence and the autocorrelation coefficient is calculated as the cumulative effect characteristic. The time delay characteristics and cumulative effect characteristics are combined according to the spatial location of the damping point to form a time-series coupling relationship.

6. The method according to claim 1, characterized in that, Based on the aforementioned temporal coupling relationship, the expected impedance response of the unprocessed unit is calculated. The impedance target value and constraint parameters in the partitioned impedance tuning strategy are corrected. The laser energy modulation sequence and scanning speed coordination scheme of the unprocessed unit are updated. This process is iteratively executed until all processing units are completed, including: The temporal coupling relationship is decomposed, and the instantaneous and delayed components of the impedance response are extracted. A two-component superposition model of the impedance response under the action of the laser energy modulation sequence is established, and the model parameters are calibrated based on the impedance change trajectory data of the processed unit. The initial laser energy modulation sequence of the unprocessed unit is input into the dual-component superposition model. The instantaneous influence of the instantaneous component on the impedance and the cumulative influence of the delayed component on the impedance are calculated. The expected impedance response value of each unprocessed unit is synthesized, and it is determined whether the impedance target value and boundary continuity constraint set by the partitioned impedance tuning strategy are met. For unprocessed units that do not meet the conditions, the joint correction vector of the impedance target value and surface quality constraint is calculated. Based on the joint correction vector, the temporal distribution characteristics of the unprocessed unit in the laser energy modulation sequence and the spatial distribution characteristics in the scanning speed coordination scheme are adjusted, and the processing parameters are updated. Process each unit sequentially according to its resistance adjustment priority, update the parameters of the impedance response dual-component superposition model synchronously, and iteratively execute the prediction, correction and update process until all processed units have completed resistance adjustment.

7. The method according to claim 6, characterized in that, The temporal coupling relationship is decomposed, and the instantaneous and delayed components of the impedance response are extracted. A two-component superposition model of the impedance response under the action of the laser energy modulation sequence is established. The model parameters are calibrated based on the impedance change trajectory data of the processed unit, including: Identify the time evolution pattern of impedance response in the time-series coupling relationship, divide the impedance response that occurs synchronously with the current laser energy modulation into an instantaneous component, and divide the impedance response that continues to evolve due to the influence of previous processing into a delayed component. For the instantaneous component, the instantaneous component is directly calculated from the laser energy modulation sequence value at the current moment. For the delayed component, the delayed component is recursively updated based on the delayed component value at the previous moment and the current laser energy modulation sequence value. The two calculation methods are combined in parallel to form a two-component superposition model of impedance response. From the impedance change trajectory data of the processed unit, the component that changes synchronously with the laser energy modulation sequence is extracted as the instantaneous component training set, and the component that lags behind the laser energy modulation sequence is extracted as the delayed component training set. The required mapping coefficients are directly calculated using the instantaneous component training set calibration, and the required recursive coefficients and initial states are recursively updated using the delayed component training set calibration, thus completing the parameter calibration of the impedance response dual-component superposition model.

8. A real-time precision control system for thin-plate impedance lines, used to implement the method as described in any one of claims 1-7, characterized in that, include: The image acquisition unit is used to acquire the surface texture image of the impedance line of the thin plate to be adjusted and the impedance distribution curve along the line; The correlation mapping unit is used to perform multi-scale feature extraction on the surface texture image, identify the surface defect region and uniform region of the impedance line, and establish a correlation mapping between surface quality and impedance stability. The partitioning strategy unit is used to divide the impedance line into multiple processing units according to the association mapping and assign impedance adjustment priority. For each processing unit, a partitioned impedance adjustment strategy including impedance target value, surface quality constraint and adjacent unit boundary continuity constraint is constructed. The processing implementation unit is used to generate a laser energy modulation sequence and scanning speed coordination scheme for each processing unit based on the partitioned impedance tuning strategy, select processing units in sequence according to impedance tuning priority and drive the laser beam to perform impedance tuning processing, and collect impedance change trajectory data after processing. The timing analysis unit is used to perform timing alignment analysis on the impedance change trajectory data and the laser energy modulation sequence, extract the time delay characteristics and cumulative effect characteristics of the impedance response, and establish a timing coupling relationship. The iterative correction unit is used to calculate the expected impedance response value of the unprocessed unit according to the time-series coupling relationship, correct the impedance target value and constraint parameters in the partitioned impedance tuning strategy, update the laser energy modulation sequence and scanning speed coordination scheme of the unprocessed unit, and iteratively execute until all processing units are completed.

9. An electronic device, characterized in that, include: processor; Memory used to store processor-executable instructions; The processor is configured to invoke instructions stored in the memory to execute the method according to any one of claims 1 to 7.

10. A computer-readable storage medium having computer program instructions stored thereon, characterized in that, When the computer program instructions are executed by the processor, they implement the method described in any one of claims 1 to 7.

Images