Copper thickness and line width collaborative uniformity modeling and parameter compensation decision-making method and system

Abstract

The invention discloses a copper thickness and line width collaborative uniformity modeling and parameter compensation decision-making method and system, and belongs to the technical field of printed circuit boards, and the method comprises the following steps: S1, obtaining comprehensive construction parameters of a control unit in a preset time window; s2, obtaining a model prediction value; s3, a temperature processing instruction is obtained, if the temperature processing instruction is to execute cooling processing, cooling processing is executed, otherwise, a parameter compensation decision instruction is obtained, if the parameter compensation decision instruction is a first instruction, layout construction parameter compensation initial adjustment is executed, and if the parameter compensation decision instruction is not a first instruction, layout construction parameter compensation initial adjustment is not executed; s4, an initial adjustment compensation label is obtained, if the initial adjustment compensation label is qualified, parameter compensation decision processing is completed, otherwise, layout construction parameter compensation secondary processing is executed, and the problems that in the prior art, due to the fact that local fluid mass transfer and heat exchange are not uniform due to circuit graph opening ratio differences, cooperative unbalance of copper thickness and line width is caused, and the layout construction parameter compensation quality is poor are solved. And thus, electrical performance fluctuation is caused.

Description

Technical Field

[0001] This invention relates to the field of printed circuit board technology, and in particular to a method and system for modeling and parameter compensation decision-making for the co-uniformity of copper thickness and linewidth. Background Technology

[0002] Existing copper thickness and linewidth co-uniformity modeling and parameter compensation decision-making systems typically control the step-by-step processes of thick copper plates or high-copper-thickness printed circuit boards to achieve overall consistency in copper thickness and linewidth. Specific methods include: first, cutting, drilling, and pre-processing the copper foil or thick copper plate; then, applying a dry film to the board surface for exposure and development to form the desired circuit pattern; subsequently, controlling the thickness of the copper plating area through steps such as pattern electroplating, copper plating, and tin plating; and using AOI scanning to detect the qualification of the circuit pattern during processes such as film removal, alkaline etching, and tin removal, adjusting the electroplating or etching time and development parameters based on the detection results; in high-copper-thickness board processes, a lamination process is also used to ensure the overlap of the circuit patterns on the front and back sides, and differential etching is used to precisely remove non-circuit metal areas, achieving linewidth control.

[0003] For example, the Chinese invention patent with announcement number CN105163502B discloses a method for controlling the etching width and spacing of fine lines on thick copper plates, which includes: cutting, drilling, and immersion copper processing on thick copper plates; applying a dry film without covering the holes to the thick copper plates after immersion copper processing; exposing and developing the plates; after the developed thick copper plates pass AOI scanning, performing pattern electroplating with copper and tin plating; and then sequentially performing the first film removal, alkaline etching, and tin removal processes; after the thick copper plates that have undergone the first etching pass AOI scanning, applying a dry film without covering the holes; exposing and developing the plates; after passing AOI scanning, performing pattern electroplating with copper and tin plating; and then sequentially performing the second film removal, alkaline etching, and tin removal processes to obtain a circuit board, which then proceeds to the next process.

[0004] For example, Chinese invention patent CN105960102B discloses a method for preparing a high-copper-thickness printed circuit board, including: (1) selecting a conductive metal foil and cutting it, and drilling positioning holes on the metal foil; (2) after photoluminescence pretreatment, constructing copper circuit patterns on the reverse side of the metal foil; (3) pressing the metal foil, prepreg, and polymer substrate together from top to bottom, wherein the reverse side of the metal foil faces the prepreg; (4) constructing copper circuit patterns on the front side of the metal foil, and the copper circuit patterns on the front and back sides of the metal foil are required to completely overlap; (5) completely etching the non-circuit area of ​​the metal foil through differential etching to obtain fine circuits; (6) performing subsequent processing on the fine circuits to obtain a printed circuit board.

[0005] The above-mentioned technology has at least the following technical problems: In the existing technology, during the manufacturing process of automotive power circuit boards (especially thick copper circuit boards), in order to meet the requirements of high current carrying capacity and heat dissipation, there are often differences in the trace density in different areas of the circuit pattern: for example, the power trace area has dense wiring, while the signal control area has sparse wiring. Although electroplating, current control, and solution ratio have been optimized, at the microscale, different pattern open ratios lead to inconsistent local mass transfer and heat exchange conditions. Therefore, there is a problem of uneven local fluid mass transfer and heat exchange caused by the difference in circuit pattern open ratios, resulting in a spatially non-uniform boundary layer thickness and temperature distribution during the bonding process. This causes deviations in the copper deposition rate in different areas, resulting in a synergistic imbalance between copper thickness and linewidth, and consequently, fluctuations in electrical performance. Summary of the Invention

[0006] To address the problem in existing technologies where uneven local mass transfer and heat exchange due to differences in the open area ratio of circuit patterns leads to spatially non-uniform boundary layer thickness and temperature distribution during the bonding process, causing deviations in copper deposition rates in different regions and resulting in a coordinated imbalance between copper thickness and linewidth, ultimately leading to fluctuations in electrical performance, this invention provides a method and system for modeling and parameter compensation decision-making regarding the coordinated uniformity of copper thickness and linewidth. The technical solution is as follows: On the one hand, a method for modeling and parameter compensation decision-making for the coordinated uniformity of copper thickness and linewidth is provided. This method includes: S1, acquiring the circuit board production design drawing and dividing it into regions to obtain each control unit, thereby obtaining the comprehensive construction parameters of the control unit under a preset time window. The comprehensive construction parameters include layout construction parameters and actual construction measurement parameters. The layout construction parameters include the electroplating time, etching time, exposure energy, and development time of the layout construction. The actual construction measurement parameters include the open ratio, surface tangential velocity, surface temperature, and electroplating set current density of the control unit; S2, constructing a copper thickness and linewidth coordinated uniformity processing model based on the comprehensive construction parameters, and outputting the model prediction values, which include predicted copper thickness and predicted linewidth. S3. Obtain the surface temperature of the circuit board and analyze the temperature processing command. If the temperature processing command is to perform cooling processing, then perform cooling processing. Otherwise, based on the model prediction value analysis, obtain the parameter compensation decision command. If the parameter compensation decision command is the first command, then perform the initial adjustment of layout construction parameter compensation. Otherwise, do not perform the initial adjustment of layout construction parameter compensation. The initial adjustment of layout construction parameter compensation includes copper thickness correction compensation and / or line width correction compensation. S4. Update the comprehensive construction parameters after obtaining the initial adjustment of layout construction parameter compensation, analyze the initial adjustment prediction value of copper thickness and the initial adjustment prediction value of line width, and obtain the initial adjustment compensation label. If the initial adjustment compensation label is qualified, then the parameter compensation decision processing is completed. Otherwise, perform the secondary processing of layout construction parameter compensation.

[0007] On the other hand, a copper thickness and linewidth collaborative uniformity modeling and parameter compensation decision-making system is provided. This system includes: a control unit partitioning module, a model building module, a construction parameter compensation initial adjustment module, and a compensation judgment module. The control unit partitioning module is used to acquire the circuit board production design drawing and divide it into regions to obtain each control unit, thereby obtaining the comprehensive construction parameters of the control unit under a preset time window. The model building module is used to construct a copper thickness and linewidth collaborative uniformity processing model based on the comprehensive construction parameters and output the model's predicted values. The construction parameter compensation initial adjustment module is used to acquire the circuit board surface temperature, partition... The system analyzes and obtains temperature processing instructions. If the temperature processing instruction is to perform cooling processing, then cooling processing is performed. Otherwise, based on the model prediction value analysis, a parameter compensation decision instruction is obtained. If the parameter compensation decision instruction is the first instruction, then initial adjustment of layout construction parameter compensation is performed; otherwise, initial adjustment of layout construction parameter compensation is not performed. The compensation determination module is used to update the comprehensive construction parameters after obtaining the initial adjustment of layout construction parameter compensation, analyze and obtain the initial adjustment prediction values ​​of copper thickness and line width, and obtain the initial adjustment compensation label. If the initial adjustment compensation label is qualified, the parameter compensation decision processing is completed; otherwise, secondary processing of layout construction parameter compensation is performed.

[0008] The beneficial effects of the technical solutions provided in the embodiments of the present invention include at least the following: 1. The copper thickness and linewidth co-uniformity modeling and parameter compensation decision-making method provided by this invention obtains the circuit board production design drawing and divides it into control units. It obtains the layout construction parameters and actual construction measurement parameters of each control unit, and constructs a copper thickness and linewidth co-uniformity processing model based on the comprehensive construction parameters. It outputs the predicted values ​​of copper thickness and linewidth, thereby accurately predicting the global copper thickness and linewidth. This enables the coordinated control of electroplating, etching, exposure and development processes. It effectively solves the problem in the prior art where uneven local fluid mass transfer and heat exchange caused by differences in the open ratio of circuit patterns leads to the formation of spatially non-uniform boundary layer thickness and temperature distribution during the bonding process, resulting in a co-imbalance between copper thickness and linewidth, and thus causing fluctuations in electrical performance.

[0009] 2. This invention analyzes the temperature range and average temperature of the surface temperature at the sampling points of the control unit, and obtains the temperature influence value by combining the weighting factor. Then, it obtains the temperature adjustment constraint factor by matching the temperature influence difference value with the database. Based on the maximum temperature processing difference and the temperature adjustment constraint factor, it obtains the comprehensive execution value of the cooling temperature, thereby performing precise cooling processing on the control unit, optimizing the local temperature distribution, stabilizing the local copper ion deposition rate and etching and development reaction rate, and solving the problem of copper thickness and linewidth deviation caused by temperature fluctuations in the prior art.

[0010] 3. This invention uses a preliminary adjustment and secondary layout construction parameter compensation method. First, based on the difference in predicted copper thickness and linewidth, a first adjustment coefficient is obtained to make preliminary adjustments to the electroplating time and etching time. Then, based on the difference in the preliminary adjustment prediction, a second adjustment coefficient is obtained to make secondary adjustments to the development time and exposure energy. This dynamically updates the layout construction parameters and combines them with the measured construction parameters to re-predict the copper thickness and linewidth, thereby achieving synergistic consistency between copper thickness and linewidth. This effectively avoids the problem of insufficient copper thickness or linewidth deviation caused by fixed parameters or delayed adjustments in the prior art.

[0011] 4. This invention spatially matches and marks the cooling inlets corresponding to high temperature, average temperature, and ordinary sampling points. It combines the comprehensive execution value of cooling temperature with the local cooling adjustment coefficient for linear transition processing. Different local cooling controls are applied to high temperature cooling inlets, low temperature cooling inlets, ordinary cooling inlets, and additional cooling inlets, thereby achieving local additional cooling and optimized resource allocation. This reduces coolant consumption and ensures precise temperature control, effectively solving the problems of uneven cooling and energy waste in the prior art. Attached Figure Description

[0012] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0013] Figure 1 A flowchart of the copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method provided in the embodiments of this application; Figure 2 A flowchart illustrating the macroscopic steps of the copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method provided in the embodiments of this application; Figure 3 A flowchart of the control unit cooling adjustment process for the copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of the copper thickness and linewidth collaborative uniformity modeling and parameter compensation decision system provided in the embodiments of this application. Detailed Implementation

[0014] The technical solution of the present invention will now be described with reference to the accompanying drawings.

[0015] In embodiments of the present invention, words such as "exemplarily," "for example," etc., are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "exemplary" in the present invention should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the word "exemplary" is intended to present the concept in a concrete manner. Furthermore, in embodiments of the present invention, the meaning expressed by "and / or" can be both, or either one.

[0016] In the embodiments of this invention, the terms "image" and "picture" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning. Similarly, the terms "of," "corresponding (relevant)," and "corresponding" may sometimes be used interchangeably. It should be noted that, without emphasizing the distinction between them, they convey the same meaning.

[0017] In this embodiment of the invention, sometimes a subscript such as W1 may be written in a non-subscript form such as W1. When the difference is not emphasized, the meaning they express is the same.

[0018] To make the technical problems, technical solutions and advantages of the present invention clearer, a detailed description will be given below in conjunction with the accompanying drawings and specific embodiments.

[0019] like Figure 1 As shown, Figure 1The flowchart below illustrates a method for modeling and parameter compensation decision-making for the coordinated uniformity of copper thickness and linewidth provided in this application. The method includes the following steps: S1. Obtaining a circuit board production design drawing and dividing it into regions to obtain each control unit. From this, comprehensive construction parameters for each control unit are obtained within a preset time window. These comprehensive construction parameters include layout construction parameters and measured construction parameters. The layout construction parameters include electroplating time, etching time, exposure energy, and development time. The measured construction parameters include the open ratio, surface tangential velocity, surface temperature, and electroplating set current density of the control unit. S2. Constructing a copper thickness and linewidth coordinated uniformity processing model based on the comprehensive construction parameters and outputting the model's predicted values. These predicted values ​​include the predicted copper thickness and linewidth. S3. Obtain the circuit board surface temperature and analyze the temperature processing command. If the temperature processing command is to execute cooling processing, then execute cooling processing; otherwise, based on the model prediction value analysis, obtain the parameter compensation decision command. If the parameter compensation decision command is the first command, then execute the initial adjustment of layout construction parameter compensation; otherwise, do not execute the initial adjustment of layout construction parameter compensation. The initial adjustment of layout construction parameter compensation includes copper thickness correction compensation and / or line width correction compensation; S4. Update the comprehensive construction parameters after obtaining the initial adjustment of layout construction parameter compensation, analyze the initial adjustment prediction value of copper thickness and the initial adjustment prediction value of line width, and obtain the initial adjustment compensation label. If the initial adjustment compensation label is qualified, then the parameter compensation decision processing is completed; otherwise, the secondary processing of layout construction parameter compensation is executed.

[0020] In this embodiment, as Figure 2 As shown, Figure 2 This document presents a flowchart outlining the macroscopic steps of the copper thickness and linewidth collaborative uniformity modeling and parameter compensation decision-making method provided in this application embodiment. The process involves obtaining a circuit board production design drawing, dividing it into regions to obtain control units, and then acquiring comprehensive construction parameters for each control unit within a preset time window, including layout construction parameters and measured construction parameters. Based on these comprehensive construction parameters, a copper thickness and linewidth collaborative uniformity processing model is constructed to obtain predicted copper thickness and linewidth values, which are used to analyze whether temperature processing is required. If the temperature processing instruction is to perform cooling processing, the control unit is further adjusted for cooling; otherwise, a parameter compensation decision instruction is directly generated, and a decision is made based on the instruction to determine whether initial adjustment of layout construction parameters should be performed. After initial adjustment, the initial predicted values ​​for copper thickness and linewidth are obtained, and an initial adjustment compensation label is generated based on the difference compared to the allowable error value. If the label is qualified, the parameter compensation decision-making process is completed; otherwise, it enters the secondary processing stage of layout construction parameter compensation, where predictions are re-performed and judgments are made based on the updated construction parameters. If the secondary processing result is qualified, the processing is completed; otherwise, a warning is issued, and the process ends.

[0021] Furthermore, a copper thickness and linewidth co-uniformity processing model is constructed. Specifically, the following methods are used: Normalization is performed based on measured construction parameters to obtain normalized values ​​for the open ratio, surface tangential velocity, and surface temperature; a copper thickness prediction sub-model is constructed based on the electroplating set current density, normalized open ratio, normalized surface tangential velocity, normalized surface temperature, and electroplating time. During training, a mass transfer constraint equation is introduced as a regularization constraint to reflect the physical dependence of copper ion transport rate on deposition thickness; based on the predicted copper thickness, electroplating set current density, and exposure... A linewidth prediction sub-model was constructed using normalized values ​​for energy, open ratio, surface tangential velocity, surface temperature, etching time, exposure energy, and development time. During training, an Arrhenius temperature dependency term was introduced as a regularization constraint to reflect the influence of temperature changes on the development rate and etching reaction rate. Joint optimization was performed by minimizing the square error between the predicted copper thickness and the target copper thickness, as well as minimizing the square error between the predicted linewidth and the target linewidth. The predicted copper thickness and linewidth were then output, thus completing the construction of the copper thickness and linewidth co-uniform processing model.

[0022] In this embodiment, a copper thickness prediction sub-model is constructed. Specifically, based on measured construction parameters, partial normalization is performed to obtain normalized values ​​for the open ratio, surface tangential velocity, and surface temperature. Using the electroplating current density, normalized open ratio, normalized surface tangential velocity, normalized surface temperature, and electroplating time as input variables, a copper thickness prediction sub-model is established. During model training, a mass transfer constraint equation is introduced as a regularization constraint. The physical dependence between copper ion transport rate and deposition thickness is embedded as an additional loss term in the model optimization objective. Model deviation is suppressed by penalizing predictions that do not meet the mass transfer constraint, ensuring both fitting accuracy and physical rationality. Through iterative training, the squared error between the predicted copper thickness and the target copper thickness is minimized, resulting in a copper thickness prediction model parameter set and a stable predicted copper thickness value.

[0023] A sub-model for linewidth prediction was constructed using the following methods: The predicted copper thickness, electroplating current density, exposure energy, normalized open ratio, normalized surface tangential velocity, normalized surface temperature, etching time, and development time were used as input variables to establish the sub-model. During model training, an Arrhenius temperature dependency term (referring to the quantitative relationship between the chemical reaction rate constant and temperature) was introduced as a regularization constraint. This mapped the influence of temperature changes on the development rate and etching reaction rate into a temperature-dependent function, enabling the model to reflect the physical impact of temperature changes on linewidth. By adding a temperature dependency term constraint to the loss function, the model's response consistency under high and low temperature conditions was achieved. The optimization objective was to minimize the squared error between the predicted linewidth and the target linewidth, resulting in a parameter set for the linewidth prediction model and stable predicted linewidth values.

[0024] The copper thickness prediction sub-model takes into account key parameters measured during construction, such as the open ratio, surface tangential velocity, and surface temperature, as well as process setting parameters such as electroplating current density and electroplating time. By establishing a nonlinear mapping relationship between the electroplated copper thickness and the input parameters, the model outputs the predicted copper thickness. During training, a mass transfer constraint equation is introduced as a regularization constraint. This constraint is based on the physical laws governing the transport of copper ions in the electrolyte and their deposition on the electrode surface; that is, there is a dependency between the diffusion rate of copper ions and the local deposition thickness. By transforming this relationship into an additional loss term embedded in the model's optimization objective, the model, during training, not only needs to fit the copper thickness distribution of historical data but also needs to satisfy the mass transfer constraint to prevent physically unrealizable abnormal thicknesses in the predicted results, thus achieving a balance between data-driven and physics-driven approaches. This is achieved through iterative optimization. In each training step, the standard mean square error (MSE) loss and the penalty loss for the mass transfer regularization constraint are calculated. The model parameters are updated using gradient descent (such as the Adam optimizer), ultimately yielding a set of copper thickness predictions and corresponding model parameters that have high fitting accuracy and are physically reasonable under historical conditions.

[0025] The linewidth prediction sub-model is the next level below the copper thickness prediction sub-model. Its inputs include the predicted copper thickness, process settings (such as current density, exposure energy, etching time, and development time), and normalized parameters from actual construction measurements (open ratio, surface tangential velocity, and surface temperature). The core objective of this model is to predict the linewidth variation of fine lines after the process and to reflect the influence of temperature and chemical reactions on the development and etching rates. During training, the model introduces an Arrhenius temperature dependency term as a regularization constraint. This constraint originates from the kinetics of chemical reactions, where the reaction rate constant changes exponentially with temperature. By embedding the temperature dependency function into the loss function, the model maintains consistent response under different temperature conditions and reflects the impact of temperature fluctuations on etching depth and linewidth contraction or expansion, thus ensuring the interpretability and stability of the linewidth prediction under actual operating conditions. During training, by minimizing the squared error between the predicted and target linewidth values, and simultaneously adding a penalty term for the temperature dependency constraint, the model parameters are gradually adjusted through gradient optimization until the loss function converges, resulting in stable linewidth prediction values ​​and generalizable model parameters.

[0026] The overall construction of the copper thickness and linewidth co-uniform processing model is a joint optimization based on the two sub-models mentioned above. Its inputs include measured construction parameters, process setting parameters, and process time series information; the outputs are predicted copper thickness and predicted linewidth. The model training process achieves co-optimization by simultaneously minimizing the prediction errors of both copper thickness and linewidth: the output of the copper thickness sub-model serves as the input of the linewidth sub-model, forming a multi-stage hierarchical relationship, so that the prediction accuracy of copper thickness directly affects the prediction accuracy of linewidth. In the training objective, the loss function includes the sum of squared errors of copper thickness and linewidth, as well as physical constraint regularization terms (including mass transfer limitation terms and temperature dependence terms), ensuring that the model maintains the constraints of physical laws while satisfying data fitting accuracy, thus making the model interpretable. During the optimization process, batch training and iterative updates are typically used, while regularization terms are added to prevent overfitting, ensuring that the model maintains reasonable predictions even under unseen operating conditions.

[0027] Coordinated uniform control of copper thickness and linewidth was achieved through joint optimization. Input data primarily includes plating current density, normalized open ratio, normalized surface tangential velocity, normalized surface temperature, exposure energy, etching time, and development time. Outputs are predicted copper thickness and linewidth values. The training process utilizes joint optimization by minimizing prediction error and introducing physical regularization constraints. Model parameters are updated iteratively with gradients to obtain a stable set, ultimately forming a coordinated uniform processing model suitable for practical process control.

[0028] By constructing a copper thickness and linewidth co-uniform processing model, multi-dimensional coupled analysis and refined prediction of key process parameters in circuit board manufacturing were achieved. Measured parameters such as electroplating current density, open area ratio, surface tangential velocity, surface temperature, and electroplating time were integrated with layout parameters to create a fusion model. Separate sub-models for copper thickness and linewidth prediction were constructed, and physical constraint terms were introduced to establish an interpretable correlation between data-driven and physical mechanisms. By introducing a mass transfer constraint equation as a regularization constraint into the copper thickness prediction sub-model, the model can accurately reflect the transport characteristics of copper ions between the solution and the board surface, as well as the dynamic changes in deposition thickness, effectively improving the physical consistency and computational accuracy of copper thickness prediction. Introducing an Arrhenius temperature dependence term as a regularization constraint into the linewidth prediction sub-model allows the model to capture the influence of temperature on the development rate and etching reaction rate, thus more accurately reflecting the correspondence between linewidth shrinkage and thermal reaction kinetics. By jointly optimizing the two sub-models, the correlation between copper thickness and linewidth can be constrained synchronously, which not only improves the model's ability to respond to changes in multiple process parameters, but also avoids parameter drift caused by single-objective optimization.

[0029] Furthermore, the temperature processing command is obtained through the following method: sampling points are set up on the circuit board control unit to obtain the surface temperature of each sampling point. Based on the surface temperature of each sampling point, temperature range processing is performed (the maximum surface temperature is subtracted from the minimum surface temperature to obtain the temperature sampling range value). The average sampling temperature is obtained by performing temperature mean processing based on the surface temperature of each sampling point. The temperature influence value is obtained by coupling the temperature sampling range value and the average sampling temperature with corresponding weighting factors. The preset temperature influence threshold in the database is obtained and compared with the temperature influence value to obtain the temperature processing command. If the temperature influence value is above the temperature influence threshold, the temperature processing command is to perform cooling processing; otherwise, the temperature processing command is not to perform cooling processing.

[0030] In this embodiment, the temperature influence value is obtained by the following method: ; In the formula, WY represents the temperature influence value, WJ represents the temperature sampling range value, WP represents the average sampling temperature, δ1 represents the weighting factor of the temperature sampling range value, and δ2 represents the weighting factor of the average sampling temperature.

[0031] By performing dual-parameter analysis (range and mean) on the surface temperature sampling data of the circuit board control unit, and introducing weighting factors to form a comprehensive temperature influence value, a quantitative assessment of local overheating risk is achieved. By comparing the temperature influence value with a preset temperature influence threshold, uneven temperature distribution areas can be dynamically identified during the electroplating process, and cooling instructions can be adaptively generated.

[0032] The weighting factors for temperature sampling range and average sampling temperature can be obtained from the database. For example, the weighting factor for temperature sampling range can be obtained by retrieving historical temperature sampling range values ​​stored in the database and comparing them with the current temperature sampling range value. The historical temperature sampling range value with the smaller deviation from the current temperature sampling range value is selected (the deviation value of the current temperature sampling range value is subtracted from the historical temperature sampling range value to obtain the historical deviation value. This historical deviation value is compared with the preset historical deviation value threshold in the database. If there is a historical deviation value that is less than the deviation value threshold, it is used as the historical reference temperature sampling range value). The weighting factor for each historical reference temperature sampling range value is obtained, and the median value of the historical temperature sampling range weighting factor is selected as the temperature sampling range weighting factor. The method for obtaining the average sampling temperature weighting factor is the same as that for the temperature sampling range weighting factor. Both can be obtained by retrieving the corresponding historical weighting factor and then taking the median value.

[0033] Further, a cooling process is performed, specifically as follows: A difference analysis is conducted between the temperature impact value and the temperature impact threshold to obtain a temperature impact difference value (the temperature impact value and the temperature impact threshold are processed to obtain a temperature impact difference value, which is then divided by the temperature impact threshold to obtain the temperature impact difference value); the temperature impact difference value is matched with the database to obtain a temperature adjustment constraint factor; the maximum surface temperature of each sampling point is obtained, and an absolute difference is processed with the preset temperature threshold in the database to obtain the maximum temperature processing difference value (wherein, if the maximum surface temperature of the sampling point is above the temperature threshold, the maximum surface temperature of the sampling point is subtracted from the temperature threshold to obtain the maximum temperature processing difference value; if the maximum temperature processing difference value is less than the temperature threshold, the preset reference temperature processing difference value in the database is used as the maximum temperature processing difference value); the maximum temperature processing difference value and the temperature adjustment constraint factor are linearly coupled to obtain a comprehensive cooling temperature execution value (the maximum temperature processing difference value is added to the product of the maximum temperature processing difference value and the temperature adjustment constraint factor to obtain the comprehensive cooling temperature execution value); the control unit is executed to adjust the cooling based on the comprehensive cooling temperature execution value.

[0034] In this embodiment, a temperature adjustment constraint factor is obtained by matching the temperature influence difference value with the database. The specific method is as follows: obtain each historical temperature influence constraint data group stored in the database, wherein each historical temperature influence constraint data group includes a set of data (historical temperature influence difference value and the historical temperature adjustment constraint factor corresponding to the historical temperature influence difference value). Compare the temperature influence difference value with the historical temperature influence difference value in each historical temperature influence constraint data group, select the historical temperature influence difference value that is closest to the temperature influence difference value as the historical control temperature influence difference value, and then use the historical temperature adjustment constraint factor corresponding to the historical control temperature influence difference value as the temperature adjustment constraint factor.

[0035] By quantitatively analyzing the difference between the temperature impact value and the temperature impact threshold, the degree of temperature impact difference is calculated. Furthermore, by combining the preset temperature adjustment constraint factors in the database to correct and linearly couple the maximum difference in temperature processing, a comprehensive cooling temperature execution value is obtained. This value can accurately quantify the cooling strategy based on a comprehensive consideration of the temperature fluctuation amplitude and physical constraints of each sampling point in the control unit, thereby avoiding the excessive influence of a single temperature point or extreme value on the overall adjustment strategy. At the same time, the comprehensive execution value achieves overall coordinated cooling of each cooling inlet, ensuring the balance and continuity of temperature regulation. This satisfies the need for rapid cooling of local hot spots while reducing unnecessary energy consumption and cooling resource waste, achieving high efficiency and economy in cooling control.

[0036] Furthermore, the cooling control unit adjusts the temperature based on the overall cooling temperature execution value. Specifically, it obtains a preset cooling temperature threshold value from the database and compares it with the overall cooling temperature execution value to obtain a cooling adjustment execution plan. If the overall cooling temperature execution value is less than the cooling temperature threshold value, the cooling adjustment execution plan is the first execution plan; otherwise, it is the second execution plan. The first execution plan specifically involves: adjusting the coolant temperature of each cooling inlet within the time window based on the overall cooling temperature execution value (specifically, subtracting the overall cooling temperature execution value from the current coolant temperature of each cooling inlet to obtain the adjusted coolant temperature). This completes the cooling adjustment of the control unit; the second execution scheme is as follows: based on the surface temperature of each sampling point, the sampling point corresponding to the maximum surface temperature is selected and marked as the high-temperature sampling point, the sampling point corresponding to the average surface temperature is selected and marked as the average temperature sampling point, and the remaining sampling points are marked as ordinary sampling points; the position of the center point of each cooling inlet is obtained, and the shortest distance analysis is performed between the center point and the high-temperature sampling point and the average temperature sampling point respectively (that is, by connecting the high-temperature sampling point and the average temperature sampling point to the center point of each cooling inlet to obtain the straight-line distance, and then selecting the minimum value of the straight-line distance to obtain the shortest distance, the shortest distance includes the distance to the high-temperature sampling point). The minimum distance between the sample point and the average temperature sampling point is used to identify the cooling inlet. The cooling inlet corresponding to the center point of the cooling inlet closest to the high-temperature sampling point is marked as the high-temperature cooling inlet, and the cooling inlet corresponding to the center point of the cooling inlet closest to the average temperature sampling point is marked as the low-temperature cooling inlet. The remaining cooling inlets are marked as ordinary cooling inlets. A difference analysis is performed between the overall cooling temperature execution value and the cooling temperature cutoff value to obtain the temperature adjustment difference value (the cooling temperature difference is obtained by subtracting the overall cooling temperature execution value from the cooling temperature cutoff value, and then divided by the cooling temperature cutoff value). The temperature adjustment difference value is then matched with the database. The local cooling adjustment coefficient is obtained. Based on the local cooling adjustment coefficient and the comprehensive execution value of the cooling temperature, a multiplicative coupling process is performed (that is, the comprehensive execution value of the cooling temperature is added to the product of the comprehensive execution value of the cooling temperature and the local cooling adjustment coefficient to obtain the local cooling adjustment execution temperature). The local cooling adjustment execution temperature is obtained. The high-temperature cooling outlet corresponding to the high-temperature cooling inlet and the low-temperature cooling outlet corresponding to the low-temperature cooling inlet are obtained. If the high-temperature cooling outlet and the low-temperature cooling outlet are the same outlet, then the cooling path of the high-temperature cooling inlet and the low-temperature cooling inlet are the same cooling path. Then, each ordinary cooling inlet in the cooling path between the high-temperature cooling inlet and the low-temperature cooling inlet is marked as the cooling inlet of each path.The local cooling adjustment execution temperature is used as the execution temperature of the high-temperature cooling inlet, and the comprehensive cooling temperature execution value is used as the execution temperature of the low-temperature cooling inlet and the normal cooling inlet. A linear transition process is performed based on the execution temperatures of the high-temperature and low-temperature cooling inlets to obtain the execution temperatures of each cooling inlet, thus completing the control unit cooling adjustment. If the high-temperature cooling outlet and the low-temperature cooling outlet are not the same outlet, sampling points with surface temperatures above the average surface temperature are selected and marked as additional cooling sampling points. The cooling inlets closest to each additional cooling sampling point are obtained (for each additional cooling sampling point, the expected nearest cooling inlet is found; if multiple additional cooling sampling points are closest to the same cooling inlet, this cooling inlet is used as the nearest cooling inlet corresponding to these multiple additional cooling sampling points; therefore, there may be a one-to-one or many-to-one correspondence between additional cooling sampling points and cooling inlets), and marked as additional cooling inlets; otherwise, they are marked as normal cooling inlets. The local cooling adjustment execution temperature is used as the execution temperature of each additional cooling cooling inlet, and the comprehensive cooling temperature execution value is used as the execution temperature of the normal cooling inlet, thus completing the control unit cooling adjustment.

[0037] In this embodiment, as Figure 3 As shown, Figure 3 The flowchart for the control unit cooling adjustment of the copper thickness and linewidth collaborative uniformity modeling and parameter compensation decision-making method provided in this application embodiment first obtains the cooling temperature segmentation threshold from the database and compares the execution value with the segmentation threshold. If the overall execution value is lower than the threshold, the first scheme is executed, which lowers the overall coolant temperature of each cooling inlet within that time window. If the overall execution value is higher than or equal to the threshold, the second scheme is executed: First, identify the high-temperature sampling point, average temperature sampling point, and ordinary sampling point of the surface temperature, and obtain the center position of the cooling inlet. Perform shortest distance analysis on each inlet and sampling point, mark the inlet closest to the high-temperature sampling point as the high-temperature cooling inlet, the inlet closest to the average temperature sampling point as the low-temperature cooling inlet, and the remaining inlets as ordinary cooling inlets; then, obtain the local cooling adjustment coefficient based on the temperature adjustment difference value and the database, calculate the local cooling adjustment execution temperature, and further classify and process it according to whether the high-temperature and low-temperature cooling outlets are the same outlet: if they are the same outlet, mark the ordinary cooling inlet in the path as the path cooling inlet, and perform linear transition processing of the high-temperature and low-temperature execution temperatures on the path cooling inlet execution temperature; if they are not the same outlet, identify the sampling point with the surface temperature higher than the average value as the additional cooling sampling point, and mark its nearest cooling inlet as the additional cooling inlet. Assign the local cooling adjustment temperature to the additional cooling inlet, and assign the overall execution value of the cooling temperature to the remaining ordinary inlets. Finally, output the execution temperature of each cooling inlet to achieve fine cooling adjustment closed-loop control of the control unit.

[0038] The local cooling adjustment coefficient is obtained by matching the temperature adjustment difference values ​​with the database. Specifically, the historical temperature adjustment difference values ​​stored in the database are retrieved and compared with the temperature adjustment difference values. If a historical temperature adjustment difference value matches the temperature adjustment difference value, the corresponding historical local cooling adjustment coefficient is used as the local cooling adjustment coefficient. Otherwise, the historical temperature adjustment difference value that is closest to and greater than the temperature adjustment difference value is selected and marked as the first temperature adjustment difference value. The historical temperature adjustment difference value that is closest to and less than the temperature adjustment difference value is selected and marked as the second temperature adjustment difference value. The historical first local cooling adjustment coefficient corresponding to the first temperature adjustment difference value and the historical second local cooling adjustment coefficient corresponding to the second temperature adjustment difference value are obtained. The historical first and second local cooling adjustment coefficients are then averaged to obtain the average historical local cooling adjustment coefficient, which is used as the local cooling adjustment coefficient.

[0039] Linear transition processing is performed based on the execution temperatures of the high-temperature cooling inlet and the low-temperature cooling inlet to obtain the execution temperatures of each path cooling inlet. Specifically, the execution temperature is obtained by subtracting the execution temperature of the low-temperature cooling inlet from the execution temperature of the high-temperature cooling inlet and then dividing by the number of cooling inlets between the high-temperature and low-temperature cooling inlets. This yields the execution temperature change, which is the execution temperature change value corresponding to each cooling inlet. Based on the positional relationship of the high-temperature cooling inlet, the path cooling inlets, and the low-temperature cooling inlet, the execution temperature is gradually increased based on the execution temperature change, i.e., the execution temperature of the high-temperature cooling inlet is the highest, and the execution temperature of the low-temperature cooling inlet is the highest. Then, the temperature adjustment step size is the execution temperature change. For example, the execution temperature value of the high-temperature cooling inlet is Ta, and the execution temperatures of each path cooling inlet are Ta+x, Ta+2x, ..., Ta+nx, where n represents the number of path cooling inlets. The execution temperature value of the low-temperature cooling inlet is Ta+(n+1)x.

[0040] By dividing the cooling adjustment of the control unit into two execution schemes, namely the first execution scheme (when the overall cooling temperature execution value is lower than the cooling temperature threshold) and the second execution scheme (when the overall cooling temperature execution value is higher than the cooling temperature threshold), a dynamic balance between temperature regulation accuracy and energy consumption efficiency under different degrees of thermal unevenness is achieved. This allows the system to adopt a matching cooling strategy when faced with different degrees of temperature deviation, avoiding energy waste or over-cooling problems caused by uniform cooling.

[0041] In the first execution scheme, the overall system temperature deviation is small. Global fine-tuning can be achieved by uniformly reducing the coolant temperature of each cooling inlet based on the comprehensive execution value of the cooling temperature. This effectively reduces the load of control calculations and coolant regulation, and achieves global temperature balance with minimal energy consumption.

[0042] To avoid the accumulation of subtle temperature gradients between regions, a linear transition regulation mechanism was designed for the cooling inlet. By implementing a linear transition temperature distribution at the ordinary cooling inlet between the high-temperature and low-temperature cooling inlets, the coolant temperature transitions gradually along the flow path. This smooths the thermal gradient between regions, prevents localized sudden cooling, reduces coolant flow fluctuations, and improves system thermal stability and cooling path utilization.

[0043] In the second implementation scheme, the system detects significant temperature differences or the presence of substantial localized high-temperature regions, necessitating localized cooling with higher spatial resolution. By analyzing the surface temperature distribution based on sampling points, cooling inlets at different locations are categorized into three types: high-temperature cooling inlets, low-temperature cooling inlets, and ordinary cooling inlets, achieving differentiated response and precise thermal compensation for heat distribution. This effectively suppresses overheating accumulation in high-temperature regions, shortens cooling response time, and ensures that low-temperature regions are not affected by excessive cooling, thereby achieving rapid equilibrium of the overall temperature field and thermal consistency during the copper thickness and linewidth formation process. Furthermore, to address complex cooling path scenarios where high-temperature and low-temperature cooling outlets are not the same, the scheme further introduces an additional cooling sampling point and additional cooling inlet calibration mechanism. In this case, the system selects sampling points with surface temperatures higher than the average value and identifies the nearest cooling inlets to these sampling points, marking them as additional cooling inlets (the correspondence can be one-to-one or many-to-one, meaning one cooling inlet can correspond to multiple additional cooling sampling points). This ensures that localized high-temperature points under different distribution patterns can receive independent or shared cooling path support, effectively improving the spatial adaptability and response efficiency of cooling resource allocation. The classification of high-temperature cooling inlets, low-temperature cooling inlets, and ordinary cooling inlets in the second execution scheme not only improves the targeting of cooling but also enables graded control of fluid temperature in the case of shared paths. This prevents the reverse accumulation of heat in the coolant during path conduction, ensuring maximum utilization of cooling energy. The introduction of an additional cooling inlet mechanism further compensates for localized cooling dead zones caused by non-unique cooling outlets or asymmetrical paths, improving the cooling coverage and energy utilization of the entire control unit. In summary, this invention, by introducing a graded adjustment strategy based on the comprehensive execution value of cooling temperature, combined with linear temperature transition at the path cooling inlets and a dynamic calibration mechanism for additional cooling inlets, achieves temperature adaptive control that balances global thermal field equilibrium with localized refined cooling. This effectively reduces cooling resource consumption and improves system operational stability and the film quality of copper thickness and linewidth uniformity.

[0044] Furthermore, the parameter compensation decision instruction is obtained, specifically through the following method: Based on the time window number, it is matched with the database to obtain the target parameter execution value under the current time window, where the target parameter execution value includes the target copper thickness and the target linewidth; a difference processing is performed between the target parameter execution value and the model prediction value, specifically by subtracting the target copper thickness from the predicted copper thickness to obtain the copper thickness prediction difference; a difference processing is also performed between the predicted linewidth and the target linewidth to obtain the linewidth prediction difference; and a preset allowable error value is obtained from the database, where the allowable error value includes the copper thickness allowable error. The difference and linewidth tolerance values ​​are compared. Based on the absolute value of the copper thickness prediction difference and the copper thickness tolerance value, and based on the absolute value of the linewidth prediction difference and the linewidth tolerance value, a parameter compensation decision instruction is obtained. If the absolute value of the copper thickness prediction difference is greater than the copper thickness tolerance value, copper thickness correction compensation is executed; otherwise, copper thickness correction compensation is not executed. If the absolute value of the linewidth prediction difference is greater than the linewidth tolerance value, linewidth correction compensation is executed; otherwise, linewidth correction compensation is not executed. If copper thickness correction compensation and / or linewidth correction compensation are executed, the parameter compensation decision instruction is the first instruction; otherwise, the parameter compensation decision instruction is the second instruction.

[0045] In this embodiment, by matching the time window number with the database, the target copper thickness and target linewidth within the current cycle can be adaptively invoked based on historical statistical patterns and production conditions, thereby achieving the temporal correlation and process continuity of the compensation parameters. By calculating the difference between the model's predicted values ​​and the executed values ​​of the target parameters, the dynamic changes in copper layer deposition rate and linewidth expansion deviations in key processes such as electroplating and exposure development are captured in real time, enabling timely identification of process deviations during non-steady-state production stages. A dual threshold comparison mechanism for the allowable error values ​​of copper thickness and linewidth is introduced, thereby realizing the parameter compensation touch... The judgment logic is quantifiable and hierarchical, thus avoiding overcompensation due to minor fluctuations and improving system response stability and energy utilization. When the absolute value of the predicted copper thickness difference exceeds the allowable error value, copper thickness correction compensation is performed. By adjusting the electroplating current density or electroplating time, the copper layer thickness deviation is corrected, thereby achieving uniform electrodeposited layer thickness. When the absolute value of the predicted linewidth difference exceeds the allowable linewidth error value, linewidth correction compensation is performed. By adjusting the exposure energy or development time, the line edge expansion or over-etching phenomenon is corrected, thereby ensuring the accuracy of the conductor width and the geometric stability of the pattern. By simultaneously judging the error states of copper thickness and linewidth, parameter compensation decision instructions are generated, thereby performing local adjustments when a single deviation occurs and performing comprehensive compensation when two deviations exist simultaneously, thus achieving global coordinated adjustment of process parameters and reducing cross-interference. Therefore, this invention establishes a linkage correction logic between copper thickness and linewidth through multi-dimensional quantitative analysis of time window, prediction difference and allowable error value, thereby realizing real-time compensation control of electroplating and exposure development processes, improving the accuracy and stability of the circuit board microstructure formation process, and effectively reducing rework rate and resource consumption.

[0046] Furthermore, the initial adjustment of the layout construction parameters is performed. Specifically, the following methods are used: First, the copper thickness prediction difference is matched with the database to obtain the first adjustment coefficient for copper thickness processing; second, the line width prediction difference is matched with the database to obtain the first adjustment coefficient for line width processing; if the copper thickness prediction difference is positive, the electroplating time is shortened based on the first adjustment coefficient for copper thickness processing (the current electroplating time is subtracted from the product of the current electroplating time and the first adjustment coefficient for copper thickness processing to obtain the adjusted electroplating time); if the copper thickness prediction difference is negative, the electroplating time is extended based on the first adjustment coefficient for copper thickness processing (the current electroplating time is added to the product of the current electroplating time and the first adjustment coefficient for copper thickness processing to obtain the extended electroplating time). The electroplating time is calculated after the initial electroplating adjustment. If the linewidth prediction difference is positive, the etching time is extended based on the first adjustment coefficient for linewidth processing (the current etching time is added to the product of the current etching time and the first adjustment coefficient for linewidth processing to obtain the adjusted etching time). If the copper thickness prediction difference is negative, the etching time is shortened based on the first adjustment coefficient for linewidth processing (the current etching time is subtracted from the product of the current etching time and the first adjustment coefficient for linewidth processing to obtain the adjusted etching time). The initial etching adjustment execution time is then obtained. The layout construction parameters are updated based on the initial electroplating adjustment execution time and the initial etching adjustment execution time, thus completing the initial adjustment of layout construction parameter compensation.

[0047] In this embodiment, the first adjustment coefficient for copper thickness processing is obtained by matching the copper thickness prediction difference with the database. The specific method is as follows: obtain each historical copper thickness prediction difference stored in the database and compare it with the copper thickness prediction difference. If there is a historical copper thickness prediction difference with the same value as the copper thickness prediction difference, mark the historical copper thickness prediction difference as the historical reference copper thickness prediction difference and obtain the historical copper thickness processing first adjustment coefficient corresponding to the historical reference copper thickness prediction difference as the first adjustment coefficient for copper thickness processing. Otherwise, select the historical copper thickness prediction difference that is closest to and greater than the copper thickness prediction difference as the historical first reference copper thickness prediction difference, and select the historical copper thickness prediction difference that is closest to and less than the copper thickness prediction difference as the historical second reference copper thickness prediction difference. Obtain the historical copper thickness processing first adjustment coefficients corresponding to the historical first reference copper thickness prediction difference and the historical second reference copper thickness prediction difference respectively, and perform mean processing to obtain the mean of the historical copper thickness processing first adjustment coefficients, and use it as the copper thickness processing first adjustment coefficient.

[0048] The first adjustment coefficient for linewidth processing is obtained by matching the linewidth prediction difference with the database. The specific method is as follows: obtain each historical linewidth prediction difference stored in the database and compare it with the linewidth prediction difference. If there is a historical linewidth prediction difference with the same value as the linewidth prediction difference, mark the historical linewidth prediction difference as the historical reference linewidth prediction difference and obtain the historical first adjustment coefficient for linewidth processing corresponding to the historical reference linewidth prediction difference as the first adjustment coefficient for linewidth processing. Otherwise, select the historical linewidth prediction difference that is closest to and greater than the linewidth prediction difference as the historical first reference linewidth prediction difference, and select the historical linewidth prediction difference that is closest to and less than the linewidth prediction difference as the historical second reference linewidth prediction difference. Obtain the historical first adjustment coefficient for linewidth processing corresponding to the historical first reference linewidth prediction difference and the historical second reference linewidth prediction difference respectively, and perform mean processing to obtain the mean of the historical first adjustment coefficient for linewidth processing, and use it as the first adjustment coefficient for linewidth processing.

[0049] By bidirectionally matching the predicted differences in copper thickness and linewidth, the first adjustment coefficients for copper thickness and linewidth processing are extracted respectively, thereby achieving dynamic preliminary compensation for the construction parameters of two core layout types: electroplating and etching. By judging the positive or negative value of the predicted copper thickness difference, conditions of excessively thick or thin copper layers can be distinguished. When the predicted difference is positive, shortening the electroplating time can effectively suppress over-deposition; when the predicted difference is negative, extending the electroplating time can compensate for insufficient copper layer thickness, thus stabilizing the electrodeposition thickness within the target range. Simultaneously, the positive or negative value of the predicted linewidth difference reflects the deviation in etching degree. When the predicted difference is positive, extending the etching time corrects insufficient conductor spacing; when the predicted difference is negative, shortening the etching time prevents over-etching of line edges, thereby achieving precise control of the geometric boundaries of the pattern.

[0050] This invention prioritizes initial adjustments to the electroplating and etching times, while placing the adjustments to exposure energy and development time in subsequent stages. This is because the former directly determines the copper layer thickness and the physical morphology of the circuitry, constituting fundamental layer parameters for layout formation and exhibiting significant geometric dimension correction effects. By adjusting the electroplating and etching time parameters first, bringing the overall copper thickness and linewidth close to the target range, and then performing refined optical corrections based on exposure energy and development time in subsequent steps, the superimposed errors caused by initial electroplating and etching fluctuations can be avoided. This reduces energy consumption and resource waste, improves the convergence of adjustments, and enhances the stability of the layout morphology. Thus, this invention establishes a difference-driven parameter initial adjustment mechanism, thereby achieving synergistic correction in the electroplating and etching stages. This provides a stable thermal and geometric benchmark for subsequent secondary fine-tuning of photolithography layer parameters, improving the uniformity of overall layout construction and the controllability of the production process.

[0051] Further, the initial adjustment compensation label is obtained through the following method: The layout construction parameters after initial adjustment are obtained and recorded as initial adjustment layout construction parameters. These parameters include the electroplating initial adjustment execution time, etching initial adjustment execution time, exposure energy, and development time. The actual measured parameters at the current moment are obtained and, combined with the initial adjustment layout construction parameters, are input into the copper thickness and linewidth collaborative uniform processing model to obtain the copper thickness initial adjustment prediction value and the linewidth initial adjustment prediction value. Difference processing is performed between the copper thickness initial adjustment prediction value and the linewidth initial adjustment prediction value and the target parameter execution value to obtain the copper thickness initial adjustment prediction difference and the linewidth initial adjustment prediction value difference (subtracting the target copper thickness from the copper thickness initial adjustment prediction value yields the copper thickness initial adjustment prediction difference, and subtracting the target linewidth from the linewidth initial adjustment prediction value yields the linewidth initial adjustment prediction difference). If both the copper thickness initial adjustment prediction difference and the linewidth initial adjustment prediction value difference are below the allowable error value, the initial adjustment compensation label is qualified; otherwise, the initial adjustment compensation label is unqualified.

[0052] In this embodiment, by introducing an "initial adjustment compensation label" judgment mechanism after the initial adjustment of layout construction parameters, closed-loop verification and dynamic screening of parameter compensation effects are achieved. The system inputs the initial adjustment layout construction parameters (including the execution time of electroplating initial adjustment, the execution time of etching initial adjustment, exposure energy, and development time) and real-time construction measured parameters into the copper thickness and linewidth collaborative uniformity processing model to obtain the predicted values ​​of copper thickness and linewidth initial adjustment. Difference analysis is then performed between these predicted values ​​and the target parameter execution values ​​to quantify the approximation degree between the post-adjustment process state and the target state. By setting the initial adjustment compensation label to "qualified" or "unqualified," the initial adjustment effect can be graded at the model level. When the differences are all within the allowable error range, it is marked as qualified, indicating that the parameter compensation has stabilized and meets the process accuracy requirements; otherwise, it is marked as unqualified, indicating the need for further correction or refinement. This improves the controllability and adaptability of the adjustment strategy and avoids the waste of time and resources caused by blindly repeating adjustments, thereby achieving intelligent closed-loop management and dynamic optimization of the copper thickness and linewidth uniformity correction process.

[0053] Furthermore, a secondary processing step is performed to compensate for the layout construction parameters. Specifically, the following steps are taken: First, the initial copper thickness prediction difference is matched with the database to obtain a second adjustment coefficient for copper thickness. Second, the initial linewidth prediction difference is matched with the database to obtain a second adjustment coefficient for linewidth. If the initial copper thickness prediction difference is positive, the development time is shortened based on the second adjustment coefficient (the current development time is subtracted from the product of the current development time and the second adjustment coefficient to obtain the adjusted development time). If the initial copper thickness prediction difference is negative, the development time is extended based on the second adjustment coefficient (the current development time is added to the product of the current development time and the second adjustment coefficient to obtain the adjusted development time). This yields the development adjustment execution time. If the initial linewidth prediction difference is positive, the exposure energy value is reduced based on the second adjustment coefficient. A minor adjustment (subtracting the product of the current exposure energy value and the second adjustment coefficient for linewidth processing to obtain the adjusted exposure energy value) is made. If the initial copper thickness adjustment prediction difference is negative, the exposure energy value is increased based on the second adjustment coefficient for linewidth processing (adding the current exposure energy value to the product of the current exposure energy value and the second adjustment coefficient for linewidth processing to obtain the adjusted exposure energy value). This yields the exposure energy adjustment execution value. The layout construction parameters are updated based on the electroplating initial adjustment execution time, etching initial adjustment execution time, development adjustment execution time, and exposure energy adjustment execution value. These updated parameters are then combined with the measured construction parameters and input into the copper thickness and linewidth collaborative uniform processing model to obtain updated copper thickness and linewidth prediction values. This yields the second processing judgment result for the layout construction parameter compensation. If the second processing judgment result for the layout construction parameter compensation is satisfactory, the processing is completed; otherwise, a warning is issued.

[0054] In this embodiment, the second adjustment coefficient for copper thickness processing is obtained by matching the initial copper thickness prediction difference with the database. Specifically, the method is as follows: Each historical initial copper thickness prediction difference stored in the database is retrieved and compared with the initial copper thickness prediction difference. If a historical initial copper thickness prediction difference has the same value as the initial copper thickness prediction difference, this historical initial copper thickness prediction difference is marked as the historical control initial copper thickness prediction difference, and the historical second adjustment coefficient for copper thickness processing corresponding to the historical control initial copper thickness prediction difference is obtained as the second adjustment coefficient for copper thickness processing. Otherwise, the coefficient is selected based on the initial copper thickness prediction difference. The historical copper thickness initial adjustment prediction difference that is closest to and greater than the initial adjustment prediction difference of copper thickness is taken as the historical second reference copper thickness initial adjustment prediction difference. The historical copper thickness initial adjustment prediction difference that is closest to and less than the initial adjustment prediction difference of copper thickness is selected as the historical second reference copper thickness initial adjustment prediction difference. The historical copper thickness processing second adjustment coefficient corresponding to the historical second reference copper thickness initial adjustment prediction difference and the historical second reference copper thickness initial adjustment prediction difference are obtained respectively and mean processing is performed to obtain the mean of the historical copper thickness processing second adjustment coefficient, which is used as the copper thickness processing second adjustment coefficient.

[0055] The second adjustment coefficient for linewidth processing is obtained by matching the initial linewidth adjustment prediction difference with the database. Specifically, the method is as follows: Retrieve the historical initial linewidth adjustment prediction differences stored in the database and compare them with the actual initial linewidth adjustment prediction difference. If a historical initial linewidth adjustment prediction difference has the same value as the actual initial linewidth adjustment prediction difference, mark this historical initial linewidth adjustment prediction difference as the historical control initial linewidth adjustment prediction difference, and obtain the corresponding historical second adjustment coefficient for linewidth processing as the second adjustment coefficient for linewidth processing. Otherwise, select the coefficient that matches the initial linewidth adjustment prediction difference. The historical linewidth initial adjustment prediction difference that is closest to and greater than the initial linewidth adjustment prediction difference is taken as the historical second reference linewidth initial adjustment prediction difference. The historical linewidth initial adjustment prediction difference that is closest to and less than the initial linewidth adjustment prediction difference is selected as the historical second reference linewidth initial adjustment prediction difference. The historical linewidth processing second adjustment coefficient corresponding to the historical second reference linewidth initial adjustment prediction difference and the historical second reference linewidth initial adjustment prediction difference are obtained respectively and mean processing is performed to obtain the mean of the historical linewidth processing second adjustment coefficient, which is used as the linewidth processing second adjustment coefficient.

[0056] The "secondary processing of layout construction parameter compensation" involves directional secondary correction of development time and exposure energy when the initial adjustment is ineffective or does not fully meet the standards. The adjusted layout construction parameters are then re-input into the copper thickness-linewidth co-factor model for re-evaluation, thus achieving refined compensation and closed-loop verification for the photolithography and development stages. Its beneficial effects are twofold: First, this step advances process compensation from "coarse adjustment (electroplating time, etching time)" to "fine adjustment (development time, exposure energy)." Utilizing the fine-tuning capability of development time on the edge profile of the photolithographic pattern and the quantitative influence of exposure energy on the photoresist's response, it can compensate for residual linewidth deviations or photolithographic coupling effects caused by copper thickness variations at the micron level, thereby significantly improving linewidth control accuracy and pattern reproducibility. Second, matching the secondary adjustment coefficients based on a historical database to form a quantitative adjustment amount avoids oscillations and overshoot caused by blindly amplifying compensation, while ensuring that the adjustment range matches the process feasibility.

[0057] By feeding back the electroplating initial adjustment execution time, etching initial adjustment execution time, and the development time and exposure energy obtained from the secondary adjustment into the collaborative prediction model for re-prediction and judgment, if the secondary correction meets the standards, compensation is completed and incorporated into the production specifications; if it still does not meet the standards, an early warning is triggered and complete adjustment records are retained for manual or higher-order algorithm intervention and analysis. This process ensures the traceability and verifiability of the compensation strategy, and achieves optimal adjustment efficiency and finished product consistency through a hierarchical compensation logic (first large-scale, then fine-tuning).

[0058] like Figure 4The diagram shown is a structural schematic of the copper thickness and linewidth collaborative uniformity modeling and parameter compensation decision system provided in this embodiment of the application. The copper thickness and linewidth collaborative uniformity modeling and parameter compensation decision system provided in this embodiment includes: a control unit partitioning module, a model building module, a construction parameter compensation initial adjustment module, and a compensation judgment module. The control unit partitioning module is used to acquire the circuit board production design drawing and divide it into regions to obtain each control unit, thereby obtaining the comprehensive construction parameters of the control unit under a preset time window. The model building module is used to construct a copper thickness and linewidth collaborative uniformity processing model based on the comprehensive construction parameters and output the model prediction value. The construction parameter compensation... The initial adjustment module is used to acquire the surface temperature of the circuit board, analyze it to obtain temperature processing instructions, and if the temperature processing instruction is to execute cooling processing, then cooling processing is executed; otherwise, based on the analysis of model prediction values, parameter compensation decision instructions are obtained. If the parameter compensation decision instructions are the first instructions, then the initial adjustment of layout construction parameter compensation is executed; otherwise, the initial adjustment of layout construction parameter compensation is not executed. The compensation determination module is used to update the comprehensive construction parameters after the initial adjustment of layout construction parameter compensation, analyze it to obtain the initial adjustment prediction values ​​of copper thickness and line width, and thus obtain the initial adjustment compensation label. If the initial adjustment compensation label is qualified, then the parameter compensation decision processing is completed; otherwise, the secondary processing of layout construction parameter compensation is executed.

[0059] Those skilled in the art will understand that embodiments of the present invention can be provided as methods, systems, or computer program products. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention can take the form of a computer program product embodied on one or more computer-usable storage media (including, but not limited to, disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0060] This invention is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of the invention. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart illustrations and / or block diagrams. Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0061] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0062] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0063] Although preferred embodiments of the invention have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including both the preferred embodiments and all changes and modifications falling within the scope of the invention.

[0064] Obviously, those skilled in the art can make various modifications and variations to this invention without departing from its spirit and scope. Therefore, if these modifications and variations fall within the scope of the claims of this invention and their equivalents, this invention also intends to include these modifications and variations.

Claims

1. A method for modeling and parameter compensation decision-making for the coordinated uniformity of copper thickness and linewidth, characterized in that, Includes the following steps: S1. Obtain the circuit board production design drawing and divide it into regions to obtain each control unit. Then, obtain the comprehensive construction parameters of the control unit under the preset time window. The comprehensive construction parameters include layout construction parameters and actual construction measurement parameters. The layout construction parameters include the electroplating time, etching time, exposure energy and development time of the layout construction. The actual construction measurement parameters include the open ratio, surface tangential speed, surface temperature and electroplating set current density of the control unit. S2. Construct a copper thickness and line width collaborative uniform processing model based on comprehensive construction parameters, and output the model prediction value, which includes the copper thickness prediction value and the line width prediction value. S3. Obtain the surface temperature of the circuit board and analyze it to obtain the temperature processing instruction. If the temperature processing instruction is to execute cooling processing, then execute cooling processing. Otherwise, based on the analysis of the model prediction value, obtain the parameter compensation decision instruction. If the parameter compensation decision instruction is the first instruction, then execute the initial adjustment of layout construction parameter compensation. Otherwise, do not execute the initial adjustment of layout construction parameter compensation. The initial adjustment of layout construction parameter compensation includes copper thickness correction compensation and / or line width correction compensation. S4. Update the comprehensive construction parameters after the initial adjustment of the layout construction parameter compensation, analyze and obtain the initial adjustment prediction values ​​of copper thickness and line width, and obtain the initial adjustment compensation label. If the initial adjustment compensation label is qualified, the parameter compensation decision processing is completed; otherwise, the layout construction parameter compensation secondary processing is performed.

2. The copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method as described in claim 1, characterized in that: The specific method for constructing the copper thickness and linewidth collaborative uniform processing model is as follows: Based on the measured construction parameters, normalization was performed to obtain normalized values ​​for open ratio, surface tangential velocity, and surface temperature. A copper thickness prediction sub-model is constructed based on the electroplating set current density, open ratio normalized value, surface tangential velocity normalized value, surface temperature normalized value and electroplating time. During the training process, a mass transfer constraint equation is introduced as a regular constraint to reflect the physical dependence of copper ion transport rate on deposition thickness. A linewidth prediction sub-model is constructed based on the predicted copper thickness, electroplating current density, exposure energy, normalized open ratio, normalized surface tangential velocity, normalized surface temperature, etching time, exposure energy, and development time. An Arrhenius temperature dependency term is introduced as a regularization constraint during the training process to reflect the influence of temperature changes on the development rate and etching reaction rate. By jointly optimizing the predicted copper thickness and the predicted linewidth, the copper thickness and linewidth are obtained, thus completing the construction of the copper thickness and linewidth co-uniform processing model.

3. The copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method as described in claim 1, characterized in that: The specific method for obtaining the temperature processing command is as follows: Sampling points are set up on the circuit board control unit to obtain the surface temperature of each sampling point. Temperature range processing is performed on the surface temperature of each sampling point to obtain the temperature sampling range value. Temperature mean processing is performed on the surface temperature of each sampling point to obtain the average sampling temperature. The temperature influence value is obtained by coupling the temperature sampling range and the average sampling temperature with the corresponding weighting factor. The system retrieves a preset temperature impact threshold from the database and compares it with the temperature impact value to obtain a temperature processing instruction. If the temperature impact value is above the temperature impact threshold, the temperature processing instruction is to perform cooling processing; otherwise, the temperature processing instruction is to not perform cooling processing.

4. The copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method as described in claim 3, characterized in that: The specific method for performing the cooling process is as follows: Based on the difference between the temperature influence value and the temperature influence threshold, the difference in temperature influence is obtained. Based on the matching of the temperature effect difference values ​​with the database, the temperature adjustment constraint factor is obtained; The maximum surface temperature of each sampling point is obtained, and the absolute difference is processed with the preset temperature threshold in the database to obtain the maximum temperature difference. A linear coupling process is performed based on the maximum temperature difference and the temperature adjustment constraint factor to obtain the comprehensive execution value of the cooling temperature. The control unit adjusts the temperature based on the overall execution value of the cooling temperature.

5. The copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method as described in claim 4, characterized in that: The method for adjusting the cooling of the control unit based on the comprehensive execution value of the cooling temperature is as follows: Obtain the preset cooling temperature threshold value from the database and compare it with the comprehensive cooling temperature execution value to obtain the cooling adjustment execution plan. If the comprehensive cooling temperature execution value is less than the cooling temperature threshold value, the cooling adjustment execution plan is the first execution plan; otherwise, the cooling adjustment execution plan is the second execution plan. The first execution scheme is as follows: based on the comprehensive execution value of cooling temperature, the coolant temperature of each cooling inlet under the time window is reduced and adjusted, thereby completing the cooling adjustment of the control unit; The second execution scheme is as follows: based on the surface temperature of each sampling point, the sampling point corresponding to the maximum surface temperature is selected and marked as a high temperature sampling point, the sampling point corresponding to the average surface temperature is selected and marked as an average temperature sampling point, and the remaining sampling points are marked as ordinary sampling points. Obtain the location of the center point of each cooling inlet, and perform shortest distance analysis with the high temperature sampling point and the average temperature sampling point respectively. Mark the cooling inlet corresponding to the center point of the cooling inlet closest to the high temperature sampling point as the high temperature cooling inlet, mark the cooling inlet corresponding to the center point of the cooling inlet closest to the average temperature sampling point as the low temperature cooling inlet, and mark the remaining cooling inlets as ordinary cooling inlets. The degree of difference between the comprehensive execution value of cooling temperature and the cooling temperature cutoff value is analyzed to obtain the degree of temperature adjustment difference value. The degree of temperature adjustment difference value is matched with the database to obtain the local cooling adjustment coefficient. The local cooling adjustment coefficient and the comprehensive execution value of cooling temperature are multiplicatively coupled to obtain the local cooling adjustment execution temperature. Obtain the high-temperature cooling outlet corresponding to the high-temperature cooling inlet and the low-temperature cooling outlet corresponding to the low-temperature cooling inlet. If the high-temperature cooling outlet and the low-temperature cooling outlet are the same outlet, then the cooling path of the high-temperature cooling inlet and the low-temperature cooling inlet are the same cooling path. Then mark each ordinary cooling inlet in the cooling path between the high-temperature cooling inlet and the low-temperature cooling inlet as the path cooling inlet. The local cooling adjustment execution temperature is used as the execution temperature of the high-temperature cooling inlet, and the comprehensive cooling temperature execution value is used as the execution temperature of the low-temperature cooling inlet and the normal cooling inlet. Based on the execution temperatures of the high-temperature cooling inlet and the low-temperature cooling inlet, a linear transition process is performed to obtain the execution temperature of each cooling inlet, thereby completing the cooling adjustment of the control unit. If the high-temperature cooling outlet and the low-temperature cooling outlet are not the same outlet, select each sampling point whose surface temperature is above the average surface temperature and mark it as each additional cooling sampling point. Obtain each cooling inlet closest to each additional cooling sampling point and mark it as each additional cooling inlet. Otherwise, mark it as a normal cooling inlet. The local cooling adjustment execution temperature is used as the execution temperature of each additional cooling inlet, and the comprehensive cooling temperature execution value is used as the execution temperature of the ordinary cooling inlet, thereby completing the cooling adjustment of the control unit.

6. The copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method as described in claim 1, characterized in that: The specific method for obtaining the parameter compensation decision instruction is as follows: Based on the time window number, the target parameter execution value under the current time window is obtained by matching with the database. The target parameter execution value includes the target copper thickness and the target line width. The difference between the target parameter execution value and the model prediction value is processed. The difference between the copper thickness prediction value and the target copper thickness is processed to obtain the copper thickness prediction difference. The difference between the line width prediction value and the target line width is processed to obtain the line width prediction difference. Obtain preset allowable error values ​​from the database, including copper thickness allowable error values ​​and line width allowable error values; Based on the comparison between the absolute value of the copper thickness prediction difference and the copper thickness allowable error value, and the comparison between the absolute value of the linewidth prediction difference and the linewidth allowable error value, a parameter compensation decision instruction is obtained. If the absolute value of the copper thickness prediction difference is greater than the copper thickness allowable error value, copper thickness correction compensation is executed; otherwise, copper thickness correction compensation is not executed. If the absolute value of the linewidth prediction difference is greater than the linewidth allowable error value, linewidth correction compensation is executed; otherwise, linewidth correction compensation is not executed. If copper thickness correction compensation and / or linewidth correction compensation are executed, the parameter compensation decision instruction is the first instruction; otherwise, the parameter compensation decision instruction is the second instruction.

7. The copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method as described in claim 6, characterized in that: The specific method for the initial adjustment of the construction parameters compensation in the execution plan is as follows: Based on the copper thickness prediction difference and matching with the database, the first adjustment coefficient for copper thickness processing is obtained; Based on the linewidth prediction difference and matching with the database, the first adjustment coefficient for linewidth processing is obtained; If the copper thickness prediction difference is positive, the electroplating time is shortened based on the first adjustment coefficient for copper thickness processing. If the copper thickness prediction difference is negative, the electroplating time is extended based on the first adjustment coefficient for copper thickness processing, thus obtaining the initial electroplating adjustment execution time. If the linewidth prediction difference is positive, the etching time is extended based on the first adjustment coefficient of linewidth processing. If the copper thickness prediction difference is negative, the etching time is shortened based on the first adjustment coefficient of linewidth processing, thus obtaining the initial etching adjustment execution time. The layout construction parameters are updated based on the initial adjustment execution time of electroplating and etching, thereby completing the initial adjustment of layout construction parameters compensation.

8. The copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method as described in claim 7, characterized in that: The method for obtaining the initial adjustment compensation label is as follows: Obtain the layout construction parameters after the initial adjustment of the layout construction parameters and record them as the initial adjustment layout construction parameters. The initial adjustment layout construction parameters include the electroplating initial adjustment execution time, etching initial adjustment execution time, exposure energy and development time of the layout construction. The actual construction parameters at the current moment are obtained and combined with the initial adjustment layout construction parameters, and then input into the copper thickness and line width collaborative uniform processing model to obtain the initial adjustment prediction values ​​of copper thickness and line width. The difference between the initial copper thickness adjustment prediction value and the initial line width adjustment prediction value and the target parameter execution value is processed to obtain the initial copper thickness adjustment prediction difference and the initial line width adjustment prediction difference. If the initial copper thickness adjustment prediction difference and the initial line width adjustment prediction difference are both below the allowable error value, the initial adjustment compensation label is qualified; otherwise, the initial adjustment compensation label is unqualified.

9. The copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method as described in claim 1, characterized in that: The specific method for the secondary processing of construction parameter compensation in the execution plan is as follows: The second adjustment coefficient for copper thickness processing is obtained by matching the initial copper thickness prediction difference with the database. The second adjustment coefficient for line width processing is obtained by matching the initial line width prediction difference with the database. If the initial copper thickness adjustment prediction difference is positive, the development time is shortened based on the second adjustment coefficient of copper thickness processing. If the initial copper thickness adjustment prediction difference is negative, the development time is extended based on the second adjustment coefficient of copper thickness processing, thus obtaining the development adjustment execution time. If the initial linewidth adjustment prediction difference is positive, the exposure energy value is reduced based on the second adjustment coefficient of the linewidth processing. If the initial copper thickness adjustment prediction difference is negative, the exposure energy value is increased based on the second adjustment coefficient of the linewidth processing, thus obtaining the exposure energy adjustment execution value. The layout construction parameters are updated based on the execution time of electroplating initial adjustment, etching initial adjustment, development adjustment, and exposure energy adjustment. These parameters are then combined with the measured parameters from the construction process and input into the copper thickness and linewidth collaborative uniformity processing model to obtain the copper thickness update prediction value and the linewidth update prediction value. From this, the result of the secondary processing of the layout construction parameter compensation is obtained. If the result of the secondary processing of the layout construction parameter compensation is qualified, the processing is completed; otherwise, an early warning is issued.

10. A system applying the copper thickness and linewidth cooperative uniformity modeling and parameter compensation decision-making method as described in any one of claims 1-9, characterized in that, include: The control unit is divided into modules, model building modules, initial adjustment modules for construction parameter compensation, and compensation determination modules. The control unit division module is used to obtain the circuit board production design drawing, divide it into regions, obtain each control unit, and thereby obtain the comprehensive construction parameters of the control unit under a preset time window. The model building module is used to construct a copper thickness and linewidth collaborative uniform processing model based on comprehensive construction parameters, and output the model prediction value. The initial adjustment module for construction parameter compensation is used to obtain the surface temperature of the circuit board, analyze and obtain the temperature processing command. If the temperature processing command is to execute cooling processing, then cooling processing is executed. Otherwise, based on the analysis of the model prediction value, the parameter compensation decision command is obtained. If the parameter compensation decision command is the first command, then the initial adjustment for layout construction parameter compensation is executed. Otherwise, the initial adjustment for layout construction parameter compensation is not executed. The compensation determination module is used to update the comprehensive construction parameters after the initial adjustment of the layout construction parameters, analyze and obtain the initial adjustment prediction values ​​of copper thickness and line width, and thus obtain the initial adjustment compensation label. If the initial adjustment compensation label is qualified, the parameter compensation decision processing is completed; otherwise, the secondary processing of layout construction parameter compensation is performed.

Images