Positioning and calibration method for high-speed and high-precision worktable of photoetching machine

Abstract

The application relates to the technical field of photoetching machine workbench, and discloses a photoetching machine high-speed high-precision workbench positioning calibration method.The method comprises the following steps: collecting workbench position data and motion parameters, analyzing position error and motion stability matching degree, and generating a position error control parameter set; then, based on the parameter set, extracting environmental interference values and gradient change amounts, analyzing the influence of environmental interference on positioning accuracy, adjusting environmental compensation parameters to generate an environmental optimization parameter set; subsequently, according to the environmental optimization parameter set, analyzing dynamic response under high-speed motion of the workbench, adjusting control input rate and feedback path distribution proportion, redistributing control parameter distribution trend and dynamic value, and obtaining a dynamic adjustment result; finally, based on the dynamic adjustment result, generating an abnormal intervention data set, and then analyzing target position distribution proportion and calibration time, adjusting target calibration path parameters, and generating a positioning calibration data table.

Description

Technical Field

[0001] This invention relates to the field of lithography machine stage technology, specifically a method for positioning and calibrating a high-speed, high-precision lithography machine stage. Background Technology

[0002] In the semiconductor manufacturing field, lithography machines are core equipment in chip production, and the positioning accuracy of their worktables directly affects the chip's process level and product yield. As chip manufacturing processes advance to more advanced nodes, the positioning requirements for lithography machine worktables are constantly increasing, while also needing to meet high-speed movement demands to adapt to production efficiency in large-scale mass production scenarios. Currently, lithography machine worktables face positioning accuracy deviations caused by various factors during actual operation, and these problems severely restrict the stable production of advanced process chips.

[0003] From the perspective of position data acquisition and error analysis, existing methods mostly rely on a single sensor to collect position information, which makes it difficult to comprehensively reflect the position changes of the worktable under complex motion conditions. This results in insufficient accuracy in the analysis of the matching degree between position error and motion stability. This imprecise analysis leads to an insufficiently targeted set of position error control parameters, which cannot effectively address error problems under different motion conditions, thus affecting the subsequent positioning calibration effect.

[0004] Regarding environmental interference, current technologies lack comprehensive monitoring capabilities, often focusing only on common environmental factors such as temperature and humidity while neglecting significant interference factors that impact the positioning accuracy of the worktable, such as vibration and electromagnetic radiation. Even when some environmental interference is detected, it is difficult to accurately extract the interference values ​​and gradient changes, making it impossible to deeply analyze the specific mechanisms by which environmental interference affects positioning accuracy. Consequently, the adjusted environmental compensation parameters lack scientific rigor, and the generated set of optimized environmental parameters cannot effectively offset the positioning deviations caused by environmental interference, resulting in significant fluctuations in the positioning accuracy of the worktable under different environmental conditions.

[0005] For dynamic response processing under high-speed motion, existing methods are ill-suited to the complex dynamic characteristics of a worktable during high-speed movement. Under high-speed motion, dynamic factors such as inertia and friction of the worktable change drastically. Existing methods cannot analyze the impact of these dynamic factors on the worktable's positioning accuracy in a timely and accurate manner. The adjusted control input rate and feedback path distribution ratio are often unreasonable, failing to reallocate the distribution trend and dynamic values ​​of control parameters according to dynamic changes. Consequently, the generated dynamic adjustment results cannot meet the positioning accuracy requirements under high-speed motion, leading to problems such as positioning overshoot and lag during high-speed movement.

[0006] Regarding real-time fluctuations and anomaly intervention, existing technologies lack sufficient monitoring frequency and accuracy for real-time position fluctuation rate and error offset during the workbench positioning process, making it impossible to detect abnormal fluctuations in a timely manner. Even if anomalies are detected, it is difficult to accurately analyze the impact of the fluctuation range on positioning accuracy, and it is impossible to dynamically adjust the control distribution path and error ratio within the target range. The generated anomaly intervention dataset cannot effectively intervene in abnormal situations, leading to further expansion of positioning errors and affecting chip production quality.

[0007] In the target position calibration and parameter adjustment stages, existing methods lack in-depth analysis of the distribution ratio of the target positions on the stage and the calibration time. They also lack a systematic approach when adjusting target calibration path parameters. The generated stage positioning calibration data tables cannot comprehensively and accurately guide actual positioning calibration operations, resulting in low calibration efficiency and difficulty in maintaining stable positioning accuracy over the long term. These problems collectively prevent existing lithography machine stage positioning calibration methods from meeting the high-speed, high-precision positioning requirements of advanced semiconductor manufacturing, becoming a bottleneck restricting the further development of chip manufacturing technology. Summary of the Invention

[0008] The purpose of this invention is to provide a high-speed, high-precision stage positioning and calibration method for lithography machines, so as to solve the problems mentioned in the background art.

[0009] To achieve the above objectives, the present invention provides a high-speed, high-precision stage positioning and calibration method for a lithography machine, the method comprising:

[0010] The position data and motion parameters of the lithography machine stage are collected. Based on the position data and motion parameters, the degree of matching between position error and motion stability is analyzed, and a set of position error control parameters is generated.

[0011] Based on the position error control parameter set, the environmental interference value and gradient change of the workbench are extracted, the impact of environmental interference on positioning accuracy is analyzed, the environmental compensation parameters are adjusted, and an environmental optimization parameter set is generated.

[0012] Based on the environmental optimization parameter set, the dynamic response of the workbench under high-speed motion is analyzed, the control input rate and feedback path distribution ratio are adjusted, the distribution trend and dynamic value of the control parameters are redistributed, and dynamic adjustment results are generated.

[0013] Based on the dynamic adjustment results, the real-time position fluctuation rate and error offset during the workbench positioning process are extracted, the impact of the fluctuation range on the positioning accuracy is analyzed, the control distribution path and error ratio within the target range are dynamically adjusted, and an abnormal intervention dataset is generated.

[0014] Based on the aforementioned abnormal intervention dataset, the distribution ratio and calibration time of the target positions on the workbench are analyzed, the parameters of the target calibration path are adjusted, and a workbench positioning calibration data table is generated.

[0015] Preferably, the step of obtaining the matching degree between the position error and motion stability specifically includes:

[0016] Based on the position data and motion parameters of the lithography machine stage, the position error parameters, motion speed data and acceleration data of the stage are extracted. A time window is set, and time points are selected to match the data. By comparing the data correlation and filtering the data, the position error parameters and motion speed data are obtained.

[0017] Based on the position error parameters and motion speed data, the motion path is matched and verified, the difference between position error and motion speed is calculated, the position error distribution and motion speed distribution are corrected by combining acceleration changes, and the path parameters are adjusted by the influence of acceleration on the data to obtain the matching situation between position error and motion speed.

[0018] Based on the matching of position error and motion speed, motion stability analysis is performed, motion stability analysis standards are set, and the error distribution under differentiated acceleration conditions is evaluated in combination with the dynamic changes of the worktable. Stability indicators are compared and acceleration conditions are optimized to obtain the degree of matching between position error and motion stability.

[0019] Preferably, the steps for obtaining the position error control parameter set are as follows:

[0020] Based on the degree of matching between the position error and motion stability, the error transmission and stability changes of the worktable under different operating conditions are analyzed, and the position error distribution of the worktable is weighted and calculated to obtain the preliminary error control requirements of the worktable.

[0021] Based on the initial error control requirements of the worktables, the error balance between worktables is analyzed, the relationship between error transmission efficiency and load distribution between worktables is identified, the error control parameters of the worktables are corrected, the position error value of the worktables after adjustment is calculated, and the error control dataset between worktables is obtained.

[0022] By combining the inter-workbench error control dataset with the motion stability matching results, the errors between the workbench are allocated, optimized, and matched to meet the required balance and stability requirements, thus obtaining the position error control parameter set.

[0023] Preferably, the steps for obtaining the environmental interference value and gradient change of the workbench are as follows:

[0024] Based on the set of position error control parameters, temperature data in the workbench environment is extracted, temperature points in each time period are filtered, and temperature fluctuations are analyzed in combination with the temperature change trends of the workbench at different positions to obtain temperature data in the workbench environment.

[0025] Based on the temperature data in the workbench environment, the vibration value of each temperature point is calculated. By analyzing the relationship between temperature and vibration, the vibration change at each measurement point is identified. Combined with the structural parameters of the workbench, the vibration changes at different locations are compared to obtain vibration distribution and gradient distribution data.

[0026] Based on the vibration distribution and gradient distribution data, the overall vibration distribution in the workbench environment is analyzed, the vibration gradient is optimized by combining temperature data, the impact of vibration changes on the workbench performance is analyzed, a stable vibration configuration is determined under differentiated operating conditions, and the workbench environment interference value and gradient change amount are obtained.

[0027] Preferably, the steps for obtaining the environmental optimization parameter set are as follows:

[0028] Based on the environmental interference value and gradient change of the workbench, the time series of environmental interference changes is determined, the current environmental interference value is compared with the original environmental interference data, the environmental gradient at each moment is analyzed, and corresponding thresholds are defined according to the workbench state partition to generate a preliminary environmental change parameter set.

[0029] The preliminary environmental change parameter set is analyzed to analyze the impact of environmental disturbances on the stability of the workbench control output, identify the correlation between environmental disturbances and control output, and calculate the section control stability influence coefficient.

[0030] By analyzing the influence coefficient of the control stability of the section, and combining it with the environmental disturbance change parameters, the environmental compensation distribution balance is adjusted, the environmental regulation data is optimized, and an environmental optimization parameter set is generated.

[0031] Preferably, the steps for obtaining the dynamic adjustment result are as follows:

[0032] Based on the environmental optimization parameter set, control adsorption data of the workbench surface is extracted, the adsorption rate of control input on the surface of differentiated materials is monitored, and the diffusion characteristics of control input are inferred by combining external environmental factors such as time and temperature. Adsorption and diffusion rate coefficients are defined, and an adsorption-diffusion dynamic parameter set is generated.

[0033] The influence of the adsorption-diffusion dynamic parameter set on the control input rate and distribution is analyzed. Based on the requirements for controlling the concentration distribution on the stage surface, the ratio between the input rate path and the control input rate is optimized, the adjustment coefficient of the control concentration distribution trend is calculated, and the control concentration regulation result is generated.

[0034] Analyze the control concentration regulation results, adjust the ratio of control input rate to input path, allocate the distribution trend of control concentration, and combine adsorption diffusion parameters and adjustment coefficients to obtain dynamic adjustment results.

[0035] Preferably, the steps for obtaining the abnormal intervention dataset are as follows:

[0036] Based on the dynamic adjustment results, the monitoring equipment monitors the position fluctuation rate and error offset in real time during the positioning process, identifies the fluctuation range, eliminates abnormal values ​​due to equipment failure, analyzes the average fluctuation rate of the data, and obtains position and error fluctuation data.

[0037] The impact of the position and error fluctuation data range on the positioning accuracy of the worktable is analyzed. Using known positioning accuracy data, the relationship between position and error is analyzed, and the positioning accuracy under the differential fluctuation range is calculated to obtain the positioning accuracy impact data.

[0038] Based on the positioning accuracy impact data, the control distribution path and error ratio within the target range are dynamically adjusted. Adjustments are made based on the relationship between the positioning accuracy impact data and the position and error fluctuation range. The control input rate and error control range are allocated, and an abnormal intervention dataset is generated.

[0039] Preferably, the steps for obtaining the workbench positioning calibration data table are as follows:

[0040] Based on the aforementioned abnormal intervention dataset, the distribution of target positions on the workbench and calibration time are analyzed. Position accuracy data at different calibration time points are collected, the accuracy time distribution is organized, the accuracy change trend is analyzed, and the data is classified to obtain the workbench position accuracy distribution data.

[0041] Based on the workbench position accuracy distribution data, the target calibration path parameters are adjusted, the optimal calibration time and accuracy distribution of the workbench position accuracy are analyzed, and the accuracy changes under different calibration conditions are compared. The calibration speed, time, and control input operating conditions are adjusted to obtain the target calibration path parameters.

[0042] Based on the target calibration path parameters, the calibration conditions are adjusted according to the current operating parameters, the calibration time, speed, and the relationship between the control input variables are controlled, and real-time calibration is performed according to the adjusted parameters to obtain the workbench positioning calibration data table.

[0043] Preferably, the adjustment steps for the target calibration path parameters specifically include:

[0044] Based on the workbench position accuracy distribution data, the key point sequence in the calibration path is determined, the cumulative path error is calculated based on the key point sequence, and the distribution density of path points is optimized in combination with calibration time constraints to generate preliminary path adjustment parameters.

[0045] Based on the preliminary path adjustment parameters, the impact of path error on overall calibration accuracy is analyzed, error-sensitive areas are identified, the position and order of path points are adjusted, the path error is recalculated, and optimized path parameters are obtained.

[0046] By combining the optimized path parameters with the target calibration time, the feasibility of the path is verified, the path parameters are adjusted to meet the accuracy requirements, and the target calibration path parameters are obtained.

[0047] Preferably, the real-time calibration execution steps specifically include:

[0048] Based on the target calibration path parameters, initialize the calibration control system, load the calibration data table, and set calibration monitoring points;

[0049] Real-time acquisition of workbench position feedback data, comparison with target position data, and calculation of real-time error value;

[0050] The control input is dynamically adjusted based on the real-time error value, and the calibration path parameters are updated by applying an error compensation algorithm.

[0051] The calibration process is continuously monitored, calibration logs are recorded, calibration result data is output, and the workbench positioning calibration is completed.

[0052] Compared with the prior art, the beneficial effects of the present invention are:

[0053] In the position data acquisition and error control parameter generation stage, this method comprehensively acquires the position information and motion state characteristics of the worktable during its movement by collecting its position data and motion parameters. Based on this, it deeply analyzes the degree of matching between position error and motion stability. This comprehensive data analysis approach can accurately identify the causes and patterns of position error under different motion conditions, thereby generating a highly targeted set of position error control parameters. This parameter set can provide precise guidance for subsequent error control based on the actual motion state of the worktable, effectively avoiding error control failure due to insufficient parameter targeting, and laying a solid foundation for subsequent positioning calibration.

[0054] In terms of environmental interference response and environmental optimization parameter generation, this method, based on a set of position error control parameters, accurately extracts the environmental interference values ​​and gradient changes of the worktable. It is no longer limited to traditional common environmental factors such as temperature and humidity, but comprehensively covers interference factors that significantly affect positioning accuracy, such as vibration and electromagnetic radiation. By deeply analyzing the specific impact of these environmental interferences on positioning accuracy, the changing trends of positioning errors under different environmental factors can be clearly understood, thereby scientifically adjusting environmental compensation parameters and generating a comprehensive set of environmental optimization parameters. This parameter set can effectively offset the impact of various environmental interferences on the worktable's positioning accuracy, reduce positioning deviations caused by environmental factors, and enable the worktable to maintain high positioning accuracy stability under different environmental conditions, reducing production quality risks caused by environmental fluctuations.

[0055] To address the dynamic response problem under high-speed motion, this method, based on an optimized set of environmental parameters, fully considers the dynamic changes in factors such as inertia and friction during high-speed table movement, and deeply analyzes the dynamic response characteristics of the table. By rationally adjusting the control input rate and feedback path distribution ratio, the distribution trend and dynamic values ​​of control parameters are real-time redistributed according to dynamic changes, generating dynamic adjustment results that meet the requirements of high-speed motion. This dynamic adjustment process can adapt to the complex dynamic changes of the table under high-speed motion in a timely manner, effectively avoiding problems such as positioning overshoot and lag, ensuring that the table can maintain accurate positioning even under high-speed motion, and meeting the dual requirements of production efficiency and positioning accuracy in large-scale mass production scenarios.

[0056] In terms of real-time fluctuation monitoring and anomaly intervention, this method, based on dynamic adjustment results, extracts the position fluctuation rate and error offset during the workbench positioning process in real time. Through high-frequency and high-precision monitoring, it promptly captures abnormal fluctuations during the positioning process. Simultaneously, it deeply analyzes the impact of the fluctuation range on positioning accuracy, dynamically adjusting the control distribution path and error ratio within the target range according to the degree of impact, generating an effective anomaly intervention dataset. This dataset enables rapid intervention in abnormal situations, curbing further expansion of positioning errors, ensuring the stability and accuracy of the positioning process, and reducing chip manufacturing defects caused by abnormal fluctuations.

[0057] In the target position calibration and calibration data table generation stage, this method comprehensively analyzes the distribution ratio and calibration time of the target positions on the workbench based on the abnormal intervention dataset. It fully considers the positioning requirements and calibration efficiency of different target positions and systematically adjusts the parameters of the target calibration path. The generated workbench positioning calibration data table comprehensively and accurately reflects the calibration requirements under different working conditions, providing detailed and reliable guidance for actual positioning calibration operations. This not only improves the efficiency of the calibration process but also ensures that the positioning accuracy after calibration remains stable over a long period, reducing the frequency of subsequent recalibrations, lowering equipment maintenance costs, and improving the overall production efficiency of the lithography machine. This helps semiconductor manufacturing companies achieve higher quality and more efficient chip production. Attached Figure Description

[0058] Figure 1 This is a schematic diagram illustrating the working principle of the high-speed, high-precision stage positioning and calibration method for a lithography machine described in this invention.

[0059] Figure 2 A flowchart for obtaining the degree of matching between position error and motion stability;

[0060] Figure 3 Flowchart for obtaining the position error control parameter set;

[0061] Figure 4 Flowchart for obtaining the environmental optimization parameter set;

[0062] Figure 5 The flowchart for obtaining the abnormal intervention dataset. Detailed Implementation

[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] Please see Figure 1 This invention provides a high-speed, high-precision stage positioning and calibration method for a lithography machine, the method comprising:

[0065] The position data and motion parameters of the lithography machine stage are collected by a sensor system. This data includes, but is not limited to, encoder feedback, laser interferometer measurements, and motor drive current. Based on the collected position data and motion parameters, the degree of matching between position error and motion stability is analyzed. This analysis involves calculating the correlation between position error parameters, motion velocity data, and acceleration data within a set time window, and generating a set of position error control parameters including error gain, integral time constant, and differential coefficients.

[0066] Based on this set of position error control parameters, the environmental interference values ​​of the workbench, such as temperature fluctuations and mechanical vibrations, as well as the gradient changes of these interferences, are extracted. The influence mechanism of these environmental factors on positioning accuracy is analyzed, and an environmental optimization parameter set is generated by adjusting the temperature compensation coefficient and vibration suppression parameters.

[0067] Based on the environmental optimization parameter set, the dynamic response characteristics of the workbench under high-speed motion conditions are analyzed. The input rate of the control system and the branch ratio of the feedback loop are adjusted, and the distribution trend of control parameters, such as the weight allocation of PID parameters, and dynamic values, such as adaptive gain, are redistributed to generate dynamic adjustment results.

[0068] Based on the dynamic adjustment results, the real-time position fluctuation rate and error offset during the worktable positioning process are extracted by high-precision displacement sensors. The impact of these fluctuation ranges on the final positioning accuracy is analyzed, and the distribution path in the control law, such as the weight of the feedforward path and the allowable error ratio, is dynamically adjusted to generate an anomaly intervention dataset containing anomaly thresholds and intervention strategies.

[0069] Based on the abnormal intervention dataset, the probability distribution ratio of the target position of the workbench and the required calibration time are analyzed. The path parameters of the calibration algorithm are adjusted, such as the sequence of calibration points and the moving speed. Finally, a workbench positioning calibration data table that can be directly called by the control system is generated to complete the entire calibration process.

[0070] Example 1: See Figure 2 The implementation of a high-speed, high-precision stage positioning and calibration method for lithography machines begins with the systematic acquisition and in-depth analysis of stage position data and motion parameters. This process relies on an integrated sensor network and data acquisition system. Position data acquisition is primarily accomplished through high-resolution grating rulers and laser interferometers mounted on the stage's motion axes. These sensors can capture nanometer-level positional changes of the stage in real time and convert analog signals into digital signals for transmission to the central processing unit. Motion parameter acquisition relies on feedback elements built into the servo drive system, which continuously monitor and record the motor's rotational speed, output torque, and the derived linear acceleration values. All this data is transmitted synchronously via a high-speed fieldbus to ensure timestamp consistency, laying a timing foundation for subsequent correlation analysis.

[0071] After data acquisition, the data filtering and correlation stage begins. The system sets a configurable movement time window, the length of which is dynamically adjusted according to the current motion characteristics of the worktable. For example, a shorter time window is used during high-speed scanning to capture faster dynamic changes, while a longer window is used for low-speed precision positioning to obtain more stable statistical characteristics. Within this time window, the system selects matching time points at fixed sampling intervals, aligning data from different sensors that have the same timestamp. By calculating the correlation between the position error data sequence and the motion velocity data sequence, the inherent relationship between the two is identified. Subsequently, digital filters, such as Kalman filters or low-pass filters, are applied to smooth the data, effectively eliminating instantaneous jumps and high-frequency noise caused by electromagnetic interference or mechanical vibration, thus obtaining a clean and reliable set of position error parameters and motion velocity data sequences. This processing flow ensures the data quality upon which subsequent analysis is based.

[0072] Based on the filtered high-quality data, the system initiates a matching and verification procedure for the preset motion path. The stage's control system stores the ideal theoretical motion path, which is pre-set according to the requirements of the photolithography process. The verification process compares the actually acquired position error data with the commanded motion speeds at corresponding points on the theoretical path, calculating the difference between the two. This difference reveals the deviation between the actual motion and the ideal model. Simultaneously, the system acquires real-time acceleration data from the drive unit, reflecting the dynamic forces and inertial effects experienced by the stage. Changes in acceleration are used to correct the distribution models of position error and motion speed. For example, during periods of high acceleration, the system anticipates a potential increase in position error and quantifies this impact using the established dynamic model, thereby fine-tuning the parameters of the path tracking controller, such as the proportional gain and feedforward coefficient. Through this series of comparisons and corrections, a quantitative evaluation report on the matching of position error and motion speed is finally obtained. This report is typically presented in the form of a matching coefficient, with values ​​closer to one indicating a higher degree of matching.

[0073] The system performs a comprehensive motion stability analysis based on a set of preset stability criteria, including the allowable range of maximum overshoot, the upper limit of settling time, and the lower limit of control system bandwidth. The analysis process closely integrates with the actual dynamic changes during table operation; for example, the dynamic response changes when the table carries silicon wafers of different masses, or the system's transient response characteristics when motion commands undergo a step change. The system evaluates the distribution and central tendency of position errors under different acceleration conditions. By comparing stability indices calculated under various acceleration settings, the system employs an optimization algorithm to find the optimal solution. This optimization is an iterative process that may involve fine adjustments to acceleration feedforward parameters, retuning of the velocity loop integration time, or reconfiguration of position loop filtering parameters. The goal is to ensure that the system meets motion accuracy requirements while possessing good stability margins, resisting certain external disturbances without oscillations or divergence. Finally, the analysis outputs a comprehensive conclusion regarding the degree of matching between position error and motion stability. This conclusion is not a single numerical value but a set of evaluations including multiple indices, comprehensively describing the system's dynamic performance under the current conditions. This conclusion will be directly used to guide the generation of subsequent control parameter sets, forming a closed-loop process from data acquisition to parameter optimization. The entire implementation process emphasizes the real-time nature of data, the dynamic nature of processing, and the adaptability of decision-making, ensuring that the lithography machine stage can maintain high-precision positioning performance under high-speed operation.

[0074] Example 2: See Figure 3 After completing the dynamic characteristic analysis of a single worktable, the system expands to the comprehensive error control and deep coupling analysis of environmental factors in a multi-worktable collaborative system. This stage begins with the in-depth utilization of the obtained matching degree between position error and motion stability. The system analyzes the performance of the worktable under differentiated operating conditions, including different scanning speed settings, carrying silicon wafer loads of different specifications and weights, and performing different motion modes such as stepping and scanning. Under these diverse operating conditions, the error propagation characteristics are not linear, and there is a complex coupling relationship between the error and system stability. The system performs weighted calculations on the position error data of the worktable throughout its entire stroke range. The weighting strategy comprehensively considers the usage frequency of each physical region in the lithography process and the pattern overlay accuracy requirements of that region. For example, the weight coefficient in the center of the exposure area is significantly higher than that in the edge areas of the stroke range. Through this weighted analysis, the system generates a preliminary error control requirement report, which clearly identifies which axial movements and which stroke intervals require priority error compensation and control enhancement. Its output is a series of weighted error distribution maps.

[0075] Based on this initial error control requirement, the system's analysis scope expands from a single workstation to multiple collaboratively operating workstations. In a dual- or multi-workstation switching system of a lithography machine, the workstations synchronize their motion and position through mechanical coupling components and a control bus network. The system needs to identify the error transmission efficiency between these workstations, i.e., the extent to which a positional deviation or vibration of one workstation is transmitted to another through the coupling structure and affects its accuracy. Simultaneously, the system analyzes the load distribution relationships between the workstations, including static load distribution and dynamic inertial force distribution. Based on these complex interrelationships, the system collaboratively corrects the independent error control parameters of each workstation. These parameters may include values ​​in the independent error compensation tables for each motion axis, bias settings for each servo drive, etc. The correction process is not isolated but solved through a coupled system of equations aimed at minimizing the overall error. After the correction, the estimated position error value for each workstation under the new collaborative settings is obtained. All this data is collected and organized into a global inter-workstation error control dataset. This dataset defines the error boundaries and control law interaction rules for multi-workstation collaborative operation.

[0076] The system fuses the aforementioned inter-workstation error control dataset with previously obtained motion stability matching results. This fusion process aims to globally distribute and optimize the errors among multiple workstations. The strategy employed is a multi-objective optimization algorithm, which simultaneously satisfies several potentially mutually restrictive objectives: overall system motion accuracy balance (avoiding excessively high accuracy on one workstation while another is too low), overall dynamic stability (preventing resonance during coordinated motion), and efficiency requirements (such as workstation exchange time). The optimization algorithm outputs a globally consistent set of position error control parameters applicable to the entire multi-workstation system. This parameter set contains comprehensive control parameters for all motion axes, which are inherently correlated and constrained, ensuring the consistency and accuracy of the coordinated motion of multiple workstations.

[0077] After generating the global control parameter set, the system focuses on the physical environment of the workbench. Based on this position error control parameter set, the system initiates the extraction and analysis of environmental disturbance data. First, it focuses on temperature data, continuously collecting temperature sampling data through multiple high-precision temperature sensors placed at key locations on the granite base of the workbench, near the linear motor stator, and around the air-bearing guide rail. The system filters temperature readings for each specific time period and, combined with the structural characteristics of the workbench, analyzes the temperature change trends at different locations (such as the cooling area near the linear motor heat source versus the cooling area away from the heat source). By calculating temperature fluctuations and spatial gradients, a temperature data set characterizing the internal thermal environment of the equipment is obtained, reflecting the distribution and dynamic changes of the thermal field.

[0078] The system couples temperature data with mechanical vibration values ​​measured at corresponding locations using low-frequency vibration sensors (such as accelerometers) for calculation. Its core is establishing a temperature-vibration relationship model that considers changes in structural stress caused by material thermal expansion and their impact on system stiffness and natural frequencies. This model analyzes how thermal disturbances induce or modulate mechanical vibrations, thereby identifying the thermodynamic causes of vibration variations at each measurement point. This analysis is closely integrated with the specific structural parameters of the workbench, such as the coefficient of thermal expansion of the ceramic material used, the heat capacity of the cast iron matrix, and the overall modal shapes of the structure. By comparing vibration variations at different locations (such as the driven and non-driven ends), the system ultimately obtains detailed vibration distribution data containing amplitude and phase information, as well as spatial gradient distribution data in different directions. This "vibration field" map reveals the distribution of environmental mechanical energy.

[0079] The system performs a higher-level integrated analysis of vibration and gradient distribution data. The aim is to analyze the overall vibration modal distribution within the workbench environment and identify potential overall structural resonant frequencies. Combined with high-precision, high-sampling-rate temperature data, the system uses a compensation algorithm to optimize the accuracy of vibration gradient estimation, distinguishing between slowly varying vibration components caused by thermal effects and transient vibration components caused by external impacts. Based on this, the system can accurately analyze the degree and mechanism of vibration changes' impact on the final positioning accuracy and dynamic response characteristics of the workbench. Ultimately, the system needs to determine the optimal vibration suppression configuration parameters that maintain stable accuracy under differentiated operating conditions (such as high-speed scanning and low-speed stepping). These parameters may include the actuation intensity of the active damper and the setting of the filter cutoff frequency. The output of the entire process is quantified environmental disturbance values ​​(such as the effective value of vibration acceleration) and gradient changes (such as the temperature change rate), which provide precise input for subsequent environmental compensation. This series of analyses and controls, from single-unit to multi-unit, and from internal control to external environmental coupling, demonstrates the systematic and comprehensive nature of this calibration method.

[0080] Example 3: See Figure 4The dynamic optimization of environmental compensation parameters and the dynamic response adjustment of the control system are the next stages. This stage begins with in-depth time-series analysis of the obtained environmental disturbance values ​​and gradient changes on the workbench. The system first determines the time-series characteristics of these environmental disturbance changes, identifying their patterns of evolution over time, such as whether temperature exhibits periodic fluctuations or trend drifts, and whether vibrations have sudden characteristics. The system compares the currently monitored environmental disturbance values ​​with the original environmental disturbance baseline data recorded during equipment initialization or the last calibration, calculating the instantaneous value and cumulative effect of the environmental gradient (such as the rate of temperature change and the rate of vibration acceleration change) at each moment. Based on different control state zones of the workbench, such as the high-acceleration coarse positioning zone, the low-speed fine positioning zone, and the static exposure zone, the system defines differentiated environmental thresholds for each zone. These thresholds are set based on the different accuracy tolerances and dynamic response requirements of each zone. Based on these comparative analyses and zone thresholds, the system generates a preliminary set of environmental change parameters. This parameter set not only includes the numerical value of the disturbance but, more importantly, its status identifier (such as whether it exceeds limits) and the information of the region to which it belongs, providing input for subsequent differentiated compensation.

[0081] The system conducts an in-depth analysis of the initial environmental change parameter set to accurately quantify the impact of various environmental disturbances on the stability of the workbench control output. The analysis focuses on identifying the dynamic correlation between environmental disturbance variables such as temperature fluctuations and mechanical vibrations and the final output variables of the control system (e.g., the thrust current of the linear motor and the control voltage of the voice coil motor). This correlation can be characterized by an influence coefficient. The system calculates the control stability influence coefficient for different segments (i.e., different working areas or different time periods). This coefficient quantitatively describes the degree of control output instability caused by a unit change in environmental disturbance. Its calculation involves regression analysis of historical data or estimation based on a physical model, and can be expressed as:

[0082] ,

[0083] in: This represents the control stability influence coefficient in the z-zone. It is a dimensionless proportionality coefficient that quantifies the sensitivity of the control output to environmental disturbances within this zone. It represents the standard deviation of the control output (such as current or voltage), reflecting the fluctuation of the output. This represents the change in environmental disturbances (such as temperature or vibration acceleration) within the z-segment. The formula means that, within a certain operating range, the change in the degree of control output fluctuation caused by each unit change in environmental disturbance.

[0084] By systematically analyzing and calculating the control stability influence coefficients for each segment, and combining this with real-time updated environmental disturbance parameters, the system initiates dynamic adjustments to the balance of environmental compensation distribution. The goal of this adjustment is to match the compensation intensity with the strength of the environmental disturbance and its impact on control stability, avoiding over-compensation or under-compensation. For example, for segments with high influence coefficients, even if the absolute value of the environmental disturbance is small, a larger compensation weight may be allocated; while for segments with severe environmental disturbances but low influence coefficients, suppression strategies such as filtering may be adopted instead of direct compensation. Based on this, the system optimizes the environmental control data, including adjusting the compensation allocation ratio for each temperature compensation zone and modifying the gain of the vibration feedforward control, ultimately generating an optimized and more adaptive set of environmental optimization parameters. This parameter set will be directly applied to the control loop to offset or mitigate the adverse effects of environmental fluctuations.

[0085] Based on this optimized parameter set, the system's focus shifts to the controlled adsorption and diffusion phenomena on the stage surface. This process involves modeling the transmission characteristics of control input signals on the mechanical structure surface. The system extracts controlled adsorption data from the stage surface by monitoring the equivalent adsorption rate of the control input signal on surfaces with differentiated material properties. The adsorption rate reflects the speed at which control energy is absorbed by the structural surface. Simultaneously, considering external environmental factors such as time and temperature, the system uses a built-in physical mechanism model to infer the diffusion characteristics of the control input, i.e., the way control energy propagates and dissipates within the structure. Based on these monitoring and inferences, the system defines two key parameters: the adsorption rate coefficient and the diffusion rate coefficient. The adsorption rate coefficient characterizes the proportion of energy absorbed by the surface per unit control input per unit time, while the diffusion rate coefficient characterizes the diffusion capacity of control energy within the material per unit time and unit distance. The combination of these two coefficients generates a set of dynamic parameters for adsorption and diffusion that describes the dynamic transfer process of control energy.

[0086] The system analyzes the impact of the adsorption-diffusion dynamic parameter set on the actual control input rate and its spatial distribution. Based on the specific requirements of the photolithography process for controlling the concentration distribution on the stage surface, the system optimizes the proportional relationship between the control input rate path and the control input rate (the amplitude of the command). For example, in regions where a rapid establishment of the control concentration is expected, a higher input rate and denser path points may be used; while in regions where stability is required, a lower input rate and a holding strategy may be employed. By calculating the adjustment coefficient of the control concentration distribution trend, which is used to dynamically adjust the output allocation of the control law, the system ultimately generates a control concentration regulation result that defines a new spatiotemporal distribution blueprint for control energy.

[0087] The system analyzes the control concentration regulation results and adjusts the ratio between the control input rate and the input path accordingly. The system redistributes the control concentration distribution trend to better match the dynamic requirements of the stage's control energy distribution under high-speed motion and high acceleration conditions. This redistribution process closely integrates previously obtained adsorption-diffusion parameters with calculated adjustment coefficients, using an iterative optimization algorithm to find the optimal solution. Ultimately, the system yields a dynamic adjustment result, which includes a comprehensively updated set of control parameter configurations, such as the PID controller's gain scheduling table, the feedforward controller's parameter matrix, and filter settings. This new configuration aims to enable the control system not only to resist environmental interference but also to adapt to its own dynamic energy transfer characteristics, thus providing a parameter basis for achieving high-speed, high-precision motion control. The entire implementation process embodies a complete chain from environmental perception to compensation generation and then to adaptive adjustment of internal control parameters, highlighting the systematic nature and closed-loop feedback characteristics of this calibration method.

[0088] Example 4: See Figure 5 The system then enters the stage of monitoring and intervening in real-time fluctuations during the positioning process, ultimately generating an execution data table for actual calibration. This stage begins with the direct application of the dynamic adjustment results. The system continuously monitors the real-time position fluctuation rate and error offset of the worktable during the positioning process using deployed high-response monitoring equipment, primarily laser interferometers and capacitive sensors with high sampling rates. Position fluctuation rate refers to the comprehensive characterization of the amplitude and frequency of high-frequency oscillations around the target position, while error offset refers to the systematic deviation between the steady-state position and the target position. The monitoring process employs statistical process control methods, calculating the moving average and standard deviation in real time. The system sets a dynamic threshold based on the 3σ principle to automatically identify and eliminate transient anomalies caused by momentary sensor malfunctions, electromagnetic interference spikes, or minute debris, ensuring the validity of the data used for subsequent analysis. After this screening process, the system obtains a reliable set of position and error fluctuation data, which includes not only statistical values ​​of the fluctuations but also their trends over time.

[0089] The system conducts an in-depth analysis of the impact of the range of position and error fluctuation data on the final positioning accuracy of the workbench. This analysis is not arbitrary but relies on a positioning accuracy database established through long-term offline measurement and characterization. This database contains the actual positioning accuracy values ​​achieved under various known operating conditions and known fluctuation levels. The system uses data mining techniques to analyze the complex mapping relationship between position fluctuation rate, error offset, and final positioning accuracy, establishing a correlation model. Based on this model, the system can calculate the expected positioning accuracy value that the workbench may achieve under a specific monitored fluctuation range, thus obtaining positioning accuracy impact data. This data clearly reveals the quantitative correspondence between the system's tolerance to fluctuations of different amplitudes and its final accuracy performance, transforming abstract fluctuation quantities into understandable accuracy prediction values.

[0090] Based on this positioning accuracy impact data, the system initiates a dynamic adjustment procedure. Its goal is to adjust the control distribution path and error ratio within the target range of the control system. The control distribution path refers to the convergence strategy adopted by the control algorithm when approaching the target position, such as whether to use an aggressive high-gain rapid approximation or a conservative smooth and gradual approach. The error ratio defines the relative magnitude of errors of different frequency components allowed in the final positioning stage. The adjustment is based on the quantitative relationship between the aforementioned positioning accuracy impact data, real-time position, and error fluctuation range. The system reallocates the control input rate and error control range through a rule engine or optimization algorithm. For example, when high-frequency fluctuations are detected and the accuracy prediction model shows a significant impact on the final accuracy, the system may automatically reduce the control bandwidth to suppress high-frequency responses, while allowing a slightly larger steady-state error in exchange for better stability. All these adjustment strategies, new parameter thresholds, and triggering conditions are collected and organized into a structured anomaly intervention dataset. This dataset serves as the direct basis for the control system to make intelligent decisions in the face of dynamic fluctuations.

[0091] Based on this abnormal intervention dataset, the system performs a comprehensive analysis of the target position distribution of the workbench and calibration time. The system collects a large amount of historical position accuracy data recorded at different calibration time points (e.g., one hour, three hours, and five hours after equipment preheating), and organizes detailed information on the distribution of accuracy over time or position. The system analyzes the long-term trends of this data and categorizes the data, identifying patterns of accuracy changes with factors such as temperature accumulation and mechanical wear. Finally, it obtains workbench position accuracy distribution data, which profoundly reflects the correlation between positioning accuracy and variables such as time and environment, providing a data foundation for predicting the optimal calibration time.

[0092] Based on the workbench position accuracy distribution data, the system performs fine-tuning of the target calibration path parameters. The analysis focuses on finding the calibration time window and accuracy distribution characteristics corresponding to the optimal workbench position accuracy. The system compares a large amount of historical data under different calibration conditions to analyze the variation patterns of accuracy results under different calibration movement speeds, calibration dwell times, and control input gains. Based on these patterns, the system adjusts key operating conditions such as calibration speed, calibration dwell time, and control input magnitude, aiming to achieve the best calibration effect within a given time constraint. For example, it may be found that within a certain temperature range, reducing the calibration speed and appropriately extending the dwell time can achieve a more uniform accuracy distribution. Finally, the system obtains a set of optimized target calibration path parameters, which defines the detailed steps and parameters of calibration execution, as shown in Table 1.

[0093] Table 1: Workbench Positioning Calibration Data Table

[0094]

[0095] Based on the finalized target calibration path parameters, the system adjusts the specific execution conditions of the calibration system according to the current workbench status and operating environment parameters. The system precisely controls various variables during the calibration process, including the calibration movement time, the movement speed at each point, the dwell time at the target point, and the magnitude of the control input, and meticulously manages the coupling relationships between these variables. Based on the adjusted complete parameter set, the system initiates the real-time online calibration process. During calibration, the system strictly follows the instruction sequence in the generated calibration data table (as shown in the example above), while fine-tuning parameters based on real-time feedback. Finally, the system generates a complete workbench positioning calibration data table, which is integrated into the lithography machine's underlying control system as a benchmark for position control during actual production operation. The entire implementation process begins with real-time monitoring, and through predictive analysis, intelligent intervention, and trend learning, ultimately generates an executable calibration plan, forming a complete closed loop from perception to decision-making to execution, ensuring the workbench continuously maintains high-precision positioning capabilities.

[0096] Example 5: The implementation of the path parameter execution and real-time calibration phase begins with the in-depth application of the obtained workbench position accuracy distribution data, which reveals the variation patterns of accuracy in spatial and temporal dimensions. Based on this distribution data, the system determines a sequence of key points in the calibration path. These key points are not uniformly distributed but carefully selected, typically corresponding to areas with the most stringent accuracy requirements, such as the center point of the exposure field, wafer edge alignment points, or locations where historical data indicates a high probability of error. Based on this key point sequence, the system calculates the amount of error that may accumulate while moving along the preset path. This accumulated error includes not only static geometric position errors but also tracking errors caused by dynamic factors such as acceleration and deceleration. The calculation process comprehensively considers the total allowable calibration time constraint, a hard requirement that calibration must be completed without affecting the overall production cycle of the equipment. To achieve a balance between pursuing high accuracy and meeting time constraints, the system uses an optimization algorithm to dynamically adjust the distribution density between path points. In accuracy-sensitive areas, the distance between points is increased; while in areas with relatively stable accuracy, the number of points is appropriately reduced to save time. Through this dynamic density adjustment, the system generates a preliminary set of path adjustment parameters, which defines the position sequence of path points and the preliminary speed planning for movement between points.

[0097] Based on the generated initial path adjustment parameters, the system conducts a more in-depth analysis to evaluate the weight of the error at each point on the path on the final overall calibration accuracy. Not all errors at all path points have an equal impact on the final result. The system identifies regions that are particularly sensitive to errors, which may be prone to generating or amplifying errors due to changes in structural stiffness, steep thermal gradients, or being located at points of motion reversal. After identifying these sensitive regions, the system adjusts the specific positions of path points within these regions, for example, slightly deviating from the theoretically optimal position to avoid known mechanical resonance points, or adjusting the order in which path points are visited, such as scheduling the calibration of high-precision points during periods of better system thermal stability. After adjustment, the system recalculates the error distribution and cumulative total of the entire path and evaluates the effect of the adjustment. This process may involve multiple iterations, each aimed at further reducing the overall path error, ultimately resulting in a set of optimized path parameters. This set of parameters includes not only the position and order of points but also suggested movement speed curves between points, striving to minimize path error.

[0098] The system integrates and verifies the optimized path parameters with the target calibration time requirements. The purpose of this verification is to ensure that the optimized path meets accuracy requirements while completing all calibration actions within the specified time. Verification may be conducted through software simulation, simulating the workbench moving according to the optimized path parameters, calculating the actual time required, and comparing it with the target time. If the simulation shows a timeout, the system adjusts the path parameters, such as appropriately increasing the movement speed in non-sensitive areas or further optimizing the acceleration curve between points to reduce acceleration and deceleration time. Conversely, if time is sufficient, a brief settling time may be added in critical areas. The goal of this adjustment is to find a set of executable parameters that meets both accuracy requirements and time constraints. After verification and necessary fine-tuning, the system finally obtains a defined set of target calibration path parameters, which serves as the blueprint for subsequent real-time calibration execution.

[0099] Based on the finalized target calibration path parameters, the system initializes the calibration control system. The initialization process includes loading the previously generated workbench positioning calibration data table, which contains detailed instructions for each calibration point, such as target position, allowable error range, movement speed, dwell time, and control gain. Simultaneously, the system sets key calibration monitoring points in the control software. These points are typically located at critical positions or state transition points along the path to trigger high-frequency data acquisition, perform real-time error assessment, or execute control actions.

[0100] After the calibration process begins, the system starts acquiring real-time feedback data of the actual position of the worktable. This data primarily comes from high-precision position feedback devices, such as laser interferometers or grating ruler reading heads. The system compares the acquired actual position data with the target position data set in the target calibration path parameters to calculate the real-time position error value. This comparison and calculation process is performed by a high-speed digital signal processor or a dedicated motion control chip to ensure real-time processing and accuracy.

[0101] Based on the calculated real-time error value, the system dynamically adjusts the control input. This adjustment is not a simple proportional amplification, but rather the application of an error compensation algorithm. The compensation algorithm may comprehensively calculate the optimal control output correction based on the magnitude, direction, and trend of the error, as well as the current motion state (such as velocity and acceleration). For example, when a certain trend in error is detected, the algorithm may predict future errors and apply compensation in advance; or when the system is in a high-speed movement state, different compensation gain strategies may be adopted. During the compensation process, the system may also update certain non-critical items in the calibration path parameters online based on real-time performance, such as fine-tuning the dwell time or local velocity at a certain point, to cope with sudden disturbances encountered in actual operation or small deviations from model predictions.

[0102] The entire calibration process is under continuous monitoring. The system records detailed calibration log data, including timestamps, the actual location of each monitoring point, the target location, calculation errors, actual control outputs, and snapshots of environmental parameters (such as temperature). This log data is valuable for subsequent analysis of calibration effectiveness, problem diagnosis, and optimization of future calibration strategies. Finally, when all calibration points have been completed as planned, the system outputs the final calibration result data, confirming the completion of the calibration process. At this point, the workbench control system loads the latest, optimized calibration parameters, and the system enters a ready state, awaiting the execution of the actual lithography production task. The entire implementation process, starting with path planning, undergoing parameter optimization and verification, and ultimately achieving real-time closed-loop calibration, demonstrates the completeness and systematic nature of the method from design to execution.

[0103] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0104] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for positioning and calibrating a high-speed, high-precision stage of a lithography machine, characterized in that, The method includes the following steps: The position data and motion parameters of the lithography machine stage are collected. Based on the position data and motion parameters, the degree of matching between position error and motion stability is analyzed, and a set of position error control parameters is generated. Based on the position error control parameter set, the environmental interference value and gradient change of the workbench are extracted, the impact of environmental interference on positioning accuracy is analyzed, the environmental compensation parameters are adjusted, and an environmental optimization parameter set is generated. Based on the environmental optimization parameter set, the dynamic response of the workbench under high-speed motion is analyzed, the control input rate and feedback path distribution ratio are adjusted, the distribution trend and dynamic value of the control parameters are redistributed, and dynamic adjustment results are generated. Based on the dynamic adjustment results, the real-time position fluctuation rate and error offset during the workbench positioning process are extracted, the impact of the fluctuation range on the positioning accuracy is analyzed, the control distribution path and error ratio within the target range are dynamically adjusted, and an abnormal intervention dataset is generated. Based on the aforementioned abnormal intervention dataset, the distribution ratio and calibration time of the target positions on the workbench are analyzed, the parameters of the target calibration path are adjusted, and a workbench positioning calibration data table is generated.

2. The high-speed, high-precision stage positioning and calibration method for a lithography machine according to claim 1, characterized in that, The specific steps for obtaining the matching degree between the position error and motion stability are as follows: Based on the position data and motion parameters of the lithography machine stage, the position error parameters, motion speed data and acceleration data of the stage are extracted. A time window is set, and time points are selected to match the data. By comparing the data correlation and filtering the data, the position error parameters and motion speed data are obtained. Based on the position error parameters and motion speed data, the motion path is matched and verified, the difference between position error and motion speed is calculated, the position error distribution and motion speed distribution are corrected by combining acceleration changes, and the path parameters are adjusted by the influence of acceleration on the data to obtain the matching situation between position error and motion speed. Based on the matching of position error and motion speed, motion stability analysis is performed, motion stability analysis standards are set, and the error distribution under differentiated acceleration conditions is evaluated in combination with the dynamic changes of the worktable. Stability indicators are compared and acceleration conditions are optimized to obtain the degree of matching between position error and motion stability.

3. The high-speed, high-precision stage positioning and calibration method for a lithography machine according to claim 2, characterized in that, The specific steps for obtaining the position error control parameter set are as follows: Based on the degree of matching between the position error and motion stability, the error transmission and stability changes of the worktable under different operating conditions are analyzed, and the position error distribution of the worktable is weighted and calculated to obtain the preliminary error control requirements of the worktable. Based on the initial error control requirements of the worktables, the error balance between worktables is analyzed, the relationship between error transmission efficiency and load distribution between worktables is identified, the error control parameters of the worktables are corrected, the position error value of the worktables after adjustment is calculated, and the error control dataset between worktables is obtained. By combining the inter-workbench error control dataset with the motion stability matching results, the errors between the workbench are allocated, optimized, and matched to meet the required balance and stability requirements, thus obtaining the position error control parameter set.

4. The high-speed, high-precision stage positioning and calibration method for a lithography machine according to claim 3, characterized in that, The specific steps for obtaining the environmental interference value and gradient change of the workbench are as follows: Based on the set of position error control parameters, temperature data in the workbench environment is extracted, temperature points in each time period are filtered, and temperature fluctuations are analyzed in combination with the temperature change trends of the workbench at different positions to obtain temperature data in the workbench environment. Based on the temperature data in the workbench environment, the vibration value of each temperature point is calculated. By analyzing the relationship between temperature and vibration, the vibration change at each measurement point is identified. Combined with the structural parameters of the workbench, the vibration changes at different locations are compared to obtain vibration distribution and gradient distribution data. Based on the vibration distribution and gradient distribution data, the overall vibration distribution in the workbench environment is analyzed, the vibration gradient is optimized by combining temperature data, the impact of vibration changes on the workbench performance is analyzed, a stable vibration configuration is determined under differentiated operating conditions, and the workbench environment interference value and gradient change amount are obtained.

5. The high-speed, high-precision stage positioning and calibration method for a lithography machine according to claim 4, characterized in that, The specific steps for obtaining the environmental optimization parameter set are as follows: Based on the environmental interference value and gradient change of the workbench, the time series of environmental interference changes is determined, the current environmental interference value is compared with the original environmental interference data, the environmental gradient at each moment is analyzed, and corresponding thresholds are defined according to the workbench state partition to generate a preliminary environmental change parameter set. The preliminary environmental change parameter set is analyzed to analyze the impact of environmental disturbances on the stability of the workbench control output, identify the correlation between environmental disturbances and control output, and calculate the section control stability influence coefficient. By analyzing the influence coefficient of the control stability of the section, and combining it with the environmental disturbance change parameters, the environmental compensation distribution balance is adjusted, the environmental regulation data is optimized, and an environmental optimization parameter set is generated.

6. The high-speed, high-precision stage positioning and calibration method for a lithography machine according to claim 5, characterized in that, The specific steps for obtaining the dynamic adjustment result are as follows: Based on the environmental optimization parameter set, control adsorption data of the workbench surface is extracted, the adsorption rate of control input on the surface of differentiated materials is monitored, and the diffusion characteristics of control input are inferred by combining external environmental factors such as time and temperature. Adsorption and diffusion rate coefficients are defined, and an adsorption-diffusion dynamic parameter set is generated. The influence of the adsorption-diffusion dynamic parameter set on the control input rate and distribution is analyzed. Based on the requirements for controlling the concentration distribution on the stage surface, the ratio between the input rate path and the control input rate is optimized, the adjustment coefficient of the control concentration distribution trend is calculated, and the control concentration regulation result is generated. Analyze the control concentration regulation results, adjust the ratio of control input rate to input path, allocate the distribution trend of control concentration, and combine adsorption diffusion parameters and adjustment coefficients to obtain dynamic adjustment results.

7. The high-speed, high-precision stage positioning and calibration method for a lithography machine according to claim 6, characterized in that, The specific steps for obtaining the abnormal intervention dataset are as follows: Based on the dynamic adjustment results, the monitoring equipment monitors the position fluctuation rate and error offset in real time during the positioning process, identifies the fluctuation range, eliminates abnormal values ​​due to equipment failure, analyzes the average fluctuation rate of the data, and obtains position and error fluctuation data. The impact of the position and error fluctuation data range on the positioning accuracy of the worktable is analyzed. Using known positioning accuracy data, the relationship between position and error is analyzed, and the positioning accuracy under the differential fluctuation range is calculated to obtain the positioning accuracy impact data. Based on the positioning accuracy impact data, the control distribution path and error ratio within the target range are dynamically adjusted. Adjustments are made based on the relationship between the positioning accuracy impact data and the position and error fluctuation range. The control input rate and error control range are allocated, and an abnormal intervention dataset is generated.

8. The high-speed, high-precision stage positioning and calibration method for a lithography machine according to claim 7, characterized in that, The specific steps for obtaining the workbench positioning calibration data table are as follows: Based on the aforementioned abnormal intervention dataset, the distribution of target positions on the workbench and calibration time are analyzed. Position accuracy data at different calibration time points are collected, the accuracy time distribution is organized, the accuracy change trend is analyzed, and the data is classified to obtain the workbench position accuracy distribution data. Based on the workbench position accuracy distribution data, the target calibration path parameters are adjusted, the optimal calibration time and accuracy distribution of the workbench position accuracy are analyzed, and the accuracy changes under different calibration conditions are compared. The calibration speed, time, and control input operating conditions are adjusted to obtain the target calibration path parameters. Based on the target calibration path parameters, the calibration conditions are adjusted according to the current operating parameters, the calibration time, speed, and the relationship between the control input variables are controlled, and real-time calibration is performed according to the adjusted parameters to obtain the workbench positioning calibration data table.

9. The high-speed, high-precision stage positioning and calibration method for a lithography machine according to claim 8, characterized in that, The adjustment steps for the target calibration path parameters specifically include: Based on the workbench position accuracy distribution data, the key point sequence in the calibration path is determined, the cumulative path error is calculated based on the key point sequence, and the distribution density of path points is optimized in combination with calibration time constraints to generate preliminary path adjustment parameters. Based on the preliminary path adjustment parameters, the impact of path error on overall calibration accuracy is analyzed, error-sensitive areas are identified, the position and order of path points are adjusted, the path error is recalculated, and optimized path parameters are obtained. By combining the optimized path parameters with the target calibration time, the feasibility of the path is verified, the path parameters are adjusted to meet the accuracy requirements, and the target calibration path parameters are obtained.

10. The high-speed, high-precision stage positioning and calibration method for a lithography machine according to claim 9, characterized in that, The specific steps for performing the real-time calibration include: Based on the target calibration path parameters, initialize the calibration control system, load the calibration data table, and set calibration monitoring points; Real-time acquisition of workbench position feedback data, comparison with target position data, and calculation of real-time error value; The control input is dynamically adjusted based on the real-time error value, and the calibration path parameters are updated by applying an error compensation algorithm. The calibration process is continuously monitored, calibration logs are recorded, calibration result data is output, and the workbench positioning calibration is completed.

Images