Control system for processing precision parts of semiconductor equipment

Abstract

The invention relates to the technical field of numerical control systems, in particular to a control system for processing precision parts of semiconductor equipment, which comprises an information entropy monitoring module for acquiring torque measurement values based on real-time data of a milling cutter and a processing path, arranging the torque measurement values according to a time sequence and selecting continuous torque data segments by using a sliding window. According to the method, through real-time monitoring and dynamic adjustment of process parameters, the precision and efficiency of semiconductor equipment precision part machining are remarkably improved, the means of combining information entropy evaluation and dynamic optimization is implemented, precise control over the stability in the machining process is achieved, it is ensured that the machining parameters are always in the optimal state, and the machining precision is improved. According to the precise control, errors in the manufacturing process are reduced, the consistency and quality of parts are improved, topological errors in multi-axis machining are effectively reduced by predicting and compensating synchronous errors in real time, higher precision and production efficiency are ensured, and the strict requirements for high performance and high reliability are met.

Description

Technical Field

[0001] The present invention relates to the technical field of numerical control systems, and in particular to a control system for machining precision parts of semiconductor equipment. Background Art

[0002] CNC system is an important branch of automated machining technology and is widely used in various precision manufacturing fields. Such systems control machine tools through computer programming, allowing the manufacture of complex and high-precision parts. CNC system can accurately control the speed, position, synchronization and motion trajectory during the machining process, greatly improving manufacturing efficiency and product quality. In addition, CNC technology also supports flexible conversion from single-piece production to large-scale production, making it play a core role in industries such as aviation, automobiles, electronics, and medical.

[0003] Among them, the control system for the processing of precision parts of semiconductor equipment refers to the use of CNC technology to accurately control the processing process of parts in semiconductor manufacturing equipment. This system can achieve extremely high processing accuracy and repeatability, and is an indispensable technology for the production of high-performance semiconductor equipment. Through precise control and operation, it can ensure that semiconductor equipment achieves optimal performance during the production process and meet the growing demand for miniaturization, high performance and reliability of electronic products.

[0004] Existing CNC systems have limitations in automatically adjusting machining parameters and responding to mechanical behavior in real time, especially in torque monitoring and synchronization error compensation. These systems usually rely on preset programs and parameters and lack the flexibility to cope with changes in the production environment. Therefore, in the face of high-precision and complex machining requirements, common problems include unstable machining accuracy and low efficiency. In addition, for the synchronization errors of multi-axis machining, existing technologies fail to effectively predict and dynamically compensate, which will lead to reduced manufacturing accuracy and affect the overall product quality. In the semiconductor industry, a field with extremely high precision requirements, these technical limitations may lead to increased production costs and higher defective product rates. Summary of the invention

[0005] The purpose of the present invention is to solve the shortcomings in the prior art and to propose a control system for processing precision parts of semiconductor equipment.

[0006] In order to achieve the above-mentioned purpose, the present invention adopts the following technical solution: A control system for processing precision parts of semiconductor equipment includes:

[0007] The information entropy monitoring module collects torque measurement values ​​based on the real-time data of the milling cutter and the processing path, arranges them in chronological order, selects continuous torque data segments using a sliding window, counts the number of occurrences of multiple torque values ​​in each data segment, calculates the probability distribution of multiple torque values, performs entropy calculation on the probability distribution, and generates an information entropy stability index;

[0008] The parameter optimization module identifies the period of low stability in the torque measurement based on the information entropy stability index, adjusts the processing speed, feed rate and cutting depth, calculates the current torque measurement value for the multi-parameter combination, compares the information entropy index, selects the parameter combination with the minimum information entropy, and generates an optimized parameter configuration;

[0009] The synchronization error prediction module collects the improved multi-axis CNC motion data and the position information of the multi-axis based on the optimization parameter configuration, constructs the point cloud data, calculates the distance between multiple points in the point cloud, extracts the topological features, analyzes the change trend of the topological features, compares the topological features of the differentiated time periods, identifies the synchronization error, and predicts the synchronization error in the future time period, and generates the synchronization error prediction result;

[0010] Based on the synchronization error prediction result, the dynamic compensation module adjusts the motion parameters of the multi-axis, modifies the axis speed and the moving distance, updates the processing path, executes the processing instructions after the parameter adjustment, monitors the processing accuracy in real time, verifies the compensation effect, and obtains the compensation adjustment result.

[0011] As a further solution of the present invention, the step of obtaining the information entropy stability index is specifically as follows:

[0012] Extract torque measurements from a real-time milling cutter and machining path monitoring system, arrange the data in chronological order, and construct a time series torque dataset;

[0013] Selecting continuous moment data segments from the time series moment data set using a sliding window technique, counting the number of occurrences of multiple moment values ​​in each data segment, and generating moment frequency statistics of multiple data segments;

[0014] According to the moment frequency statistics of the multiple data segments, the probability distribution of multiple moment values ​​is calculated using the formula:

[0015]

[0016] Get the moment probability distribution;

[0017] Among them, p i represents the probability of the i-th moment value, indicating the frequency of the moment value appearing in the selected data segment, n i is the number of occurrences of the torque value in the data segment, reflecting the statistical quantity of the target torque value, and ∑n is the sum of all measurements in the data segment;

[0018] The moment probability distribution is used to calculate information entropy to measure the randomness and uncertainty of data using the formula:

[0019] H=-∑(p i log(p i ))

[0020] The information entropy stability index is generated by calculating the information entropy by taking the negative logarithm of the weighted probability;

[0021] Where H represents information entropy, which is used to measure the randomness and uncertainty of moment data and is calculated by the negative logarithm sum of the moment value probability, p i is the probability of the i-th moment value, log(p i ) is the logarithm of the probability of the i-th moment value and is used to calculate each contribution in the information entropy.

[0022] As a further solution of the present invention, the step of obtaining the optimization parameter configuration is specifically:

[0023] Based on the information entropy stability index, identify the torque measurement period with low stability, verify the start and end positions of the unstable period by analyzing the torque data fluctuation, and generate the identified unstable period;

[0024] For the identified unstable period, the machining parameter adjustment rule is called to optimize the machining speed, feed rate and cutting depth, and the current parameter set is formed after combined adjustment, thereby generating an adjusted multi-parameter experimental combination;

[0025] The current torque measurement value under the adjusted multi-parameter experimental combination is calculated using the formula:

[0026]

[0027] Calculate the torque measurement value under each parameter combination and generate the torque measurement value under multiple parameter combinations;

[0028] Among them, T i represents the torque measurement value under parameter combination i, v i Represents the adjustment value of the processing speed, f i Represents the feed rate adjustment value, d i represents the adjustment value of cutting depth, k represents the adjustment coefficient for the combination of machining parameters, which is used to weigh the influence of multiple parameters on torque measurement, and a represents the additional parameter that affects the torque measurement value;

[0029] The torque measurement values ​​under the multi-parameter combination are used to compare the information entropy index, select the parameter combination with the smallest information entropy, and generate an optimized parameter configuration.

[0030] As a further solution of the present invention, the step of acquiring the point cloud data is specifically as follows:

[0031] Starting a multi-axis CNC machine tool, collecting the dynamics of the machine tool during the machining process according to the optimized parameter configuration, recording the motion data of the multi-axis, and generating a CNC motion data set;

[0032] Using the position information in the numerical control motion data set, a three-dimensional trajectory model of the machine tool motion is constructed, the data represents the position changes of multiple axes during the machining process, and three-dimensional trajectory data is generated;

[0033] The three-dimensional trajectory data is analyzed, the spatial distance between multiple points in the point cloud is calculated, the motion accuracy and stability of the machine tool are described by the distance, and the point cloud data is generated.

[0034] As a further solution of the present invention, the step of obtaining the synchronization error prediction result is specifically:

[0035] Extracting key topological features from the point cloud data, the features reflecting the spatial relationship and dynamic changes of multiple parts of the machine tool during the machining process, evaluating the movement synchronization of the parts by measuring the distance and angle difference between the points, and generating topological feature analysis results;

[0036] By comparing the topological feature analysis results in differentiated time periods, analyzing and determining the changing trend of the synchronization error of the multi-component movement, identifying the synchronization deviation of the components in the critical period based on the changing trend, and generating the synchronization error time trend analysis result;

[0037] Based on the synchronization error time trend analysis results, the prediction model is used to predict the synchronization error in the future time period to ensure that the model matches the operating conditions and data characteristics of the machine tool target. The formula is:

[0038] E future =αE current +βΔE+γΣE past

[0039] Generate synchronization error prediction results;

[0040] Among them, E future is the future prediction synchronization error, E current is the current synchronization error, ΔE is the synchronization error change rate, ΣE past is the synchronization error accumulation, α represents the weight of the current error, β represents the influence of the error change rate, and γ represents the influence of the error accumulation.

[0041] As a further solution of the present invention, the step of obtaining the compensation adjustment result is specifically:

[0042] According to the synchronization error prediction result, optimizing the operation of the machine tool based on the prediction error data, adjusting the motion parameters of the multi-axis CNC machine tool, and generating an adjusted motion parameter configuration;

[0043] Applying the adjusted motion parameter configuration, updating the machining path and re-issuing machining instructions, monitoring real-time data during the machining process through the numerical control system, and generating real-time machining monitoring records;

[0044] According to the real-time processing monitoring record, the influence of the adjusted parameter configuration on the processing accuracy is evaluated using the formula:

[0045] P new =P current +ΔP

[0046] generating compensation adjustment results;

[0047] Among them, P new Indicates the adjusted machining accuracy, P current is the machining accuracy before adjustment. ΔP quantifies the change in machining accuracy caused by parameter adjustment and is a direct indicator to measure the adjustment effect.

[0048] Compared with the prior art, the advantages and positive effects of the present invention are:

[0049] In the present invention, the precision and efficiency of the processing of precision parts of semiconductor equipment are significantly improved by real-time monitoring and dynamic adjustment of process parameters. The combination of information entropy evaluation and dynamic optimization is implemented to achieve precise control of stability during the processing process, ensuring that the processing parameters are always in the optimal state. This precise control reduces errors in the manufacturing process and improves the consistency and quality of parts. By predicting and compensating for synchronization errors in real time, the topological error in multi-axis processing is effectively reduced, ensuring higher precision and production efficiency, and meeting the strict requirements for high performance and high reliability. BRIEF DESCRIPTION OF THE DRAWINGS

[0050] Figure 1 is a system flow chart of the present invention;

[0051] Figure 2 This is a flow chart of the steps for obtaining the information entropy stability index of the present invention;

[0052] Figure 3 A flowchart of the acquisition steps for optimizing parameter configuration of the present invention;

[0053] Figure 4 This is a flow chart of the steps for obtaining point cloud data of the present invention;

[0054] Figure 5 A flowchart of the steps for obtaining the synchronization error prediction result of the present invention;

[0055] Figure 6 The figure is a flow chart of the steps for obtaining the compensation adjustment results of the present invention. DETAILED DESCRIPTION

[0056] In order to make the purpose, technical solution and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the present invention.

[0057] In the description of the present invention, it should be understood that the terms "length", "width", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside" and the like indicate positions or positional relationships based on the positions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as limiting the present invention. In addition, in the description of the present invention, "multiple" means two or more, unless otherwise clearly and specifically defined.

[0058] Embodiment 1

[0059] See also Figure 1 , a control system for processing precision parts of semiconductor equipment includes:

[0060] The information entropy monitoring module collects torque measurement values ​​based on the real-time data of the milling cutter and the processing path, arranges them in chronological order, selects continuous torque data segments using a sliding window, counts the number of occurrences of multiple torque values ​​in each data segment, calculates the probability distribution of multiple torque values, performs entropy calculation on the probability distribution, and generates an information entropy stability index;

[0061] The parameter optimization module identifies the period of low stability in torque measurement based on the information entropy stability index, adjusts the processing speed, feed rate and cutting depth, calculates the current torque measurement value for multiple parameter combinations, compares the information entropy index, selects the parameter combination with the smallest information entropy, and generates the optimized parameter configuration;

[0062] The synchronization error prediction module collects the improved multi-axis CNC motion data and the position information of the multi-axis based on the optimized parameter configuration, constructs the point cloud data, calculates the distance between multiple points in the point cloud, extracts the topological features, analyzes the changing trend of the topological features, compares the topological features of the differentiated time periods, identifies the synchronization error, and predicts the synchronization error in the future time period to generate the synchronization error prediction result;

[0063] Based on the synchronization error prediction results, the dynamic compensation module adjusts the motion parameters of multiple axes, modifies the axis speed and moving distance, updates the processing path, executes the processing instructions after parameter adjustment, monitors the processing accuracy in real time, verifies the compensation effect, and obtains the compensation adjustment results.

[0064] The information entropy stability index is specifically the probability distribution of the torque data segment and the entropy value calculation result. The optimization parameter configuration includes processing speed, feed rate, and cutting depth. The synchronization error prediction result includes the predicted error value, error trend analysis, and topological change characteristics. The compensation adjustment result includes the adjusted motion parameters, updated path data, and verified accuracy indicators.

[0065] See also Figure 2 , the specific steps for obtaining the information entropy stability index are:

[0066] Extract torque measurements from a real-time milling cutter and machining path monitoring system, arrange the data in chronological order, and construct a time series torque dataset;

[0067] The torque data is extracted from the real-time monitoring system of the milling cutter and the processing path, and the data is sorted in time series to form a time-based torque data set. The data set not only reflects the torque changes of the milling cutter during the processing process, but also retains the timestamp information for subsequent analysis. Through the data set, the trend of torque changes over time can be monitored, which provides a data basis for the use of sliding window technology. Each measured value of the torque is stored in time sequence to ensure the continuity and integrity of the data. This process ensures the accuracy and real-time nature of data acquisition and generates a time series torque data set.

[0068] The sliding window technique is used to select continuous moment data segments in the time series moment data set, and the number of occurrences of multiple moment values ​​in each data segment is counted to generate moment frequency statistics of multiple data segments;

[0069] The sliding window technology is applied to process the time series torque data set, and continuous data segments are selected for analysis. This technology allows dynamic monitoring of the changes in torque data. The window slides on the data set to capture the torque value of each segment and count the number of times multiple torque values ​​appear. This method assists in analyzing the distribution of torque in the differentiated processing stage. The statistical frequency distribution can reflect the performance of the machine under differentiated operating conditions, generate torque frequency statistics for multiple data segments, and provide the necessary preliminary data for the calculation of probability distribution.

[0070] According to the moment frequency statistics of multiple data segments, the probability distribution of multiple moment values ​​is calculated using the formula:

[0071]

[0072] Get the moment probability distribution;

[0073] Among them, p i represents the probability of the i-th moment value, indicating the frequency of the moment value appearing in the selected data segment, n i is the number of occurrences of the torque value in the data segment, reflecting the statistical quantity of the target torque value, and ∑n is the sum of all measurements in the data segment;

[0074] formula:

[0075]

[0076] Detailed explanation of the formula and the process of formula calculation and derivation:

[0077] The formula is used to calculate the probability of the target torque value appearing within a certain time window. i represents the probability of the target torque value, n i represents the number of times the moment value appears in the window, and ∑n is the total number of times all moment values ​​appear in the window. To ensure accuracy, n i and ∑n are directly monitored by the real-time data acquisition system.

[0078] Refer to actual data: In a milling process, assume that a time window of 5 minutes is selected for data acquisition, and the total number of torque measurements in the window is 500. Assume that among these 500 measurements, the number of measurements with a torque value of 10 Nm is 75.

[0079] The calculation process is as follows:

[0080] First determine n i =75,∑n=500.

[0081] Then substitute into the formula to calculate p i Values:

[0082]

[0083] The results show that within the monitoring window, the probability of a torque value of 10 N·m is 15%, which reflects the commonness of the torque value within this time period and provides basic data for iterative analysis of information entropy calculation.

[0084] Use moment probability distribution to calculate information entropy and measure the randomness and uncertainty of data using the formula:

[0085] H=-∑(p i log(p i ))

[0086] The information entropy stability index is generated by calculating the information entropy by taking the negative logarithm of the weighted probability;

[0087] Where H represents information entropy, which is used to measure the randomness and uncertainty of moment data and is calculated by the negative logarithm sum of the moment value probability, p i is the probability of the i-th moment value, log(p i ) is the logarithm of the probability of the i-th moment value and is used to calculate each contribution in the information entropy.

[0088] formula:

[0089] H=-∑(p i log(p i ))

[0090] The benefits of the formula are: the information entropy formula can quantify the randomness and uncertainty of the torque data, providing an indicator to measure the stability of the machining process. Through calculation, irregular fluctuations in the data can be identified, which is helpful for machine maintenance and parameter optimization.

[0091] Detailed explanation of the formula and the process of formula calculation and derivation:

[0092] Set the moment probability distribution known from the previous step to have three probability values ​​p 1 =0.15, p 2 =0.35, p 3 = 0.50. These probability values ​​reflect the frequency of occurrence of the three differentiated torque values ​​within the target time window.

[0093] The calculation process of information entropy is as follows:

[0094] The calculation is done using the natural logarithm, calculating the logarithm of the multiple probability values ​​one by one and multiplying them by their probabilities:

[0095] -(p 1 log(p 1 )+p 2 log(p 2 )+p 3 log(p 3 ))

[0096] =-(0.15log(0.15)+0.35log(0.35)+0.50log(0.50))

[0097] =-(-0.274+(-0.360)+(-0.301))=0.935

[0098] The results show that the total information entropy of the torque value in the monitored window is 0.935, which is between 0 (completely certain) and 1 (completely random). The higher the value of information entropy, the greater the uncertainty of the system, indicating that the machining process needs to be adjusted to reduce uncertainty.

[0099] See also Figure 3 , the specific steps for obtaining the optimization parameter configuration are:

[0100] Based on the information entropy stability index, the torque measurement period with low stability is identified. By analyzing the torque data fluctuation, the start and end positions of the unstable period are verified, and the identified unstable period is generated;

[0101] Based on the information entropy stability index, the data fluctuation is analyzed and the unstable periods are eliminated. By performing time series analysis on the data, the torque data in each time period are grouped to obtain the torque fluctuation and identify the intervals with unstable trends, including the fluctuation amplitude and change rate of the data points. The qualified time periods are screened by setting thresholds and the start and end positions of the unstable periods are verified. After data cleaning, it is convenient to effectively optimize the processing parameters and generate identified unstable periods.

[0102] For the identified unstable period, the machining parameter adjustment rules are called to optimize the machining speed, feed rate and cutting depth. After combined adjustment, the current parameter set is formed, and the adjusted multi-parameter experimental combination is generated;

[0103] After identifying the unstable period, adjusting the machining parameters including machining speed, feed rate and cutting depth is the key. These parameters affect the stability of the torque by adjustment. The machining speed is adjusted from 50 to 200m / min, the feed rate is adjusted from 0.1 to 1m / rev, and the cutting depth is from 1 to 5mm. By combining differentiated parameter settings, actual machining tests are carried out to obtain the optimal parameter combination. The adjustment is optimized based on previous machining experience and real-time feedback on site to ensure that each parameter combination can test its effect during the unstable period, thereby capturing the most effective parameter setting for reducing torque fluctuations.

[0104] The current torque measurement value under the adjusted multi-parameter experimental combination is calculated using the formula:

[0105]

[0106] Calculate the torque measurement value under each parameter combination and generate the torque measurement value under multiple parameter combinations;

[0107] Among them, T i represents the torque measurement value under parameter combination i, v i Represents the adjustment value of the processing speed, f i Represents the feed rate adjustment value, d i represents the adjustment value of cutting depth, k represents the adjustment coefficient for the combination of machining parameters, which is used to weigh the influence of multiple parameters on torque measurement, and a represents the additional parameter that affects the torque measurement value;

[0108] formula:

[0109]

[0110] The benefit of the formula is that by introducing the combination of speed, feed rate and cutting depth, it can more fully reflect the changes in the torque measurement, thereby providing a more accurate measurement result.

[0111] Detailed explanation of the formula and the process of formula calculation and derivation:

[0112] In the formula, T i represents the torque measurement value under parameter combination i; v i 、f i and d i Respectively represent the speed, feed rate and cutting depth under the target machining conditions. The values ​​are obtained through experimental monitoring, with a speed range of 50-200m / min, a feed rate of 0.1-1m / rev, and a cutting depth of 1-5mm. Set the speed v i =100m / min, feed rate f i =0.5m / rev, cutting depth d i =3mm, adjustment coefficient k=10, additional parameter a=1.

[0113] After substituting into the formula:

[0114]

[0115] The results show that the measured value of the torque is 3.87, which indicates that the torque has good stability under the target parameter combination, providing a data basis for subsequent optimization.

[0116] The torque measurement values ​​under multiple parameter combinations are used to compare the information entropy index, and the parameter combination with the smallest information entropy is selected to generate the optimized parameter configuration.

[0117] By calculating and comparing the torque measurement values ​​under multiple parameter combinations, the parameter combination with the smallest information entropy is selected for optimization. This process involves recalculating the torque values ​​under multiple combinations of parameters and comparing the information entropy. For example, by monitoring the torque measurement values ​​corresponding to multiple combinations and calculating their corresponding information entropy, this series of data analysis can determine which parameter combination can effectively reduce the instability of the torque and select the parameter combination with the smallest information entropy. This is not only based on theoretical calculations, but also needs to be verified in the actual processing process to ensure that the selected parameter combination can indeed improve the stability and efficiency of the processing and generate an optimized parameter configuration. This configuration will directly affect the subsequent production efficiency and product quality.

[0118] See also Figure 4 , the specific steps for obtaining point cloud data are:

[0119] Start the multi-axis CNC machine tool, collect the dynamics of the machine tool during the processing according to the optimized parameter configuration, record the motion data of the multi-axis, and generate the CNC motion data set;

[0120] The optimized parameter configuration is started in the multi-axis monitoring center of the CNC machine tool to monitor the response of the multi-axis and the operating status of the machine tool in actual operation. The data including the speed, position and acceleration of the axis are recorded. The data set will directly reflect the dynamic performance of the machine tool during the processing and provide basic data for the next step of analysis. The real-time data acquisition system is used to ensure the real-time and accuracy of the obtained data. The monitoring data is preliminarily processed by the digital signal processor to filter out noise and interference to ensure the purity and availability of the data. The motion data of each axis will be recorded by the system and integrated into a complete data set, which reflects the behavior of the machine tool under the target operating conditions.

[0121] Using the position information in the CNC motion data set, a three-dimensional trajectory model of the machine tool motion is constructed. The data is expressed in the position changes of multiple axes during the machining process, and three-dimensional trajectory data is generated;

[0122] Using data obtained from the monitoring system of the CNC machine tool, the data records the position information of the multiple axes of the machine tool during the processing process. The data includes the X, Y, and Z coordinates of the axes. The coordinate information will be used to construct the motion trajectory model of the machine tool. The point cloud is constructed through the data captured by the position sensor. This process involves converting continuous position data into a discrete point set to form a three-dimensional trajectory view. Each point represents the spatial position of the machine tool at the target time point. The collected points can intuitively display the trajectory and range of motion of the machine tool.

[0123] Analyze the 3D trajectory data, calculate the spatial distance between multiple points in the point cloud, describe the motion accuracy and stability of the machine tool through the distance, and generate point cloud data.

[0124] By analyzing the data extracted from the 3D trajectory dataset, the data reveals the motion characteristics of the machine tool during operation. The spatial distance between points is calculated through measurement technology, which ensures the accuracy and reliability of the measurement. The distance between each pair of points is calculated using the Euclidean distance formula, which not only improves the efficiency of data processing, but also can more accurately capture the slight changes in the movement of the machine tool, which is crucial for optimizing machine tool performance and preventing faults. Through calculation, engineers can understand the performance of the machine tool in actual work, as well as the interaction and motion coordination between multiple components. The generated point cloud data provides a scientific basis for the iterative adjustment and performance evaluation of the machine tool.

[0125] See also Figure 5 , the steps for obtaining the synchronization error prediction results are as follows:

[0126] Extract key topological features from point cloud data. The features reflect the spatial relationship and dynamic changes of multiple parts of the machine tool during the processing. By measuring the distance and angle differences between points, the movement synchronization of the parts is evaluated to generate topological feature analysis results.

[0127] Extract key topological features from point cloud data. In this step, the motion synchronization of components is evaluated by measuring the distance and angle difference between points, and the topological feature analysis results are generated. This process involves analyzing the spatial coordinates of each point to determine the geometric relationship and interaction between points. By measuring the actual distance between points and calculating the angle difference between each other, it is possible to effectively evaluate how the machine tool components move in space. In addition, by analyzing the point cloud data, any errors or deviations can be identified, which is crucial for adjusting the accuracy of the machine tool and improving its synchronization. The point cloud data analysis results can intuitively display the dynamic performance of the machine tool during the machining process and provide a reliable basis for error prediction.

[0128] By comparing the topological feature analysis results in different time periods, the changing trend of the synchronization error of the multi-component movement is analyzed and determined. Based on the changing trend, the synchronization deviation of the components in the critical period is identified, and the synchronization error time trend analysis results are generated.

[0129] By comparing the topological feature analysis results within differentiated time periods, the changing trend of the synchronization error of multi-component movement is analyzed and determined. This step involves data comparison and time series analysis. First, it is necessary to extract key data from the topological feature analysis results, including the position information of each component at differentiated time points and the relative movement speed and direction between each other. Then, statistical methods are used to compare the changes in data within differentiated time periods to identify synchronization errors and their development trends. This process requires not only data processing technology, but also analysis of machine tool operation and performance to ensure the accuracy of data comparison and the reliability of analysis results. The analysis results of this step can determine in which time periods the machine tool's synchronization performance is poor, providing a basis for subsequent preventive measures and adjustment strategies.

[0130] Based on the results of the synchronization error time trend analysis, the prediction model is used to predict the synchronization error in the future time period to ensure that the model matches the operating conditions and data characteristics of the machine tool target. The formula is:

[0131] E future =αE current +βΔE+γΣE past

[0132] Generate synchronization error prediction results;

[0133] Among them, E future is the future prediction synchronization error, E currentis the current synchronization error, ΔE is the synchronization error change rate, ΣE past is the synchronization error accumulation, α represents the weight of the current error, β represents the influence of the error change rate, and γ represents the influence of the error accumulation.

[0134] formula:

[0135] E future =αE current +βΔE+γΣE past

[0136] The benefit of the formula is that it can predict future error trends based on historical and current synchronous error data, providing proactive support for machine tool maintenance and error adjustment.

[0137] Detailed explanation of the formula and the process of formula calculation: First, set the known parameters, E current Indicates that the currently measured synchronization error is 0.02, ΔE indicates that the synchronization error change rate is 0.005, ΣE past Indicates that the accumulated synchronization error in the past is 0.15; the parameters α, β, and γ are set to 0.5, 0.3, and 0.2 respectively. Based on these values, the calculation formula is:

[0138] E future =0.5×0.02+0.3×0.005+0.2×0.15=0.01+0.0015+0.03

[0139] =0.0415

[0140] The results show that if the current trend continues, the future synchronization error is expected to be 0.0415, which suggests that the machine tool needs to be adjusted to reduce the error and maintain efficient and accurate machining performance.

[0141] See also Figure 6 , the specific steps for obtaining the compensation adjustment results are:

[0142] According to the synchronization error prediction result, the operation of the machine tool is optimized based on the prediction error data, the motion parameters of the multi-axis CNC machine tool are adjusted, and the adjusted motion parameter configuration is generated;

[0143] According to the synchronization error prediction results, engineers adjusted the axis speed and moving distance of the CNC machine tool. This adjustment is based on error analysis, which refers to the performance of the machine tool under differentiated operating conditions and the error data recorded in the previous production process. During the adjustment process, the effect of parameter adjustment is monitored through a real-time feedback system. Real-time data is collected by sensors and transmitted to the central processing unit through a data interface. Each parameter adjustment is based on the previous processing results to ensure that each modification is accurate. In this way, engineers can gradually reduce synchronization errors and improve processing accuracy, thereby achieving the purpose of optimizing the production process. The purpose of this adjustment is to reduce errors in production and improve product quality. Through this series of adjustments, engineers have formulated the current motion parameter configuration, laying the foundation for the next step of processing instruction update and execution.

[0144] Apply the adjusted motion parameter configuration, update the processing path and re-issue the processing instructions, monitor the real-time data during the processing through the CNC system, and generate real-time processing monitoring records;

[0145] The current motion parameter configuration is applied to the CNC machine tool, the processing path is updated and the current processing instructions are issued. This process involves reprogramming the processing program and optimizing the path. Engineers redraw the processing path according to the current parameter settings to ensure the optimization of the path. At the same time, the real-time monitoring system during the processing process continuously tracks the performance of the machine tool and records each motion data and processing results. The monitoring system uses high-speed cameras and sensors to capture the situation during the processing in real time. These data are used to iteratively analyze the motion stability and processing accuracy of the machine tool. In this way, engineers can get instant feedback on the processing process and make adjustments as needed to ensure that the processing quality meets the predetermined standards.

[0146] According to the real-time processing monitoring records, the influence of the adjusted parameter configuration on the processing accuracy is evaluated using the formula:

[0147] P new =P current +ΔP

[0148] generating compensation adjustment results;

[0149] Among them, P new Indicates the adjusted machining accuracy, P current is the machining accuracy before adjustment. ΔP quantifies the change in machining accuracy caused by parameter adjustment and is a direct indicator to measure the adjustment effect.

[0150] formula:

[0151] P new =P current +ΔP

[0152] The benefit of the formula is that it provides a direct calculation method to predict the machining accuracy after adjusting the parameters, and can make real-time adjustments based on changes in the machining process to achieve the optimal machining effect.

[0153] Detailed explanation of the formula and the process of formula calculation and derivation:

[0154] Set the current processing precision P current is 0.005mm, and the expected adjustment precision ΔP can be 0.001mm. The current machining precision P is calculated according to the formula new It will be: P new =0.005+0.001=0.006mm.

[0155] The results show that the machining accuracy is expected to be improved from 0.005mm to 0.006mm through parameter adjustment, which is of key significance for improving product quality and reducing material waste.

[0156] The above are only preferred embodiments of the present invention and are not intended to limit the present invention in other forms. Any technician familiar with the profession may use the technical contents disclosed above to change or modify them into equivalent embodiments with equivalent changes and apply them to other fields. However, any simple modification, equivalent change and modification made to the above embodiments based on the technical essence of the present invention without departing from the technical solution of the present invention still falls within the protection scope of the technical solution of the present invention.

Claims

1. A control system for processing precision parts of semiconductor equipment, characterized in that: The system comprises: The information entropy monitoring module collects torque measurement values ​​based on the real-time data of the milling cutter and the processing path, arranges them in chronological order, selects continuous torque data segments using a sliding window, counts the number of occurrences of multiple torque values ​​in each data segment, calculates the probability distribution of multiple torque values, performs entropy calculation on the probability distribution, and generates an information entropy stability index; The parameter optimization module identifies the period of low stability in the torque measurement based on the information entropy stability index, adjusts the processing speed, feed rate and cutting depth, calculates the current torque measurement value for the multi-parameter combination, compares the information entropy index, selects the parameter combination with the minimum information entropy, and generates an optimized parameter configuration; The synchronization error prediction module collects the improved multi-axis CNC motion data and the position information of the multi-axis based on the optimization parameter configuration, constructs the point cloud data, calculates the distance between multiple points in the point cloud, extracts the topological features, analyzes the change trend of the topological features, compares the topological features of the differentiated time periods, identifies the synchronization error, and predicts the synchronization error in the future time period, and generates the synchronization error prediction result; Based on the synchronization error prediction result, the dynamic compensation module adjusts the motion parameters of the multi-axis, modifies the axis speed and the moving distance, updates the processing path, executes the processing instructions after the parameter adjustment, monitors the processing accuracy in real time, verifies the compensation effect, and obtains the compensation adjustment result.

2. The control system for processing precision parts of semiconductor equipment according to claim 1, characterized in that: The steps for obtaining the information entropy stability index are specifically as follows: Extract torque measurements from a real-time milling cutter and machining path monitoring system, arrange the data in chronological order, and construct a time series torque dataset; Selecting continuous moment data segments from the time series moment data set using a sliding window technique, counting the number of occurrences of multiple moment values ​​in each data segment, and generating moment frequency statistics of multiple data segments; According to the moment frequency statistics of the multiple data segments, the probability distribution of multiple moment values ​​is calculated using the formula: Get the moment probability distribution; Among them, p i represents the probability of the i-th moment value, indicating the frequency of the moment value appearing in the selected data segment, n i is the number of occurrences of the torque value in the data segment, reflecting the statistical quantity of the target torque value, and ∑n is the sum of all measurements in the data segment; The moment probability distribution is used to calculate information entropy to measure the randomness and uncertainty of data using the formula: H=-∑(p i log(p i )) The information entropy stability index is generated by calculating the information entropy by taking the negative logarithm of the weighted probability; Where H represents information entropy, which is used to measure the randomness and uncertainty of moment data and is calculated by the negative logarithm sum of the moment value probability, p i is the probability of the i-th moment value, log(p i ) is the logarithm of the probability of the i-th moment value and is used to calculate each contribution in the information entropy.

3. The control system for processing precision parts of semiconductor equipment according to claim 2, characterized in that: The steps for obtaining the optimization parameter configuration are specifically as follows: Based on the information entropy stability index, identify the torque measurement period with low stability, verify the start and end positions of the unstable period by analyzing the torque data fluctuation, and generate the identified unstable period; For the identified unstable period, the machining parameter adjustment rule is called to optimize the machining speed, feed rate and cutting depth, and the current parameter set is formed after combined adjustment, thereby generating an adjusted multi-parameter experimental combination; The current torque measurement value under the adjusted multi-parameter experimental combination is calculated using the formula: Calculate the torque measurement value under each parameter combination and generate the torque measurement value under multiple parameter combinations; Among them, T i represents the torque measurement value under parameter combination i, v i Represents the adjustment value of the processing speed, f i Represents the feed rate adjustment value, d i represents the adjustment value of cutting depth, k represents the adjustment coefficient for the combination of machining parameters, which is used to weigh the influence of multiple parameters on torque measurement, and a represents the additional parameter that affects the torque measurement value; The torque measurement values ​​under the multi-parameter combination are used to compare the information entropy index, select the parameter combination with the smallest information entropy, and generate an optimized parameter configuration.

4. The control system for processing precision parts of semiconductor equipment according to claim 3, characterized in that: The steps for obtaining the point cloud data are specifically as follows: Starting a multi-axis CNC machine tool, collecting the dynamics of the machine tool during the machining process according to the optimized parameter configuration, recording the motion data of the multi-axis, and generating a CNC motion data set; Using the position information in the numerical control motion data set, a three-dimensional trajectory model of the machine tool motion is constructed, the data represents the position changes of multiple axes during the machining process, and three-dimensional trajectory data is generated; The three-dimensional trajectory data is analyzed, the spatial distance between multiple points in the point cloud is calculated, the motion accuracy and stability of the machine tool are described by the distance, and the point cloud data is generated.

5. The control system for processing precision parts of semiconductor equipment according to claim 4, characterized in that: The steps for obtaining the synchronization error prediction result are specifically as follows: Extract key topological features from the point cloud data, the features reflecting the spatial relationship and dynamic changes of multiple parts of the machine tool during the machining process, evaluate the movement synchronization of the parts by measuring the distance and angle difference between the points, and generate topological feature analysis results; By comparing the topological feature analysis results in differentiated time periods, analyzing and determining the changing trend of the synchronization error of the multi-component movement, identifying the synchronization deviation of the components in the critical period based on the changing trend, and generating the synchronization error time trend analysis result; Based on the synchronization error time trend analysis results, the prediction model is used to predict the synchronization error in the future time period to ensure that the model matches the operating conditions and data characteristics of the machine tool target. The formula is: E future =αE current +βΔE+γΣE past Generate synchronization error prediction results; Among them, E future is the future prediction synchronization error, E current is the current synchronization error, ΔE is the synchronization error change rate, ΣE past is the synchronization error accumulation, α represents the weight of the current error, β represents the influence of the error change rate, and γ represents the influence of the error accumulation.

6. The control system for processing precision parts of semiconductor equipment according to claim 5, characterized in that: The steps for obtaining the compensation adjustment result are specifically as follows: According to the synchronization error prediction result, optimizing the operation of the machine tool based on the prediction error data, adjusting the motion parameters of the multi-axis CNC machine tool, and generating an adjusted motion parameter configuration; Applying the adjusted motion parameter configuration, updating the machining path and re-issuing machining instructions, monitoring real-time data during the machining process through the numerical control system, and generating real-time machining monitoring records; According to the real-time processing monitoring record, the influence of the adjusted parameter configuration on the processing accuracy is evaluated using the formula: P new =P current +ΔP generating compensation adjustment results; Among them, P new Indicates the adjusted machining accuracy, P current is the machining accuracy before adjustment. ΔP quantifies the change in machining accuracy caused by parameter adjustment and is a direct indicator to measure the adjustment effect.

Images