Method for milling high-precision sealing surface of machining center by using numerical control gantry

Abstract

The invention discloses a method for machining a high-precision sealing surface of a numerical control gantry milling center, which relates to the technical field of numerical control machining and comprises the following steps of: fixing a workpiece on a workbench of a numerical control gantry milling machine, and setting machining parameters through automatic tool setting; based on the machining parameters, a random forest algorithm is adopted to generate a milling path; according to the milling path, the workpiece is milled, meanwhile, machining parameters in the milling process are collected in real time, and machining state monitoring data are generated; dynamically optimizing a milling path by combining hyper-parameters of quantum reinforcement learning on the basis of machining state monitoring data; according to the optimized milling path and the optimal parameter combination, sealing face finish machining operation is executed; according to the method, when the milling path is dynamically optimized through the quantum reinforcement learning method, path adjustment is more accurate through optimization of hyper-parameters, and the method can better adapt to machining tasks with high complexity and high precision requirements.

Description

Technical Field

[0001] The invention relates to the technical field of numerical control machining, in particular to a method for machining a high-precision sealing surface of a numerical control gantry milling center. Background Art

[0002] In recent years, CNC gantry milling processing technology has been widely used, especially in the field of high-precision sealing surface processing. As the requirements for high precision and high efficiency in industrial manufacturing continue to increase, CNC processing technology has gradually introduced advanced intelligent algorithms and automated control. At present, the commonly used CNC processing technology improves the processing accuracy by precisely controlling cutting parameters, path planning and processing status monitoring. However, traditional CNC technology still faces some challenges, especially in tool path optimization and multivariable processing parameter correlation optimization. Although existing technologies use limited algorithms (such as genetic algorithms, simulated annealing, etc.) for path optimization and parameter adjustment, these methods generally have problems such as large computational complexity, slow convergence speed, and inability to dynamically adapt to complex processing environments.

[0003] The shortcomings of existing technologies are mainly reflected in the real-time monitoring of processing status and path adjustment. Traditional path optimization methods are usually based on static processing parameters, and fail to make full use of real-time feedback data for dynamic adjustment. In addition, although some traditional algorithms can optimize the processing process, their efficiency and accuracy are often difficult to meet the needs of high-precision sealing surface processing when processing high-dimensional complex data. Therefore, it is particularly urgent to develop an efficient processing method that can combine machine learning, quantum computing and dynamic adjustment algorithms. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a method for machining high-precision sealing surfaces in a CNC gantry milling center to solve the problems of high-precision sealing surface path optimization and dynamic adjustment of machining parameters in the prior art.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a method for high-precision sealing surface of a CNC gantry milling machining center, which comprises fixing a workpiece on a worktable of the CNC gantry milling machine and setting machining parameters by automatic tool setting; Based on the machining parameters, the random forest algorithm is used to generate the milling path; According to the milling path, the workpiece is milled, and the processing parameters of the milling process are collected in real time to generate processing status monitoring data; Based on the processing status monitoring data, the milling path is dynamically optimized by combining the hyperparameters of quantum reinforcement learning; The particle swarm optimization algorithm is used to calculate the correlation between machining parameters in the milling process and solve the optimal parameter combination; The sealing surface finishing operation is performed according to the optimized milling path and the optimal parameter combination.

[0007] As a preferred solution of the method for high-precision sealing surface of CNC gantry milling machining center of the present invention, wherein: the workpiece is fixed on the workbench of the CNC gantry milling machine, and the machining parameters are set by automatic tool setting. The specific steps are as follows: Use intelligent positioning fixture to calibrate the placement of workpieces on the workbench of CNC gantry milling machine; Activate the automatic tool setting function of the CNC equipment, measure the distance between the tool and the workpiece surface through the tool setting detector, and correct the initial position of the tool; Based on the initial position of the tool and the material properties of the workpiece, the CNC sets the feed rate; Input the workpiece material, processing technology and tool information into the CNC equipment and set the cutting depth; Based on the set cutting depth and feed rate, the CNC equipment calculates the tool speed by combining the cutting characteristics in the material properties of the workpiece; Integrate cutting depth, feed rate and tool speed to generate machining parameters.

[0008] As a preferred solution of the method for high-precision sealing surface of CNC gantry milling machining center of the present invention, wherein: based on the machining parameters, the random forest algorithm is used to generate the milling path, and the specific steps are as follows: Clean, denoise, standardize and handle outliers of processing parameters; Extract machining parameters, material properties of workpieces and tool information from the historical machining records of CNC machining equipment as feature variables; The processing performance index and path evaluation index are used as target variables; Construct a training data set through feature variables and target variables, and use the training data set to train the random forest model; Cross-validation is used to optimize the parameters of the random forest model, and the performance of the random forest model is tested by comparing the predicted values ​​with the actual values ​​to obtain a trained random forest model; The machining parameters and workpiece machining targets are used as input parameters and fed into the trained random forest model to predict the initial path; The initial path is simulated, the local characteristics of the path are optimized according to the simulation feedback, and the optimized milling path is output.

[0009] As a preferred solution of the method for high-precision sealing surface of CNC gantry milling machining center of the present invention, wherein: the milling operation is performed on the workpiece according to the milling path, and the machining parameters of the milling process are collected in real time to generate machining status monitoring data. The specific steps are as follows: Use the optimized milling path as the processing instruction of the CNC equipment; Start the CNC equipment according to the processing instructions and use the segmented processing strategy to gradually complete the workpiece processing; After each section of workpiece is processed, sensors are used to check the milling path status and evaluate the tool status; If a deviation in the milling path or an abnormality in the tool is detected, the central controller adjusts the parameters of the next milling path or prompts to replace the tool based on the evaluation results; Collect cutting force, vibration and temperature data in real time during milling process and integrate them into feedback data set; transmitting feedback data sets to a central controller for storage; The central controller performs structured processing on the stored feedback data set, records the processing timestamp, parameter change trend and abnormal status, and generates processing status monitoring data.

[0010] As a preferred solution of the method for high-precision sealing surface of CNC gantry milling machining center described in the present invention, wherein: the milling path is dynamically optimized based on the machining state monitoring data combined with the hyperparameters of quantum reinforcement learning, and the specific steps are as follows: Standardize the processing state monitoring data and map it into the quantum state space; Each processing state monitoring data in the quantum state space is represented by a quantum bit, and the expression is: ; in, Indicates qubits, represents the normalized cutting force , Represents the normalized cutting force vibration data , represents the normalization operation, represents the zero state of the quantum bit, represents a state of a quantum bit, An index variable representing a qubit; Map different processing state monitoring data to corresponding quantum bits respectively, and associate the quantum bits through quantum entanglement operation to form a joint quantum state of the overall processing state , the expression is: ; in, Representing the cutting force in the quantum bit The quantum state of Representing vibration data in qubits The quantum state of Representing temperature data in quantum bits The quantum state of Representing tool wear data in quantum bits The quantum state of represents the tensor product of quantum states; Using joint quantum states Participate in the calculation of the immediate reward function, the expression is: ; in, Indicates at time and hyperparameters The quantitative index value of the machining path optimization effect under Indicates time and hyperparameters The reward operator under Representing the joint quantum state The conjugate transpose of Representing the joint quantum state original state, Indicates time Processing data collected at all times; Loss Function Based on processing state data, quantum state and immediate reward function , weighted feedback is given to the processing state, and the expression is: ; in, represents the upper limit of the integral interval, For time from arrive The integral of is the symbol of the second-order partial derivative, Indicates time The second-order partial derivative of represents a non-negative weight coefficient, is the immediate reward function For hyperparameters The gradient of is the square of the gradient magnitude, represents a small increment of the integral variable, is the regularization coefficient, is the total number of hyperparameters, The index variable representing the hyperparameter, represents a nonlinear periodic function, Indicates The square of the hyperparameters, Represents the form of hyperparameter regularization; According to real-time processing data , calculate the second-order change trend of the reward , evaluate the dynamic trend of processing performance; Based on the dynamic trend of the evaluated processing performance, combined with the immediate reward function For hyperparameters The gradient of , evaluates the path optimization direction; Based on the joint quantum state , the quantum state influence factor when calculating the path optimization parameter update is expressed as: ; in, represents the quantum state influence factor, Represents the excited state in the quantum state The square of the probability magnitude, Represents the excited state in the quantum state The square of the probability magnitude; The square sum quantum state influence factor based on gradient norm , calculate the optimized hyperparameter set, the expression is: ; in, Represents the time step The optimized hyperparameter set is represents the learning rate, Represents the logarithmic transformation of the absolute value of the current parameter, represents the smoothness term; Generate multiple candidate paths based on the optimized hyperparameter set; Use simulation tools to evaluate candidate paths with an instant reward function The highest path is used as the milling path.

[0011] As a preferred solution of the method for high-precision sealing surface of CNC gantry milling machining center of the present invention, wherein: based on the optimized hyperparameter set, multiple candidate paths are generated, and the specific steps are as follows: Using the optimized hyperparameter set and real-time processing status data, determine the upper and lower limits of the path parameters; The search space for path generation is limited by the upper and lower limits of the path parameters, and several candidate paths are generated within the search range using the Monte Carlo method.

[0012] As a preferred solution of the method for high-precision sealing surface of CNC gantry milling machining center described in the present invention, wherein: the particle swarm optimization algorithm is used to calculate the correlation between machining parameters in the milling process and solve the optimal parameter combination. The specific steps are as follows: The machining parameters in the milling process are defined as the dimensions of the particle group, the particle positions and velocities are initialized, and the initial parameter combinations are generated; Based on the cutting force, vibration and temperature data in the feedback data set, the fitness value is calculated using the weighted summation method; Update the particle velocity and adjust the particle position by using the inertia weight and acceleration factor in the initial parameter combination; Re-evaluate the fitness value according to the adjusted new particle position, and update the historical optimal position and global optimal position of each particle; Iteration continues until the maximum number of iterations is reached, and the global optimal particle position is output as the optimal combination of processing parameters.

[0013] As a preferred solution of the method for high-precision sealing surface of CNC gantry milling center of the present invention, wherein: the sealing surface finishing operation is performed according to the optimized milling path and the optimal parameter combination, and the specific steps are as follows: Load the optimized milling path and the best combination of machining parameters, and verify the path through CNC simulation software; Activate the CNC equipment, adjust the tool path according to the simulation verification results, start preliminary cutting, and monitor the cutting force, vibration and temperature data in real time; Based on real-time monitoring of cutting force, vibration and temperature data, the initial cutting results are evaluated and the path corrected; After the initial cutting is correct, the optimized milling path is used for finishing operations; During the milling process, machining parameters are fine-tuned based on real-time data feedback.

[0014] In a second aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the method for high-precision sealing surface of a CNC gantry milling machining center as described in the first aspect of the present invention is implemented.

[0015] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the method for high-precision sealing surface of a CNC gantry milling machining center as described in the first aspect of the present invention is implemented.

[0016] The beneficial effects of the present invention are as follows: the present invention generates a preliminary milling path through a random forest algorithm, and performs dynamic optimization in combination with processing state monitoring data, ensuring that the processing path can adapt in real time to changes in feedback information such as tool status, cutting force, vibration and temperature during the processing process. Secondly, when the quantum reinforcement learning method dynamically optimizes the milling path, the path adjustment is made more accurate through the optimization of hyperparameters, and it can better adapt to processing tasks with high complexity and high precision requirements. In addition, when the particle swarm optimization algorithm solves the correlation between processing parameters, it uses an iterative optimization method to not only improve the efficiency of parameter selection, but also realize the dynamic adaptation of multidimensional data during the processing process, ensuring high-quality control and efficiency in high-precision sealing surface processing, reducing human intervention and errors, and improving processing accuracy and intelligence. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 This is a flow chart of the method for high-precision sealing surface of a CNC gantry milling machining center in Example 1.

[0019] Figure 2 This is a flow chart for solving the optimal parameter combination in Example 1. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0022] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0023] Example 1, reference Figure 1 and Figure 2, which is the first embodiment of the present invention, provides a method for high-precision sealing surface of a CNC gantry milling machining center, comprising the following steps: S1. Fix the workpiece on the workbench of the CNC gantry milling machine and set the processing parameters through automatic tool setting.

[0024] Furthermore, an intelligent positioning fixture is used to calibrate the placement of the workpiece on the worktable of the CNC gantry milling machine to ensure the accuracy and stability of the workpiece positioning; Activate the automatic tool setting function of the CNC equipment, measure the distance between the tool and the workpiece surface through the tool setting detector, correct the initial position of the tool, and ensure that the tool position is accurate; Among them, CNC equipment refers to a CNC machine tool with automatic control, real-time monitoring, processing program analysis and optimization functions, usually equipped with hardware and software components such as computer numerical control, automatic tool setting, sensors and feedback units.

[0025] Based on the initial position of the tool, combined with the material properties of the workpiece and the cutting characteristics of the workpiece, the CNC equipment calculates the feed rate to ensure reasonable cutting forces between the tool and the workpiece during machining; It should be noted that the material properties of the workpiece refer to the inherent characteristics of the material used in the workpiece. These characteristics determine how the material interacts with the tool during the machining process, thereby affecting the cutting efficiency and machining quality. Common material properties include hardness, elastic modulus, yield strength, ductility, etc. These properties directly affect the magnitude of the cutting force and the wear rate of the tool. For example, high-hardness materials usually require lower feed rates and higher tool speeds to avoid excessive tool wear. The rigidity and thermal conductivity of the material will also affect the temperature change during the cutting process, thereby affecting the machining accuracy. By analyzing the material properties of the workpiece, CNC equipment can optimize the cutting conditions and select appropriate tools and parameters to achieve efficient and precise machining.

[0026] The cutting characteristics of a workpiece refer to the behavioral characteristics of the material during the cutting process. These characteristics are usually evaluated by analyzing factors such as cutting force, cutting temperature, and surface roughness. The cutting characteristics are not only affected by the workpiece material, but are also closely related to factors such as processing conditions, tool information, and cutting methods. For example, some materials may generate a large amount of cutting heat during the cutting process, causing rapid tool wear, while some materials may generate a large cutting force during the cutting process due to their high toughness, thereby increasing the difficulty of processing. According to the cutting characteristics of the workpiece, the CNC system can dynamically adjust the processing parameters, such as tool speed, feed rate, cutting depth, etc., to optimize the cutting process, reduce tool wear and thermal damage, and improve processing efficiency and accuracy.

[0027] Input the workpiece material, processing technology and tool information into the CNC equipment, set the cutting depth, ensure the stability of the cutting process, and avoid excessive load on the tool or workpiece; Specifically, first, basic information such as the material, processing technology and tool information of the workpiece is input through the operation interface of the CNC equipment. These inputs can be completed through machine vision systems, sensors or manual input to ensure that the CNC equipment accurately obtains the basic information of the workpiece. Next, the CNC equipment selects appropriate tool information based on the input workpiece material (such as hardness, rigidity, etc.) and processing technology (such as roughing or finishing requirements). Based on this information, the CNC equipment will call the built-in cutting database, and refer to the mechanical properties, thermal conductivity and other characteristics of the workpiece material, combined with the geometry and material properties of the tool, to calculate the preliminary range of the cutting depth. In this process, the equipment will also evaluate factors such as tool load capacity and cutting force distribution to ensure that the cutting depth can ensure efficient processing without causing excessive pressure or damage to the tool. If the processing technology has special precision requirements, the equipment will further adjust the cutting depth according to the precision standard to avoid processing errors. Finally, the calculated and adjusted cutting depth value will be determined and fed back to the main control program as a key parameter for the subsequent processing process for actual milling operations.

[0028] Based on the cutting depth and feed rate, the CNC equipment calculates the tool speed to ensure an efficient and stable cutting process; Specifically, the tool speed is automatically calculated based on the previously set cutting depth and feed rate, combined with the material properties of the workpiece (such as hardness, thermal conductivity, etc.) and cutting characteristics (such as cutting force, heat generation, etc.). Through the built-in cutting force model and thermodynamic model, the CNC equipment comprehensively evaluates the required tool speed based on the impact of cutting depth on tool load and the requirements of feed rate on cutting force. This calculation process ensures that the tool can maintain efficient cutting performance at the set cutting depth and feed rate, avoiding insufficient cutting due to too low a speed, or excessive tool wear or workpiece damage due to too high a speed. Finally, the tool speed is automatically adjusted to meet the optimized processing requirements to ensure the smooth progress of the processing process.

[0029] Integrate cutting depth, feed rate and tool speed to generate machining parameters, providing a basis for subsequent milling operations.

[0030] S2. Based on the machining parameters, the random forest algorithm is used to generate the milling path.

[0031] Furthermore, the processing parameters are cleaned, denoised, standardized, and outliers are processed; Extract machining parameters, material properties of workpieces and tool information from the historical machining records of CNC machining equipment as feature variables; Among them, the historical processing records are extracted and sorted from a large amount of raw data generated by CNC processing equipment during the actual processing process. The historical processing records include the tool motion trajectory, processing parameters (such as cutting depth, feed rate, tool speed, etc.), processing environment conditions (such as temperature, vibration) and workpiece performance evaluation results after processing (such as surface roughness, dimensional accuracy) in each processing task.

[0032] The processing performance index and path evaluation index are used as target variables; It should be noted that machining performance indicators are key parameters for measuring workpiece quality, machining efficiency and tool usage during CNC machining, mainly including surface roughness, machining time, tool wear rate, cutting temperature, etc. Surface roughness reflects the machining quality of the workpiece, machining time measures machining efficiency, and tool wear rate and cutting temperature are directly related to tool life and stability. Path evaluation indicators are used to evaluate the rationality and execution effect of the milling path, including path smoothness, path length, machining stability, etc. Path smoothness reflects the continuity of the tool motion trajectory, path length directly affects machining time, and machining stability reflects the adaptability of path design to vibration and force distribution.

[0033] Construct a training data set through feature variables and target variables, and use the training data set to train the random forest model; Cross-validation is used to optimize the parameters of the random forest model, and the performance of the random forest model is tested by comparing the predicted values ​​with the actual values ​​to obtain a trained random forest model; The machining parameters and workpiece machining targets are used as input parameters and fed into the trained random forest model to predict the initial path; Specifically, the preprocessed processing parameters (such as cutting depth, feed rate, tool speed) and workpiece processing goals (such as surface roughness requirements, processing time constraints, etc.) are uniformly formatted into numerical feature vectors to ensure that they match the input format of the random forest model; then, the feature vectors are gradually input into the trained random forest model, and the model splits and classifies the input parameters through multiple decision trees within it. Each decision tree independently predicts the optimal path characteristics based on the processing data rules learned during the training process; then, the random forest model performs weighted voting or average calculation on the output results of all decision trees to comprehensively generate a preliminary milling path plan, including the geometric path of the tool, speed distribution and feed modes in different areas; finally, the preliminary path plan is formatted and output for use in subsequent simulation and optimization steps to ensure that the path is closely matched with actual processing requirements.

[0034] The initial path is simulated, the local characteristics of the path are optimized according to the simulation feedback, and the optimized milling path is output.

[0035] It should be noted that when simulating the initial path, the path is input into the simulation program of the CNC equipment to simulate the actual machining process of the tool moving along the path. By analyzing the performance indicators such as cutting force changes, tool load distribution, and workpiece surface roughness, possible problem areas in the path such as tool cutting too deep, tool vibration, or path unevenness are identified. For these problems, the local characteristics of the path are adjusted, such as reducing the cutting depth or reducing the feed rate in the overcut area, optimizing the tool acceleration at the path reversal point to avoid vibration, and increasing the path density in complex curved areas to improve the surface quality. After optimization, the simulation program is run again to verify the improvement effect of the adjusted path on cutting smoothness, machining efficiency, and surface quality, to ensure that the path can ultimately meet the machining requirements and be directly used as a machining instruction for the CNC equipment, thereby improving the overall machining efficiency and significantly improving the quality stability of the workpiece.

[0036] S3. Perform milling operation on the workpiece according to the milling path, and collect the processing parameters of the milling process in real time to generate processing status monitoring data.

[0037] Furthermore, the optimized milling path is used as a processing instruction for CNC equipment; Start the CNC equipment according to the processing instructions and use the segmented processing strategy to gradually complete the workpiece processing; It should be noted that the "segmented processing strategy" refers to dividing the entire milling path into several processing segments, and performing processing operations segment by segment to improve the stability and efficiency of the processing process, while achieving real-time status monitoring and adjustment. The specific implementation process is as follows: After loading and verifying the processing instructions, the complete processing path is divided into several logical segments according to the complexity of the milling path and the geometric features of the workpiece. Each segment contains a continuous processing area or specific geometric features. After starting the CNC equipment, the tool movement is controlled by the path instruction, the first cutting operation is executed, and the tool movement trajectory and processing effect are monitored in real time. After completing the processing of this segment, the processing operation is paused, and the path execution effect of the processed area is checked by using sensors, while the tool wear is evaluated and the relevant processing data is recorded. If path deviation or abnormality is found, the processing parameters of the subsequent segments (such as feed rate, cutting depth, etc.) are adjusted according to the detection results, or a prompt is given to replace the tool. After completing the necessary adjustments, the next processing path is started, and the above steps are repeated until all processing segments are completed. This segment-by-segment execution and adjustment method ensures the stability and accuracy of the processing process, while providing real-time data support for subsequent processing status monitoring and path optimization.

[0038] After each section of workpiece is processed, sensors are used to check the milling path status and evaluate the tool status; If a deviation in the milling path or an abnormality in the tool is detected, the central controller adjusts the cutting path parameters for the next section of processing or prompts to replace the tool based on the evaluation results; It should be noted that the central controller is the core control unit of the CNC equipment. It is responsible for receiving real-time data collected by sensors (such as path status, tool status, etc.), analyzing and evaluating processing parameters and path execution, and dynamically adjusting processing strategies based on the evaluation results; at the same time, the central controller manages the instruction execution, data storage and feedback generation of the CNC equipment to ensure the stability, accuracy and efficiency of the processing process.

[0039] Specifically, the central controller receives the milling path status data and tool working status data collected by the sensor in real time, calculates the error value between the actual path and the optimized path through the built-in deviation analysis module, and evaluates the wear of the tool at the same time; if the error value exceeds the preset threshold, the central controller adjusts the local parameters of the path according to the position and degree of the current path deviation through the optimization algorithm, such as correcting the cutting depth, optimizing the path smoothness or adjusting the feed rate; if the tool wear reaches the safety threshold, the controller sends a prompt signal to notify the tool to be replaced, and updates the new tool parameters to the processing system at the same time; after completing the path parameter adjustment or tool replacement, the controller stores the modified path data and tool status as feedback data for call when the next processing path is executed, and simultaneously generates a processing report for analyzing the effectiveness of the adjustment results, thereby ensuring the stability and accuracy of subsequent processing.

[0040] Collect cutting force, vibration and temperature data in real time during milling process and integrate them into feedback data set; Specifically, based on the milling path status check and tool evaluation results, the cutting force, vibration and temperature sensor network is activated, covering the tool spindle, worktable and processing area, and the cutting force, vibration and temperature data are collected in real time with high-frequency sampling, and transmitted to the central controller through a high-speed communication bus; the central controller filters, denoises and standardizes the data, integrates it into multi-dimensional time series data, and stores it in a structured manner, annotating timestamps, processing segment paths and status labels, and finally forming a feedback data set to provide support for path adjustment and parameter optimization.

[0041] transmitting feedback data sets to a central controller for storage; The central controller performs structured processing on the stored feedback data set, records the processing timestamp, parameter change trend and abnormal status, and generates processing status monitoring data.

[0042] Specifically, hierarchical cluster analysis and multivariate time series modeling methods are used to perform structured processing on the collected parameters such as cutting force, vibration and temperature; cluster analysis is used to group the parameter data according to the workpiece area, tool position and processing time, and abnormal data points are marked; at the same time, a parameter change trend curve is generated based on the time series model, and the timestamp of the abnormal state and the associated processing path information are extracted; finally, the processed structured data is used to generate processing status monitoring data, which provides a clear decision-making basis for path optimization and parameter adjustment, and is transmitted back to the path optimization module or fed back to the CNC equipment in real time.

[0043] S4. Based on the processing state monitoring data, the milling path is dynamically optimized by combining the hyperparameters of quantum reinforcement learning.

[0044] Furthermore, the processing state monitoring data is standardized and mapped into the quantum state space; Preferably, the processing state monitoring data is standardized and mapped to the quantum state space, which can unify the data scale, eliminate dimensional differences, and ensure that different processing state data are calculated in a consistent form; the data is compactly represented through the superposition of quantum states, reducing the dimension and enhancing the representation efficiency, providing an efficient basis for the construction of joint quantum states and the calculation of immediate reward functions, thereby significantly improving the accuracy and real-time nature of the optimization process and the ability to handle complex processing tasks.

[0045] Each processing state monitoring data in the quantum state space is represented by a quantum bit, and the expression is: ; in, Indicates qubits, represents the normalized cutting force , Represents the normalized cutting force vibration data , represents the normalization operation, represents the zero state of the quantum bit, represents a state of a quantum bit, An index variable representing a qubit; It should be noted that by utilizing the superposition and entanglement characteristics of quantum computing, complex processing state information is compactly encoded into high-dimensional quantum states. This representation method can significantly improve the efficiency of data processing and the flexibility of path optimization.

[0046] Specifically, the expression ; The normalized cutting force and vibration data Through this process, the correlation between the two processing state variables can be captured in the same quantum bit and quantum computing operations can be performed, while meeting the normalization requirements of the quantum state, so that each bit state can accurately express the current characteristics of the processing state. This processing not only reduces the representation redundancy of the data dimension, but also lays the foundation for the subsequent construction of the joint quantum state. The joint quantum state further associates the processing states through the tensor product operation, which can effectively express the global state information of the entire processing process, and provide more efficient and compact data structure support for the calculation of the immediate reward function and loss function, thereby accelerating the calculation process of path optimization and improving the overall optimization effect.

[0047] Map different processing state monitoring data to corresponding quantum bits respectively, and associate the quantum bits through quantum entanglement operation to form a joint quantum state of the overall processing state , the expression is: ; in, Representing the cutting force in the quantum bit The quantum state of Representing vibration data in qubits The quantum state of Representing temperature data in quantum bits The quantum state of Representing tool wear data in quantum bits The quantum state of represents the tensor product of quantum states; Using joint quantum states Participate in the calculation of the immediate reward function, the expression is: ; in, Indicates at time and hyperparameters The quantitative index value of the machining path optimization effect under Indicates time and hyperparameters The reward operator under Representing the joint quantum state The conjugate transpose of Representing the joint quantum state original state, Indicates time Processing data collected at all times; Loss Function Based on processing state data, quantum state and immediate reward function , weighted feedback is given to the processing state, and the expression is: ; in, represents the upper limit of the integral interval, For time from arrive The integral of represents the cumulative optimization effect during the entire processing process. is the symbol of the second-order partial derivative, Indicates time The second-order partial derivative of Represents a non-negative weight coefficient, which is used to control the influence of the gradient modulus on the loss function. is the immediate reward function For hyperparameters The gradient indicates the optimization direction and strength. is the square of the gradient modulus, which is used to reflect the sensitivity of reward changes to parameter adjustments. represents a small increment of the integral variable, is the regularization coefficient, is the total number of hyperparameters, The index variable representing the hyperparameter, represents a nonlinear periodic function, Indicates The square of the hyperparameters, Represents the form of hyperparameter regularization; It should be noted that the purpose of weighted feedback on the processing state is to more accurately reflect the changing trend of the processing state and its contribution to the final processing quality by quantifying the dynamic influence of the processing state in path optimization, thereby improving the sensitivity and adaptability of the loss function to processing path optimization.

[0048] Specifically, the dynamic changes of machining performance are captured by the second-order change trend of the immediate reward function, and the exponential decay term of the square of the gradient modulus is used to dynamically adjust the weight ratio of the machining state, thereby enhancing the optimization response ability of the model when the machining state changes drastically, and reducing the adjustment amplitude when the change is stable, thereby improving the stability of the optimization process; the regularization term is used to suppress overfitting, ensure the rationality of parameter adjustment and improve the versatility of the machining path; by weighted feedback of the machining state, this step can not only significantly improve the accuracy and stability of the machining path optimization, but also adaptively adjust the optimization strategy under complex machining conditions, ensuring that the machining process is always efficient and stable, and ultimately realizing the intelligent and refined milling path optimization.

[0049] According to real-time processing data , calculate the second-order change trend of the reward , evaluate the dynamic trend of processing performance; Preferably, evaluating the dynamic change trend of machining performance can quantify the degree of accelerated change of machining performance over time, which is used to identify the optimization direction and adjustment timing of the machining path, and guide the dynamic adjustment of subsequent hyperparameters to improve the sensitivity of path planning and machining efficiency, thereby ensuring that the performance indicators are kept in the optimal state during real-time machining.

[0050] Based on the dynamic trend of the evaluated processing performance, combined with the immediate reward function For hyperparameters The gradient of , evaluates the path optimization direction; Preferably, the evaluation path optimization direction is used to identify potential improvement areas for the current path optimization and provide directional guidance for subsequent hyperparameter updates, thereby further improving the accuracy and efficiency of the machining path.

[0051] Based on the joint quantum state , the quantum state influence factor when calculating the path optimization parameter update is expressed as: ; in, represents the quantum state influence factor, Represents the excited state in the quantum state The square of the probability magnitude, Represents the excited state in the quantum state The square of the probability magnitude; The square sum quantum state influence factor based on gradient norm , calculate the optimized hyperparameter set, the expression is: ; in, Represents the time step The optimized hyperparameter set is represents the learning rate, Represents the logarithmic transformation of the absolute value of the current parameter, represents the smoothness term; Specifically, the learning rate is adjusted in real time through quantum state measurement and smoothing term , adapt to the dynamic processing environment. The learning rate adjustment balances the exploration and convergence efficiency of path optimization and responds quickly to sudden states; the smoothing term optimization enhances numerical stability and avoids gradient oscillation problems. Furthermore, the adaptability and accuracy of processing path planning are improved, the processing error and tool loss are significantly reduced, and the processing efficiency is improved.

[0052] Generate multiple candidate paths based on the optimized hyperparameter set; Specifically, the upper and lower limits of the path parameters are determined using the optimized hyperparameter set and real-time processing status data; The search space for path generation is limited by the upper and lower limits of the path parameters, and several candidate paths are generated within the search range using the Monte Carlo method.

[0053] The candidate paths are evaluated using simulation tools, and the path with the highest quantitative index value in the immediate reward function is selected as the milling path.

[0054] It should be noted that simulation tools are software or modules used to simulate real machining processes in a virtual environment. By inputting machining paths, workpiece models, machine tool parameters and material properties, simulation tools can predict performance indicators such as cutting force, vibration, temperature and surface roughness during machining, evaluate the machining effects of different paths, and provide data support for optimizing and selecting the best machining paths. The highest path refers to the path with the highest comprehensive score after simulation evaluation. The comprehensive score is calculated based on the performance of multiple quantitative indicators (such as cutting force stability, vibration amplitude control, temperature balance and surface roughness optimization), and the weight is assigned based on the importance of each indicator. The final score value is that the higher the score, the better the comprehensive performance of the path in terms of machining efficiency, quality and stability.

[0055] Specifically, multiple candidate paths generated by the optimized hyperparameter set are input into the simulation tool. The simulation tool establishes a machining simulation environment and loads machining status data according to the workpiece geometry model, machine tool parameters and material properties, simulates each candidate path, and calculates in real time the quantitative indicators in the immediate reward function corresponding to the path, including the average value of cutting force, the average value of vibration amplitude, the average value of temperature and the average value of surface roughness, etc., and then assigns weights according to the importance of these quantitative indicators, and combines the weights with the inverse values ​​of these quantitative indicators to calculate a comprehensive score. By comparing the comprehensive scores of each path, the path with the highest comprehensive score is selected as the optimal path. The parameters of the optimal path are used to generate milling processing instructions and load them into the CNC equipment, and then the actual processing is performed. At the same time, the processing status is monitored in real time during the processing to ensure the adaptability and optimization effect of the path.

[0056] S5. The particle swarm optimization algorithm is used to calculate the correlation between the machining parameters in the milling process and solve the optimal parameter combination.

[0057] Furthermore, the machining parameters in the milling process are defined as the dimensions of the particle group, and the particle positions and velocities are initialized to generate the initial parameter combinations; Based on the cutting force, vibration and temperature data in the feedback data set, the fitness value is calculated using the weighted summation method; It should be noted that based on the cutting force, vibration and temperature data in the feedback data set, these data are first standardized and converted to a unified range to eliminate the differences between different dimensions and values. Then, according to the degree of influence of each processing parameter on the processing performance, weights are assigned to the cutting force, vibration and temperature respectively. Usually, these weight values ​​are set in previous experiments or experience. Next, the fitness value of each particle is calculated using the weighted summation method using the standardized cutting force, vibration and temperature data combined with the corresponding weight values.

[0058] Specifically, the fitness value is obtained by multiplying the cutting force, vibration and temperature data with their corresponding weights and then summing them up. This fitness value reflects the quality of the processing parameter combination corresponding to the particle. The higher the fitness value, the greater the contribution of the combination to the processing performance. Finally, based on the calculated fitness value, the particles are sorted and a basis is provided for subsequent particle position updates to ensure that excellent particles can be retained in iterations and approach the global optimal solution.

[0059] Update the particle velocity and adjust the particle position by using the inertia weight and acceleration factor in the initial parameter combination; It should be noted that the inertia weight is an important parameter in the particle swarm optimization algorithm. It controls the inertia of the particle at the current speed and determines the degree of inertia retention of the particle at the current position. By setting the inertia weight appropriately, the local search ability and the global search ability of the particle search can be balanced to avoid premature convergence or excessive fluctuation of particles in the search space. A larger inertia weight helps particles explore a wider search space, while a smaller inertia weight helps particles quickly concentrate near the local optimal solution. The inertia weight helps particles determine their new direction of movement based on their current speed to more effectively adjust the parameter combination.

[0060] It should also be noted that the acceleration factor is used to guide the particles to their historical optimal position and the global optimal position. It adjusts the particle speed to move it in the direction of the optimal solution. Specifically, the acceleration factor affects the guidance of particles from the local optimal solution and the global optimal solution. A larger acceleration factor will make the particles approach the optimal solution more quickly, while a smaller acceleration factor will allow the particles to explore the search space more carefully. By properly selecting the acceleration factor, a balance can be found between global search and local search to ensure that the particles can gradually converge to the optimal combination of processing parameters during the optimization process.

[0061] Re-evaluate the fitness value according to the adjusted new particle position, and update the historical optimal position and global optimal position of each particle; Iteration continues until the maximum number of iterations is reached, and the global optimal particle position is output as the optimal combination of processing parameters.

[0062] S6. Perform sealing surface finishing operations according to the optimized milling path and optimal parameter combination.

[0063] Furthermore, the optimized milling path and the best combination of machining parameters are loaded, and the path is verified through CNC simulation software to ensure that the path is collision-free, the cutting load is balanced, and it meets the machining requirements; Activate the CNC equipment, adjust the tool path according to the simulation verification results, start preliminary cutting, and monitor the cutting force, vibration and temperature data in real time to ensure stable cutting conditions; After the initial cutting is correct, continue to perform finishing operations according to the optimized milling path to further improve the machining accuracy of the sealing surface; During the milling process, machining parameters such as cutting depth, feed rate, etc. are adjusted according to real-time feedback data to optimize the tool path and ensure machining accuracy and efficiency.

[0064] Specifically, during the milling process, the central controller analyzes the status of each processing stage through real-time collection of cutting force, vibration and temperature data to evaluate whether the current cutting conditions are within the expected range. Based on real-time feedback data, the controller automatically adjusts processing parameters such as cutting depth and feed rate to reduce tool vibration and temperature, ensure the best working state of the tool, and avoid over-cutting or excessive tool wear. When the tool is in good condition, the controller passes the adjusted processing parameters to the CNC equipment to ensure the optimization of the cutting path. At the same time, it judges the changing trend of the processing status through real-time monitoring data, further corrects the path, and improves processing accuracy and stability.

[0065] This embodiment also provides a computer device, which is suitable for the method of high-precision sealing surface of a CNC gantry milling machining center, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the method of high-precision sealing surface of a CNC gantry milling machining center as proposed in the above embodiment.

[0066] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0067] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for realizing a high-precision sealing surface of a CNC gantry milling machining center as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as a static random access memory (SRAM), an electrically erasable programmable read-only memory (EEPROM), an erasable programmable read-only memory (EPROM), a programmable read-only memory (PROM), a read-only memory (ROM), a magnetic storage device, a flash memory, a disk or an optical disk.

[0068] In summary, the present invention generates a preliminary milling path through a random forest algorithm, and dynamically optimizes it in combination with the processing state monitoring data to ensure that the processing path can adapt in real time to changes in feedback information such as tool status, cutting force, vibration and temperature during the processing process. Secondly, when the quantum reinforcement learning method dynamically optimizes the milling path, it optimizes the hyperparameters to make the path adjustment more accurate and better adapt to processing tasks with high complexity and high precision requirements. In addition, when the particle swarm optimization algorithm solves the correlation between processing parameters, it uses an iterative optimization method to not only improve the efficiency of parameter selection, but also realize the dynamic adaptation of multidimensional data during the processing process, ensuring high-quality control and efficiency in high-precision sealing surface processing, reducing human intervention and errors, and improving processing accuracy and intelligence.

[0069] Example 2, referring to Table 1, is the second example of the present invention. To further verify the technical solution of the present invention, experimental simulation data of a method for high-precision sealing surface of a CNC gantry milling machining center are provided.

[0070] The experiment selected a typical aluminum alloy workpiece, which is made of 6061 aluminum alloy, with high machining performance and good stability. The CNC gantry milling machine uses a high-precision CNC machine tool of the XYZ model, equipped with an automatic tool setting function and a high-precision sensor, where XYZ represents any brand or model of CNC machine tools. The specific experimental steps are as follows: The workpiece is first fixed on the workbench of the CNC gantry milling machine by an intelligent positioning fixture. After the fixture is calibrated, the automatic tool setting function is activated, and the distance between the tool and the workpiece surface is measured by the tool setting detector to correct the initial position of the tool. This step can ensure the accuracy of the initial position of the tool, thereby providing accurate processing parameters for the subsequent cutting process. The workpiece material is 6061 aluminum alloy, and the CNC equipment inputs the material information, processing requirements and tool type of the workpiece. Based on this information, the system sets the feed rate to 150mm / min and the cutting depth to 2mm, and calculates the tool speed to 3500RPM based on the cutting depth and feed rate.

[0071] Next, based on the set processing parameters, the random forest algorithm is used to generate the milling path. During this process, historical data of multiple workpiece processing are collected, including information such as cutting force, vibration, temperature, and tool wear. These data are cleaned, standardized, and used to train the random forest model to finally generate the milling path. When performing the milling operation, processing data such as cutting force, vibration, and temperature are collected in real time, and the central control system is used to monitor the processing status in real time. If a deviation occurs in a certain section of cutting, the system will automatically adjust the tool path and processing parameters according to the preset rules.

[0072] Secondly, the quantum reinforcement learning algorithm is used to dynamically optimize the milling path based on the real-time collected processing status monitoring data. In this process, the real-time data is mapped to quantum bits, the immediate reward function is calculated, and the hyperparameters are adjusted through the quantum state influencing factor to optimize the milling path. During the milling process, the correlation between cutting force, vibration and temperature is calculated by the particle swarm optimization algorithm, and the optimal combination of processing parameters is finally obtained. This process is iterated multiple times based on real-time data feedback until the global optimal solution is found.

[0073] Finally, the sealing surface finishing operation is performed using the optimized milling path and optimal parameter combination. During this process, the cutting force, vibration and temperature data are monitored in real time, and the tool path and processing parameters are fine-tuned based on the data feedback to ensure high-precision sealing surface processing.

[0074] The following is part of the data collected during the experiment, including two groups of comparisons: one group is the traditional method under the existing technology (a standard processing method based on conventional CNC milling path planning and traditional cutting parameter setting), and the other group is the experimental data after using the method of the present invention. All data are based on the actual experimental environment and workpiece. The details are shown in Table 1 below: Table 1 Experimental data comparison table parameter Existing technology (traditional method) Method of the present invention Cutting force(N) 35.6 31.2 Cutting vibration (mm / s) 0.018 0.011 Cutting temperature (°C) 95 85 Tool wear (mm) 0.004 0.002 Processing time (minutes) 90 75 Processing accuracy (μm) 12 4 Cutting path deviation (μm) 20 5 According to the experimental data table, it can be clearly seen that the method of the present invention has significant advantages in many aspects. First, the reduction in cutting force (from 35.6N to 31.2N) shows that the cutting resistance generated during the contact between the tool and the workpiece is effectively reduced, which helps to extend the tool life and reduce possible tool damage during the cutting process.

[0075] Secondly, the cutting vibration is reduced from 0.018mm / s to 0.011mm / s, indicating that the path optimization method of the present invention effectively reduces the vibration during milling, can improve the stability of the machining process, and reduce unnecessary errors. In terms of cutting temperature, the cutting temperature under the traditional method is 95°C, while the temperature under the present invention is reduced to 85°C. This means that the generation of heat during the machining process is effectively controlled, thereby avoiding material deformation and tool wear caused by overheating, and improving the machining accuracy and tool durability.

[0076] In addition, the significant reduction in tool wear (from 0.004mm to 0.002mm) also reflects the optimization of the method in controlling cutting conditions, especially after combining real-time feedback data and quantum reinforcement learning algorithms, tool wear has been effectively suppressed. Regarding processing time, the method of the present invention shortens the processing time from 90 minutes to 75 minutes, improving production efficiency, which is particularly important for high-precision processing, especially in industrial production, and can significantly increase the output of the production line.

[0077] The present invention has obvious advantages in terms of machining accuracy. The machining accuracy of the traditional method is 12 μm, while the accuracy of the present invention reaches 4 μm, which shows the accuracy of the present invention in optimizing the milling path and machining parameters.

[0078] Finally, the cutting path deviation was reduced from 20 μm to 5 μm, which further verified that the optimization of the path through quantum reinforcement learning and particle swarm optimization algorithm greatly reduced the path error and ensured the processing quality of high-precision sealing surface.

[0079] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A method for high-precision sealing surface of a CNC gantry milling center, characterized in that: include, Fix the workpiece on the workbench of the CNC gantry milling machine and set the processing parameters through automatic tool setting; Based on the machining parameters, the random forest algorithm is used to generate the milling path; According to the milling path, the workpiece is milled, and the processing parameters of the milling process are collected in real time to generate processing status monitoring data; Based on the processing status monitoring data, the milling path is dynamically optimized by combining the hyperparameters of quantum reinforcement learning; The particle swarm optimization algorithm is used to calculate the correlation between machining parameters in the milling process and solve the optimal parameter combination; The sealing surface finishing operation is performed according to the optimized milling path and the optimal parameter combination.

2. The method for high-precision sealing surface of a CNC gantry milling machining center as claimed in claim 1, characterized in that: The workpiece is fixed on the workbench of the CNC gantry milling machine, and the processing parameters are set by automatic tool setting. The specific steps are as follows: Use intelligent positioning fixture to calibrate the placement of workpieces on the workbench of CNC gantry milling machine; Activate the automatic tool setting function of the CNC equipment, measure the distance between the tool and the workpiece surface through the tool setting detector, and correct the initial position of the tool; Based on the initial position of the tool, combined with the material properties of the workpiece, the CNC sets the feed rate; Input the workpiece material, processing technology and tool information into the CNC equipment and set the cutting depth; Based on the set cutting depth and feed rate, the CNC equipment calculates the tool speed by combining the cutting characteristics in the material properties of the workpiece; Integrate cutting depth, feed rate and tool speed to generate machining parameters.

3. The method for high-precision sealing surface of a CNC gantry milling machining center as claimed in claim 2, characterized in that: Based on the processing parameters, the random forest algorithm is used to generate the milling path. The specific steps are as follows: Clean, denoise, standardize and handle outliers of processing parameters; Extract machining parameters, material properties of workpieces and tool information from the historical machining records of CNC machining equipment as feature variables; The processing performance index and path evaluation index are used as target variables; Construct a training data set through feature variables and target variables, and use the training data set to train the random forest model; Cross-validation is used to optimize the parameters of the random forest model, and the performance of the random forest model is tested by comparing the predicted values ​​with the actual values ​​to obtain a trained random forest model; The machining parameters and workpiece machining targets are used as input parameters and fed into the trained random forest model to predict the initial path; The initial path is simulated, the local characteristics of the path are optimized according to the simulation feedback, and the optimized milling path is output.

4. The method for high-precision sealing surface of a CNC gantry milling machining center as claimed in claim 3, characterized in that: According to the milling path, the workpiece is milled, and the processing parameters of the milling process are collected in real time to generate processing status monitoring data. The specific steps are as follows: Use the optimized milling path as the processing instruction of the CNC equipment; Start the CNC equipment according to the processing instructions and use the segmented processing strategy to gradually complete the workpiece processing; After each section of workpiece is processed, sensors are used to check the milling path status and evaluate the tool status; If a deviation in the milling path or an abnormality in the tool is detected, the central controller adjusts the parameters of the next milling path or prompts to replace the tool based on the evaluation results; Collect cutting force, vibration and temperature data in real time during milling process and integrate them into feedback data set; transmitting feedback data sets to a central controller for storage; The central controller performs structured processing on the stored feedback data set, records the processing timestamp, parameter change trend and abnormal status, and generates processing status monitoring data.

5. The method for high-precision sealing surface of a CNC gantry milling machining center as claimed in claim 4, characterized in that: The specific steps of dynamically optimizing the milling path based on the processing state monitoring data and combining the hyperparameters of quantum reinforcement learning are as follows: Standardize the processing state monitoring data and map it into the quantum state space; Each processing state monitoring data in the quantum state space is represented by a quantum bit, and the expression is: ； in, Indicates qubits, represents the normalized cutting force , Represents the normalized cutting force vibration data , represents the normalization operation, represents the zero state of the quantum bit, represents a state of a quantum bit, An index variable representing a qubit; Map different processing state monitoring data to corresponding quantum bits respectively, and associate the quantum bits through quantum entanglement operation to form a joint quantum state of the overall processing state , the expression is: ； in, Representing the cutting force in the quantum bit The quantum state of Representing vibration data in qubits The quantum state of Representing temperature data in quantum bits The quantum state of Representing tool wear data in quantum bits The quantum state of represents the tensor product of quantum states; Using joint quantum states Participate in the calculation of the immediate reward function, the expression is: ； in, Indicates at time and hyperparameters The quantitative index value of the machining path optimization effect under Indicates time and hyperparameters The reward operator under Representing the joint quantum state The conjugate transpose of Representing the joint quantum state original state, Indicates time Processing data collected at all times; Loss Function Based on processing state data, quantum state and immediate reward function , weighted feedback is given to the processing state, and the expression is: ； in, represents the upper limit of the integral interval, For time from arrive The integral of is the symbol of the second-order partial derivative, Indicates time The second-order partial derivative of represents a non-negative weight coefficient, is the immediate reward function For hyperparameters The gradient of is the square of the gradient magnitude, represents a small increment of the integral variable, is the regularization coefficient, is the total number of hyperparameters, The index variable representing the hyperparameter, represents a nonlinear periodic function, Indicates The square of the hyperparameters, Represents the form of hyperparameter regularization; According to real-time processing data , calculate the second-order change trend of the reward , evaluate the dynamic trend of processing performance; Based on the dynamic trend of the evaluated processing performance, combined with the immediate reward function For hyperparameters The gradient of , evaluates the path optimization direction; Based on the joint quantum state , the quantum state influence factor when calculating the path optimization parameter update is expressed as: ； in, represents the quantum state influence factor, Represents the excited state in the quantum state The square of the probability magnitude, Represents the excited state in the quantum state The square of the probability magnitude; The square sum quantum state influence factor based on gradient norm , calculate the optimized hyperparameter set, the expression is: ； in, Represents the time step The optimized hyperparameter set is represents the learning rate, Represents the logarithmic transformation of the absolute value of the current parameter, represents the smoothing term; Generate multiple candidate paths based on the optimized hyperparameter set; Use simulation tools to evaluate candidate paths with an instant reward function The highest path is used as the milling path.

6. The method for high-precision sealing surface of a CNC gantry milling machining center as claimed in claim 5, characterized in that: The specific steps of generating multiple candidate paths based on the optimized hyperparameter set are as follows: Using the optimized hyperparameter set and real-time processing status data, determine the upper and lower limits of the path parameters; The search space for path generation is limited by the upper and lower limits of the path parameters, and several candidate paths are generated within the search range using the Monte Carlo method.

7. The method for high-precision sealing surface of a CNC gantry milling machining center as claimed in claim 6, characterized in that: The particle swarm optimization algorithm is used to calculate the correlation between the machining parameters in the milling process and solve the optimal parameter combination. The specific steps are as follows: The machining parameters in the milling process are defined as the dimensions of the particle group, the particle positions and velocities are initialized, and the initial parameter combinations are generated; Based on the cutting force, vibration and temperature data in the feedback data set, the fitness value is calculated using the weighted summation method; Update the particle velocity and adjust the particle position by using the inertia weight and acceleration factor in the initial parameter combination; Re-evaluate the fitness value according to the adjusted new particle position, and update the historical optimal position and global optimal position of each particle; Iteration continues until the maximum number of iterations is reached, and the global optimal particle position is output as the optimal combination of processing parameters.

8. The method for high-precision sealing surface of a CNC gantry milling machining center as claimed in claim 7, characterized in that: The sealing surface finishing operation is performed according to the optimized milling path and the optimal parameter combination. The specific steps are as follows: Load the optimized milling path and the best combination of machining parameters, and verify the path through CNC simulation software; Activate the CNC equipment, adjust the tool path according to the simulation verification results, start preliminary cutting, and monitor the cutting force, vibration and temperature data in real time; Based on real-time monitoring of cutting force, vibration and temperature data, the initial cutting results are evaluated and the path corrected; After the initial cutting is correct, the optimized milling path is used for finishing operations; During the milling process, machining parameters are fine-tuned based on real-time data feedback.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for high-precision sealing surface of a CNC gantry milling machining center described in any one of claims 1 to 8 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for high-precision sealing surface of a CNC gantry milling machining center described in any one of claims 1 to 8 are implemented.

Images