Automatic control method and system for milling cutter machining

Abstract

The invention relates to an automatic control method and system for milling cutter machining, and the method comprises the steps: obtaining geometric feature data and technological parameter information of a milling cutter, and carrying out the machining analysis of the geometric feature data based on the technological parameter information, and obtaining the corresponding original machining flow information; acquiring machining flow sensing data of the milling cutter, and performing state association on the machining flow sensing data and the original machining flow information to obtain a corresponding real-time machining state group; real-time cutting parameters of the milling cutter are obtained, machining prediction is conducted on the original machining process information according to the real-time cutting parameters, and a corresponding machining quality trend is obtained; performing intelligent control optimization on the real-time processing state group and the processing quality trend to obtain a corresponding initial control scheme; and performing global parameter optimization on the initial control scheme according to the compensation control parameters to obtain a corresponding global processing optimization scheme. According to the invention, the real-time processing state group and the processing quality trend can be intelligently controlled and optimized.

Description

Technical Field

[0001] The invention relates to the technical field of milling cutter processing, and in particular to an automatic control method and system for milling cutter processing. Background Art

[0002] With the rapid development of the manufacturing industry, milling has been widely used in the field of modern manufacturing as an important mechanical processing method. In order to improve processing efficiency and processing quality, how to realize the intelligent and automated control of the milling cutter processing process has become one of the key research topics. Existing milling cutter processing control methods are mostly limited to the adjustment of a single process parameter, such as simple control of the spindle speed or feed speed, while ignoring the comprehensive influence of multi-dimensional factors such as milling cutter geometric characteristics, cutting parameters, and processing status during the processing. This isolated control method is difficult to accurately grasp the dynamic changes of the processing process, which often leads to problems such as unstable processing accuracy and surface quality fluctuations, and cannot meet the needs of modern manufacturing for high-precision and high-efficiency processing. Summary of the invention

[0003] The main purpose of the present invention is to provide an automated control method and system for milling cutter processing, which can formulate a more reasonable processing control plan by intelligently controlling and optimizing the real-time processing state group and processing quality trend.

[0004] To achieve the above object, the present invention provides an automatic control method for milling cutter processing, comprising: Acquire geometric feature data and process parameter information of the milling cutter, perform processing analysis on the geometric feature data based on the process parameter information, and obtain corresponding original processing flow information; Acquire the processing flow sensor data of the milling cutter, and associate the data with the original processing flow information to obtain a corresponding real-time processing state group; Acquire the real-time cutting parameters of the milling cutter, perform processing prediction on the original processing flow information according to the real-time cutting parameters, and obtain the corresponding processing quality trend; Performing intelligent control optimization on the real-time processing state group and the processing quality trend to obtain a corresponding initial control solution; Based on the machining quality trend, error identification is performed on the real-time cutting parameters, and cutting correction compensation analysis is performed to obtain corresponding compensation control parameters; The initial control scheme is globally optimized according to the compensation control parameters to obtain a corresponding global processing optimization scheme.

[0005] Furthermore, the acquiring of the geometric feature data and process parameter information of the milling cutter, and the processing analysis of the geometric feature data based on the process parameter information to obtain the corresponding original processing flow information include: Acquire cutting edge angle data, cutting edge length data and tool diameter data obtained by optically scanning and measuring the milling cutter; Performing geometric feature vector matrix operation on the cutting edge angle data, the cutting edge length data and the tool diameter data to obtain the geometric feature data; Searching and matching the geometric feature data with a preset process parameter database to obtain corresponding process parameter information; Perform machining process analysis according to the process parameter information and the geometric feature data to obtain a corresponding original machining trajectory; Segmenting the original processing trajectory to obtain a corresponding processing path point sequence; Calculating the path point cutting force components of the processing path point sequence according to the geometric feature data to obtain original force distribution data; Performing critical value analysis on the original force distribution data to obtain an original processing parameter range; A processing control instruction sequence is generated according to the original processing parameter range and the processing path point sequence to obtain the original processing flow information.

[0006] Furthermore, the processing flow sensor data of the milling cutter is obtained, and the processing flow sensor data is associated with the original processing flow information to obtain a corresponding real-time processing state group, including: Monitoring the machining process of the milling cutter to obtain original sensor monitoring data; Performing processing data screening on the original sensor monitoring data to obtain corresponding processing flow sensor data; Performing processing state clustering on the processing flow sensor data to obtain corresponding processing state feature clusters; Performing information mapping on the processing state feature cluster to obtain corresponding processing state mapping information; Performing process clustering on the original process information according to the process sensing data to obtain corresponding process type information; The process type information is state-combined and associated according to the processing state mapping information to obtain the real-time processing state group.

[0007] Furthermore, the acquiring of the real-time cutting parameters of the milling cutter and the performing of processing prediction on the original processing flow information according to the real-time cutting parameters to obtain the corresponding processing quality trend include: Acquiring an acoustic emission signal and a cutting acceleration signal of the milling cutter during a cutting process, and performing multi-source heterogeneous processing on the acoustic emission signal and the cutting acceleration signal to obtain an original cutting data stream; Extracting real-time parameters from the original cutting data stream to obtain the real-time cutting parameters; Performing feature decomposition on the real-time cutting parameters to obtain corresponding feature frequency band groups; Performing adaptive threshold analysis on the characteristic frequency band group to obtain a corresponding cutting stability index; Perform a prediction function according to the cutting stability index and the real-time cutting parameters to obtain a corresponding cutting force prediction mapping function; Calculating the workpiece surface quality according to the cutting force prediction mapping function to obtain a corresponding machining accuracy evaluation value; Performing error compensation analysis on the machining accuracy evaluation value to obtain a corresponding dynamic error compensation amount; The dynamic error compensation amount is predicted by time series regression according to the cutting force prediction mapping function to obtain a corresponding machining quality trend.

[0008] Furthermore, the prediction function is performed according to the cutting stability index and the real-time cutting parameter to obtain the corresponding cutting force prediction mapping function, including: Performing Fourier transform on the cutting stability index to obtain a corresponding frequency domain feature sequence; Performing least square fitting on the real-time cutting parameters according to the frequency domain feature sequence to obtain a corresponding parameter fitting curve; Performing piecewise linearization processing on the parameter fitting curve to obtain a corresponding piecewise function group; Performing a polynomial interpolation operation on the frequency domain feature sequence according to the piecewise function group to obtain a corresponding interpolation coefficient matrix; Performing singular value decomposition on the interpolation coefficient matrix to obtain a corresponding eigenvalue sequence; Performing orthogonal basis function expansion on the piecewise function group according to the eigenvalue sequence to obtain a corresponding basis function set; Optimizing the basis function set using the Lagrange multiplier method to obtain a corresponding constraint condition group; Solving the basis function set by the calculus of variations according to the constraint condition group to obtain a corresponding cutting force prediction mapping function; Wherein, the cutting force prediction mapping function is: ; represents the predicted cutting force, is a set of basis functions obtained by expanding the orthogonal basis functions. is the coefficient corresponding to the eigenvalue sequence obtained by singular value decomposition, is the Lagrange multiplier, h is the constraint function, n is the number of orthogonal basis functions, and t is any time in the cutting process.

[0009] Furthermore, the intelligent control optimization of the real-time processing state group and the processing quality trend to obtain a corresponding initial control scheme includes: Performing multi-dimensional feature decomposition on the real-time processing state group to obtain a corresponding state feature matrix; Performing a time series correlation analysis on the processing quality trend according to the state characteristic matrix to obtain a corresponding state correlation vector; Performing nonlinear mapping transformation on the state association vector to obtain a corresponding control feature space; Performing dynamic optimization calculation on the processing parameters according to the control feature space to obtain a corresponding parameter optimization matrix; Performing machining trajectory planning on the parameter optimization matrix to obtain a corresponding trajectory control sequence; Adaptively adjust and calculate the cutting force according to the trajectory control sequence to obtain corresponding force control parameters; The force control parameters are subjected to multi-objective collaborative optimization to obtain a corresponding initial control solution.

[0010] Furthermore, the error identification of the real-time cutting parameters based on the machining quality trend and the cutting correction compensation analysis to obtain corresponding compensation control parameters include: Extracting the features of the real-time cutting parameters in sections to obtain a corresponding cutting feature sequence; Performing time-frequency domain analysis on the machining quality trend according to the cutting feature sequence to obtain a corresponding frequency spectrum feature matrix; Performing error clustering on the frequency spectrum feature matrix to obtain a corresponding error category set; Performing membership evaluation on the error category set to obtain a corresponding error membership matrix; Calculating the compensation amount of the cutting parameters according to the error membership matrix to obtain the corresponding initial compensation parameters; Constructing a response curve for the initial compensation parameters to obtain a corresponding response characteristic curve; Calculating the correlation degree of the response characteristic curve to obtain a corresponding correlation degree sequence; The correlation degree sequence and the initial compensation parameter are collaboratively optimized to obtain the corresponding compensation control parameter.

[0011] Furthermore, the initial control scheme is globally optimized according to the compensation control parameters to obtain a corresponding global processing optimization scheme, including: Parameter stratification is performed on the compensation control parameters to obtain corresponding process layer parameter groups and control layer parameter groups; Calculate the target cutting parameter of the initial control scheme according to the process layer parameter group to obtain a corresponding cutting optimization parameter matrix; Performing parameter sensitivity analysis on the control layer parameter group to obtain corresponding parameter influence weight coefficients; Performing weighted calculation on the cutting optimization parameter matrix according to the parameter influence weight coefficient to obtain a corresponding weighted optimization parameter group; Performing constraint boundary analysis on the weighted optimization parameter group to obtain the corresponding parameter feasible domain range; According to the feasible domain range of the parameters, nonlinear programming is performed on the weighted optimization parameter group to obtain a corresponding target control parameter group; Performing stability verification on the target control parameter group to obtain corresponding stability index data; Perform dynamic compensation according to the stability index data and the target control parameter group to obtain corresponding dynamic compensation parameters; The initial control scheme is globally optimized according to the dynamic compensation parameters and the target control parameter group to obtain the global processing optimization scheme.

[0012] The present invention further provides an automated control system for milling cutter processing, characterized in that the automated control method for milling cutter processing applied to any one of the above-mentioned methods comprises: An acquisition module, the acquisition module is used to obtain geometric feature data and process parameter information of the milling cutter, and perform processing analysis on the geometric feature data based on the process parameter information to obtain corresponding original processing flow information; An analysis module, the analysis module is used to obtain the processing flow sensor data of the milling cutter, and associate the data with the original processing flow information to obtain a corresponding real-time processing state group; An association module, the association module is used to obtain the real-time cutting parameters of the milling cutter, perform processing prediction on the original processing flow information according to the real-time cutting parameters, and obtain a corresponding processing quality trend; A processing module, the processing module is used to perform intelligent control optimization on the real-time processing state group and the processing quality trend to obtain a corresponding initial control solution; A control module, the control module is used to perform error identification on the real-time cutting parameters based on the machining quality trend, and perform cutting correction compensation analysis to obtain corresponding compensation control parameters; An execution module is used to perform global parameter optimization on the initial control scheme according to the compensation control parameters to obtain a corresponding global processing optimization scheme.

[0013] The present invention provides an automatic control method and system for milling cutter processing, which has the following beneficial effects: By acquiring the geometric feature data and process parameter information of the milling cutter for processing analysis, it is possible to more comprehensively evaluate the various parameter requirements of the processing process, thereby improving the accuracy of the processing plan formulation and providing a reliable data basis for automatic control. By associating the processing process sensor data with the original processing process information, real-time monitoring and analysis of the processing process can be achieved, which helps to timely discover and deal with abnormal situations in the processing process and ensure the stability of processing quality. Processing prediction of the original processing process information based on real-time cutting parameters can ensure that good processing effects can be maintained under different processing conditions, reducing processing quality fluctuations. By intelligently controlling and optimizing the real-time processing state group and processing quality trend, a more reasonable processing control plan can be formulated, and the precise adjustment of processing parameters can be achieved through cutting correction compensation analysis, thereby effectively improving processing accuracy. At the same time, global parameter optimization based on compensation control parameters can flexibly adjust the control strategy according to the processing characteristics and requirements of different workpieces, making the processing system more adaptable to diverse processing needs. BRIEF DESCRIPTION OF THE DRAWINGS

[0014] Figure 1 It is a flow chart of an automated control method for milling cutter processing provided by the present invention; Figure 2 The invention discloses a structure diagram of an automatic control system for milling cutter processing.

[0015] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0016] 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.

[0017] The present invention is further described below in conjunction with the accompanying drawings and specific implementation methods.

[0018] Reference Figure 1 As shown, the present invention provides 1. an automatic control method for milling cutter processing, comprising: Step S1: acquiring geometric feature data and process parameter information of the milling cutter, and performing processing analysis on the geometric feature data based on the process parameter information to obtain corresponding original processing flow information; Step S2: Acquire the processing flow sensor data of the milling cutter, and associate the data with the original processing flow information to obtain a corresponding real-time processing state group; Step S3: obtaining the real-time cutting parameters of the milling cutter, and performing processing prediction on the original processing flow information according to the real-time cutting parameters to obtain the corresponding processing quality trend; Step S4: performing intelligent control optimization on the real-time processing state group and processing quality trend to obtain a corresponding initial control solution; Step S5: Based on the machining quality trend, the real-time cutting parameters are subjected to error identification, and cutting correction compensation analysis is performed to obtain corresponding compensation control parameters; Step S6: Optimize the global parameters of the initial control scheme according to the compensation control parameters to obtain the corresponding global processing optimization scheme.

[0019] Based on the above steps, the detailed process is as follows: Step S1: Collect the key geometric feature data of the milling cutter, including the basic parameters of the tool such as diameter, number of blades, helix angle, rake angle, and back angle. At the same time, obtain the process parameter information, such as cutting speed, feed speed, cutting depth, cooling method and other processing requirements. This information is stored in a preset database, and these parameters are analyzed by a special algorithm. During the analysis process, factors such as the machinability of the material, the wear resistance of the tool, and the shape characteristics of the workpiece are considered, and the key parameters in the processing process such as cutting force and cutting temperature are predicted by establishing a mathematical model. Based on these analysis results, a detailed original processing process information is generated, including recommended cutting parameters, tool path planning, feed strategy, etc. This process information will serve as the basic basis for subsequent processing control, and also provide a reference standard for real-time monitoring. The generation of the original processing process information needs to consider the balance of multiple aspects such as the quality requirements of the workpiece, production efficiency, and cost, to ensure that the generated processing plan can meet the process requirements and has high economic efficiency.

[0020] Step S2: Various physical signals in the processing process are collected in real time through various sensors arranged on the machine tool, such as force sensors, vibration sensors, temperature sensors, acoustic emission sensors, etc. These sensor data are collected and preprocessed by the data acquisition system to remove noise and perform signal conditioning. The system compares and correlates these real-time collected data with the original processing flow information. By establishing a data mapping relationship, it is judged in real time whether the current processing status meets expectations, and a real-time processing status group is formed. This status group contains status information in multiple dimensions such as cutting force, vibration, temperature, etc., which can fully reflect the current processing status. Through the comprehensive analysis of these status information, abnormal conditions in the processing process can be discovered in time, providing a basis for subsequent control optimization.

[0021] Step S3: Real-time collection of current cutting parameters, including actual operating parameters such as spindle speed, feed speed, and cutting depth. These parameters are directly obtained through the CNC system of the machine tool to ensure the accuracy and real-time nature of the data. These real-time parameters are compared and analyzed with the original processing flow information, and the influence of the current processing state on the final processing quality is evaluated through the established prediction model. Predict the change trend of quality indicators such as the surface roughness, dimensional accuracy, and shape accuracy of the workpiece.

[0022] Step S4: Based on the real-time processing state group and processing quality trend, start intelligent control optimization. Establish a multi-objective optimization model to comprehensively consider multiple objective functions such as processing efficiency, processing quality and tool life. By adopting intelligent optimization algorithms, such as genetic algorithms, particle swarm algorithms or ant colony algorithms, global search and optimization of various control parameters are performed. In the optimization process, the changing trends of physical quantities such as cutting force, vibration, temperature, etc. reflected in the real-time processing state group are involved, and the optimal control parameter combination is determined in combination with the processing quality prediction results. This optimization process is a dynamic adjustment process. According to the characteristics of different processing stages, the weights of the optimization objectives are dynamically adjusted to ensure the rationality of the control scheme. The final initial control scheme includes adjustment strategies for key parameters such as spindle speed, feed speed, and cutting depth, and also includes special control requirements for different processing stages.

[0023] Step S5: Perform error analysis on the real-time cutting parameters. Based on the processing quality trend obtained in step S3, compare the deviation between the actual cutting parameters and the theoretical parameters in real time. Various types of errors, including systematic errors, random errors, and periodic errors, are identified through signal processing methods such as Fourier analysis and wavelet analysis. Based on error identification, the compensation control parameters are calculated. The compensation analysis process takes into account the dynamic characteristics of the machine tool, the deformation characteristics of the workpiece, and the influence of environmental factors, and determines the appropriate compensation amount through an adaptive control algorithm. Establish a compensation parameter database to record the compensation effects under different working conditions, and continuously optimize the compensation strategy through machine learning methods. The compensation control parameters finally generated include compensation amounts in multiple dimensions such as position compensation, velocity compensation, and acceleration compensation.

[0024] Step S6: Integrate the obtained initial control scheme with the obtained compensation control parameters to establish a global optimization model. This model uses the analytic hierarchy process or the fuzzy comprehensive evaluation method to perform a global evaluation of all control parameters. The optimization process takes into account the stability, reliability and economy of the machining system. By establishing the state space equation, modern control theory is used to optimize the parameters. Perform multiple iterative calculations to find the optimal parameter combination under the premise of meeting various constraints. The global machining optimization scheme not only includes the specific control parameter values, but also includes the timing arrangement of parameter adjustment and the emergency treatment strategy. This scheme can ensure the stability of the entire machining process and the consistency of machining quality.

[0025] The present invention provides an automatic control method for milling cutter processing. By acquiring the geometric feature data and process parameter information of the milling cutter for processing analysis, it is possible to more comprehensively evaluate the various parameter requirements of the processing process, thereby improving the accuracy of the processing plan formulation and providing a reliable data basis for automatic control. By linking the processing process sensor data with the original processing process information, real-time monitoring and analysis of the processing process is achieved, which helps to timely discover and handle abnormal situations in the processing process and ensure the stability of the processing quality. Based on the real-time cutting parameters, the original processing process information is processed and predicted, which can ensure that good processing effects can be maintained under different processing conditions, reducing the fluctuation of processing quality. By intelligently controlling and optimizing the real-time processing state group and processing quality trend, a more reasonable processing control plan is formulated, and the precise adjustment of processing parameters is achieved through cutting correction compensation analysis, thereby effectively improving the processing accuracy. At the same time, global parameter optimization is performed based on the compensation control parameters, and the control strategy can be flexibly adjusted according to the processing characteristics and requirements of different workpieces, so that the processing system is more adaptable to diversified processing needs.

[0026] In one embodiment, the geometric feature data and process parameter information of the milling cutter are obtained, and the geometric feature data is processed and analyzed based on the process parameter information to obtain the corresponding original processing flow information, including: In the process of automated control of milling cutter processing, the milling cutter is subjected to a full range of three-dimensional optical scanning. The milling cutter surface is scanned and imaged at multiple angles by using a high-precision CCD camera array in conjunction with a ring light source. The scanning measurement accuracy reaches ±0.001mm, and the milling cutter's cutting edge angle data, cutting edge length data, and tool diameter data are obtained by scanning. The cutting edge angle data includes the front angle, back angle, and cutting edge inclination angle; the cutting edge length data includes the actual processing length of each cutting edge; and the tool diameter data includes the nominal diameter and actual processing diameter of the tool.

[0027] A geometric feature vector matrix is ​​established for the acquired cutting edge angle data, cutting edge length data and tool diameter data. The matrix is ​​based on the tool coordinate system and converts each measurement data into a standardized feature vector. The spatial geometric feature data of the tool is obtained through matrix operation, including the tool contour curve equation, cutting edge spatial distribution function and cutting edge trajectory equation.

[0028] The built-in process parameter database stores the recommended cutting parameters under different processing conditions. The database is classified according to the workpiece material, processing method and precision requirements, and includes parameters such as cutting speed, feed rate, cutting depth, etc. The system matches and searches the calculated geometric feature data with the database to extract the applicable process parameter information.

[0029] Based on the acquired process parameter information and geometric feature data, the system uses numerical simulation methods to analyze the machining process. The stress and strain distribution during the cutting process is calculated using the finite element method to obtain the relative motion trajectory between the milling cutter and the workpiece, forming the original machining trajectory data.

[0030] The original machining trajectory is segmented, and the path points are arranged on the trajectory curve according to the principle of equal arc length. The density of the path points is adaptively adjusted according to the change of the trajectory curvature, and the density of the path points increases where the curvature is larger. This generates a sequence of machining path points containing spatial coordinates and tangent vector information.

[0031] For each path point, the cutting force components are calculated based on the cutting force prediction model. The model comprehensively considers material properties, cutting parameters and tool geometry to calculate tangential force, radial force and axial force. The cutting force calculation results are superimposed on the path point sequence to obtain the original force distribution data of the entire machining process.

[0032] Set safety thresholds for the original force distribution data, including maximum cutting force, maximum torque and maximum power limits. Determine the reasonable range of each processing parameter through critical value analysis to form the original processing parameter range. This parameter range ensures that the cutting force, torque and power during the processing are within the tolerance of the equipment.

[0033] According to the original machining parameter range, the machining path point sequence is converted into CNC machining code. The code sequence contains process parameters such as feed speed, spindle speed and tool position and posture information, forming a complete original machining process information. This process information is used as the input instruction of the CNC system to realize the automatic control of the milling cutter machining process.

[0034] This embodiment uses a high-precision optical scanning system to measure the milling cutter in all directions and obtain complete geometric feature data, which overcomes the defects of insufficient accuracy and incomplete data acquisition of traditional measurement methods. Based on the geometric feature vector matrix operation method, the accurate digital expression of tool features is realized, providing an accurate data basis for subsequent processing analysis. The process parameter database matching method is adopted, so that the system can quickly select appropriate processing parameters and improve the efficiency of processing plan formulation. Through numerical simulation analysis of the cutting process, the processing trajectory and force distribution are predicted, which avoids the trial cutting link in actual processing and reduces the waste of materials and time. The introduction of an adaptive density adjustment mechanism in path planning ensures the processing accuracy of complex contours and optimizes the processing efficiency. By setting the safety thresholds of cutting force, torque and power, a complete parameter optimization system is established to ensure the safety and reliability of the processing process and extend the service life of equipment and tools. The entire method realizes the full process automation control from measurement to processing, significantly improving the processing quality and efficiency.

[0035] In one embodiment, the processing flow sensor data of the milling cutter is obtained, and the state is associated with the original processing flow information to obtain the corresponding real-time processing state group, including: The milling cutter is monitored through multiple sensors, including temperature sensors, pressure sensors, vibration sensors and displacement sensors. These sensors collect various data information generated by the milling cutter during the processing in real time to form original sensor monitoring data. The sampling frequency of the sensor is set to 100 times per second to ensure the continuity and accuracy of data collection.

[0036] The collected raw sensor monitoring data is processed and screened to remove abnormal values ​​and invalid data. The abnormal value judgment standard is: the data exceeds the preset threshold range, or the deviation from the adjacent data point exceeds 3 times the standard deviation. The processing process sensor data obtained after screening is more representative and reliable, providing an effective data basis for subsequent analysis.

[0037] The K-means clustering algorithm is used to cluster the processing state of the processing process sensor data, and the data points with similar characteristics are classified into the same cluster to form a processing state feature cluster. The optimal value of the cluster number K is determined by the silhouette coefficient method, and the feature vector contains data in four dimensions: temperature, pressure, vibration, and displacement.

[0038] The information mapping of the processing state feature cluster is carried out, the corresponding relationship between the feature cluster and the processing state is established, and the processing state mapping information is generated. The neural network model includes an input layer, two hidden layers and an output layer, and uses the ReLU activation function to optimize the network parameters through the back propagation algorithm.

[0039] Based on the hierarchical clustering method, the original processing process information is clustered according to the processing process sensor data. The cluster distance uses the Euclidean distance, and the link method uses the hierarchical clustering link method. The optimal number of clusters is determined through dendrogram analysis to obtain the process type information.

[0040] The association rule mining algorithm is used to associate the process type information with the state combination according to the processing state mapping information. The minimum support is set to 0.3 and the minimum confidence is set to 0.8. The association rules between the state and the process are mined to finally form a real-time processing state group. The real-time processing state group contains the current processing state, processing parameters and corresponding process type information, providing a decision-making basis for the automated control of the milling cutter processing process.

[0041] This embodiment deploys a variety of sensors for real-time monitoring during the milling cutter processing, combined with a data screening mechanism, effectively eliminates outliers and invalid data, and improves data quality and reliability. The K-means clustering algorithm is used to perform state clustering on the processing process sensor data, accurately identify data groups with similar characteristics, and lay a good foundation for subsequent analysis. By establishing a mapping relationship between feature clusters and processing states, the accuracy and robustness of state recognition are improved. The original processing process information is classified by a hierarchical clustering method, and the association rules between states and processes are established in combination with an association rule mining algorithm, which realizes accurate identification of processing states and intelligent matching of processes. This method significantly improves the level of automated control of the milling cutter processing process, reduces manual intervention, optimizes processing efficiency, and provides a strong guarantee for improving the quality of milling cutter processing.

[0042] In one embodiment, the real-time cutting parameters of the milling cutter are obtained, and the original processing flow information is processed according to the real-time cutting parameters to obtain the corresponding processing quality trend, including: During the milling process, a microphone and an accelerometer are set to collect the acoustic emission signal and cutting acceleration signal of the milling cutter during the cutting process. The frequency range of the acoustic emission signal collected by the microphone is 20kHz-2MHz, and the sampling frequency is set to 2MHz; the frequency range of the cutting acceleration signal collected by the accelerometer is 0-10kHz, and the sampling frequency is set to 20kHz. The collected acoustic emission signals and cutting acceleration signals are synchronized in the time domain and normalized in the frequency domain to construct a multi-source heterogeneous data stream to form the original cutting data stream.

[0043] Real-time parameter extraction is performed based on the original cutting data stream, and the extracted parameters include key parameters such as cutting force, cutting vibration, and cutting temperature. The cutting force is obtained by quadratic integration of the acceleration signal; the cutting vibration is obtained by spectrum analysis of the acceleration signal; and the cutting temperature is obtained by the energy characteristics of the acoustic emission signal. The extracted real-time cutting parameters are decomposed by wavelet packets, and the signal is decomposed into 8 frequency bands. Each frequency band corresponds to different cutting state characteristics, forming a characteristic frequency band group.

[0044] For each frequency band of the characteristic frequency band group, calculate its energy entropy value and set the adaptive threshold judgment criterion. When the frequency band energy entropy value exceeds the threshold, the cutting state corresponding to the frequency band is determined to be unstable. Combining the stability judgment results of all frequency bands, the cutting stability index is obtained. The value range of this index is 0-1, 0 means completely unstable and 1 means completely stable.

[0045] The cutting force prediction mapping function is constructed through a deep neural network. The input of this function is the cutting stability index and the real-time cutting parameters, and the output is the predicted cutting force. The neural network adopts a 5-layer structure, including 3 hidden layers. The input layer receives the cutting stability index and the real-time cutting parameter data. The number of neurons in the input layer is the same as the input feature dimension, including parameters such as cutting force, cutting vibration, cutting temperature, and stability index. The first hidden layer is composed of 128 neurons, and the ReLU activation function is used to perform nonlinear transformation on the input data. The mathematical expression of the ReLU function is f(x)=max(0,x), which can effectively solve the gradient vanishing problem of the traditional activation function. The second hidden layer contains 64 neurons and also uses the ReLU activation function. This layer further extracts and transforms the output of the first hidden layer to enhance the nonlinear expression ability of the network. The third hidden layer is set with 32 neurons and uses the ReLU activation function. As the last hidden layer, this layer is responsible for extracting high-level feature representation and providing key information for the final cutting force prediction. The output layer has the same number of neurons as the prediction target dimension, and outputs the predicted cutting force value. This layer uses a linear activation function to directly output the regression prediction result. The layers are connected through full connection, and the weight parameters are optimized and trained based on 2000 sets of cutting condition data through the back propagation algorithm. The batch normalization technology is used in the training process to improve the convergence speed and generalization ability of the network.

[0046] The ReLU activation function is used and the training is performed through the back propagation algorithm. The training dataset contains 2000 sets of cutting condition data.

[0047] Based on the predicted cutting force, combined with the workpiece material characteristics and processing parameters, a workpiece surface quality calculation model is established. The model considers the influence of cutting force on the workpiece surface morphology and calculates processing accuracy indicators such as surface roughness and profile. The model uses a polynomial fitting method to establish a mapping relationship between cutting force and surface quality parameters.

[0048] For the calculated machining accuracy evaluation value, an error compensation model based on Kalman filtering is established. This model takes into account the influence of factors such as machine tool thermal deformation and tool wear, calculates various error components in real time, and obtains the comprehensive error compensation amount. The state equation of the error compensation model contains two state variables: position error and attitude error.

[0049] The dynamic error compensation amount is input into the long short-term memory network for time series prediction to predict the quality change trend in the future processing process. The network contains 64 LSTM units and uses a sliding window with a time step of 100 for prediction. The prediction results are used to guide the real-time adjustment of processing parameters and realize closed-loop control of processing quality.

[0050] This embodiment collects acoustic emission signals and cutting acceleration signals during the milling process, and performs multi-source heterogeneous processing on the signals, effectively realizing the fusion collection of multi-dimensional information during the cutting process, and improving the integrity and accuracy of data collection. The wavelet packet decomposition method is used to perform feature decomposition of real-time cutting parameters, and combined with adaptive threshold analysis, the judgment of cutting stability is more accurate and reliable. The cutting force prediction mapping function constructed based on the deep neural network can accurately predict the dynamic changes in the cutting process, providing a reliable basis for the control of the surface quality of the workpiece. By establishing an error compensation model based on Kalman filtering, the influence of various factors such as thermal deformation of machine tools and tool wear is comprehensively considered, and accurate compensation of processing errors is achieved. Using long short-term memory networks for time series prediction can effectively predict the quality change trend in the future processing process, provide guidance for the real-time adjustment of processing parameters, and finally realize the closed-loop control of processing quality, significantly improving the automation level and processing accuracy of milling processing.

[0051] In one embodiment, a prediction function is performed according to the cutting stability index and the real-time cutting parameters to obtain a corresponding cutting force prediction mapping function, including: The cutting stability indexes include cutting vibration amplitude, cutting force fluctuation rate and cutting temperature change rate. When Fourier transforming the cutting stability indexes, the fast Fourier transform algorithm FFT is used to convert the time domain signal into a frequency domain feature sequence. The frequency domain feature sequence consists of two parts: the amplitude spectrum and the phase spectrum. The amplitude spectrum reflects the energy distribution of each frequency component, and the phase spectrum reflects the phase relationship of each frequency component.

[0052] The real-time cutting parameters include spindle speed, feed rate, cutting depth and tool wear. When performing least squares fitting on the real-time cutting parameters, the frequency domain feature sequence is used as the dependent variable and the cutting parameters are used as the independent variable to establish a multivariate nonlinear regression model. By minimizing the sum of squares of the error and solving the regression coefficient, the parameter fitting curve is obtained. The fitting curve reflects the mapping relationship between the cutting parameters and the frequency domain features.

[0053] When performing piecewise linearization on a parameter fitting curve, it is divided into multiple linear intervals according to the change of the curve slope. In each interval, a piecewise linear function is used for approximation to form a piecewise function group. This piecewise function group retains the main features of the original fitting curve and simplifies the mathematical description.

[0054] When performing polynomial interpolation operations on frequency domain feature sequences based on piecewise function groups, Lagrange interpolation polynomials are selected as basis functions. Interpolation coefficients are calculated at each sampling point to form an interpolation coefficient matrix. This matrix describes the local variation characteristics of frequency domain features at each sampling point.

[0055] When performing singular value decomposition on the interpolation coefficient matrix, it is decomposed into the product form of three matrices. By calculating the eigenvalues ​​and their corresponding eigenvectors, the eigenvalue sequence is obtained. This sequence reflects the main characteristic components of the system and their importance.

[0056] When the orthogonal basis functions of the piecewise function group are expanded according to the eigenvalue sequence, Legendre polynomials are selected as orthogonal basis functions. The piecewise function is expressed as a linear combination of orthogonal basis functions to obtain a basis function set. This set constitutes a complete orthogonal function system that describes the dynamic characteristics of the system.

[0057] When the Lagrange multiplier method is used to optimize the basis function set, the objective function is to minimize the prediction error, and the physical constraints of the system are used as constraints to construct the Lagrangian function. The stationary point of the function is solved to obtain the optimal solution that satisfies the constraints and form a constraint group.

[0058] When solving the basis function set by the calculus of variations according to the constraint condition group, the optimization problem is transformed into the Euler-Lagrange equation. By solving the equation, the cutting force prediction mapping function is obtained.

[0059] Among them, the cutting force prediction mapping function is: ; represents the predicted cutting force, is a set of basis functions obtained by expanding the orthogonal basis functions. is the coefficient corresponding to the eigenvalue sequence obtained by singular value decomposition, is the Lagrange multiplier, h is the constraint function, n is the number of orthogonal basis functions, and t is any time in the cutting process.

[0060] This embodiment achieves accurate extraction of the dynamic characteristics of the cutting process by performing Fourier transform and least squares fitting on the cutting stability index, laying the foundation for subsequent parameter optimization. The combination of piecewise linearization and polynomial interpolation operations effectively reduces the complexity of the mathematical model and improves the calculation efficiency. The introduction of singular value decomposition and orthogonal basis function expansion methods successfully constructs a complete description of the system characteristics and enhances the robustness of the model. Combining Lagrange multiplier optimization and variational solution technology, the optimal cutting force prediction under the conditions of satisfying physical constraints is achieved, and the processing accuracy is significantly improved. This method not only overcomes the problem of insufficient accuracy of traditional cutting force prediction methods, but also realizes real-time adaptive control of the cutting process, providing effective guarantee for improving processing quality and efficiency.

[0061] In one embodiment, intelligent control optimization is performed on the real-time processing state group and the processing quality trend to obtain a corresponding initial control solution, including: By performing feature decomposition on the real-time machining state group data including machining parameters such as cutting force, spindle speed, feed speed, etc., the main feature vectors are extracted to construct the state feature matrix. The state feature matrix reflects the relationship between various parameters in the machining process and their changing rules.

[0062] Based on the obtained state feature matrix, the dynamic time warping algorithm is used to perform time series correlation analysis on the processing quality trend to obtain the state association vector. The algorithm calculates the correlation coefficient between the processing state and the processing quality index at different times, establishes the state-quality mapping relationship, and forms a state association vector that characterizes the dynamic characteristics of the processing process.

[0063] When the state association vector is nonlinearly mapped, the radial basis function neural network is used to realize the mapping from high-dimensional feature space to low-dimensional control feature space. By selecting appropriate kernel functions and network structure parameters, the state association vector is converted into a feature expression form that is convenient for control optimization.

[0064] After obtaining the control feature space, the particle swarm optimization algorithm is used to dynamically optimize the processing parameters. The algorithm takes processing efficiency and processing quality as the optimization goals, searches for the optimal solution within the set parameter constraints, and generates a parameter optimization matrix. The parameter optimization matrix contains the optimal combination of key processing parameters such as spindle speed and feed speed.

[0065] When planning the machining trajectory for the parameter optimization matrix, an improved B-spline curve interpolation algorithm is used to generate a smooth and continuous tool motion trajectory. The algorithm ensures the continuity of trajectory curvature by optimizing the control point position and node vector distribution, avoids impact during acceleration and deceleration, and forms a trajectory control sequence.

[0066] Based on the trajectory control sequence, the model predictive control method is used to perform adaptive adjustment calculation of the cutting force. This method establishes a dynamic prediction model between the cutting force and the processing parameters, predicts and compensates the cutting force fluctuation in real time, and calculates the force control parameters. The force control parameters include control quantities such as the feed speed correction amount and the spindle speed adjustment amount.

[0067] In the final stage, a multi-objective genetic algorithm is used to coordinately optimize the force control parameters. The algorithm uses machining efficiency, surface quality and tool life as optimization targets, and iteratively searches for the optimal solution through population evolution to obtain an initial control solution that takes into account multiple performance indicators. The initial control solution includes a complete machining parameter setting and control strategy, providing a control basis for subsequent machining processes.

[0068] This embodiment performs multi-dimensional feature decomposition and time-series correlation analysis on the real-time processing state group, so that the system can accurately capture the key features and dynamic change rules of the processing process, and effectively improves the perception and understanding of the processing state. By combining nonlinear mapping transformation and dynamic optimization calculation, adaptive adjustment of processing parameters is achieved, which significantly enhances the control accuracy and stability of the system over the processing process. The improved B-spline curve interpolation algorithm is used for trajectory planning to ensure the smooth continuity of tool movement and effectively reduce vibration and impact during the processing process. The real-time adjustment of cutting force is achieved through the model predictive control method, and the multi-objective genetic algorithm is combined for parameter collaborative optimization. While ensuring processing efficiency, it effectively improves the processing surface quality and extends the tool life, realizing intelligent control of the processing process and overall improvement of comprehensive performance.

[0069] In one embodiment, based on the machining quality trend, the real-time cutting parameters are subjected to error identification, and cutting correction compensation analysis is performed to obtain corresponding compensation control parameters, including: By setting the cutting parameter sampling period T and the number of sampling points N, the cutting force, cutting vibration, spindle current and other parameters are collected in real time. The sliding time window method is used to divide the collected data into several feature segments, each of which contains M data points. The time domain features of each feature segment are extracted based on wavelet transform to obtain statistical features such as mean, variance, peak value, etc., which constitute the cutting feature sequence.

[0070] In the time-frequency domain analysis stage, the short-time Fourier transform method is used to convert the cutting feature sequence into the frequency domain. The windowing process is performed by the Hanning window function, and the window length is L. The spectrum amplitude and phase information are calculated to form a P×Q-dimensional spectrum feature matrix. This matrix contains the main frequency components and their energy distribution characteristics in the cutting process.

[0071] Error clustering analysis uses the fuzzy C-means clustering algorithm to cluster the spectrum feature matrix into K error categories. The cluster center is determined by minimizing the objective function, and the error category set {E1, E2, ..., EK} is obtained. Each category corresponds to a certain error mode in the cutting process.

[0072] The membership evaluation is based on the fuzzy comprehensive evaluation criteria. An evaluation index system is established, including factors such as error amplitude, duration, and change rate. The expert scoring method is used to determine the weight of each index and construct an R×S dimensional error membership matrix. This matrix reflects the severity of each error category.

[0073] In the compensation calculation stage, the compensation strategy is designed based on the error membership matrix. The compensation amount is calculated for cutting parameters such as feed speed and spindle speed to obtain the initial compensation parameter set {δ1, δ2, ..., δm}. The compensation amount is determined taking into account the process constraints and machining accuracy requirements.

[0074] The response characteristic curve is constructed by the least square method, with the initial compensation parameters as independent variables and the cutting quality index as dependent variables, and the n-order polynomial response curve is fitted. This curve describes the mapping relationship between the compensation parameters and the machining quality.

[0075] The grey correlation analysis method is used to calculate the correlation degree. The reference series and the comparison series are selected to calculate the correlation coefficient. By taking the average correlation degree, the correlation degree sequence {r1, r2, ..., rp} between the compensation parameters and the cutting quality is obtained.

[0076] In the collaborative optimization stage, genetic algorithms are used to optimize the correlation sequence and initial compensation parameters. The population size is set to G and the evolutionary generation is set to H. Through operations such as selection, crossover, and mutation, the optimal solution is searched to obtain the final compensation control parameters. This parameter realizes precise compensation control of the cutting process.

[0077] This embodiment realizes dynamic segmentation of data through the sliding time window method, ensuring continuous and effective monitoring of the machining process. The time-frequency domain analysis method based on short-time Fourier transform deeply mines the frequency domain information in the cutting feature sequence, realizing accurate grasp of the changing trend of machining quality. The fuzzy C-means clustering algorithm is used for error identification, and the evaluation index system is established in combination with the fuzzy comprehensive evaluation criteria, thereby improving the recognition accuracy of different types of machining errors. The response characteristic curve is constructed by the least squares method, and the mapping relationship between the compensation parameters and the machining quality is established. The gray correlation analysis and genetic algorithm are combined for optimization and solution, realizing accurate compensation of the cutting parameters. While ensuring the machining quality, this method effectively improves the level of automation control of the milling cutter machining process and enhances the stability and reliability of the machining process.

[0078] In one embodiment, the initial control scheme is globally optimized according to the compensation control parameters to obtain a corresponding global processing optimization scheme, including: When the compensation control parameters are used to optimize the global parameters of the initial control scheme, the compensation control parameters include processing parameters such as feed rate, spindle speed, and cutting depth. These parameters are layered according to their roles in the processing process and divided into process layer parameter groups and control layer parameter groups. The process layer parameter group mainly includes parameters such as feed rate and cutting depth that directly affect the processing quality, and the control layer parameter group includes parameters such as spindle speed and feed rate that affect the processing efficiency.

[0079] After the parameter stratification of the compensation control parameters is completed, the target cutting parameters of the initial control scheme are calculated based on the process layer parameter group. The calculation process uses the cutting force model, substitutes the process parameters such as feed rate and cutting depth into the model, obtains the processing quality indicators such as cutting force, cutting temperature, surface roughness, etc., and forms a cutting optimization parameter matrix. The matrix contains the predicted values ​​of the processing quality indicators under all process parameter combinations.

[0080] When performing sensitivity analysis on the control layer parameter group, the orthogonal test method is used to observe the changes in the processing quality indicators by changing the values ​​of each control parameter. By calculating the influence of each parameter on the processing quality index, the parameter influence weight coefficient is obtained. This coefficient reflects the influence of different control parameters on the processing results.

[0081] According to the obtained parameter influence weight coefficient, each index in the cutting optimization parameter matrix is ​​weighted and calculated. During the calculation, the weight coefficient is multiplied and summed with the index value of the corresponding parameter to obtain a comprehensive evaluation index to form a weighted optimization parameter group. This parameter group comprehensively considers the influence of each control parameter on the processing quality.

[0082] When performing constraint boundary analysis on the weighted optimization parameter group, the range of each parameter is determined by combining the machine tool performance parameters, tool performance parameters and workpiece material characteristics. By establishing a set of constraint equations, the feasible domain range of the parameters is obtained. This range ensures that the optimized parameters meet the actual processing condition constraints.

[0083] Within the feasible range of the determined parameters, the nonlinear programming method is used to optimize the weighted optimization parameter group. The optimization goal is to maximize the processing efficiency while ensuring the processing quality. The target control parameter group is obtained by solving the nonlinear programming model. This parameter group optimizes the processing indicators under the constraints.

[0084] After the target control parameter group is determined, its stability is verified through simulation analysis. During the simulation process, the dynamic characteristics of the machining system are simulated, and the stability index of the system under the parameter group is calculated, including vibration amplitude, frequency response and other data. These stability index data are used to evaluate the reliability of the parameter group.

[0085] Based on the stability index data, a dynamic compensation model is established. This model takes into account the real-time changing factors in the machining process, such as tool wear, temperature changes, etc., and calculates the corresponding dynamic compensation parameters. The dynamic compensation parameters are used to adjust the target control parameters in real time to ensure the stability of the machining process.

[0086] The dynamic compensation parameters are combined with the target control parameter group to globally optimize the initial control scheme. The optimization process comprehensively considers the static parameter optimization results and dynamic compensation effects to generate a global processing optimization plan.

[0087] This embodiment achieves reasonable classification of process parameters and control parameters by layering the compensation control parameters, effectively improving the pertinence and accuracy of parameter optimization. The cutting force model is used to calculate the target cutting parameters, and the orthogonal test method is combined to perform parameter sensitivity analysis to accurately identify the degree of influence of each control parameter, providing reliable data support for subsequent optimization. By establishing constraint boundaries and nonlinear programming models, the optimal configuration of parameters is achieved while ensuring processing quality. The stability verification and dynamic compensation mechanism based on simulation analysis effectively respond to real-time changing factors in the processing process and enhance the stability and reliability of the processing process. The final global processing optimization solution not only ensures the improvement of processing efficiency, but also ensures the stability of processing quality, and significantly improves the overall processing performance.

[0088] The present invention also provides an automated control system for milling cutter processing, which is applied to any of the automated control methods for milling cutter processing described above, comprising: An acquisition module is used to obtain the geometric feature data and process parameter information of the milling cutter, and to perform processing analysis on the geometric feature data based on the process parameter information to obtain the corresponding original processing flow information; An analysis module is used to obtain the processing flow sensor data of the milling cutter and associate the data with the original processing flow information to obtain a corresponding real-time processing state group; The association module is used to obtain the real-time cutting parameters of the milling cutter, and perform processing prediction on the original processing flow information according to the real-time cutting parameters to obtain the corresponding processing quality trend; A processing module, which is used to perform intelligent control optimization on the real-time processing state group and processing quality trend to obtain a corresponding initial control solution; A control module, which is used to identify errors of real-time cutting parameters based on machining quality trends, and to perform cutting correction compensation analysis to obtain corresponding compensation control parameters; The execution module is used to optimize the global parameters of the initial control scheme according to the compensation control parameters to obtain the corresponding global processing optimization scheme.

[0089] The present invention provides an automatic control method and system for milling cutter processing, which has the following beneficial effects: By acquiring the geometric feature data and process parameter information of the milling cutter for processing analysis, it is possible to more comprehensively evaluate the various parameter requirements of the processing process, thereby improving the accuracy of the processing plan formulation and providing a reliable data basis for automatic control. By associating the processing process sensor data with the original processing process information, real-time monitoring and analysis of the processing process can be achieved, which helps to timely discover and deal with abnormal situations in the processing process and ensure the stability of processing quality. Processing prediction of the original processing process information based on real-time cutting parameters can ensure that good processing effects can be maintained under different processing conditions, reducing processing quality fluctuations. By intelligently controlling and optimizing the real-time processing state group and processing quality trend, a more reasonable processing control plan can be formulated, and the precise adjustment of processing parameters can be achieved through cutting correction compensation analysis, thereby effectively improving processing accuracy. At the same time, global parameter optimization based on compensation control parameters can flexibly adjust the control strategy according to the processing characteristics and requirements of different workpieces, making the processing system more adaptable to diverse processing needs.

[0090] It should be noted that technicians in the relevant technical field can clearly understand that for the convenience and conciseness of description, the specific working process of the system and each module described above can refer to the corresponding process in the aforementioned method embodiment, and will not be repeated here.

[0091] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. An automated control method for milling cutter processing, characterized in that: include: Acquire geometric feature data and process parameter information of the milling cutter, perform processing analysis on the geometric feature data based on the process parameter information, and obtain corresponding original processing flow information; Acquire the processing flow sensor data of the milling cutter, and associate the data with the original processing flow information to obtain a corresponding real-time processing state group; Acquire the real-time cutting parameters of the milling cutter, perform processing prediction on the original processing flow information according to the real-time cutting parameters, and obtain the corresponding processing quality trend; Performing intelligent control optimization on the real-time processing state group and the processing quality trend to obtain a corresponding initial control solution; Based on the machining quality trend, error identification is performed on the real-time cutting parameters, and cutting correction compensation analysis is performed to obtain corresponding compensation control parameters; The initial control scheme is globally optimized according to the compensation control parameters to obtain a corresponding global processing optimization scheme.

2. The automated control method for milling cutter processing according to claim 1, characterized in that: The step of acquiring geometric feature data and process parameter information of the milling cutter, and performing processing analysis on the geometric feature data based on the process parameter information to obtain corresponding original processing flow information includes: Acquire cutting edge angle data, cutting edge length data and tool diameter data obtained by optically scanning and measuring the milling cutter; Performing geometric feature vector matrix operation on the cutting edge angle data, the cutting edge length data and the tool diameter data to obtain the geometric feature data; Searching and matching the geometric feature data with a preset process parameter database to obtain corresponding process parameter information; Perform machining process analysis according to the process parameter information and the geometric feature data to obtain a corresponding original machining trajectory; Segmenting the original processing trajectory to obtain a corresponding processing path point sequence; Calculating the path point cutting force components of the processing path point sequence according to the geometric feature data to obtain original force distribution data; Performing critical value analysis on the original force distribution data to obtain an original processing parameter range; A processing control instruction sequence is generated according to the original processing parameter range and the processing path point sequence to obtain the original processing flow information.

3. The automated control method for milling cutter processing according to claim 1, characterized in that: The step of acquiring the processing flow sensor data of the milling cutter and associating the data with the original processing flow information to obtain a corresponding real-time processing state group includes: Monitoring the machining process of the milling cutter to obtain original sensor monitoring data; Performing processing data screening on the original sensor monitoring data to obtain corresponding processing flow sensor data; Performing processing state clustering on the processing flow sensor data to obtain corresponding processing state feature clusters; Performing information mapping on the processing state feature cluster to obtain corresponding processing state mapping information; Performing process clustering on the original process information according to the process sensing data to obtain corresponding process type information; The process type information is state-combined and associated according to the processing state mapping information to obtain the real-time processing state group.

4. The automated control method for milling cutter processing according to claim 1, characterized in that: The acquiring of the real-time cutting parameters of the milling cutter, and performing processing prediction on the original processing flow information according to the real-time cutting parameters to obtain the corresponding processing quality trend, includes: Acquiring an acoustic emission signal and a cutting acceleration signal of the milling cutter during a cutting process, and performing multi-source heterogeneous processing on the acoustic emission signal and the cutting acceleration signal to obtain an original cutting data stream; Extracting real-time parameters from the original cutting data stream to obtain the real-time cutting parameters; Performing feature decomposition on the real-time cutting parameters to obtain corresponding feature frequency band groups; Performing adaptive threshold analysis on the characteristic frequency band group to obtain a corresponding cutting stability index; Perform a prediction function according to the cutting stability index and the real-time cutting parameters to obtain a corresponding cutting force prediction mapping function; Calculating the workpiece surface quality according to the cutting force prediction mapping function to obtain a corresponding machining accuracy evaluation value; Performing error compensation analysis on the machining accuracy evaluation value to obtain a corresponding dynamic error compensation amount; The dynamic error compensation amount is predicted by time series regression according to the cutting force prediction mapping function to obtain a corresponding machining quality trend.

5. The automated control method for milling cutter processing according to claim 4, characterized in that: The prediction function is performed according to the cutting stability index and the real-time cutting parameter to obtain a corresponding cutting force prediction mapping function, including: Performing Fourier transform on the cutting stability index to obtain a corresponding frequency domain feature sequence; Performing least square fitting on the real-time cutting parameters according to the frequency domain feature sequence to obtain a corresponding parameter fitting curve; Performing piecewise linearization processing on the parameter fitting curve to obtain a corresponding piecewise function group; Performing a polynomial interpolation operation on the frequency domain feature sequence according to the piecewise function group to obtain a corresponding interpolation coefficient matrix; Performing singular value decomposition on the interpolation coefficient matrix to obtain a corresponding eigenvalue sequence; Performing orthogonal basis function expansion on the piecewise function group according to the eigenvalue sequence to obtain a corresponding basis function set; Optimizing the basis function set using the Lagrange multiplier method to obtain a corresponding constraint condition group; Solving the basis function set by the calculus of variations according to the constraint condition group to obtain a corresponding cutting force prediction mapping function; Wherein, the cutting force prediction mapping function is: ； represents the predicted cutting force, is a set of basis functions obtained by expanding the orthogonal basis functions. is the coefficient corresponding to the eigenvalue sequence obtained by singular value decomposition, is the Lagrange multiplier, h is the constraint function, n is the number of orthogonal basis functions, and t is any time in the cutting process.

6. The automated control method for milling cutter machining according to claim 1, characterized in that: The intelligent control optimization of the real-time processing state group and the processing quality trend to obtain a corresponding initial control scheme includes: Performing multi-dimensional feature decomposition on the real-time processing state group to obtain a corresponding state feature matrix; Performing a time series correlation analysis on the processing quality trend according to the state characteristic matrix to obtain a corresponding state correlation vector; Performing nonlinear mapping transformation on the state association vector to obtain a corresponding control feature space; Performing dynamic optimization calculation on the processing parameters according to the control feature space to obtain a corresponding parameter optimization matrix; Performing machining trajectory planning on the parameter optimization matrix to obtain a corresponding trajectory control sequence; Adaptively adjust and calculate the cutting force according to the trajectory control sequence to obtain corresponding force control parameters; The force control parameters are subjected to multi-objective collaborative optimization to obtain a corresponding initial control solution.

7. The automated control method for milling cutter processing according to claim 1, characterized in that: The error identification of the real-time cutting parameters based on the machining quality trend and the cutting correction compensation analysis to obtain corresponding compensation control parameters include: Extracting the features of the real-time cutting parameters in sections to obtain a corresponding cutting feature sequence; Performing time-frequency domain analysis on the machining quality trend according to the cutting feature sequence to obtain a corresponding frequency spectrum feature matrix; Performing error clustering on the frequency spectrum feature matrix to obtain a corresponding error category set; Performing membership evaluation on the error category set to obtain a corresponding error membership matrix; Calculating the compensation amount of the cutting parameters according to the error membership matrix to obtain the corresponding initial compensation parameters; Constructing a response curve for the initial compensation parameters to obtain a corresponding response characteristic curve; Calculating the correlation degree of the response characteristic curve to obtain a corresponding correlation degree sequence; The correlation degree sequence and the initial compensation parameter are collaboratively optimized to obtain the corresponding compensation control parameter.

8. The automated control method for milling cutter machining according to claim 1, characterized in that: The initial control scheme is globally optimized according to the compensation control parameters to obtain a corresponding global processing optimization scheme, including: Parameter stratification is performed on the compensation control parameters to obtain corresponding process layer parameter groups and control layer parameter groups; Calculate the target cutting parameter of the initial control scheme according to the process layer parameter group to obtain a corresponding cutting optimization parameter matrix; Performing parameter sensitivity analysis on the control layer parameter group to obtain corresponding parameter influence weight coefficients; Performing weighted calculation on the cutting optimization parameter matrix according to the parameter influence weight coefficient to obtain a corresponding weighted optimization parameter group; Performing constraint boundary analysis on the weighted optimization parameter group to obtain a corresponding parameter feasible domain range; According to the feasible domain range of the parameters, nonlinear programming is performed on the weighted optimization parameter group to obtain a corresponding target control parameter group; Performing stability verification on the target control parameter group to obtain corresponding stability index data; Perform dynamic compensation according to the stability index data and the target control parameter group to obtain corresponding dynamic compensation parameters; The initial control scheme is globally optimized according to the dynamic compensation parameters and the target control parameter group to obtain the global processing optimization scheme.

9. An automatic control system for milling cutter processing, characterized in that: The automated control method for milling cutter machining as described in any one of claims 1 to 8 above comprises: An acquisition module, the acquisition module is used to obtain geometric feature data and process parameter information of the milling cutter, and perform processing analysis on the geometric feature data based on the process parameter information to obtain corresponding original processing flow information; An analysis module, the analysis module is used to obtain the processing flow sensor data of the milling cutter, and associate the data with the original processing flow information to obtain a corresponding real-time processing state group; An association module, the association module is used to obtain the real-time cutting parameters of the milling cutter, perform processing prediction on the original processing flow information according to the real-time cutting parameters, and obtain a corresponding processing quality trend; A processing module, the processing module is used to perform intelligent control optimization on the real-time processing state group and the processing quality trend to obtain a corresponding initial control solution; A control module, the control module is used to perform error identification on the real-time cutting parameters based on the machining quality trend, and perform cutting correction compensation analysis to obtain corresponding compensation control parameters; An execution module is used to perform global parameter optimization on the initial control scheme according to the compensation control parameters to obtain a corresponding global processing optimization scheme.

Images