Bidirectional milling force prediction method and system

Abstract

The invention discloses a bidirectional milling force prediction method and system, and the system comprises a data obtaining module which is used for obtaining a current and milling force data set; a preprocessing module; a feature extraction module; the feature screening module is used for screening out current features meeting a preset condition to form a target feature matrix according to the stability of the current features, the sensitivity of the current features to the rotating speed and the correlation between the current features and the milling force; the initial module is used for constructing a neural network model based on a CNN-ResNet deep learning algorithm; the training module is used for obtaining a final prediction model for predicting the bidirectional milling force; the input module is used for inputting the target characteristic matrix of the current of the cutter with the milling force to be monitored into the prediction model; and the output module is used for outputting the bidirectional milling force predicted by the prediction model. The bidirectional milling force Fx and the bidirectional milling force Fy in the feeding direction and the direction perpendicular to the feeding direction can be predicted, and the prediction and monitoring accuracy is high; and the method can be suitable for complex cutting conditions and material changes.

Description

Technical Field

[0001] The present invention relates to a prediction method and system, and in particular to a bidirectional milling force prediction method and system. Background Art

[0002] Milling force is a key parameter in machine tool design and machining process optimization, reflecting the interaction between the tool and the workpiece during machining. Milling force not only includes the size of the cutting load, but also involves a lot of information such as tool status, machining quality, and chatter. Accurate milling force monitoring can directly affect tool life, vibration stability, cutting deformation, thermal effects, and machining surface quality. It can directly reflect the changes in the machine tool machining state and is of great significance to the study of CNC machine tool adaptive control technology.

[0003] Effective monitoring of milling force is the key to accurately reflect the machining status of machine tools. At present, there are two main methods for monitoring milling force: the first is the direct monitoring method, which usually directly measures the force in the cutting area through a dedicated force sensor, so that high-precision cutting force data can be obtained in real time, which helps to adjust the machining parameters and ensure the machining quality; however, this method is costly and complex to install. The second is the indirect monitoring method, which infers the size of the cutting force by analyzing vibration, power or current signals. The vibration signal analysis method uses vibration sensors to reflect the dynamic changes in the cutting process in real time, but it is affected by the vibration characteristics of the machine tool structure and the data analysis is complicated; the power signal analysis method indirectly infers the cutting force through the change of motor power. It has wide applicability but slow response speed and limited accuracy. In addition, most of the existing monitoring and prediction models can only predict the resultant cutting force, and it is difficult to accurately predict the bidirectional milling force. Summary of the invention

[0004] Objective of the invention: The first objective of the present invention is to provide a method for predicting bidirectional milling force that can predict bidirectional milling force more accurately.

[0005] A second object of the present invention is to provide a bidirectional milling force prediction system.

[0006] Technical solution: A bidirectional milling force prediction method disclosed in the present invention comprises the following steps:

[0007] The experiment obtains the current and milling force data sets of the tool when the same workpiece is milled at two or more different speeds; the current and milling force data sets include the equivalent DC current I rms , bidirectional milling force F x and F y , where F x and F y The cutting force F pointing in the instantaneous cutting direction t Decomposed, and F x and Fy The directions are respectively along the milling feed direction and perpendicular to the milling feed direction;

[0008] The time domain, frequency domain and time-frequency domain characteristics of the current signal in the current and milling force data set are extracted. According to the stability of the current characteristics, the sensitivity of the current characteristics to the rotation speed, and the correlation between the current characteristics and the milling force, the current characteristics that meet the preset conditions are screened out to form the target feature matrix.

[0009] Build a neural network model based on the CNN-ResNet deep learning algorithm;

[0010] The data contained in the target feature matrix is ​​divided into a training set and a validation set, and the neural network model is trained with the training set to obtain a preliminary model; the validation set is imported, the preliminary model is used, the loss value is calculated, and then back propagation and parameter update are performed until the loss function value is minimized and tends to be stable, and the iteration is stopped to obtain the final prediction model for predicting bidirectional milling force;

[0011] The target characteristic matrix of the current of the tool to be monitored for milling force is obtained and input into the prediction model to predict the bidirectional milling force.

[0012] Furthermore, before extracting the time domain, frequency domain and time-frequency domain features of the current signal, the current data in the current and milling force data sets are subjected to wavelet denoising, and the cutting-in and cutting-out parts are removed for the milling force data in the current and milling force data sets, and the signal data of the stable milling stage are retained.

[0013] Furthermore, the workpiece is composed of two different materials; the steps of obtaining the current and milling force data sets are as follows:

[0014] The tool first idles on one side of the workpiece, then moves along the length of the workpiece and mills two different materials in succession;

[0015] The tool is made to perform the aforementioned milling operation at different rotation speeds;

[0016] The milling operation at the same speed is repeated for multiple times, and the current and milling force data obtained from the milling operation at the same speed constitute a data group;

[0017] All data sets obtained at different rotational speeds constitute the current and milling force data set.

[0018] Furthermore, the time domain, frequency domain and time-frequency domain features of the current signal in the extracted current and milling force data set include each I rms Multiple features in idling state, milling carbon steel state, and milling aluminum alloy state.

[0019] Furthermore, the multiple features include maximum value, minimum value, average value, peak-to-peak value, absolute mean value, variance, standard deviation, root mean square value, root amplitude, skewness index, kurtosis index, waveform factor, peak factor, pulse factor, margin factor, center of gravity frequency, mean square frequency, root mean square frequency, frequency variance and 8-layer wavelet decomposition, and all the extracted features constitute the current signal feature M C0 .

[0020] Furthermore, the stability of the current characteristic is determined by the milling state and the milling speed. C0 As an evaluation index, the calculation formula and screening method of stability are as follows:

[0021] Solve the experimental characteristic dispersion coefficient c of each data group separately v , the calculation formula is as follows:

[0022] c v =σ fea / μ fea

[0023] Among them, σ fea and μ fea are the standard deviation and mean of the same feature in multiple millings at the same speed,

[0024] According to the calculated experimental characteristic dispersion coefficient c v As a result, the threshold is set and the experimental characteristic dispersion coefficient c v The features below the threshold are constructed into the feature matrix M C1 .

[0025] Furthermore, the sensitivity of the current characteristic to the rotation speed is used to characterize the characteristic current signal characteristic M C0 Response to speed changes, sensitivity Var i The calculation formula and screening method are as follows:

[0026]

[0027] Among them, μ i is the characteristic average value of the i-th data group; n i is the rotation speed of the i-th data group;

[0028] The features that are highly sensitive to the changes in milling force are recorded as the feature matrix M C2 .

[0029] Furthermore, the correlation between the current characteristics and the milling force is screened as follows:

[0030] Take the feature matrix M C1 and the feature matrix M C2 The intersection ofC3 ;

[0031] Calculate F separately x and F y The average value of the feature matrix M C3 With F x The mean and F y The average values ​​of M were analyzed by Pearson correlation. C3 Medium milling force F x and F y The features with correlation greater than 0.8 are recorded as the target feature matrix.

[0032] Furthermore, the neural network model includes a convolutional layer of a CNN and a ResNet residual block.

[0033] Based on the same inventive concept, the present invention also discloses a bidirectional milling force prediction system, comprising:

[0034] A data acquisition module is used to acquire a data set of current and milling force of a tool when the same workpiece is milled at two or more different speeds;

[0035] A preprocessing module is used to perform wavelet denoising on the current signal in the current and milling force data sets, remove the cutting-in and cutting-out parts of the milling force data in the current and milling force data sets, and retain the signal data of the stable milling stage;

[0036] A feature extraction module is used to extract the time domain, frequency domain and time-frequency domain features of the current signal in the current and milling force datasets;

[0037] The feature screening module selects the current features that meet the preset conditions to form a target feature matrix based on the stability of the current features, the sensitivity of the current features to the rotation speed, and the correlation between the current features and the milling force;

[0038] The initial module is used to build a neural network model based on the CNN-ResNet deep learning algorithm;

[0039] The training module divides the data contained in the target feature matrix into a training set and a validation set, and uses the training set to train the neural network model to obtain a preliminary model; imports the validation set, uses the preliminary model, calculates the loss value, and then performs back propagation and parameter updates until the loss function value is minimized and tends to be stable, and the iteration stops to obtain the final prediction model for predicting bidirectional milling force;

[0040] An input module, used for inputting a target characteristic matrix of the current of the tool to be monitored for milling force into the prediction model;

[0041] The output module is used to output the bidirectional milling force predicted by the prediction model.

[0042] Beneficial effects: Compared with the prior art, the present invention has the following significant advantages: the present invention can predict the bidirectional milling force F along the feed direction and perpendicular to the feed direction x and F y Compared with only being able to predict the milling force, the bidirectional milling force is more useful for cutting state analysis or tool wear state monitoring; the present invention utilizes feature analysis to screen features with high stability, high sensitivity to the milling force, and high correlation with the milling force, establishes a current signal feature matrix, and establishes a mapping relationship between the current signal feature matrix and the bidirectional milling force through the CNN-ResNet deep learning network, verifies the accuracy of current-based milling force monitoring, and realizes intelligent monitoring of current-based bidirectional milling force, and the accuracy of predictive monitoring is high; the workpiece of the present invention's experiment is composed of two materials, so that the prediction model of the present invention can be applied to complex cutting conditions and material changes, and has strong versatility. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Figure 1 is a flow chart of the method of the present invention;

[0044] Figure 2 A power transmission chain from the motor to the tool in the present invention;

[0045] Figure 3 It is the design drawing of the experimental workpiece material in the present invention;

[0046] Figure 4 This is a diagram of the experimental platform in the present invention;

[0047] Figure 5 This is a current characteristic stability analysis diagram in the present invention;

[0048] Figure 6 is the ratio of the characteristics of the milling forces Fx and Fy and the rotation speed under different cutting materials in the present invention;

[0049] Figure 7 It is the sensitivity analysis of the current characteristics to the rotation speed in the present invention;

[0050] Figure 8 is a correlation diagram between current characteristics and milling force in the present invention;

[0051] Fig. 9 It is a flow chart of the CNN-ResNet algorithm in the present invention;

[0052] Fig.10 It is the training and test loss graph of the CNN-ResNet model in the present invention;

[0053] Fig.11 The milling force prediction results of different algorithms in the present invention for the milling material being carbon steel (0.45% C);

[0054] Fig.12 The milling force prediction results of different algorithms in the present invention for the milling material being aluminum alloy (7050);

[0055] Fig.13 It is a schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0056] The technical solution of the present invention is further described below in conjunction with the accompanying drawings.

[0057] Example 1

[0058] In order to predict the milling force through current, the theoretical relationship between the milling force and the motor spindle current is first established: the power transmission chain from the motor to the tool is as follows: Figure 2 As shown in the figure, the original power comes from the spindle motor, and then is transmitted to the tool through the gear box and the machine tool spindle. During the transmission process, the original power motor torque T is decomposed into cutting torque T q , friction torque T f and moment of inertia T J The current I measured in the U, V, and W three-phase wires of the input spindle motor u ,I v ,I w Convert to equivalent DC current I rms . Using I rms Predicted cutting force F t (F x and F y The combined force of Figure 2 As shown, edge 1, edge 2, edge 3 and edge 4 represent the four cutting edges of the tool respectively.

[0059] During milling operation, the current I measured in the U, V, and W three-phase wires of the input spindle motor is u ,I v ,I w Convert to equivalent DC current I rms , the three-phase current of the motor is described as follows:

[0060]

[0061] Among them, I u ,I v ,I w are the phase currents of U, V, and W respectively; I m ,ω i are the amplitude and angular frequency of the three-phase current respectively;

[0062] Equivalent DC current I rms The solution is as follows:

[0063]

[0064] Repeat the above steps to calculate the equivalent DC current I of the motor in idling state rms0 .

[0065] Establish the theoretical relationship between milling force and spindle current:

[0066] According to the principle of conservation of momentum:

[0067] T=T f +T J +T q (3)

[0068] Among them, T is the motor torque, representing the original power;

[0069] Cutting torque T q Proportional to the cutting force, that is, T q =F t ·r, where r is the tool radius, F t is the cutting force pointing in the instantaneous cutting direction; cutting force F t The change of cutting torque T q The change of affects the motor torque T; therefore, in theory, the motor torque T can be used for the cutting force F t prediction.

[0070] Therefore, equation (3) can be expressed in another form:

[0071]

[0072] Where K is the motor torque constant, T f =KI rms0 , T J Depends on the inertia of the transmission system J, T J =J(dw / dt), w is the speed of the spindle motor, t is the time. So far, the relationship between input current and cutting force has been established.

[0073] Considering that the spindle speed is constant during the milling process, that is, dw / dt = 0, T J =J(dw / dt)=0. Equation (4) can be further deduced as:

[0074] KI rms =KI rms0 +rF t (5)

[0075] It can be expressed in another form:

[0076] F t =KI rms / r-KI rms0 / r(6)

[0077] Now, we can use equation (6) to calculate the equivalent DC current I of the spindle motor. rms To predict the cutting force F t , I rms0 , K and r are all constants, indicating that F t and I rms There is a linear relationship between them; however, equation (6) has two problems. On the one hand, in practical applications, F t and I rms There is not a simple linear relationship between them, because the friction torque T f It may change with the increase of spindle speed, that is, K is no longer a constant. Therefore, the cutting force F cannot be accurately predicted using equation (6) t On the other hand, equation (6) can only predict their resultant force F t , but cannot predict F separately x and F y , so along the milling feed direction F x and bidirectional milling force F perpendicular to the feed direction y It is more useful for cutting condition analysis or tool wear condition monitoring.

[0078] Therefore, the present invention provides a bidirectional milling force prediction method, such as Figure 1 As shown, the following steps are included:

[0079] The experiment obtains the current and milling force data sets of the tool when the same workpiece is milled at two or more different speeds. The current and milling force data sets include the equivalent DC current I rms , bidirectional milling force F x and F y , where F x and F y The cutting force F pointing in the instantaneous cutting direction t Decomposed, and F x and F y The directions are along the milling feed direction and perpendicular to the milling feed direction respectively; the experimental workpiece is composed of two different materials spliced ​​together, such as Figure 3 As shown, half of the workpiece materials in this experiment are carbon steel (0.45% C) and the other half are aluminum alloy (7050); the steps for obtaining the current and milling force data sets are as follows: the tool first idles on one side of the workpiece, and then the tool moves along the length of the workpiece and mills the carbon steel and aluminum alloy successively; the tool performs the aforementioned milling operation at different rotation speeds; the milling operation at the same rotation speed is repeated multiple times, and the current and milling force data obtained from the milling operation at the same rotation speed constitute a data set; all data sets obtained at different rotation speeds constitute the current and milling force data sets. In this embodiment, Figure 4As shown, the experiment uses a DX-650 high-speed CNC milling machine. The cutting tool is a 4-edge carbide vertical milling cutter with a diameter of 6mm and a helix angle of 48 degrees. Down milling, dry cutting; force signal is measured by force sensor Kistler9119AA2; current signal is measured by CPL8100B current probe and DL950 oscilloscope; the sampling frequency of the signal is 50kHz. Milling operation is carried out at four different speeds, and the same speed is repeated 10 times. The four data groups obtained are recorded as T1, T2, T3 and T4 respectively. The data groups of T1-T4 are shown in Table 1 below:

[0080] Table 1

[0081]

[0082]

[0083] In Table 1, n refers to the motor's spindle speed, RPM refers to the motor's spindle speed unit (rotation per minute), and v f Refers to the tool feed speed, a e Refers to the cutting width, a p Refers to the cutting depth. The current and milling force data sets are processed: the current signal is subjected to wavelet denoising, the wavelet basis is selected as db, the decomposition level is 4, and the current time domain monitoring signal is obtained; for the milling force signal, the cut-in and cut-out parts of the collected signal are deleted, and the signal of the stable milling stage is retained to obtain the cutting force time domain monitoring signal.

[0084] A theoretical milling force prediction model based on experimental data is constructed. The milling force prediction model based on the theoretical model uses the experimental data of T1-T4 to substitute into equation (6), and after linear fitting, the following results are obtained:

[0085] Ft=1.099 6 (Irms-Irms0) 3 -1.754 6 (Irms-Irms0) 2 +9.315 5 (Irms-Irms0)-1.646(7)

[0086] The result is not a strict linear polynomial relationship, which verifies that the theoretical relationship between milling force and motor spindle current, equation (6), cannot accurately predict the cutting force F. t This may be because during the milling process, the tool may experience different degrees of runout, which leads to dynamic changes in the milling radial force and affects the friction torque. Therefore, this equation cannot accurately represent the mathematical relationship between the spindle current and the cutting force. t Perpendicular to the tool axis plane, it can be decomposed into F x and Fy , along the milling feed direction F x and bidirectional milling force F perpendicular to the feed direction y It is more useful for machining status analysis or tool wear monitoring, so pay attention to the spindle current and F x and F y The relationship between them is more meaningful.

[0087] Extract the time domain, frequency domain and time-frequency domain features of the current signal in the current and milling force data set, and select the current features that meet the preset conditions to form the target feature matrix based on the stability of the current features, the sensitivity of the current features to the rotation speed, and the correlation between the current features and the milling force. Extract the time domain, frequency domain and time-frequency domain features of the current signal in the current and milling force data set, including: extracting each I rms The maximum value, minimum value, average value, peak-to-peak value, absolute mean value, variance, standard deviation, root mean square value, root square amplitude, skewness index, kurtosis index, waveform factor, peak factor, pulse factor, margin factor, center of gravity frequency, mean square frequency, root mean square frequency, frequency variance and 8-layer wavelet decomposition in the idling state, milling carbon steel state and milling aluminum alloy state, respectively, a total of 81 features, all the extracted features constitute the current signal feature M C0 , current signal characteristics M C0 I used to represent 40 samples rms All 81 features, current signal feature M C0 The structure is as follows:

[0088]

[0089] Among them, F 1_K1 Refers to when the motor is in idling state I rms Feature 1 of sample 1; F 1_S1 When the workpiece material is carbon steel rms Feature 1 of sample 1; F 1_A1 When the workpiece material is aluminum alloy rms Feature 1 of sample 1; F 27_K40 Refers to when the motor is in idling state I rms Features 27 of sample 40; F 27_S40 When the workpiece material is carbon steel rms Features 27 of sample 40; F 27_A40 When the workpiece material is aluminum alloy rms Features 27 of sample 40.

[0090] The stability of the current characteristic is based on the milling state and milling speed as variables, and the current signal characteristic M C0 As an evaluation index, the calculation formula and screening method of stability are as follows:

[0091] Solve the experimental characteristic dispersion coefficient c of ten cuttings of T1, T2, T3 and T4 respectively v , the calculation formula is as follows:

[0092] c v =σ fea / μ fea

[0093] Among them, σ fea and μ fea They are the standard deviation and mean of the same feature in multiple millings at the same speed;

[0094] According to the calculated experimental characteristic dispersion coefficient c v As a result, the threshold is set and the experimental characteristic dispersion coefficient c v The features below the threshold are constructed into the feature matrix M C1 , that is, the experimental characteristic dispersion coefficient c v Features below the threshold are defined as features that are not affected by the rotation speed. Preferably, the threshold can be set according to actual conditions. In this embodiment, the threshold is set to 0.1. Figure 5 As shown, Figure 5 (a), 5(b), 5(c) and 5(d) show the results of the coefficient of dispersion of 27 features for the T1, T2, T3 and T4 tests, respectively, where feature 1 represents the maximum value feature, feature 2 represents the minimum value feature, feature 3 represents the average value feature, feature 4 represents the peak-to-peak value feature, feature 5 represents the absolute mean value feature, feature 6 represents the variance feature, feature 7 represents the standard deviation feature, feature 8 represents the root mean square value feature, feature 9 represents the root amplitude feature, feature 10 represents the skewness index feature, feature 11 represents the kurtosis index feature, and feature 12 represents the waveform factor feature. The characteristic 13 represents the peak factor characteristic, the characteristic 14 represents the pulse factor characteristic, the characteristic 15 represents the margin factor characteristic, the characteristic 16 represents the center of gravity frequency characteristic, the characteristic 17 represents the mean square frequency characteristic, the characteristic 18 represents the root mean square frequency characteristic, the characteristic 19 represents the frequency variance characteristic, and the characteristic 20-27 represents the 1st-8th layer characteristics of the wavelet decomposition; from the calculation results of the discrete coefficient, it can be seen that the discrete coefficients of the same characteristics are basically the same under different milling materials; under different speeds, the characteristics with discrete coefficients less than 0.1 are 1, 3-16, and 20, that is, these characteristics are not affected by the speed. It can be seen that the maximum value, average value, peak-peak value, absolute mean, variance, standard deviation, root mean square value, root amplitude, skewness index, kurtosis index, waveform factor, peak factor, pulse factor, margin factor, center of gravity frequency and the first layer subband of wavelet decomposition have high stability.

[0095] The obtained feature matrix M C1 for:

[0096]

[0097] The sensitivity of the current characteristic to the rotation speed is used to characterize the characteristic current signal characteristic M C0 Response to speed changes, sensitivity Var i The calculation formula and screening method are as follows:

[0098]

[0099] Among them, μ i is the characteristic average value of the i-th data group; n i is the rotation speed of the i-th data group;

[0100] The characteristics that change with the change of milling force are recorded as the characteristic matrix M C2 , during the cutting process, F x and F y The change of is consistent with the change of the speed ratio, which means that the milling force F x and F y The magnitude of is positively correlated with the change of spindle speed, that is, its characteristic is defined as a characteristic that is highly sensitive to the change of milling force; Figure 6 As shown, the milling force F under different milling materials is shown. x and F y The ratio of the change in and the spindle speed change, from Figure 6 It can be seen that in the process of cutting carbon steel, F x The change of is consistent with the change of the speed ratio, indicating that the milling force F x The magnitude of F is positively correlated with the change of spindle speed; however, y No consistent changes are shown, therefore, this rule does not apply when the workpiece material is aluminum alloy.

[0101] The characteristic changes between different materials are consistent, increasing or decreasing with the increase of rotation speed, such as Figure 7 As shown in the figure, the horizontal axes 1, 2, and 3 represent the T1-T2 test group, the T2-T3 test group, and the T3-T4 test group, respectively, and feai refers to feature i. The experimental results show that I rms The characteristics of: mean value, absolute mean value, variance, standard deviation, root mean square value, root square amplitude, skewness index, kurtosis index, waveform factor, peak factor, pulse factor, margin factor, center of gravity frequency and the first and second sub-bands of wavelet decomposition are highly sensitive to the change of milling force. These characteristics change accordingly with the change of milling force and can be used as the prediction characteristics of milling force, that is, the obtained characteristic matrix M C2 for:

[0102]

[0103] The correlation between current characteristics and milling force is screened as follows:

[0104] Take the feature matrix M C1 and the feature matrix M C2 The intersection of C3 ;

[0105] Calculate F separately x and F y The average value of the feature matrix M C3 With F x The mean and F y The average values ​​of M were analyzed by Pearson correlation. C3 Medium milling force F x and F y The features with correlation greater than 0.8 are recorded as the target feature matrix;

[0106] In this embodiment, the F of 40 samples is calculated. x Mean and F y Average value, such as Figure 8 As shown, Figure 8 When cutting different workpiece materials, the characteristic matrix M C3 and milling force F x Mean and F y Pearson correlation analysis results between mean values. C3 The characteristics of the milling force correlation greater than 0.8 are: mean value, absolute mean value, root mean square value, root square amplitude, centroid frequency and the first layer subband of wavelet decomposition, which are recorded as the target feature matrix. rms The feasibility of the target feature matrix to monitor the milling force. The features contained in the target feature matrix are shown in Table 2 (for the convenience of listing and matrix, the target feature matrix is ​​recorded as feature matrix M C4 ):

[0107] Table 2

[0108]

[0109]

[0110] Construct a neural network model based on the CNN-ResNet deep learning algorithm; the neural network model includes the convolutional layer of CNN and the residual block of ResNet; the core idea of ​​this neural network model is to solve the gradient vanishing or gradient exploding problem in the deep network by introducing residual learning, thereby improving the training efficiency and generalization ability of the model. Fig. 9 As shown, Fig. 9The structure of the CNN-ResNet deep learning algorithm is shown. Input features are extracted through the convolutional layers of the CNN. The input data is initially fed into the convolutional layers of the feature extraction part for preliminary processing and then propagated through the two structured residual blocks of the ResNet residual block part. These residual blocks effectively promote the flow of information within the deep neural network by combining the recognition mapping technique, which helps to alleviate the gradient vanishing problem that is prevalent in deep learning models. In addition, the batch normalization operation integrated in these residual blocks further enhances the stability of the training process. By standardizing the output of the intermediate layers, the effective propagation of the gradient during the network training phase is ensured. After being processed by the residual blocks, the data flow is directed to the first fully connected layer of the regression part. The task of this layer is to transform the high-level features extracted by the deep network into the final output prediction. Through the further synthesis and processing of the features in this stage, the model is able to produce accurate prediction results based on the input data. This whole process reflects the effective mechanism of deep learning models using a hierarchical structure approach for feature extraction and transformation when processing complex data.

[0111] The function of the convolution layer in the feature extraction part is to perform preliminary feature extraction on the input data. This layer can be represented by the following convolution operation formula: z = Dx + b, where z represents the output of the convolution layer, D and b represent the weight of the convolution kernel and the weight of the bias term respectively, and x represents the current feature matrix.

[0112] Residual blocks form a key component of the model and are designed to enable the network to directly learn the residual between input and output. Each residual block consists of two one-dimensional convolutional layers, each followed by batch normalization and a ReLU activation function. The output of the residual block can be expressed as the following residual learning formula:

[0113] y=F(z,{D j})+z

[0114] Among them, y represents the output of the residual block, and x represents the input. j}) represents the combination of convolution operation and nonlinear activation function, where {D j} represents the weight in the convolutional layer, j represents the index of the weight in the convolutional layer, that is, it represents the jth weight parameter. This architecture helps to solve the problem of information transmission in deep networks.

[0115] After each convolution operation, batch normalization is used to stabilize the learning process and accelerate convergence. Batch normalization can be expressed as follows:

[0116]

[0117] Where m (k) represents the output of the convolutional layer, s (k)represents the output of the batch normalization layer, and E[m (k) ] and Var[m (k) ] represent the mean and variance of the batch data, respectively, and τ is a very small number to prevent division by zero. In addition, γ and β are learnable parameters used to recover some information that may be lost during the normalization process.

[0118] After each batch normalization, the Relu activation function is applied to introduce nonlinearity to improve the expressiveness of the model. The model is trained using the Adam optimizer and the mean square error (MSE) loss function. The parameters are adjusted in each training period through the back-propagation algorithm to minimize the difference between the predicted output and the actual label.

[0119] During the training process, the Xavier uniform initialization method is used to initialize the model parameters to ensure the stability of the training process. The model is evaluated on the validation set to test its generalization ability.

[0120] The uniqueness of the method of the present invention is that it can effectively predict the milling force signal without inputting any cutting parameters, relying only on the screened I rms Features, that is, the target feature matrix. The data contained in the target feature matrix is ​​divided into a training set and a validation set, and the neural network model is trained with the training set to obtain a preliminary model; the validation set is imported, the preliminary model is used to calculate the loss value, and then backpropagation and parameter updates are performed until the loss function value is minimized and tends to be stable, and the iteration stops to obtain the final prediction model for predicting bidirectional milling force; in this embodiment, first, a target feature matrix is ​​constructed and screened for milling test samples T1, T2, T3 and T4, and the features protected by the target feature matrix constitute a complete data set. After random encoding, the data set is divided into a training set (accounting for 80%) and a validation set (accounting for 20%). Subsequently, the constructed neural network model is trained using the data set obtained from the experiment. Fig.10 As shown in , the training process is configured to iterate for a maximum of 100 times, and it automatically terminates when no further reduction in the iteration loss is observed. To ensure the stability of the model, the training scheme is repeated 5 times, and the details of each iteration are shown in Fig.10 (a) to 10(e). The average iterative loss calculated in these five training stages is shown in Fig.10 (f) shows the reliability of the model. After optimization, the model can be used to predict the milling force under different spindle speeds and material conditions.

[0121] Verify the accuracy of the prediction model. The steps to verify the accuracy of the prediction model are as follows: use the prediction model, CNN model, RNN model and LSTM model to predict the milling force at the same time, compare the four prediction results with the actual measurement results, and judge the accuracy of the predictions of the four models. In this embodiment, three different models, CNN, RNN and LSTM, are used for comparison. The specific parameters set by the comparison model method are as follows: (1) CNN has two convolutional layers, two maximum pooling layers, one flat layer, two fully connected layers, one input layer and one output layer, and the activation function of the fully connected layer is Relu; (2) RNN has a GRU layer as the input layer, the output layer is a fully connected layer, and the activation function of the output layer is Relu; (3) LSTM has 1 LSTM layer, 2 dropout layers, 1 fully connected layer as the output layer, 1 input layer, the activation function of the fully connected layer is Relu, the dropout layer is used to prevent overfitting, and the dropout ratio is 0.2; each method is trained at least 5 times to obtain the final prediction result.

[0122] The final prediction results of different models are as follows: Fig.11 and Fig.12 The test results show that this method is better than the other three model methods in predicting milling force. When the cutting material is steel, the average F x The prediction accuracy reached 96%, and the average F y The prediction accuracy reached 90%, which verified that the method has good generalization performance and can effectively capture the force variation law during steel milling. The experimental results show that the prediction effect of the two different materials of the workpiece is very good, so it can be proved that the prediction effect of the prediction model constructed in this paper is little affected by the change of workpiece material.

[0123] When in use, the target characteristic matrix of the current of the tool whose milling force is to be monitored is obtained and input into the prediction model, and the prediction model can output the predicted bidirectional milling force.

[0124] Example 2

[0125] The present invention discloses a bidirectional milling force prediction system, such as Fig.13 As shown, it includes a data acquisition module, a preprocessing module, a feature extraction module, a feature screening module, an initial module and a training module.

[0126] The data acquisition module is used to obtain the current and milling force data sets of the tool when the same workpiece is milled at two or more different speeds; the current and milling force data sets include the equivalent direct current I rms , bidirectional milling force F x and F y , where F x and F yThe cutting force F pointing in the instantaneous cutting direction t Decomposed, and F x and F y The directions are respectively along the milling feed direction and perpendicular to the milling feed direction; the workpiece is composed of two different materials. In this embodiment, half of the workpiece is carbon steel and the other half is aluminum alloy; the steps of obtaining the current and milling force data sets are as follows:

[0127] The tool first idles on one side of the workpiece, then moves along the length of the workpiece and mills carbon steel and aluminum alloy in turn;

[0128] The tool is made to perform the aforementioned milling operation at different rotation speeds;

[0129] The milling operation at the same speed is repeated for multiple times, and the current and milling force data obtained from the milling operation at the same speed constitute a data group;

[0130] All data sets obtained at different rotational speeds constitute the current and milling force data set.

[0131] A preprocessing module is used to perform wavelet denoising on the current signal in the current and milling force data sets, remove the cutting-in and cutting-out parts of the milling force data in the current and milling force data sets, and retain the signal data of the stable milling stage;

[0132] The feature extraction module is used to extract the time domain, frequency domain and time-frequency domain features of the current signal in the current and milling force data set; extracting the time domain, frequency domain and time-frequency domain features of the current signal in the current and milling force data set includes: extracting each I rms The maximum value, minimum value, average value, peak-to-peak value, absolute mean value, variance, standard deviation, root mean square value, root square amplitude, skewness index, kurtosis index, waveform factor, peak factor, pulse factor, margin factor, center of gravity frequency, mean square frequency, root mean square frequency, frequency variance and 8-layer wavelet decomposition in the idling state, milling carbon steel state and milling aluminum alloy state respectively. All the extracted features constitute the current signal feature M C0 .

[0133] The feature screening module selects the current features that meet the preset conditions to form a target feature matrix based on the stability of the current features, the sensitivity of the current features to the rotation speed, and the correlation between the current features and the milling force;

[0134] The stability of the current characteristic is based on the milling state and milling speed as variables, and the current signal characteristic M C0 As an evaluation index, the calculation formula and screening method of stability are as follows:

[0135] Solve the experimental characteristic dispersion coefficient c of each data group separately v , the calculation formula is as follows:

[0136] c v =σ fea / μ fea

[0137] Among them, σ fea and μ fea are the standard deviation and mean of the same feature in multiple millings at the same speed,

[0138] According to the calculated experimental characteristic dispersion coefficient c v As a result, the threshold is set and the experimental characteristic dispersion coefficient c v The features below the threshold are constructed into the feature matrix M C1 , that is, the experimental characteristic dispersion coefficient c v Features below the threshold are defined as features that are not affected by the rotation speed. Preferably, the threshold can be set according to actual conditions. In this embodiment, the threshold is set to 0.1.

[0139] The sensitivity of the current characteristic to the rotation speed is used to characterize the characteristic current signal characteristic M C0 Response to speed changes, sensitivity Var i The calculation formula and screening method are as follows:

[0140]

[0141] Among them, μ i is the characteristic average value of the i-th data group; n i is the rotation speed of the i-th data group;

[0142] The characteristics that change with the change of milling force are recorded as the characteristic matrix M C2 ,During the cutting process, the changes of Fx and Fy are consistent with the changes of the speed ratio, which means that the magnitude of the milling forces Fx and Fy is positively correlated with the change of the spindle speed, that is, its characteristics are defined as those that are highly sensitive to the changes in the milling forces.

[0143] The correlation between current characteristics and milling force is screened as follows:

[0144] Take the feature matrix M C1 and the feature matrix M C2 The intersection of C3 ;

[0145] Calculate F separately x and F y The average value of the feature matrix M C3 With F x The average value and F y The average values ​​of M were subjected to Pearson correlation analysis. C3 Medium milling force F x and Fy The features with correlation greater than 0.8 are recorded as the target feature matrix.

[0146] The initial module is used to build a neural network model based on the CNN-ResNet deep learning algorithm; the neural network model includes the convolutional layer of CNN and the residual block of ResNet; the core idea of ​​this neural network model is to solve the gradient vanishing or gradient exploding problem in the deep network by introducing residual learning, thereby improving the training efficiency and generalization ability of the model. Fig. 9 As shown, Fig. 9 The structure of the CNN-ResNet deep learning algorithm is shown. Input features are extracted through the convolutional layers of the CNN. The input data is initially fed into the level convolutional layers of the feature extraction part for preliminary processing and then propagated through the two structured residual blocks of the ResNet residual block part. These residual blocks effectively promote the flow of information within the deep neural network by combining the recognition mapping technique, which helps to alleviate the gradient vanishing problem that is prevalent in deep learning models. In addition, the batch normalization operation integrated in these residual blocks further enhances the stability of the training process. By standardizing the output of the intermediate layers, the effective propagation of the gradient during the network training phase is ensured. After being processed by the residual blocks, the data flow is directed to the first fully connected layer of the regression part. The task of this layer is to transform the high-level features extracted by the deep network into the final output prediction. Through the further synthesis and processing of the features in this stage, the model is able to produce accurate prediction results based on the input data. This whole process reflects the effective mechanism of deep learning models using a hierarchical structure approach for feature extraction and transformation when processing complex data.

[0147] The primary convolution layer of the feature extraction part is used to perform preliminary feature extraction on the input data. This layer can be represented by the following convolution operation formula: z = Dx + b, where z represents the output of the convolution layer, D and b represent the weight of the convolution kernel and the weight of the bias term respectively, and x represents the current feature matrix.

[0148] Residual blocks form a key component of the model and are designed to enable the network to directly learn the residual between input and output. Each residual block consists of two one-dimensional convolutional layers, each followed by batch normalization and a ReLU activation function. The output of the residual block can be expressed as the following residual learning formula:

[0149] y=F(z,{D j})+z

[0150] Where y represents the output of the residual block and x represents the input. j}) represents the combination of convolution operation and nonlinear activation function, where {D j} represents the weight in the convolutional layer, j represents the index of the weight in the convolutional layer, that is, it represents the jth weight parameter. This architecture helps to solve the problem of information transmission in deep networks.

[0151] After each convolution operation, batch normalization is used to stabilize the learning process and accelerate convergence. Batch normalization can be expressed as follows:

[0152]

[0153] Where m (k) represents the output of the convolutional layer, s (k) represents the output of the batch normalization layer, and E[m (k) ] and Var[m (k) ] represent the mean and variance of the batch data, respectively, and τ is a very small number to prevent division by zero. In addition, γ and β are learnable parameters used to recover some information that may be lost during the normalization process.

[0154] After each batch normalization, the Relu activation function is applied to introduce nonlinearity to improve the expressiveness of the model. The model is trained using the Adam optimizer and the mean square error (MSE) loss function. The parameters are adjusted in each training period through the back-propagation algorithm to minimize the difference between the predicted output and the actual label.

[0155] During the training process, the Xavier uniform initialization method is used to initialize the model parameters to ensure the stability of the training process. The model is evaluated on the validation set to test its generalization ability.

[0156] The training module divides the data contained in the target feature matrix into a training set and a validation set, and uses the training set to train the neural network model to obtain a preliminary model; imports the validation set, uses the preliminary model, calculates the loss value, and then performs back propagation and parameter updates until the loss function value is minimized and tends to be stable, the iteration stops, and the final prediction model for predicting bidirectional milling force is obtained.

[0157] The input module is used to input the target characteristic matrix of the current of the tool to be monitored for milling force into the prediction model.

[0158] The output module is used to output the bidirectional milling force predicted by the prediction model.

[0159] In order to further verify the accuracy of the prediction model, the prediction model, CNN model, RNN model and LSTM model were used to predict the milling force at the same time. The four prediction results were compared with the actual measurement results to judge the accuracy of the predictions of the four models.

Claims

1. A bidirectional milling force prediction method, characterized in that: The following steps are involved: The experiment obtains the tool current and milling force data sets when the same workpiece is milled at two or more different speeds; The current and milling force data sets include the equivalent DC current I rms , bidirectional milling force F x and F y , where F x and F y The cutting force F pointing in the instantaneous cutting direction t Decomposed, and F x and F y The directions are respectively along the milling feed direction and perpendicular to the milling feed direction; The time domain, frequency domain and time-frequency domain characteristics of the current signal in the current and milling force data set are extracted. According to the stability of the current characteristics, the sensitivity of the current characteristics to the rotation speed, and the correlation between the current characteristics and the milling force, the current characteristics that meet the preset conditions are screened out to form the target feature matrix. Build a neural network model based on the CNN-ResNet deep learning algorithm; The data contained in the target feature matrix is ​​divided into a training set and a validation set, and the neural network model is trained with the training set to obtain a preliminary model; the validation set is imported, the preliminary model is used, the loss value is calculated, and then back propagation and parameter update are performed until the loss function value is minimized and tends to be stable, and the iteration is stopped to obtain the final prediction model for predicting bidirectional milling force; The target characteristic matrix of the current of the tool to be monitored for milling force is obtained and input into the prediction model to predict the bidirectional milling force.

2. The bidirectional milling force prediction method according to claim 1, characterized in that: Before extracting the time domain, frequency domain and time-frequency domain features of the current signal, the current data in the current and milling force data sets are subjected to wavelet denoising. For the milling force data in the current and milling force data sets, the cutting-in and cutting-out parts are removed, and the signal data in the stable milling stage is retained.

3. The bidirectional milling force prediction method according to claim 1, characterized in that: The workpiece is composed of two different materials. The steps for obtaining the current and milling force data sets are as follows: The tool first idles on one side of the workpiece, then moves along the length of the workpiece and mills two different materials one after the other; The tool is made to perform the aforementioned milling operation at different rotation speeds; The milling operation at the same speed is repeated for multiple times, and the current and milling force data obtained from the milling operation at the same speed constitute a data group; All data sets obtained at different rotational speeds constitute the current and milling force data set.

4. The bidirectional milling force prediction method according to claim 3, characterized in that: The time domain, frequency domain and time-frequency domain features of the current signal in the current and milling force data set are extracted, including each I rms Multiple features in idling state, milling carbon steel state, and milling aluminum alloy state.

5. The bidirectional milling force prediction method according to claim 4, characterized in that: The multiple features include maximum value, minimum value, average value, peak-to-peak value, absolute mean value, variance, standard deviation, root mean square value, root amplitude, skewness index, kurtosis index, waveform factor, peak factor, pulse factor, margin factor, center of gravity frequency, mean square frequency, root mean square frequency, frequency variance and 8-layer wavelet decomposition. All extracted features constitute the current signal feature M C0 .

6. The bidirectional milling force prediction method according to claim 5, characterized in that: The stability of the current characteristic is based on the milling state and the milling speed as variables, and the current signal characteristic M C0 As an evaluation index, the calculation formula and screening method of stability are as follows: Solve the experimental characteristic dispersion coefficient c of each data group separately v , the calculation formula is as follows: c v =s fea / m fea Among them, σ fea and μ fea are the standard deviation and mean of the same feature in multiple millings at the same speed, According to the calculated experimental characteristic dispersion coefficient c v As a result, the threshold is set and the experimental characteristic dispersion coefficient c v The features below the threshold are constructed into the feature matrix M C1 .

7. The bidirectional milling force prediction method according to claim 6, characterized in that: The sensitivity of the current characteristic to the rotation speed is used to characterize the characteristic current signal characteristic M C0 Response to speed changes, sensitivity Var i The calculation formula and screening method are as follows: Among them, μ i is the characteristic average value of the i-th data group; n i is the rotation speed of the i-th data group; The features that are highly sensitive to the changes in milling force are recorded as the feature matrix M C2 .

8. The bidirectional milling force prediction method according to claim 7, characterized in that: The correlation between the current characteristics and the milling force is screened as follows: Take the feature matrix M C1 and the feature matrix M C2 The intersection of C3 ; Calculate F separately x and F y The average value of the feature matrix M C3 With F x The average value and F y The average values ​​of M were subjected to Pearson correlation analysis. C3 Medium milling force F x and F y The features with correlation greater than 0.8 are recorded as the target feature matrix.

9. The bidirectional milling force prediction method according to claim 1, characterized in that: The neural network model includes a convolutional layer of a CNN and a ResNet residual block.

10. A bidirectional milling force prediction system, characterized in that: include, A data acquisition module is used to acquire a data set of current and milling force of a tool when the same workpiece is milled at two or more different speeds; A preprocessing module is used to perform wavelet denoising on the current signal in the current and milling force data sets, remove the cutting-in and cutting-out parts of the milling force data in the current and milling force data sets, and retain the signal data of the stable milling stage; A feature extraction module is used to extract the time domain, frequency domain and time-frequency domain features of the current signal in the current and milling force datasets; The feature screening module selects the current features that meet the preset conditions to form a target feature matrix based on the stability of the current features, the sensitivity of the current features to the rotation speed, and the correlation between the current features and the milling force; The initial module is used to build a neural network model based on the CNN-ResNet deep learning algorithm; The training module divides the data contained in the target feature matrix into a training set and a validation set, and uses the training set to train the neural network model to obtain a preliminary model; imports the validation set, uses the preliminary model, calculates the loss value, and then performs back propagation and parameter update until the loss function value is minimized and tends to be stable, and the iteration stops to obtain the final prediction model for predicting the bidirectional milling force; An input module, used for inputting a target characteristic matrix of the current of the tool to be monitored for milling force into a prediction model; The output module is used to output the bidirectional milling force predicted by the prediction model.

Images