Numerical control machine tool machining state monitoring method

Abstract

The invention discloses a numerical control machine tool machining state monitoring method, belongs to the technical field of numerical control machine tool monitoring, and is used for monitoring the vibration state of a machine tool milling cutter in the milling process in real time, namely pre-judging the vibration influence on a workpiece. Dividing the complete milling process of the milling cutter into a cut-in stage, a stable stage and a cut-out stage along the timeline, and respectively obtaining self-power spectral density values of the milling cutter in the cut-in stage, the stable stage and the cut-out stage; and S20, processing the auto-power spectral density values of the cut-in stage and the cut-out stage in a compensation manner to maximize the auto-power spectral density values to be close to the auto-power spectral density value of the stationary stage, respectively obtaining a first processing value and a third processing value, and calculating and outputting the auto-power spectral density value of the stationary stage, the average of the first processing value and the third processing value, vibration state information of the machine tool milling cutter is obtained. According to the numerical control machine tool machining state monitoring method, the vibration state information of the milling cutter of the machine tool can be accurately and completely monitored.

Description

Technical Field

[0001] The present invention relates to the technical field of CNC machine tool monitoring, and in particular to a CNC machine tool machining state monitoring method for real-time monitoring of the vibration state of a machine tool milling cutter during a milling process, that is, predicting the vibration impact on a workpiece. Background Art

[0002] In the advanced manufacturing industry, CNC machine tools, as core equipment, play an indispensable and important role. The stability of the milling cutter during the milling process is crucial to the processing quality, especially in the high-speed milling process at a speed of 500m / min. It has important theoretical and practical guiding significance for the promotion and application of high-speed milling.

[0003] In the prior art, there are many methods for monitoring the machining state of CNC machine tools, such as the CNC machine tool machining state monitoring method and device with application number CN116880356A, which can only roughly monitor the parameters of key parts of the machine tool based on sensors, such as vibration parameters, noise parameters and temperature parameters, and then use CAE tools to establish CAE models for simulation analysis. However, there are multiple operation stages in the milling process of the milling cutter, including the cutting-in stage, the stable stage and the cutting-out stage, among which the stable stage lasts the longest. Directly monitoring the vibration parameters through sensors has the problem of inaccurate monitoring data. If the early monitoring data as the data source has a large deviation, no matter how it is adjusted and optimized in the later stage, it is difficult to reflect the actual vibration state of the milling cutter.

[0004] For example, the method for monitoring faults during the milling cutter cutting process based on vibration and acoustic emission sensors with application number CN202311437936.6 collects sensor data containing information such as tool status, processing progress, and fault form through a three-axis accelerometer and acoustic emission sensor installed on a CNC machine tool, monitors the cutting state of the milling cutter, and uses a CNN neural network to improve the training speed and accuracy of the tool fault identification model to prevent production risks that may be caused by the milling cutter continuing to process after it is damaged. It only monitors the cutting process of the milling cutter, and there is a problem of incomplete monitoring data, which cannot reflect the complete vibration state of the milling cutter, and it is also difficult to analyze the stability of the milling process of the milling cutter. Summary of the invention

[0005] The embodiment of the present application provides a method for monitoring the machining status of a CNC machine tool, which can overcome the defects of inaccurate and incomplete monitoring data in the prior art, and thus can obtain accurate and complete information on the vibration status of the machine tool milling cutter, so that problems in the milling cutter machining process can be discovered and adjusted in a timely manner, thereby improving the machining accuracy of the workpiece, and helping to determine the optimal cutting parameters of the milling cutter, thereby achieving an efficient and stable cutting process.

[0006] The embodiment of the present application provides a method for monitoring the machining state of a CNC machine tool, which is used to monitor the vibration state of a machine tool milling cutter in real time during the milling process, that is, to predict the vibration impact on a workpiece, and includes the following steps:

[0007] S10, dividing the complete milling process of the milling cutter into three stages along the time line, namely, a cutting-in stage, a stable stage, and a cutting-out stage, and obtaining the auto-power spectrum density values ​​of the milling cutter in the cutting-in stage, the stable stage, and the cutting-out stage respectively;

[0008] S20, processing the auto-power spectrum density values ​​of the cutting-in stage and the cutting-out stage in a compensatory manner to maximize the auto-power spectrum density value close to the auto-power spectrum density value of the stable stage, and obtaining the first processing value and the third processing value respectively, calculating and outputting the average of the auto-power spectrum density value of the stable stage, the first processing value, and the third processing value, to obtain the vibration state information of the machine tool milling cutter.

[0009] In a possible implementation, in step S10, the self-power spectrum density values ​​of the milling cutter in the cutting-in stage, the stable stage and the cutting-out stage are respectively obtained through the machine tool milling cutter vibration model, wherein the method for establishing the machine tool milling cutter vibration model includes:

[0010] Based on the predetermined milling direction, the milling area is divided into n segments, and the characteristic parameters of the feed speed direction, milling width direction, and milling depth direction of the milling cutter in each segment are extracted respectively to form a sub-segment milling area database, where n is greater than or equal to 50;

[0011] In the sub-segment milling area database, the n-segment counts are 1, 2, 3, ... n as horizontal columns, and the feed speed parameter, the milling width parameter and the milling depth parameter are parallel vertical columns to form a milling data matrix;

[0012] Compare the feed speed parameter, milling width parameter and milling depth parameter corresponding to each transverse column respectively, and arrange them in ascending or descending order according to the numerical values ​​to form a first data set, a second data set and a third data set respectively;

[0013] Respectively remove the largest and smallest numbers in the first data set, the second data set and the third data set, and then remove the entire column corresponding to the sub-paragraph milling area database, including repeated and non-repeated columns in the first data set, the second data set and the third data set, to form a screened milling area database, including a cut-in phase milling database, a stable phase milling database and a cut-out phase milling database, wherein the number of transverse columns in the screened milling area database is greater than or equal to 30, and when the number of transverse columns in the sub-paragraph milling area database is less than 30, reduce the value of the several numbers or increase the value of n, and rearrange and screen so that the number of transverse columns is greater than or equal to 30;

[0014] Feature parameters are extracted from the milling database in the cutting-in stage, the milling database in the stable stage and the milling database in the cutting-out stage respectively, and a vibration model of the machine tool milling cutter with three stages is constructed using the Gaussian process model.

[0015] In one possible implementation, in step S20, the auto-power spectral density values ​​of the cut-in phase and the cut-out phase are processed in a compensatory manner to maximize the auto-power spectral density values ​​close to the auto-power spectral density values ​​of the stable phase, specifically: a compensation model is established based on the output of the neural network to minimize the difference between the auto-power spectral density values ​​of the cut-in phase and the cut-out phase and the auto-power spectral density values ​​of the stable phase, wherein the neural network is a multilayer perceptron (MLP) or a radial basis function network (RBFN).

[0016] In a possible implementation, the compensation model is established based on the output of the neural network, specifically, the compensation model is established through the output of a multi-layer perceptron (MLP);

[0017] The multilayer perceptron (MLP) includes three input layers, two hidden layers and one output layer. The three input layers correspond to the feed speed parameters, milling width parameters and milling depth parameters in the milling database of the cutting stage respectively. The three input layers are input into the fully connected network of the two hidden layers, and then activated by the activation function ReLU and output through the output layer to establish a cutting stage compensation model, and then calculate the cutting stage compensation coefficient. Then, the cutting out stage compensation model is established in the same way, and the cutting out stage compensation coefficient is calculated. Then, the cutting parameters are adjusted based on the vibration conditions monitored in the actual cutting process to continuously and dynamically update and optimize the cutting-in stage compensation model and the cutting-out stage compensation model, that is, update and optimize the cutting-in stage compensation coefficient and the cutting-out stage compensation coefficient.

[0018] In a possible implementation, in the process of establishing the compensation model based on the output of the neural network, firstly, frequency features of the feed speed parameters, milling width parameters and milling depth parameters in the milling database of the cutting-in phase and the feed speed parameters, milling width parameters and milling depth parameters in the milling database of the cutting-out phase are extracted to determine the frequency range that needs to be compensated;

[0019] An IIR filter is designed according to the extracted frequency characteristic results to determine the frequency range, order and bandwidth of the filter. The milling database in the cutting-in stage and the milling database in the cutting-out stage are filtered respectively and then input into the fully connected network.

[0020] In one possible implementation, the compensation effects of the cut-in stage compensation coefficient and the cut-out stage compensation coefficient are verified respectively by the correlation coefficient method. If the compensation effect exceeds a predetermined threshold, the filter is additionally filtered, band-stop filtering and adaptive filtering are performed in sequence in a cyclic manner until the compensation effect is within a predetermined threshold range.

[0021] Beneficial effect: Compared with the prior art, the CNC machine tool processing state monitoring method provided by the present application obtains the self-power spectrum density values ​​of the milling cutter in the cutting-in stage, the stable stage and the cutting-out stage respectively, and processes the self-power spectrum density values ​​of the cutting-in stage and the cutting-out stage in a compensation manner to maximize the self-power spectrum density value close to the stable stage, and then obtains the average value of the three processed values ​​to obtain the vibration state information of the machine tool milling cutter, thereby fully considering the vibration information of the milling cutter at different stages in the cutting process, and then ensuring the integrity and accuracy of the monitoring data, so that various problems occurring in the milling cutter processing process can be discovered and adjusted in time, the processing accuracy and processing stability of the workpiece can be improved, the wear and damage of the milling cutter can be monitored and the milling cutter can be replaced in time, which is helpful to determine the optimal cutting parameters of the milling cutter, realize an efficient and stable cutting process, and help optimize the machine tool results, improve the rigidity and stability of the machine tool, thereby improving the overall performance of the machine tool;

[0022] The milling area is divided into n segments, and the characteristic parameters of the milling cutter in the feed speed direction, milling width direction, and milling depth direction in each segment are extracted respectively. Then, a specific method is used to strip and filter the data set to form a milling database for the cutting-in stage, a milling database for the stable stage, and a milling database for the cutting-out stage. Finally, based on these databases, characteristic parameters are extracted to construct a machine tool milling cutter vibration model with three stages. This can effectively strip the unreliable factors in the milling area database of the sub-segment segment, and ensure the accuracy of the machine tool milling cutter vibration model.

[0023] Combining the neural network algorithm and the compensation model to calculate the compensation coefficient of the milling cutter in the cutting-in stage and the cutting-out stage can make full use of the powerful nonlinear mapping ability and self-learning ability of the neural network, as well as the accuracy of the compensation model, to more accurately estimate the vibration compensation amount of the milling cutter in the cutting-in stage and the vibration compensation amount of the milling cutter in the cutting-out stage, thereby ensuring the accuracy of the estimated vibration state of the milling cutter and further ensuring the integrity and accuracy of the CNC machine tool processing status monitoring data.

[0024] These and other objects, features and advantages of the present invention will be fully reflected in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0025] Figure 1 The schematic diagram shows the principle of the CNC machine tool processing status monitoring method of the present application.

[0026] Figure 2 An example schematic diagram of screening the milling area database in the present application is shown. DETAILED DESCRIPTION

[0027] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art can think of other obvious variations. The basic principles of the present invention defined in the following description can be applied to other embodiments, variations, improvements, equivalents, and other technical solutions that do not deviate from the spirit and scope of the present invention.

[0028] Those skilled in the art should understand that, in the disclosure of the specification, the orientation or position relationship indicated by the terms "longitudinal", "lateral", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", etc. are based on the orientation or position relationship shown in the drawings, which are only for the convenience of describing the present invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, be constructed and operate in a specific orientation. Therefore, the above terms should not be understood as limiting the present invention.

[0029] It is to be understood that the term "one" should be understood as "at least one" or "one or more", that is, in one embodiment, the number of an element may be one, while in another embodiment, the number of the element may be multiple, and the term "one" should not be understood as a limitation on the quantity.

[0030] During the experimental study of the stability of the machine tool milling cutter vibration system, it was found that the milling process of the machine tool milling cutter can be roughly divided into three stages, namely the cutting-in stage when the milling cutter and the workpiece just come into contact, the stable stage after the contact is completed, and the cutting-out stage when the milling cutter moves closer to the sensor position. The changes in the self-power spectrum density value of the milling cutter vibration response correspond to these three stages, namely, in the cutting-in stage, the vibration is significant due to the cutting-in impact, the self-power spectrum density value is the largest, and then it decreases rapidly. In the stable stage, the spectrum value does not fluctuate significantly. In the cutting-out stage, the vibration response gradually increases again due to the cutting-out impact, until the milling cutter completely leaves the workpiece and stops processing, and the vibration of the workpiece will reach an extreme value. Therefore, there are significant differences in the self-power spectrum density values ​​corresponding to different stages, among which the maximum power spectrum density value will be several times the minimum power spectrum density value, which also indicates that the vibration impact on the workpiece during the milling process has a large difference, and the maximum vibration energy value is several times the minimum vibration energy value.

[0031] Therefore, in the prior art of monitoring the machining state of CNC machine tools, there is an inaccurate problem in directly monitoring the vibration information of the milling cutter through sensors (such as displacement sensors, velocity sensors, acceleration sensors, and piezoelectric sensors). For example, different monitoring times may monitor data of different stages. For example, even through multiple monitoring and averaging methods, it is still difficult to make up for this deviation, because there is a large irrational deviation in the data of the cut-in stage and the cut-out stage. Even if this irrational deviation is balanced by substituting data from more stable stages, it still affects the accuracy of the data. In addition, only monitoring the state of the milling cutter at a certain stage, such as the state of the cut-in stage disclosed in the comparative document, or even monitoring the states of the three stages of the cut-in stage, stable stage, and cut-out stage at the same time, will still cause incomplete and inaccurate monitoring data, because these three stages are difficult to accurately and clearly divide, and the states of these three stages are not integrated. In addition, in the actual operation process, real-time monitoring of the states of the three stages at the same time will seriously increase the monitoring task, high monitoring cost, and high maintenance cost.

[0032] Therefore, how to fully and effectively integrate the three stages of the milling process to ensure the accuracy and completeness of monitoring the vibration state of the milling cutter during the milling process, that is, the accuracy and completeness of predicting the impact of workpiece vibration, is a major problem in the current advanced manufacturing CNC machine tool field.

[0033] refer to Figure 1 The embodiment of the present application provides a method for monitoring the machining state of a CNC machine tool, which is used to monitor the vibration state of a machine tool milling cutter in real time during the milling process, that is, to predict the vibration impact on a workpiece, and includes the following steps:

[0034] S10, dividing the complete milling process of the milling cutter into three stages along the time line, namely, a cutting-in stage, a stable stage, and a cutting-out stage, and obtaining the auto-power spectrum density values ​​of the milling cutter in the cutting-in stage, the stable stage, and the cutting-out stage respectively;

[0035] S20, processing the auto-power spectrum density values ​​of the cutting-in stage and the cutting-out stage in a compensatory manner to maximize the auto-power spectrum density value close to the auto-power spectrum density value of the stable stage, and obtaining the first processing value and the third processing value respectively, calculating and outputting the average of the auto-power spectrum density value of the stable stage, the first processing value, and the third processing value, to obtain the vibration state information of the machine tool milling cutter.

[0036] Therefore, the CNC machine tool processing state monitoring method provided in the present application compensates for the self-power spectrum density values ​​in the cutting-in stage and the cutting-out stage to maximize the self-power spectrum density values ​​close to the self-power spectrum density values ​​in the stable stage, so that in the actual monitoring process, the vibration information of the milling cutter monitored in any stage is basically accurate vibration information, the monitoring is convenient, and the monitoring accuracy is high. At the same time, in the monitoring method provided in the present application, the first processing value, the self-power spectrum density value in the stable stage, and the third processing value are further averaged, which can maximize the accuracy and completeness of the vibration state information of the machine tool milling cutter, and can easily and accurately analyze the vibration impact of the machine tool milling cutter on the workpiece, and then can timely discover and adjust various problems that occur in the milling cutter during the processing process, improve the processing accuracy and processing stability of the workpiece, and at the same time can also monitor the wear and damage of the milling cutter in order to replace the milling cutter in time, and help determine the optimal cutting parameters of the milling cutter to achieve an efficient and stable cutting process.

[0037] In one embodiment, in step S10, the self-power spectrum density values ​​of the milling cutter in the cutting-in stage, the stable stage and the cutting-out stage are respectively obtained through the machine tool milling cutter vibration model, wherein the method for establishing the machine tool milling cutter vibration model includes:

[0038] Based on the predetermined milling direction, the milling area is divided into n segments, and the characteristic parameters of the feed speed direction, milling width direction, and milling depth direction of the milling cutter in each segment are extracted respectively to form a sub-segment milling area database, where n is greater than or equal to 50;

[0039] Then, in the sub-segment milling area database, the n-segment counts are 1, 2, 3, ... n as horizontal columns, and the feed speed parameter, the milling width parameter and the milling depth parameter are parallel vertical columns to form a milling data matrix;

[0040] Then, the feed speed parameter, the milling width parameter and the milling depth parameter corresponding to each transverse column are compared respectively, and arranged in ascending or descending order according to the numerical values ​​to form a first data set, a second data set and a third data set respectively;

[0041] Then, the largest and smallest numbers in the first data set, the second data set, and the third data set are removed respectively, and then the entire column corresponding to the sub-paragraph milling area database is removed, including the repeated and non-repeated columns in the first data set, the second data set, and the third data set, and the remaining data are recombined (the order of combination is not important, because the final average is obtained, and the combination can be performed by column, or by increasing or decreasing a row of data), to form a screened milling area database, including a cut-in stage milling database, a stable stage milling database, and a cut-out stage milling database, wherein the transverse columns in the screened milling area database are greater than or equal to 30, and when the transverse columns in the sub-paragraph milling area database are less than 30, the values ​​of the several numbers are reduced or the value of n is increased, and the rearrangement and screening are performed so that the number of transverse columns is greater than or equal to 30, so as to ensure that there are enough data to support the accuracy of the monitoring results;

[0042] Finally, feature parameters are extracted from the milling database of the cutting-in stage, the milling database of the stable stage, and the milling database of the cutting-out stage, and a machine tool milling cutter vibration model with three stages is constructed using the Gaussian process model. In this way, the milling area can be accurately divided into the cutting-in stage, the stable stage, and the cutting-out stage through sufficient data support. At the same time, the irrational data of the feed speed parameters, the milling width parameters, and the milling depth parameters in the three stages can be filtered out respectively to ensure the accuracy of the data. Finally, feature parameters are extracted from the milling database of the cutting-in stage, the milling database of the stable stage, and the milling database of the cutting-out stage, and a machine tool milling cutter vibration model with three stages is constructed using the Gaussian process model. This is a complete and accurate machine tool milling cutter vibration model, which can accurately obtain the self-power spectral density values ​​of the milling cutter in the cutting-in stage, the stable stage, and the cutting-out stage during the milling process.

[0043] Gaussian Process (GP) is a kind of stochastic process in probability theory and mathematical statistics. It is a combination of a series of random variables that obey the normal distribution in an index set.

[0044] Any linear combination of random variables in a Gaussian process obeys a normal distribution, and each finite-dimensional distribution is a joint normal distribution. Its probability density function on a continuous exponential set is the Gaussian measure of all random variables, so it is considered an infinite-dimensional generalized extension of the joint normal distribution. The Gaussian process is completely determined by its mathematical expectation and covariance function, and inherits many properties of the normal distribution.

[0045] An example of screening the milling area database is as follows. Based on the predetermined milling direction (such as the cutting y direction), the milling area is divided into n segments to form n horizontal columns. The feed speed parameters, milling width parameters and milling depth parameters corresponding to each horizontal column are arranged in ascending or descending order according to the numerical value to form the first data set, the second data set and the third data set respectively. Then, the two largest numbers and the two smallest numbers in each data set are removed (the number of the largest and smallest numbers removed may be the same or different), a total of 3*4=12 numbers, and then the entire column corresponding to the 12 numbers just removed is removed. If two of the 12 numbers correspond to the same column at the same time, a total of 11 columns are removed. If six of the 12 numbers correspond to two columns at the same time, a total of 9 columns of data are removed. For example, refer to Figure 2 In the feed speed parameters, the 3rd, 6th, 10th, and nth columns are removed as the two largest and two smallest numbers, in the milling width parameters, the 2nd, 6th, 8th, and 14th columns are removed as the two largest and two smallest numbers, and in the milling depth parameters, the 2nd, 6th, 8th, and 14th columns are removed as the two largest and two smallest numbers. Then, the data in the 2nd, 3rd, 5th, 6th, 8th, 10th, 13th, 14th, and nth columns are all removed, thereby ensuring the accuracy of the feed speed parameters, milling width parameters, and milling depth parameters, and then the milling area of ​​the milling cutter can be divided as accurately as possible into the cutting-in stage, the stable stage, and the cutting-out stage by means of big data calculation, so that the self-power spectral density values ​​of the milling cutter in the cutting-in stage, the stable stage, and the cutting-out stage can be accurately obtained by the machine tool milling cutter vibration model, providing data support for the accurate determination of the subsequent cutting-in stage compensation coefficient and the cutting-out stage compensation coefficient.

[0046] In addition, it should be noted that in order to ensure the accuracy of the data, there must be enough basic data. Therefore, it is necessary to ensure that the number of horizontal columns in the milling area database after screening is greater than or equal to 30. When the number of horizontal columns in the sub-paragraph milling area database is less than 30, the value of the number can be reduced or the value of n can be increased, and then the data can be rearranged and screened again so that the number of horizontal columns is greater than or equal to 30.

[0047] Among them, the initially formed sub-segment milling area database, where n is greater than or equal to 50, can be implemented by a precision robot or by computer simulation.

[0048] In one embodiment, in step S20, the self-power spectral density values ​​of the cut-in stage and the cut-out stage are processed in a compensatory manner to maximize the self-power spectral density values ​​close to the self-power spectral density values ​​of the stable stage, specifically: a compensation model is established based on the output of the neural network to minimize the difference between the self-power spectral density values ​​of the cut-in stage and the cut-out stage and the self-power spectral density values ​​of the stable stage, wherein the neural network is a multilayer perceptron (MLP) or a radial basis function network (RBFN), and thus the complex patterns in the data can be captured through the powerful nonlinear mapping ability of the neural network, thereby improving the prediction accuracy of the compensation model, and minimizing the difference between the self-power spectral density values ​​of the cut-in stage and the cut-out stage and the self-power spectral density values ​​of the stable stage. At the same time, the neural network can also make predictions on unseen data and has good generalization ability. In addition, it can also automatically adjust the model parameters according to the changes in the input data, so that the compensation model can adapt to different milling conditions, and can reduce the time of human intervention and debugging through automatic prediction and compensation, and has high work efficiency.

[0049] Further preferably, the compensation model is established based on the output of the neural network, specifically, the compensation model is established through the output of a multi-layer perceptron (MLP);

[0050] The multilayer perceptron (MLP) includes three input layers, two hidden layers and one output layer, wherein the three input layers correspond to the feed speed parameters, milling width parameters and milling depth parameters in the milling database of the cutting stage, respectively. The three input layers are input into the fully connected network of the two hidden layers, and then output through the output layer after being activated by the activation function ReLU, to establish a cutting stage compensation model, and then various parameters (including the feed speed parameters, milling width parameters and milling depth parameters of the milling cutter and related information of the workpiece) can be directly introduced in the actual milling process to calculate the cutting stage compensation coefficient, and then the cutting-out stage compensation model is established in the same way to calculate the cutting-out stage compensation coefficient, and then the cutting parameters are adjusted based on the vibration conditions monitored in the actual cutting process, wherein the cutting parameters include feed speed, milling width and milling depth, and the cutting-in stage compensation model and the cutting-out stage compensation model can be continuously and dynamically updated and optimized, that is, the cutting-in stage compensation coefficient and the cutting-out stage compensation coefficient are updated and optimized, wherein the activation formula of the activation function is: Therefore, in the process of establishing the compensation model, the feed speed, milling width and milling depth of the milling cutter can be fully considered through the multi-layer perceptron to ensure the accuracy of the compensation coefficient of the cutting-in stage and the compensation coefficient of the cutting-out stage. Of course, this is a generalized compensation model based on fixed workpiece information. In the actual working process, the situation of the workpiece needs to be considered specifically. Once the information of the workpiece to be processed is determined, the vibration state information of the machine tool milling cutter can be directly monitored accurately through the machine tool milling cutter vibration model and compensation model. The information of the workpiece to be processed includes the hardness, toughness, thermal conductivity and related mechanical performance parameters (such as yield strength, elongation, etc.) of the workpiece.

[0051] For example, in the process of determining the compensation coefficient in the cut-in phase, refer to the simplified formula Kv=f(V,W,D,T), where:

[0052] Kv is the compensation coefficient of the cut-in phase.

[0053] V is the feed speed parameter.

[0054] W is the cutting width parameter.

[0055] D is the cutting depth parameter.

[0056] T is the workpiece information. When the workpiece toughness is mainly considered, the workpiece toughness parameter ∫σdε can be substituted into the formula. When the workpiece hardness, toughness and yield strength need to be considered at the same time, the specific proportion of each parameter can be determined based on actual needs, and then the T value is calculated, such as T = hardness value * 30% + toughness value * 30% + yield strength value * 40%.

[0057] f is a function that represents the relationship between the compensation coefficient of the cutting stage and the cutting parameters and workpiece information, and is directly obtained from the cutting stage compensation model.

[0058] In one embodiment, in the process of establishing a compensation model based on the output of the neural network, firstly, frequency feature extraction is performed on the feed speed parameter, milling width parameter and milling depth parameter in the milling database of the cutting-in phase and the feed speed parameter, milling width parameter and milling depth parameter in the milling database of the cutting-out phase to determine the frequency range that needs to be compensated;

[0059] An IIR filter is designed according to the extracted frequency characteristic results to determine the frequency range, order and bandwidth of the filter, and the milling database of the cutting-in stage and the milling database of the cutting-out stage are filtered respectively, and then input into the fully connected network. In this way, the signals of the milling cutter in the cutting-in stage and the cutting-out stage can be processed by the IIR filter specially designed for the feed speed parameters, milling width parameters and milling depth parameters, and the compensated signals are input into the multi-layer perceptron (MLP), which can further improve the accuracy of the compensation coefficients of the cutting-in stage and the cutting-out stage.

[0060] In one embodiment, the compensation effects of the cut-in stage compensation coefficient and the cut-out stage compensation coefficient are verified respectively by the correlation coefficient method. If the compensation effect exceeds a predetermined threshold, the filter is cyclically added with additional filtering stages, band-stop filtering and adaptive filtering are performed in sequence until the compensation effect is within a predetermined threshold range. If the ideal effect cannot be achieved after n cycles, the calculation and simulation verification are performed again, where n is greater than or equal to 50.

[0061] It should be understood by those skilled in the art that the embodiments of the present invention described above and shown in the accompanying drawings are only examples and do not limit the present invention. The advantages of the present invention have been fully and effectively achieved. The functional and structural principles of the present invention have been demonstrated and explained in the embodiments, and the embodiments of the present invention may be deformed or modified in any way without departing from the principles.

Claims

1. A method for monitoring the machining state of a CNC machine tool, used for real-time monitoring the vibration state of a machine tool milling cutter during the milling process, that is, predicting the vibration impact on a workpiece, characterized in that: The following steps are involved: S10, dividing the complete milling process of the milling cutter into three stages along the time line, namely, a cutting-in stage, a stable stage, and a cutting-out stage, and obtaining the auto-power spectrum density values ​​of the milling cutter in the cutting-in stage, the stable stage, and the cutting-out stage respectively; S20, processing the auto-power spectrum density values ​​of the cutting-in stage and the cutting-out stage in a compensatory manner to maximize the auto-power spectrum density value close to the auto-power spectrum density value of the stable stage, and obtaining the first processing value and the third processing value respectively, calculating and outputting the average of the auto-power spectrum density value of the stable stage, the first processing value, and the third processing value, to obtain the vibration state information of the machine tool milling cutter.

2. The method for monitoring the machining state of a CNC machine tool according to claim 1, wherein: In step S10, the self-power spectrum density values ​​of the milling cutter in the cutting-in stage, the stable stage and the cutting-out stage are respectively obtained through the machine tool milling cutter vibration model, wherein the method for establishing the machine tool milling cutter vibration model includes: Based on the predetermined milling direction, the milling area is divided into n segments, and the characteristic parameters of the feed speed direction, milling width direction, and milling depth direction of the milling cutter in each segment are extracted respectively to form a sub-segment milling area database, where n is greater than or equal to 50; In the sub-segment milling area database, the n-segment counts are 1, 2, 3, ... n as horizontal columns, and the feed speed parameter, the milling width parameter and the milling depth parameter are parallel vertical columns to form a milling data matrix; Compare the feed speed parameter, milling width parameter and milling depth parameter corresponding to each transverse column respectively, and arrange them in ascending or descending order according to the numerical values ​​to form a first data set, a second data set and a third data set respectively; Respectively remove the largest and smallest numbers in the first data set, the second data set and the third data set, and then remove the entire column corresponding to the sub-paragraph milling area database, including repeated and non-repeated columns in the first data set, the second data set and the third data set, to form a screened milling area database, including a cut-in phase milling database, a stable phase milling database and a cut-out phase milling database, wherein the number of transverse columns in the screened milling area database is greater than or equal to 30, and when the number of transverse columns in the sub-paragraph milling area database is less than 30, reduce the value of the several numbers or increase the value of n, and rearrange and screen so that the number of transverse columns is greater than or equal to 30; Feature parameters are extracted from the milling database in the cutting-in stage, the milling database in the stable stage and the milling database in the cutting-out stage respectively, and a vibration model of the machine tool milling cutter with three stages is constructed using the Gaussian process model.

3. The method for monitoring the machining state of a CNC machine tool according to claim 2, characterized in that: In step S20, the auto-power spectral density values ​​of the cut-in phase and the cut-out phase are processed in a compensatory manner to maximize the auto-power spectral density values ​​close to the auto-power spectral density values ​​of the stable phase, specifically: a compensation model is established based on the output of the neural network to minimize the difference between the auto-power spectral density values ​​of the cut-in phase and the cut-out phase and the auto-power spectral density values ​​of the stable phase, wherein the neural network is a multilayer perceptron (MLP) or a radial basis function network (RBFN).

4. The method for monitoring the machining state of a CNC machine tool according to claim 3, characterized in that: The compensation model is established based on the output of the neural network, specifically, the compensation model is established through the output of a multi-layer perceptron (MLP); The multilayer perceptron (MLP) includes three input layers, two hidden layers and one output layer. The three input layers correspond to the feed speed parameters, milling width parameters and milling depth parameters in the milling database of the cutting stage respectively. The three input layers are input into the fully connected network of the two hidden layers, and then activated by the activation function ReLU and output through the output layer to establish a cutting stage compensation model, and then calculate the cutting stage compensation coefficient. Then, the cutting out stage compensation model is established in the same way, and the cutting out stage compensation coefficient is calculated. Then, the cutting parameters are adjusted based on the vibration conditions monitored in the actual cutting process to continuously and dynamically update and optimize the cutting-in stage compensation model and the cutting-out stage compensation model, that is, update and optimize the cutting-in stage compensation coefficient and the cutting-out stage compensation coefficient.

5. The method for monitoring the machining state of a CNC machine tool according to claim 4, characterized in that: In the process of establishing the compensation model based on the output of the neural network, firstly, the frequency characteristics of the feed speed parameters, milling width parameters and milling depth parameters in the milling database of the cutting-in phase and the feed speed parameters, milling width parameters and milling depth parameters in the milling database of the cutting-out phase are extracted to determine the frequency range that needs to be compensated; An IIR filter is designed according to the extracted frequency characteristic results to determine the frequency range, order and bandwidth of the filter. The milling database in the cutting-in stage and the milling database in the cutting-out stage are filtered respectively and then input into the fully connected network.

6. The method for monitoring the machining state of a CNC machine tool according to claim 5, characterized in that: The compensation effects of the cut-in stage compensation coefficient and the cut-out stage compensation coefficient are verified respectively by the correlation coefficient method. If the compensation effect exceeds the predetermined threshold, the filter is additionally filtered, band-stop filtering and adaptive filtering are performed in sequence until the compensation effect is within the predetermined threshold range.

Images