Deep hole machining chatter monitoring method, device and system

Abstract

The present invention provides a method, device, and system for monitoring chatter during deep-hole machining. The method comprises: preprocessing collected raw signals using empirical mode decomposition to obtain a signal containing machining state information; establishing a feature vector matrix based on the obtained signal using nonlinear evaluation indicators; constructing a boring chatter recognition model based on a support vector machine to identify early-stage chatter states; establishing a mapping relationship between the boring machine's operating state determination results and output feedback information; establishing a multi-manifold spatial embedding intelligent decision-making mechanism based on manifold learning and support vector machines by analyzing the mapping relationship between the boring machine's operating state determination results and output feedback information; and establishing performance evaluation indicators and a model optimization strategy for the dynamic pattern recognition model based on a dynamic recognition model with continuous adaptive learning capabilities to track and describe changes in the deep-hole boring machine's machining state. The present invention enables online, real-time, and precise monitoring of deep-hole machining chatter.

Description

Technical Field

[0001] The present invention relates to the technical field of mechanical processing detection, and in particular to a deep hole processing chatter monitoring method, device and system. Background Art

[0002] Large, deep-hole parts are widely used in industries such as metallurgy, nuclear power, weapons, aviation, petroleum, and coal. These parts typically have high requirements for size and weight, high material strength and hardness, large hole depth-to-diameter ratios, and demanding geometric accuracy and surface topography. These requirements place extremely high demands on precise and efficient machining processes, specialized tool design and manufacturing, and online detection and control of machining errors.

[0003] Tool-workpiece interaction, a unique physical effect of machining processes, is widely present in processes such as turning, milling, and grinding. It is particularly prominent in deep-hole machining. For example, in typical deep-hole precision machining processes such as floating boring and bore machining, the tool-workpiece interaction effect significantly impacts the machining process and surface finish. The main reasons for this are as follows.

[0004] (1) The geometric shape of the boring tool is much more complex than that of the turning tool and the milling cutter. The rake angle and chip thickness of the boring tool change dynamically along the cutting edge direction. The extrusion, polishing, friction, plowing and process damping effects of the cutting edge during the closed processing are significant, and there are large-scale vibrations in various forms such as torsional flutter, axial torsion, lateral torsion and bending torsion.

[0005] (2) The boring tool has a large aspect ratio and a large overhang during operation. The operating speed is usually higher than the first-order bending critical speed and behaves as a flexible rotor. The stiffness of the process system decreases sharply with the increase of the boring tool overhang. The vibration form is complex and it is easy to generate machining vibration and chatter, resulting in a poor machining surface.

[0006] (3) The relative vibration between the boring tool and the workpiece is introduced by the different dynamic responses of the various components of the machine tool to the machine tool's own disturbance sources and external vibrations, and its vibration conditions will be directly reflected on the workpiece surface.

[0007] (4) When the machine tool vibrates, the boring cutter teeth move slightly in a direction perpendicular to the cutting motion, affecting the actual rake angle, back angle and cutting thickness, and thus affecting the cutting force and the machined surface profile.

[0008] Machining status monitoring is an effective way to suppress machining chatter online, adjust machining status in real time, improve machining process reliability, and ensure deep hole machining accuracy and surface topography requirements. However, the current existing technology still lacks an effective method for boring machining chatter monitoring. Summary of the Invention

[0009] In order to overcome the above-mentioned defects of the prior art, the technical problem to be solved by the embodiments of the present invention is to provide a deep hole machining vibration monitoring method, device and system, which are designed based on the characteristics of deep hole parts such as large volume, high weight, large aspect ratio, and demanding geometric accuracy and surface morphology, to achieve online, real-time and precise monitoring of deep hole machining vibration.

[0010] The above-mentioned object of the present invention can be achieved by adopting the following technical solutions. The present invention provides a method for monitoring chatter during deep hole machining, comprising:

[0011] The empirical mode decomposition method is used to preprocess the collected original signal to obtain a signal containing processing status information;

[0012] Based on the obtained signal containing the processing state information, a characteristic vector matrix is ​​established using a nonlinear evaluation index;

[0013] Construct a boring chatter recognition model based on support vector machine to identify early chatter states;

[0014] Based on the identified early chatter state, a mapping relationship between the boring machine's operating state determination results and output feedback information is established;

[0015] By analyzing the mapping relationship between the boring machine's operating status determination results and output feedback information, a multi-manifold space embedding intelligent decision-making mechanism based on manifold learning and support vector machine is established.

[0016] Based on a dynamic recognition model with continuous adaptive learning capabilities, the performance evaluation indicators and model optimization strategies of the dynamic pattern recognition model are established to track and describe the changes in the machining state of the deep hole boring machine.

[0017] In a preferred embodiment, the signal containing the processing status information includes any one of the following or a combination thereof: a vibration acceleration signal, a displacement signal, and a current signal.

[0018] In a preferred embodiment, the nonlinear evaluation index includes any one of the following or a combination thereof: fractal dimension, power spectrum entropy, standard deviation, and complexity.

[0019] In a preferred embodiment, before constructing the boring chatter recognition model based on the support vector machine, the method further comprises: performing dimensionality reduction processing on the eigenvector matrix using manifold learning.

[0020] In a preferred embodiment, the performance evaluation indicators of the dynamic pattern recognition model include: fractal dimension D and LZ complexity.

[0021] In a preferred embodiment, the process of establishing the dynamic pattern recognition model includes:

[0022] The fractal dimension D and LZ complexity are used as performance evaluation indicators of the dynamic pattern recognition model. The relationship between the processing state and the fractal dimension D is established by calculating the fractal dimension in each time period; the LZ complexity characteristics in each time period are calculated respectively to obtain the relationship between the processing state and the LZ complexity;

[0023] The training set features are subjected to manifold learning feature dimensionality reduction, and the optimal parameter combination is selected using grid search. The combination is then substituted into the support vector machine to establish a flutter monitoring training model.

[0024] The signal after the processing parameters are adjusted is used as a test set, and signal processing analysis and feature dimensionality reduction are performed. The signal is substituted into the chatter monitoring training model to determine the accuracy of the chatter monitoring training model. If the accuracy requirements are met, the model is output; if not, the model establishment method is adjusted to form an optimization strategy for the dynamic pattern recognition model.

[0025] A deep hole machining chatter monitoring device, comprising:

[0026] The signal acquisition module uses the empirical mode decomposition method to pre-process the collected original signal to obtain a signal containing processing status information;

[0027] an eigenvector matrix establishment module, configured to establish an eigenvector matrix using a nonlinear evaluation index based on the obtained signal containing the processing state information;

[0028] Chatter recognition model building module, used to build a boring chatter recognition model based on support vector machine to identify early chatter states;

[0029] A mapping relationship establishment module is used to establish a mapping relationship between the boring machine operation state determination result and the output feedback information based on the identified early chatter state;

[0030] The decision-making mechanism establishment module is used to establish a multi-manifold space embedding intelligent decision-making mechanism based on manifold learning and support vector machine by analyzing the mapping relationship between the boring machine operation status judgment results and output feedback information;

[0031] The processing state tracking module is used to establish the performance evaluation index and model optimization strategy of the dynamic pattern recognition model based on the dynamic recognition model with continuous adaptive learning ability, and to track and describe the changes in the processing state of the deep hole boring machine.

[0032] A data processing server comprises: a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements any one of the methods of claims 1 to 4 when executing the computer program.

[0033] A deep hole machining chatter monitoring system comprises: the data processing server mentioned above, and a controller capable of communicating with the data processing server.

[0034] A deep hole machining chatter monitoring method, comprising:

[0035] The empirical mode decomposition method is used to preprocess the collected original signal to obtain a signal containing processing status information;

[0036] Based on the obtained signal containing the processing state information, a characteristic vector matrix is ​​established using a nonlinear evaluation index;

[0037] A boring chatter recognition model based on support vector machine is constructed to identify the early chatter state.

[0038] In a preferred embodiment, after identifying the early flutter state, the method further includes:

[0039] Based on the identified early chatter state, a mapping relationship between the boring machine's operating state determination results and output feedback information is established;

[0040] By analyzing the mapping relationship between the boring machine's operating status determination results and output feedback information, a multi-manifold space embedding intelligent decision-making mechanism based on manifold learning and support vector machine is established.

[0041] Based on a dynamic recognition model with continuous adaptive learning capabilities, the performance evaluation indicators and model optimization strategies of the dynamic pattern recognition model are established to track and describe the changes in the machining state of the deep hole boring machine.

[0042] In a preferred embodiment, before constructing the boring chatter recognition model based on the support vector machine, the method further comprises: performing dimensionality reduction processing on the eigenvector matrix using manifold learning.

[0043] The technical solution of the present invention has the following significant beneficial effects:

[0044] Overall, the deep hole machining chatter monitoring method provided in this application comprehensively applies sensor testing, signal processing, and information fusion technology to carry out chatter monitoring during deep hole machining, realizes active suppression of machining chatter, and improves deep hole machining efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0046] The drawings described herein are for illustrative purposes only and are not intended to limit the scope of the present invention in any way. In addition, the shapes and proportional dimensions of the various components in the drawings are merely illustrative and are used to help understand the present invention, and are not intended to specifically limit the shapes and proportional dimensions of the various components of the present invention. Those skilled in the art can select various possible shapes and proportional dimensions to implement the present invention according to specific circumstances under the guidance of the present invention.

[0047] Figure 1 A flowchart of the steps of a deep hole machining chatter monitoring method provided in an embodiment of the present invention;

[0048] Figure 2 This is a flow chart of establishing and optimizing a dynamic pattern recognition model provided in an embodiment of the present invention. DETAILED DESCRIPTION

[0049] The details of the present invention can be more clearly understood in conjunction with the accompanying drawings and the description of the specific embodiments of the present invention. However, the specific embodiments of the present invention described herein are for illustrative purposes only and are not to be construed as limiting the present invention in any way. Based on the teachings of the present invention, skilled artisans can conceive of any possible variations based on the present invention, all of which should be considered within the scope of the present invention. It should be noted that when an element is referred to as being "disposed on" another element, it can be directly on the other element or there can be an intermediate element. When an element is considered to be "connected to" another element, it can be directly connected to the other element or there can be an intermediate element. The terms "mounted," "connected," and "connected" should be interpreted broadly. For example, they can refer to mechanical or electrical connections, internal communication between two elements, direct connection, or indirect connection through an intermediary. Those skilled in the art will understand the specific meanings of these terms based on the specific circumstances. The terms "vertical," "horizontal," "upper," "lower," "left," "right," and similar expressions used herein are for illustrative purposes only and do not represent the only embodiments.

[0050] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art to which this application pertains. The terms used herein in the specification of this application are intended only to describe specific embodiments and are not intended to limit this application. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0051] The present application is a boring machining chatter monitoring method based on multi-sensor feature fusion for deep hole machining processes. It is specially designed for deep hole parts with large volume, high weight, large aspect ratio, and demanding geometric accuracy and surface morphology, to achieve online, real-time and precise monitoring of deep hole machining chatter.

[0052] Please refer to Figure 1 To address the problem of difficulty in identifying early-stage chatter during deep hole machining, an embodiment of the present invention provides a method for monitoring deep hole machining chatter. Specifically, the method may include the following steps:

[0053] Step S10: preprocessing the collected original signal using the empirical mode decomposition method to obtain a signal containing processing state information;

[0054] Step S12: Based on the obtained signal containing the processing state information, a characteristic vector matrix is ​​established using a nonlinear evaluation index;

[0055] Step S14: performing dimensionality reduction processing on the eigenvector matrix using manifold learning, constructing a boring chatter recognition model based on a support vector machine, and identifying an early chatter state;

[0056] Step S16: establishing a mapping relationship between the boring machine operation state determination result and the output feedback information based on the identified early chatter state;

[0057] Step S18: By analyzing the mapping relationship between the boring machine operation status determination result and the output feedback information, a multi-manifold space embedding intelligent decision-making mechanism based on manifold learning and support vector machine is established;

[0058] Step S20: Based on the dynamic recognition model with continuous adaptive learning capability, performance evaluation indicators and model optimization strategies of the dynamic pattern recognition model are established to track and describe the changes in the processing state of the deep hole boring machine.

[0059] During deep hole machining, the interaction between the tool and the workpiece and its dynamic evolution lead to significant changes in the machining characteristics and mechanisms of the cutting zone. Factors such as the tool tip action position fluctuating with the tool holder, changes in the friction state of the cutting zone, fluctuations in the actual cutting angle, fluctuations in the instantaneous undeformed chip thickness, and irregular contact between the tool flank wear zone and the deep hole surface will directly or indirectly lead to instability in the deep hole machining process.

[0060] The deep hole machining chatter monitoring method provided in this application comprehensively applies sensor testing, signal processing, and information fusion technology to carry out chatter monitoring during the deep hole machining process, realize active suppression of machining chatter, and improve deep hole machining efficiency and accuracy.

[0061] During deep-hole machining, the key to stability behavior control lies in timely detecting early-stage chatter and rapidly adjusting machining parameters to eliminate it before it's too late. Therefore, establishing an effective early-stage chatter identification and feedback mechanism is central to stability behavior control.

[0062] Chatter can cause changes in signal energy and frequency during machining. When the boring process is stable, the frequency components are widely distributed across the frequency band, and the signal energy is relatively low. When chatter occurs, as the intensity of the chatter increases, the frequency components gradually converge toward the chatter frequency, and the signal energy gradually increases. Rapidly and effectively identifying changes in signal frequency and energy is key to early detection of chatter.

[0063] In an embodiment of the present application, the collected original signal is first preprocessed by adopting the empirical mode decomposition method to obtain a signal containing processing state information; based on the obtained signal containing processing state information, a eigenvector matrix is ​​established using a nonlinear evaluation index; the eigenvector matrix is ​​subjected to dimensionality reduction processing using manifold learning, and a boring chatter recognition model based on a support vector machine is constructed to identify the early chatter state.

[0064] Specifically, in order to obtain a signal that can reflect the processing status, the empirical mode decomposition method is used to preprocess the collected original signal, filter out useless information, and retain the signal containing the processing status information.

[0065] Then, based on vibration acceleration, displacement, current and other signals, the eigenvector matrix can be established using nonlinear evaluation indicators such as fractal dimension, power spectrum entropy, standard deviation, and complexity.

[0066] In addition, in order to reduce the dimension of the eigenvector matrix and improve the computational efficiency, manifold learning is used to reduce the dimensionality of the eigenvector matrix, and a boring chatter recognition model based on support vector machine is constructed to identify the early chatter state.

[0067] Furthermore, the inventors discovered that during deep-hole boring, power consumption varies across different machining states, and this variation can be reflected in the current signal. Due to the low rigidity of the boring bar, chatter easily occurs, damaging the boring tool and exacerbating tool wear. A worn tool consumes higher power. Therefore, by analyzing the deep-hole machining state through changes in current and vibration signals, a mapping relationship can be established between the evolution of machining chatter and the characteristic signals of the machining process.

[0068] By analyzing the mapping relationship between the boring machine's operating status determination results and output feedback information, a multi-manifold space embedding intelligent decision-making mechanism based on manifold learning and support vector machine is established.

[0069] A dynamic recognition model with continuous adaptive learning capabilities is designed, and performance evaluation indicators and model optimization strategies of the dynamic pattern recognition model are established to track and describe the changes in the processing state of the deep hole boring machine and solve the adaptive control problem of the boring process.

[0070] In the embodiment of the present application, a specific process for establishing and optimizing a dynamic pattern recognition model is proposed (see Figure 2 ), and the implementation process of performance evaluation indicators and model optimization strategies for establishing dynamic pattern recognition models is given.

[0071] In view of the changes in the process system stiffness k, natural frequency p and damping ratio ζ caused by reducing or increasing the tool bar overhang, reasonable machining parameters (spindle speed, feed speed, cutting depth) are selected in combination with preliminary experiments to obtain the stable machining stage and the vibration generation stage, and the machining status is monitored online.

[0072] Under different cutting parameter conditions, the displacement signal, acceleration signal and sound pressure signal of the boring process are collected in real time by eddy current sensor, vibration acceleration sensor and sound pressure sensor respectively, and the influence of different machining states on the signal characteristics is analyzed.

[0073] The chatter signals are collected and feature analyzed to obtain 10 chatter-related features of different sensors. The obtained features are preliminarily analyzed and screened, and N (N≤10) features are selected and saved for each signal.

[0074] The EMD (Empirical Mode Decomposition) method is used to denoise the displacement signal, acceleration signal and sound pressure signal, extract flutter-related features from nine different aspects, and construct a high-dimensional observation space.

[0075] Based on the number of fused sensors, we obtain an xN (x = 2, 3, 4, … n)-dimensional observation space. To avoid the dimensionality crisis, we use a manifold learning algorithm for dimensionality reduction. The reduced feature vectors are processed using a support vector machine to identify the machining state. The accuracy of the monitoring models using different dimensionality reduction methods and different signal combinations is compared to determine the dynamic recognition model.

[0076] Fractal dimension D and LZ complexity are used as performance evaluation indicators of dynamic pattern recognition models.

[0077] By calculating the fractal dimension in each time period, the relationship between the processing state and the fractal dimension D is established; the LZ complexity characteristics in each time period are calculated respectively to obtain the relationship between the processing state and the LZ complexity.

[0078] The training set features are subjected to manifold learning feature dimensionality reduction, and the optimal parameter combination is selected by grid search. The parameters are then substituted into the support vector machine to establish a flutter monitoring training model.

[0079] The signal after the processing parameters are adjusted is used as a test set for signal processing analysis and feature dimensionality reduction. The signal is substituted into the training model to judge the accuracy of the model. If the accuracy requirements are met, the model is output. If the accuracy requirements are not met, the model establishment method is adjusted to form an optimization strategy for the dynamic pattern recognition model.

[0080] The deep hole machining chatter monitoring method provided in this application can at least achieve the following technical effects:

[0081] (1) For the boring processing of small-diameter, slender deep holes, a deep hole machining chatter monitoring method based on multi-sensor feature fusion is constructed to suppress the adverse effects of machining chatter on surface topography and surface position error, and achieve precise control of deep hole roundness and straightness;

[0082] (2) In order to solve the problem of deep hole closure and difficulty in directly observing the cutting state of the tool and the surface contour and morphology of the workpiece, a processing stability detection system is built using piezoelectric acceleration sensors and capacitive sensors to achieve rapid and accurate detection and evaluation of the dynamic characteristics of the process system.

[0083] Based on the deep hole machining chatter monitoring method provided in the above embodiment, the present application further provides a deep hole machining chatter monitoring device, which may include:

[0084] The signal acquisition module uses the empirical mode decomposition method to pre-process the collected original signal to obtain a signal containing processing status information;

[0085] an eigenvector matrix establishment module, configured to establish an eigenvector matrix using a nonlinear evaluation index based on the obtained signal containing the processing state information;

[0086] Chatter recognition model building module, used to build a boring chatter recognition model based on support vector machine to identify early chatter states;

[0087] A mapping relationship establishment module is used to establish a mapping relationship between the boring machine operation state determination result and the output feedback information based on the identified early chatter state;

[0088] The decision-making mechanism establishment module is used to establish a multi-manifold space embedding intelligent decision-making mechanism based on manifold learning and support vector machine by analyzing the mapping relationship between the boring machine operation status judgment results and output feedback information;

[0089] The processing state tracking module is used to establish the performance evaluation index and model optimization strategy of the dynamic pattern recognition model based on the dynamic recognition model with continuous adaptive learning ability, and to track and describe the changes in the processing state of the deep hole boring machine.

[0090] It should be noted that the functions implemented by the signal acquisition module, eigenvector matrix establishment module, chatter identification model construction module, mapping relationship establishment module, decision mechanism establishment module, and machining state tracking module correspond to the various steps in the deep-hole machining chatter monitoring method provided in the aforementioned embodiment. Furthermore, the deep-hole machining chatter monitoring device addresses similar technical issues as the aforementioned deep-hole machining chatter monitoring method. Therefore, the use of the deep-hole machining chatter monitoring device can be referenced to the deep-hole machining chatter monitoring method in the aforementioned embodiment and will not be further elaborated here.

[0091] Based on the deep hole machining vibration monitoring method provided in the above embodiment, a data processing server is also provided in the embodiment of the present application, which may include: a memory, a processor, and a computer program stored in the memory and run on the processor. When the processor executes the computer program, the deep hole machining vibration monitoring method in the above embodiment is implemented.

[0092] The data processing server may specifically include a processor, memory, a communications interface, and a communications bus. The processor, memory, and communications interface communicate with each other via the communications bus, and the communications interface is used to transmit information between related devices. The processor is used to invoke a computer program stored in the memory, and when the processor executes the computer program, it implements the deep hole machining chatter monitoring method described in the above embodiment.

[0093] Based on the data processing server provided in the above embodiment, this application also provides a deep hole machining chatter monitoring system, which may include: the above data processing server and a controller capable of communicating with the data processing server. The data processing server and the controller may communicate via a switch, or other methods, which are not specifically limited in this application.

[0094] The controller may specifically be a PLC controller, and other forms of controllers may also be used. Specifically, this application does not make any specific limitations here.

[0095] It will be understood by those skilled in the art that embodiments of the present invention may be provided as methods, systems, or computer program products. Thus, the present invention may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0096] The present invention is described with reference to flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to embodiments of the present invention. It should be understood that each process and / or block in the flowcharts and / or block diagrams, as well as combinations of processes and / or blocks in the flowcharts and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowcharts and / or block diagrams. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0097] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0098] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0099] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. A deep hole machining chatter monitoring method, characterized in that: include: The collected original signal is preprocessed by using an empirical mode decomposition method to obtain a signal containing processing state information, wherein the signal containing processing state information includes: a vibration acceleration signal, a displacement signal, a current signal, and a sound pressure signal; Based on the obtained signal containing the processing state information, a characteristic vector matrix is ​​established using a nonlinear evaluation index, wherein the nonlinear evaluation index includes any one of the following or a combination thereof: fractal dimension, power spectrum entropy, standard deviation, and complexity; Construct a boring chatter recognition model based on support vector machine to identify early chatter states; Based on the identified early chatter state, a mapping relationship between the boring machine's operating state determination results and output feedback information is established; By analyzing the mapping relationship between the boring machine's operating status determination results and output feedback information, a multi-manifold space embedding intelligent decision-making mechanism based on manifold learning and support vector machine is established. Based on a dynamic recognition model with continuous adaptive learning capabilities, the performance evaluation index and model optimization strategy of the dynamic pattern recognition model are established to track and describe the changes in the processing state of the deep hole boring machine; The performance evaluation indicators of the dynamic pattern recognition model include: fractal dimension D and LZ complexity. The establishment process of the dynamic pattern recognition model includes: The fractal dimension D and LZ complexity are used as performance evaluation indicators of the dynamic pattern recognition model. The relationship between the processing state and the fractal dimension D is established by calculating the fractal dimension in each time period; the LZ complexity characteristics in each time period are calculated to obtain the relationship between the processing state and the LZ complexity; The training set features are subjected to manifold learning feature dimensionality reduction, and the optimal parameter combination is selected using grid search. The combination is then substituted into the support vector machine to establish a flutter monitoring training model. The signal after the processing parameters are adjusted is used as a test set, and signal processing analysis and feature dimensionality reduction are performed. The signal is substituted into the chatter monitoring training model to determine the accuracy of the chatter monitoring training model. If the accuracy requirements are met, the model is output; if not, the model establishment method is adjusted to form an optimization strategy for the dynamic pattern recognition model.

2. The deep hole machining chatter monitoring method according to claim 1, characterized in that: Before constructing the boring chatter recognition model based on the support vector machine, the method further includes: performing dimensionality reduction processing on the eigenvector matrix by using manifold learning.

3. A deep hole machining vibration monitoring device, characterized in that: include: A signal acquisition module pre-processes the collected original signal using an empirical mode decomposition method to obtain a signal containing processing state information, wherein the signal containing the processing state information includes: a vibration acceleration signal, a displacement signal, a current signal, and a sound pressure signal; an eigenvector matrix establishment module, configured to establish an eigenvector matrix based on the obtained signal containing the processing state information using a nonlinear evaluation index, wherein the nonlinear evaluation index includes any one of the following or a combination thereof: fractal dimension, power spectrum entropy, standard deviation, and complexity; Chatter recognition model building module, used to build a boring chatter recognition model based on support vector machine to identify early chatter states; A mapping relationship establishment module is used to establish a mapping relationship between the boring machine operation state determination result and the output feedback information based on the identified early chatter state; The decision-making mechanism establishment module is used to establish a multi-manifold space embedding intelligent decision-making mechanism based on manifold learning and support vector machine by analyzing the mapping relationship between the boring machine operation status judgment results and output feedback information; The processing state tracking module is used to establish performance evaluation indicators and model optimization strategies for the dynamic pattern recognition model based on a dynamic recognition model with continuous adaptive learning capabilities, and to track and describe the changes in the processing state of the deep hole boring machine. The performance evaluation indicators of the dynamic pattern recognition model include fractal dimension D and LZ complexity. The process of establishing the dynamic pattern recognition model includes: The fractal dimension D and LZ complexity are used as performance evaluation indicators of the dynamic pattern recognition model. The relationship between the processing state and the fractal dimension D is established by calculating the fractal dimension in each time period; the LZ complexity characteristics in each time period are calculated to obtain the relationship between the processing state and the LZ complexity; The training set features are subjected to manifold learning feature dimensionality reduction, and the optimal parameter combination is selected using grid search. The combination is then substituted into the support vector machine to establish a flutter monitoring training model. The signal after the processing parameters are adjusted is used as a test set, and signal processing analysis and feature dimensionality reduction are performed. The signal is substituted into the chatter monitoring training model to determine the accuracy of the chatter monitoring training model. If the accuracy requirements are met, the model is output; if not, the model establishment method is adjusted to form an optimization strategy for the dynamic pattern recognition model.

4. A data processing server, characterized in that: include: A memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the method according to any one of claims 1 to 2 is implemented when the processor executes the computer program.

5. A deep hole machining chatter monitoring system, characterized in that: include: The data processing server according to claim 4, and a controller capable of communicating with the data processing server.

Images