A numerical control machine tool health diagnosis method and system based on vibration spectrum coupling

Abstract

The present application relates to a kind of numerical control machine tool health diagnosis method and system based on vibration spectrum coupling, belong to the field of fault diagnosis.Therein, the method includes obtaining original electric energy by energy self-adaptive capture and converting into stable electric energy, obtain coupling coefficient matrix;Based on coupling coefficient matrix, stability evaluation is carried out, and dynamic stability margin is obtained;Based on dynamic stability margin, early fault risk prediction model is constructed, and fault development time prediction value is obtained;Based on three-order resonance modal coupling coefficient, fault coupling frequency combination is screened out, dynamic parameter compensation is carried out based on fault coupling frequency combination, and targeted parameter compensation amount is obtained;Based on targeted parameter compensation amount, corresponding parameter is compensated, and global knowledge evolution is carried out.The present application realizes vibration spectrum coupling analysis, energy self-capture, dynamic stability evaluation and knowledge evolution mechanism, realizes sensitive capture of early fault, predictive maintenance and adaptive compensation, and improves the intelligent level of machine tool health management.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of fault diagnosis, and particularly relates to a numerical control machine tool health diagnosis method and system based on vibration spectrum coupling. BACKGROUND

[0002] In the current wave of intelligent transformation of manufacturing industry, the health monitoring technology of numerical control machine tools still faces severe challenges. Traditional methods mainly rely on external sensor networks (such as accelerometers and power meters) to collect vibration and energy consumption data, and transmit the data to the central system through wired or wireless networks for threshold alarm or spectrum analysis. Although this external scheme can achieve basic state monitoring, it has several fundamental defects. First, its deployment requires complex wiring or battery maintenance, which is unreliable and costly in industrial sites with serious oil pollution and electromagnetic interference. Second, traditional vibration analysis only focuses on amplitude or linear spectrum characteristics, and cannot capture weak precursors of non-linear interactions between components, resulting in alarm lagging behind actual damage. Third, existing systems lack deep analysis of fault mechanisms, and can only determine abnormalities but cannot accurately locate where the abnormalities are and what the abnormalities are, making maintenance decisions dependent on human experience. More importantly, these monitoring systems themselves constitute additional energy consumption units, which deviate from the green industrialization goal. SUMMARY

[0003] To solve the above problems in the prior art, the application provides a numerical control machine tool health diagnosis method and system based on vibration spectrum coupling.

[0004] The purpose of the application can be achieved by the following technical solutions:

[0005] A numerical control machine tool health diagnosis method based on vibration spectrum coupling, the implementation of the numerical control machine tool health diagnosis method based on vibration spectrum coupling includes the following steps:

[0006] Step S1: obtaining original electric energy through energy self-adaptive capture and converting the original electric energy into stable electric energy, charging an energy storage battery, and obtaining a third-order resonance modal coupling coefficient, and constructing a coupling coefficient matrix based on the third-order resonance modal coupling coefficient;

[0007] Step S2: performing stability evaluation based on the coupling coefficient matrix to obtain a dynamic stability margin;

[0008] Step S3: constructing an early fault risk prediction model based on the dynamic stability margin, and obtaining a fault development time prediction value;

[0009] Step S4: when the fault development time prediction value exceeds a preset threshold range, screening a fault coupling frequency combination based on the third-order resonance modal coupling coefficient, performing dynamic parameter compensation based on the fault coupling frequency combination, and obtaining a targeted parameter compensation amount;

[0010] Step S5: Compensate the corresponding parameters based on the target parameter compensation amount, and perform global knowledge evolution.

[0011] Preferably, step S1 specifically includes:

[0012] The energy adaptive capture specifically involves converting machine tool vibration into the original electrical energy through piezoelectric power generation and electromagnetic power generation.

[0013] The stable electrical energy drives the microprocessor to collect the vibration acceleration signal of the CNC machine tool, and the vibration acceleration signal is converted into a vibration velocity signal through digital integration.

[0014] A set of key resonance frequencies is predefined, and a set of core resonance frequencies is locked through knocking tests and idle tests. Multiple sets of key monitoring triplets are formed based on any combination of the core resonance frequencies.

[0015] The vibration velocity signal is processed in segments. A window function is applied to each segment of the vibration velocity signal and a Fourier transform is performed to obtain the Fourier coefficients of each segment of the vibration velocity signal at a specific core resonant frequency.

[0016] For each satisfying The key monitoring ternary group ( The coupling coefficients of the third-order resonant modes are obtained, and mathematically described as follows: ,in, Here, M represents the coupling coefficient of the third-order resonant mode, and M represents the total number of signal segments. For the m-th signal at the core resonant frequency Fourier coefficients at the location, For the m-th signal at the core resonant frequency Fourier coefficients at the location, For the m-th signal at the core resonant frequency Fourier coefficients at the specified location, * indicates complex conjugate;

[0017] The coupling coefficients of the three-order resonant modes are combined to construct a three-dimensional coupling coefficient matrix, and then the three-dimensional coupling coefficient matrix is ​​flattened into a two-dimensional form.

[0018] Preferably, the stability assessment in step S2 specifically includes:

[0019] Get matrix The maximum eigenvalue, based on which the dynamic stability margin is obtained, is mathematically described as follows: ,in, For dynamic stability margin, The largest eigenvalue, The coupling coefficient matrix, Let f be the Frobenius norm of the matrix.

[0020] Preferably, the construction of the early failure risk prediction model in step S specifically involves:

[0021] Based on the dynamic stability margin, the generalized Lyapunov exponent is obtained, mathematically described as follows: ,in, For the generalized Lyapunov index, The k-th dynamic stability margin is collected according to the time series. For the (k+1)th dynamic stability margin collected according to the time series, Where N is the sampling time interval, and N is the number of dynamic stability margin samples collected. In a stable state The natural fluctuation range;

[0022] The predicted fault development time is obtained based on the dynamic stability margin and the generalized Lyapunov exponent, mathematically described as follows: ,in, This is the predicted value for the fault development time. Critical value, For the present value.

[0023] Preferably, the dynamic parameter compensation in step S4 specifically includes:

[0024] The phase information of the fault coupling frequency combination is obtained, and the targeting parameter compensation amount is obtained based on the phase information and the third-order resonant mode coupling coefficient. Mathematically, this is described as follows: ,in, For targeted parameter compensation amount, To compensate for the gain matrix, It is a set of combinations of fault coupling frequencies. Phase information for fault coupling frequency combinations. The coupling coefficient of the third-order resonant mode is the combination of fault coupling frequencies.

[0025] Preferably, the global knowledge evolution in step S5 specifically refers to:

[0026] Initialize the globally shared knowledge model and distribute it to each CNC machine tool;

[0027] Acquire the core knowledge models of each CNC machine tool;

[0028] The global shared knowledge model is updated based on the core knowledge model, mathematically described as follows: ,in, is a global shared knowledge model at t+1 time, is a global shared knowledge model at t time, is a fusion learning rate, G is the number of numerical control machine tools participating in this round of learning, is the core knowledge model of the gth machine tool.

[0029] A numerical control machine tool health diagnosis system based on vibration spectrum coupling is used to execute the numerical control machine tool health diagnosis method based on vibration spectrum coupling described above, comprising a vibration spectrum coupling module, a stability evaluation module, a fault risk prediction module, a parameter compensation module and a global evolution module.

[0030] The vibration spectrum coupling module is used to obtain original electric energy through energy self-adaptive capture and convert the original electric energy into stable electric energy, charge the energy storage battery, and obtain a third-order resonance modal coupling coefficient, and a coupling coefficient matrix is constructed based on the third-order resonance modal coupling coefficient;

[0031] The stability evaluation module is used to perform stability evaluation based on the coupling coefficient matrix to obtain a dynamic stability margin;

[0032] The fault risk prediction module is used to construct an early fault risk prediction model based on the dynamic stability margin, and obtain a fault development time prediction value;

[0033] The parameter compensation module is used to filter out a fault coupling frequency combination based on the third-order resonance modal coupling coefficient when the fault development time prediction value exceeds a preset threshold range, perform dynamic parameter compensation based on the fault coupling frequency combination, and obtain a targeted parameter compensation amount;

[0034] The global evolution module is used to compensate the corresponding parameters based on the targeted parameter compensation amount, and perform global knowledge evolution.

[0035] The beneficial effects of the present application are:

[0036] (1) The piezoelectric-electromagnetic composite power generation unit driven by the inherent vibration of the machine tool converts the power generation process into a high-precision sensing means while realizing energy self-sufficiency, and constructs a resonance modal coupling coefficient matrix through vibration signals to capture specific nonlinear coupling characteristics representing early damage of key components such as bearings and spindles from chaotic frequency spectrum.

[0037] (2) The dynamic stability margin quantifies the degree of system health degradation, and predicts the fault evolution time window based on the generalized Lyapunov index, realizing the leap from post-maintenance to pre-warning.

[0038] (3) According to the fault coupling mode, a targeted parameter compensation strategy is generated, and parameters such as speed and feed rate are fine-tuned to suppress fault development, creating a buffer period for planned maintenance.

[0039] (4) Cross-machine knowledge sharing is achieved through a federal learning architecture, enabling swarm intelligence to continuously optimize the diagnostic model. BRIEF DESCRIPTION OF DRAWINGS

[0040] For the convenience of those skilled in the art, the present application will be further described below with reference to the accompanying drawings.

[0041] Figure 1 A step flow chart of a numerical control machine tool health diagnosis method based on vibration spectrum coupling. DETAILED DESCRIPTION

[0042] For a better understanding of the present application, various aspects of the present application will be described in more detail with reference to the accompanying drawings. It is to be understood that the detailed description is merely descriptive of illustrative embodiments of the application and is not intended, in any way, to limit the scope of the application. Throughout this document, the expression "and / or" includes any and all combinations of one or more of the associated listed items. As used herein, the terms "substantially", "approximately", and similar terms are used as terms of approximation and not as terms of degree, and are intended to account for the inherent deviations in a measuring or computing process that would be recognized by those of ordinary skill in the art. Additionally, in the present application, the order of the steps of the processes described does not necessarily indicate the order in which the processes occur in actual operation, unless explicitly otherwise limited or derivable from the context.

[0043] It should also be understood that expressions such as "include", "including", "have", "has", "contain" and / or "containing" are open-ended terms that are intended to mean that the listed item is present, but that one or more additional items can also be present. Additionally, when expressions such as "at least one of" appear after a list of two or more items, it is intended that the expression include any one of the listed items individually, as well as any combination of one or more of the listed items. Furthermore, when describing embodiments of the application, the use of "may" means "one or more embodiments of the application". Also, the use of the term "exemplary" is intended to present examples or illustrations.

[0044] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. It will be further understood that terms, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and not be interpreted in an idealized or overly formal sense unless expressly so defined herein.

[0045] It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the drawings and in combination with the embodiments.

[0046] Embodiment 1:

[0047] Please refer to Figure 1 A numerical control machine tool health diagnosis method based on vibration spectrum coupling, comprising:

[0048] Step S1: The original electric energy is obtained by energy self-adaptive capture, and the original electric energy is converted into stable and controllable stable electric energy (rectification, filtering, voltage stabilization / reduction, etc.), the energy storage battery is charged, and the three-order resonance modal coupling coefficient is obtained, and a coupling coefficient matrix is constructed based on the three-order resonance modal coupling coefficient ;

[0049] Step S2: Based on the coupling coefficient matrix, the stability evaluation is carried out, and the dynamic stability margin is obtained;

[0050] Step S3: Based on the dynamic stability margin, an early fault risk prediction model is constructed, and a fault development time prediction value is obtained;

[0051] Step S4: When the fault development time prediction value exceeds the preset threshold range, the three-order resonance modal coupling coefficient is used to screen the most critical three-element group from the key monitoring three-element group, which causes the current instability, as a fault coupling frequency combination (i.e. one / multiple three-element groups whose three-order resonance modal coupling coefficient exceeds the threshold value), and based on the fault coupling frequency combination, the dynamic parameter compensation is carried out, and the target parameter compensation amount is obtained;

[0052] Step S5: Based on the target parameter compensation amount, the corresponding parameter is compensated, and the global knowledge evolution is carried out.

[0053] In the present embodiment, the step S1 can be implemented by the following steps:

[0054] S101: The energy self-adaptive capture is specifically: converting the machine tool vibration into the original electric energy by piezoelectric power generation and electromagnetic power generation, the piezoelectric power generation is specifically: converting the vibration of the machine tool into electric energy by extrusion or bending of piezoelectric material, the electromagnetic power generation is specifically: installing a micro electromagnetic generator on the machine tool to convert the vibration of the machine tool into electric energy, the piezoelectric power generation is more suitable for high-frequency micro-amplitude vibration, and the electromagnetic power generation is more suitable for large-amplitude low-frequency shaking, and through the combination of the two methods, the energy generated by different vibration modes of the machine tool can be captured as much as possible;

[0055] S102: Collect vibration acceleration signal of the numerical control machine tool by the stable electric energy driven microprocessor, and convert the vibration acceleration signal into vibration speed signal by digital integration (high-pass filter can be applied to eliminate direct current offset and low frequency drift);

[0056] S103: Predefine a group of key resonance frequencies (such as spindle rotation frequency, bearing characteristic frequency, inner ring fault frequency, etc.), lock a group of core resonance frequencies by knocking test and idling test, the knocking test is to knock the key components such as spindle under the condition of machine tool shutdown and health state with a hammer, and measure the response with an accelerometer, and do FFT on the response signal, the peak value on the frequency spectrum corresponds to the resonance frequency of the machine tool structure, the idling test is to let the machine tool spindle idle at a typical speed, and measure the vibration frequency spectrum, those peaks that are independent of the speed and exist continuously are the structural resonance frequencies; according to the type of fault to be monitored, form a plurality of key monitoring triplets based on any combination of the core resonance frequencies (for example, the core resonance frequencies are f1, f2, f3, f4, f5 and f6, and the key monitoring triplets (f1, f2, f3), (f5, f2, f3) … can be formed);

[0057] S104: Process the vibration speed signal in sections (50% overlap), apply window function to each section of the vibration speed signal to reduce frequency spectrum leakage, and perform Fourier transform to obtain the Fourier coefficients of each section of the vibration speed signal at a specific core resonance frequency;

[0058] S105: For each key monitoring triplet that meets (the third-order resonance mode coupling coefficient is 0 if it does not meet), obtain the third-order resonance mode coupling coefficient, which is mathematically described as wherein, is the third-order resonance mode coupling coefficient, M is the total number of signal segments, is the Fourier coefficient of the mth segment of signal at the core resonance frequency is the Fourier coefficient of the mth segment of signal at the core resonance frequency is the Fourier coefficient of the mth segment of signal at the core resonance frequency is the Fourier coefficient of the mth segment of signal at the core resonance frequency is the Fourier coefficient of the mth segment of signal at the core resonance frequency is the Fourier coefficient of the mth segment of signal at the core resonance frequency, and * is complex conjugate; the third-order resonance mode coupling coefficient close to 0 indicates no coupling between frequencies, system health, and close to 1 indicates strong nonlinear coupling, indicating failure;

[0059] S106: Collect the third-order resonance mode coupling coefficients to construct a three-dimensional coupling coefficient matrix, and flatten the three-dimensional coupling coefficient matrix into two-dimensional form for subsequent data processing.

[0060] ​In this embodiment, the stability assessment in step S2 specifically includes:

[0061] Get matrix The maximum eigenvalue (whose magnitude reflects the strength of the strongest nonlinear coupling link), based on which the dynamic stability margin is derived, is mathematically described as follows: ,in, For dynamic stability margin, The largest eigenvalue, The coupling coefficient matrix, is the Frobenius norm of the matrix (which reflects the overall strength of all coupling relationships); The closer a value is to 1, the less dominant, strong nonlinear coupling it represents. A decline indicates that a specific failure coupling mode is forming and becoming dominant.

[0062] In this embodiment, the construction of the early failure risk prediction model in step S3 specifically involves:

[0063] S301: Based on the aforementioned dynamic stability margin, the generalized Lyapunov exponent is obtained, mathematically described as follows: ,in, For the generalized Lyapunov index, The k-th dynamic stability margin is collected according to the time series. For the (k+1)th dynamic stability margin collected according to the time series, Where N is the sampling time interval, and N is the number of dynamic stability margin samples collected. In a stable state The natural fluctuation range; the generalized Lyapunov index quantifies A divergence rate of a time series that is greater than 0 means that small disturbances (or the development of faults) will be amplified exponentially, and the system is moving towards instability. The larger it is, the faster it becomes unstable.

[0064] S302: Based on the dynamic stability margin and the generalized Lyapunov exponent, the predicted fault development time is obtained, mathematically described as follows: ,in, This is the predicted fault development time (i.e., the predicted time required for the current state to develop into a critical fault state). Critical points identified based on historical data as areas where significant failures are imminent. value, For the present value.

[0065] In this embodiment, the dynamic parameter compensation in step S4 specifically refers to:

[0066] obtaining phase information (by double-coherence spectrum) of the fault coupling frequency combination, obtaining the targeting parameter compensation amount based on the phase information and the third-order resonance modal coupling coefficient, and mathematically describing as wherein, is a targeting parameter compensation amount (such as a fine adjustment amount of a main shaft rotating speed, a correction amount of a feed rate, etc.), is a compensation gain matrix, is a set of fault coupling frequency combinations, is phase information of the fault coupling frequency combination, is a third-order resonance modal coupling coefficient of the fault coupling frequency combination.

[0067] In the embodiment, the step S5 of evolving the global knowledge is specifically:

[0068] S501: initializing a general and basic global shared knowledge model, and delivering the global shared knowledge model to each numerical control machine tool;

[0069] S502: obtaining a core knowledge model of each numerical control machine tool, and the core knowledge model includes a reference mode of a coupling coefficient matrix, a corresponding relationship of a dynamic stability margin and a generalized Lyapunov index, a compensation gain matrix, etc.

[0070] S503: updating the global shared knowledge model based on the core knowledge model, and mathematically describing as wherein, is the global shared knowledge model at t+1 time, is the global shared knowledge model at t time, is a fusion learning rate, G is a number of numerical control machine tools participating in this round of learning, is a core knowledge model of the gth machine tool after running for a period of time locally.

[0071] Embodiment 2:

[0072] A numerical control machine tool health diagnosis system based on vibration spectrum coupling includes a vibration spectrum coupling module, a stability evaluation module, a fault risk prediction module, a parameter compensation module and a global evolution module.

[0073] The vibration spectrum coupling module is used to obtain original electric energy through energy self-adaptive capture, and convert the original electric energy into stable and controllable stable electric energy (rectification, filtering, voltage stabilization / reduction, etc.), charge the energy storage battery, drive subsequent microprocessors, sensors and communication modules, etc., and obtain a third-order resonance modal coupling coefficient, and construct a coupling coefficient matrix based on the third-order resonance modal coupling coefficient .

[0074] The stability evaluation module is configured to perform stability evaluation based on the coupling coefficient matrix to obtain a dynamic stability margin;

[0075] The fault risk prediction module is configured to construct an early fault risk prediction model based on the dynamic stability margin to obtain a fault development time prediction value;

[0076] The parameter compensation module is configured to, when the fault development time prediction value exceeds a preset threshold range, screen out, based on the third-order resonance modal coupling coefficient, a three-element group that is most critical to current instability from the key monitoring three-element group as a fault coupling frequency combination (i.e., one / multiple three-element groups whose third-order resonance modal coupling coefficient exceeds the threshold value), perform dynamic parameter compensation based on the fault coupling frequency combination to obtain a targeted parameter compensation amount;

[0077] The global evolution module is configured to compensate corresponding parameters based on the targeted parameter compensation amount and perform global knowledge evolution.

[0078] The above is only a preferred embodiment of the present application, and does not limit the present application in any form. Although the present application has been disclosed as above with a preferred embodiment, it is not intended to limit the present application. Any person skilled in the art can make minor changes or modifications to the above disclosed technical content to obtain equivalent embodiments with equivalent changes, without departing from the technical solution of the present application. Any simple modification, equivalent change and modification of the above embodiments made according to the technical essence of the present application are still within the scope of the technical solution of the present application.

Claims

1. A method for health diagnosis of CNC machine tools based on vibration spectrum coupling, characterized in that, Includes the following steps: Step S1: Obtain raw electrical energy through adaptive energy capture and convert it into stable electrical energy to charge the energy storage battery, and obtain the third-order resonant mode coupling coefficient. Construct a coupling coefficient matrix based on the third-order resonant mode coupling coefficient. The adaptive energy capture specifically involves: converting machine tool vibration into raw electrical energy through piezoelectric and electromagnetic power generation; using the stable electrical energy to drive a microprocessor to collect the vibration acceleration signal of the CNC machine tool, and converting the vibration acceleration signal into a vibration velocity signal through digital integration; predefining a set of key resonant frequencies, locking a set of core resonant frequencies through impact and idle tests, and forming multiple sets of key monitoring triplets based on any combination of the core resonant frequencies; processing the vibration velocity signal in segments, applying a window function to each segment of the vibration velocity signal and performing a Fourier transform to obtain the Fourier coefficients of each segment of the vibration velocity signal at a specific core resonant frequency; for each segment satisfying... The key monitoring ternary group ( The coupling coefficients of the third-order resonant modes are obtained, and mathematically described as follows: ,in, Here, M represents the coupling coefficient of the third-order resonant mode, and M represents the total number of signal segments. For the m-th signal at the core resonant frequency Fourier coefficients at the location, For the m-th signal at the core resonant frequency Fourier coefficients at the location, For the m-th signal at the core resonant frequency The Fourier coefficients at the given location, * represents the complex conjugate; the coupling coefficients of the three-order resonant modes are used to construct the three-dimensional coupling coefficient matrix, and the three-dimensional coupling coefficient matrix is ​​flattened into a two-dimensional form; Step S2: Perform a stability assessment based on the coupling coefficient matrix to obtain the dynamic stability margin; Step S3: Construct an early failure risk prediction model based on the dynamic stability margin, and obtain the predicted failure development time; Step S4: When the predicted fault development time exceeds the preset threshold range, the fault coupling frequency combination is selected based on the third-order resonant mode coupling coefficient, and dynamic parameter compensation is performed based on the fault coupling frequency combination to obtain the targeted parameter compensation amount. Step S5: Compensate the corresponding parameters based on the targeted parameter compensation amount, and perform global knowledge evolution. Global knowledge refers to the shared model parameter set of the intelligent monitoring system deployed on all similar machine tools. The global knowledge evolution specifically involves: initializing the global shared knowledge model and distributing it to each CNC machine tool; obtaining the core knowledge model of each CNC machine tool; and updating the global shared knowledge model based on the core knowledge model. Mathematically, this can be described as follows: ,in, This is the global shared knowledge model at time t+1. Let be the global shared knowledge model at time t. To represent the fusion learning rate, G is the number of CNC machine tools participating in this round of learning. This is the core knowledge model for the g-th machine tool.

2. The method for health diagnosis of CNC machine tools based on vibration spectrum coupling according to claim 1, characterized in that, The stability assessment in step S2 specifically involves: Get matrix The maximum eigenvalue, based on which the dynamic stability margin is obtained, is mathematically described as follows: ,in, For dynamic stability margin, The largest eigenvalue, The coupling coefficient matrix, Let f be the Frobenius norm of the matrix.

3. The method for health diagnosis of CNC machine tools based on vibration spectrum coupling according to claim 1, characterized in that, The construction of the early failure risk prediction model in step S is specifically as follows: Based on the dynamic stability margin, the generalized Lyapunov exponent is obtained, mathematically described as follows: ,in, For the generalized Lyapunov index, The k-th dynamic stability margin is collected according to the time series. For the (k+1)th dynamic stability margin collected according to the time series, Where N is the sampling time interval, and N is the number of dynamic stability margin samples collected. In a stable state The natural fluctuation range; The predicted fault development time is obtained based on the dynamic stability margin and the generalized Lyapunov exponent, mathematically described as follows: ,in, This is the predicted value for the fault development time. Critical value, For the present value.

4. The method for health diagnosis of CNC machine tools based on vibration spectrum coupling according to claim 1, characterized in that, The dynamic parameter compensation in step S4 specifically refers to: The phase information of the fault coupling frequency combination is obtained, and the targeting parameter compensation amount is obtained based on the phase information and the third-order resonant mode coupling coefficient. Mathematically, this is described as follows: ,in, For targeted parameter compensation amount, To compensate for the gain matrix, It is a set of combinations of fault coupling frequencies. Phase information for fault coupling frequency combinations. The coupling coefficient of the third-order resonant mode is the combination of fault coupling frequencies.

5. A health diagnosis system for CNC machine tools based on vibration spectrum coupling, characterized in that, The system is applied to the CNC machine tool health diagnosis method based on vibration spectrum coupling as described in any one of claims 1-4, including a vibration spectrum coupling module, a stability assessment module, a fault risk prediction module, a parameter compensation module, and a global evolution module; The vibration spectrum coupling module is used to capture raw electrical energy through energy adaptive capture and convert the raw electrical energy into stable electrical energy to charge the energy storage battery, and to obtain the third-order resonance mode coupling coefficients, and to construct a coupling coefficient matrix based on the third-order resonance mode coupling coefficients; The stability assessment module is used to perform stability assessment based on the coupling coefficient matrix to obtain the dynamic stability margin. The fault risk prediction module is used to construct an early fault risk prediction model based on the dynamic stability margin and obtain the fault development time prediction value. The parameter compensation module is used to select a combination of fault coupling frequencies based on the third-order resonant mode coupling coefficient when the predicted fault development time exceeds a preset threshold range, and to perform dynamic parameter compensation based on the combination of fault coupling frequencies to obtain a targeted parameter compensation amount. The global evolution module is used to compensate the corresponding parameters based on the targeted parameter compensation amount and to perform global knowledge evolution.

Images