A CNC machine tool tool wear monitoring method and system based on acoustic vibration sensing technology

Abstract

The application discloses a kind of CNC machine tool tool wear monitoring method and system based on acoustic vibration sensing technology, which comprises: obtaining tool number, processing state from machine tool;Cutting vibration signals in the machining process of machine tool are collected by acoustic vibration sensor;The cutting vibration signals are preprocessed, time domain characteristic indexes and frequency domain characteristic indexes are extracted, and a comprehensive wear coefficient is calculated;According to the interval where the comprehensive wear coefficient is located, the state of the tool is judged;If the tool reaches the failure level, a prompt signal is sent out.The method is suitable for detecting common tool failure states and can be used in machine tools with different tool tooth numbers, different speeds and different machining processes.These indicators can be effectively used to evaluate tool wear in actual manufacturing processes, and play a crucial role in improving production capacity and promoting high-quality production.

Description

A method and system for monitoring CNC machine tool tool wear based on acoustic vibration sensing technology Technical Field

[0001] This invention relates to the field of machining tool inspection, and specifically to a method and system for monitoring CNC machine tool wear based on acoustic vibration sensing technology. Background Technology

[0002] With the rise of smart manufacturing technologies, industry is entering a new era focused on improving efficiency, profitability, and sustainability. In this context, employing CNC equipment equipped with novel sensing technologies has become a key strategy for real-time assessment of machining reliability. Especially under conditions such as intermittent cutting, high-speed operation, and heavy loads, cutting tools may suddenly fail due to mechanical or thermal stress, severely impacting machining quality and efficiency, and even leading to economic losses and safety issues. Therefore, the development and application of Tool Condition Monitoring (TCM) technology is crucial for effectively preventing catastrophic accidents and ensuring production safety.

[0003] To achieve this goal, most current solutions involve installing various types of sensors on machine tools and monitoring using fusion and big data technologies. Notably, the main focus of TCM research is monitoring tool wear, including identifying wear conditions, monitoring side wear width (VB), and predicting remaining service life (RUL). TCM employs two main approaches: direct monitoring and indirect monitoring. Direct monitoring relies on machine vision, which presents challenges in effectively detecting sudden tool breakage during machining. Furthermore, it is susceptible to interference from cutting fluid, lighting conditions, and chips, thus requiring interrupted measurements. Conversely, indirect monitoring methods in TCM utilize data-driven approaches and embedded sensors within the manufacturing system, providing a cost-effective solution for industrial applications. During the cutting process, a large amount of system feedback regarding tool condition is generated. The physical signals used for data analysis can be collected by various sensors, such as cutting forces, acoustic emissions (AE), vibration, sound, spindle motor power, and current. Tool life prediction is achieved by extracting relevant features from the physical signals and performing parameter optimization, model training, and deep learning.

[0004] The aforementioned existing methods have numerous risks and drawbacks in their use:

[0005] Option 1: Optical methods using external detection probes such as infrared probes. This typically involves using lasers to measure surface finish or fiber optic sensors to measure the reflectivity of worn and unworn areas, calculating the blade breakage value based on surface finish and reflectivity. This method is susceptible to interference from light and chips, generally requires stopping the machine for monitoring, and the detection process is complex. It cannot obtain accurate wear values, only a relative range.

[0006] Option Two: Cutting Force Method. Different degrees of tool wear cause changes in cutting force, therefore, the cutting force signal is often used as a monitoring signal. This method commonly uses stress plate sensors, piezoelectric elements, and a three-dimensional force measuring stage to collect cutting force signals. The signal changes are significant, unaffected by the machining environment, and has strong anti-interference capabilities. However, sensor installation is difficult, sometimes requiring machine tool modification, and the cost is relatively high.

[0007] Method Three: Current and Power Method. In machining, when machining parameters remain constant, different degrees of tool wear will alter the cutting force, thus causing changes in the current and power of the spindle motor. Therefore, changes in current and power can predict cutting force and tool wear. This method requires no sensor installation or machine tool modification and is unaffected by other machining factors, but its sensitivity is relatively low and its applicability is limited. It is often used in conjunction with cutting force signals to improve the accuracy of the cutting force signals. Summary of the Invention

[0008] Based on the defects and shortcomings of the above-mentioned solutions, this invention uses an acoustic emission method with strong anti-interference and wide adaptability to construct a low-cost CNC tool monitoring system, create characteristic indicators for the wear degree of the tool at different stages, realize full-cycle life monitoring of the tool, and overcome the above-mentioned technical problems.

[0009] To achieve the above objectives, this invention provides a CNC machine tool tool wear monitoring method based on acoustic vibration sensing technology, comprising: obtaining the tool number and machining status from the machine tool; collecting cutting vibration signals during the machining process using an acoustic vibration sensor; preprocessing the cutting vibration signals to extract time-domain and frequency-domain feature indicators, and calculating a comprehensive wear coefficient; determining the tool status based on the range of the comprehensive wear coefficient; and issuing a warning signal if the tool reaches a failure level.

[0010] A further improvement of the present invention is that:

[0011] In calculating the amplitude spillover ratio (ASR), a tolerable working range is constructed based on sample data of cutting vibration signals from normal tools. Its expression is:

[0012]

[0013] Among them, S w S indicates the tolerable working range of the cutting tool. env S represents the average value of the time signal envelope of the cutting vibration signal in the sample data. std The standard deviation of the time envelope of the cutting vibration signal in the sample data, t s This indicates the time-domain tolerance; it also indicates the tolerable level of tool wear, which can be customized.

[0014] Due to the highly complex nonlinear vibration time response during tool machining, tool chipping and breakage can occur instantaneously and within a very short time. Strong acoustic emission vibration signals are emitted from within the structure, resulting in significant spikes in the time-domain signal. An amplitude spillover ratio (ASR) metric is constructed, which is the ratio of the spillover portion integrated with exponential weights to the value obtained from the working area.

[0015]

[0016] Where γ represents amplitude sensitivity, and A is the envelope range of the vibration signal of the tool to be monitored; A w It is the tolerable working range of the tool vibration signal envelope; the ASR index can simultaneously identify the acoustic emission energy caused by cracks and chipping from the time and amplitude dimensions, and realize the tool damage monitoring function through weight amplification.

[0017] Frequency domain characteristic indicators include peak offset coefficient (POC), spectrum noise ratio (SNR), and wear energy coefficient (WEC).

[0018] The Peak Offset Coefficient (POC) reflects energy dissipation during machining, and its significance only becomes apparent after significant tool wear. From the acoustic-vibration response signal spectrum, frequency components fluctuate with the degree of tool wear. The dominant frequency is characterized by higher energy and is more sensitive to tool condition. Tool wear can lead to reduced cutting force, cutting point slippage, or increased friction, all of which increase machining damping and nonlinearity, resulting in energy dissipation and offset of the dominant frequency. Therefore, the Peak Offset Coefficient (POC) is constructed, considering two dimensions of change—amplitude and frequency—and establishing a normal operating frequency range. This range is conceptualized as a circular region, and the POC is defined by the following formula:

[0019]

[0020] Among them, POC i A i and f i Let A represent the characteristic value, peak amplitude, and peak frequency of the cutting vibration signal in the i-th process (here, process refers to the machining process of the monitored tool), respectively. d and f d This represents the amplitude and frequency of the average spectrum of the dominant frequency in the sample data; POC m This represents the average value of the peak offset coefficient (POC) in the sample data. std The standard deviation of the peak offset coefficient (POC) in the sample data, t fSpectrum tolerance represents the acceptable degree of tool wear; POC calculates the deviation between the eigenvalue and the average value of the i-th process.

[0021] The spectral noise ratio (SNR) is used to identify gradual changes in tool wear as the number of interference peaks increases with tool wear. The characteristics of the differential spectrum lie not only in the dominant frequency but also in the noise generated by unstable vibrations caused by tool wear.

[0022] Tool wear begins with single-tooth chipping and gradually develops into multi-tooth wear. In the early stages of wear, the cutting force of the tool exhibits instability due to changes in the depth of cut. Amplitude modulation occurs in the tool, and slippage during the cutting process causes frequency components to deviate from the rotational frequency. These deviated frequency components are also considered interference peaks. Compared with the spectrum in the initial state, some interference peaks show increased energy or the appearance of new frequency components.

[0023] When quantifying tool wear, a threshold amplitude needs to be defined, and amplitudes above this threshold are considered interference peaks. The degree of tool wear can be quantified by calculating the number of interference peaks observed in the differential spectrum. Generally, tool wear causes energy to disperse from the dominant frequency and transfer to other characteristic frequencies or interference peaks, resulting in a decrease in amplitude. By selecting a threshold amplitude with a fixed coefficient based on the dominant frequency, the threshold amplitude adjusts proportionally as the dominant frequency decreases. Therefore, this adjustment amplifies the number of interference peaks, thereby improving the accuracy of damage identification. The expression for the spectral noise ratio (SNR) is:

[0024]

[0025] Among them, SNR i This represents the characteristic value of the i-th process. This represents an indicator function, where x > 0 and the function is 1, otherwise 0; A if The amplitude of the cutting vibration signal spectrum corresponding to the i-th process for each frequency component is represented by T; T is the threshold coefficient, i.e., A d / T represents the threshold; SNR m It is the average value of the spectral noise ratio (SNR) in the sample data; SNR std It is the standard deviation of the spectral noise ratio (SNR) in the sample data;

[0026] The Wear Energy Coefficient (WEC) index primarily reflects the initial wear of the tool. As tool wear intensifies, energy dissipation gradually increases, leading to interference peaks falling below a threshold. At higher wear levels, the number of interference peaks increases significantly. With later tool wear, the cutting force on the cutting edge decreases, resulting in increased frictional damping. This energy transfer from the rotational frequency leads to an increase in the number of interference peaks. However, the energy becomes more dispersed, and even interference peaks with amplitudes below the threshold may become undetectable. A WEC index is constructed to summarize the energy of interference peaks exceeding the threshold: the expression for the Wear Energy Coefficient (WEC) index is:

[0027]

[0028] Among them, WEC i Let A represent the characteristic value of the i-th process. n Indicates the amplitude of the interference peak exceeding the threshold amplitude; A d / T represents the threshold; WEC m It is the average value of the wear energy coefficient (WEC) in the sample data. std It is the standard deviation of the wear energy coefficient index WEC in the sample data.

[0029] To achieve full-cycle tool life monitoring, it is necessary to integrate the characteristics of the above indicators to identify the three stages of tool wear: initial wear, uniform wear, and accelerated wear. The formula for calculating the comprehensive wear coefficient (CWI) is:

[0030] CWI=α·ASR+β·WEC+γ·POC+δ·SNR

[0031] Wherein, α, β, γ, and δ are the weighting coefficients for the amplitude spillover ratio (ASR), peak offset coefficient (POC), spectral noise ratio (SNR), and wear energy coefficient (WEC), respectively. The weighting coefficients α, β, γ, and δ are determined through weight allocation, parameter optimization, or deep learning methods.

[0032] A further improvement of the present invention is that when a tool reaches a failure level, the form of the warning signal issued includes: illuminating the system alarm light, controlling the machine tool to stop, and displaying the tool number of the failed tool through the user interface.

[0033] This method introduces four innovatively designed metrics, each with varying sensitivity, suitable for detecting common tool failure conditions and applicable to machine tools with different tooth counts, rotational speeds, and machining processes. These metrics are effectively used to assess tool wear in actual manufacturing processes, playing a crucial role in improving productivity and promoting high-quality production.

[0034] This invention also provides a CNC machine tool tool wear monitoring system based on acoustic vibration sensing technology, used to execute the above-described CNC machine tool tool wear monitoring method based on acoustic vibration sensing technology. The system includes:

[0035] A sound and vibration sensor is fixedly mounted on a machine tool to detect cutting vibration signals;

[0036] An analog signal acquisition card is electrically connected to the acoustic vibration sensor and is used to receive cutting vibration signals, preprocess the cutting vibration signals, and convert them into digital signals.

[0037] The signal processing module communicates with both the machine tool and the analog signal acquisition card. It is used to obtain the tool number and machining status from the machine tool and to calculate the comprehensive wear coefficient. After signal processing, it provides real-time tool status feedback and wear warning.

[0038] The signal processing module is also used to initialize the entire system, including sampling parameters, machine tool information table, time-domain and frequency-domain characteristic parameters and indices, and reading the tool sample library. The tool sample library stores sample data used to calculate time-domain and frequency-domain characteristic indices.

[0039] Furthermore, to minimize environmental interference and improve the signal-to-noise ratio, coaxial cables and I / O buses facilitated the integration of the data acquisition module and the high efficiency of dual-module data communication. This integration method allows for real-time monitoring of tool wear during machining.

[0040] The design and implementation of this system greatly simplifies the installation process and is extremely cost-effective. At the same time, by acquiring the machining information of the machine tool, it can monitor the wear status of multiple processes and multiple tools, and provides an effective data acquisition method and monitoring system structure for data-driven monitoring methods such as acoustic vibration technology or other sensing technologies.

[0041] A further improvement of the present invention is that the acoustic vibration sensor is an encapsulated acoustic vibration sensor, which is fitted and mounted on the spindle or spindle mount of the machine tool.

[0042] In this invention, based on theoretical case analysis of a cutting vibration model, the frequency variation law is obtained, leading to the construction of four tool wear indices with different sensitivities. These indices are then integrated into a comprehensive wear factor, enabling full-cycle tool life prediction with minimal data. Simultaneously, the monitoring system, through the fusion of system, machine tool, and sensor information, achieves low-cost, simplified, and real-time production requirements for multi-process, multi-tool fully automated manufacturing. This provides strong theoretical and technical support for the current field of tool wear monitoring, possessing significant social and economic value.

[0043] The advantages of this invention are:

[0044] 1. Low cost and simplified system: The system structure of this invention is clear and simple. It only requires a packaged acoustic vibration sensor, an analog signal acquisition card and a low cost signal processing system to realize the monitoring function.

[0045] 2. High scalability and customizability: The system of this invention can be integrated with multi-sensor signal fusion for optimized wear monitoring and improved accuracy. For machine tools of the same model but different processes, only the tool feature sample library needs to be changed to run. For different machine tool models, collaborative operation can be achieved by adjusting the communication protocol between the machine tool and the system.

[0046] 3. Multiple perception of damage coefficients: Indicators obtained through theoretical simulation are more targeted, with different damage indicators showing sensitivity to different stages of tool development. This allows for customization of the monitoring focus based on industrial needs. Utilizing big data and machine learning technologies can optimize the fitting and allocation of metric weights, thereby improving the accuracy and reliability of wear diagnosis.

[0047] 4. Future advancements in quantifying and improving the accuracy of tool wear analysis will facilitate the adjustment of machining parameters and tool specifications. This, in turn, will ensure the stable production of qualified products, thereby improving yield and tool life. These developments are of forward-looking and strategic significance to the field of intelligent manufacturing. Attached Figure Description

[0048] Figure 1 is a schematic diagram of the CNC machine tool tool wear monitoring system based on acoustic vibration sensing technology in this invention.

[0049] Figure 2 is a simplified flowchart of the tool wear monitoring system in this invention;

[0050] Figure 3 is a simplified diagram of the algorithm for extracting time-domain features;

[0051] Figure 4 is a simplified diagram of the algorithm for extracting spectral features.

[0052] Figure 5 is a schematic diagram of the actual application of the CNC machine tool tool wear monitoring system based on acoustic vibration sensing technology in this invention;

[0053] Figure 6 is a schematic diagram of three wear states of cutting tools;

[0054] Figure 7 is a schematic diagram of the workpiece surface roughness corresponding to the three wear states of the cutting tool;

[0055] Figure 8 is a schematic diagram of the quantitative results of the amplitude spillover ratio index;

[0056] Figure 9 is a schematic diagram of the quantitative results of the peak offset coefficient;

[0057] Figure 10 is a schematic diagram of the quantitative results of the spectral noise ratio;

[0058] Figure 11 is a schematic diagram of the quantitative results of the wear energy coefficient. Detailed Implementation

[0059] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be noted that, unless otherwise specified, the following embodiments and features described therein can be combined with each other.

[0060] Some exemplary embodiments of the invention have been described for illustrative purposes. It should be understood that the invention may be implemented in other ways not specifically shown in the accompanying drawings.

[0061] This embodiment only takes the identification of the wear state of a single tool as an example to illustrate the CNC machine tool wear monitoring system based on acoustic vibration sensing technology. In actual application, the system can monitor and provide feedback on the wear of multiple tools in multiple processes in real time.

[0062] (1) CNC machine tool tool wear monitoring system based on acoustic vibration sensing technology:

[0063] As shown in Figures 1 and 5, the CNC machine tool tool wear monitoring system based on acoustic vibration sensing technology in this embodiment consists of a machining center (Mazark), an NI data acquisition card (NI-6366), and a LabVIEW-based monitoring system. The strategy of mounting the aluminum-encapsulated piezoelectric acoustic vibration sensor near the machine tool spindle maximizes the system's vibration response and improves the data signal-to-noise ratio and feature sensitivity.

[0064] Data exchange between the FANUC machine tool system and the LabVIEW monitoring system is achieved through I / O cable interfaces and the MODBUS communication protocol. Machine tool data includes basic information such as tool machining coordinates, the current tool number, and the current machining status. When the spindle displacement along the Z-axis drops below 1, the monitoring system triggers the data acquisition card to begin recording the response signal.

[0065] Subsequently, the system's nonlinear response is converted into an analog electrical signal using an aluminum-encapsulated piezoelectric vibration sensor. An NI data acquisition card acquires the vibration response signal at a sampling frequency of 50 kHz and transmits it to the signal processing module.

[0066] As shown in Figures 6 and 7, during long-term tool processing, 15 sets of processing data for each of the three tool states in the intermediate section were selected: initial state, wear 1, and wear 2. The initial state was used as the benchmark sample for comparative analysis. Wear 1 and wear 2 are mainly experimental states. Wear 2 indicates accelerated wear, and the appearance of burrs on the later product indicates a failed tool. (The specific failure level is mainly determined based on the definition of non-conformity of the sample.) In specific implementation cases, only very obvious burrs constitute failure; in industrial applications, wear 1 might already indicate failure. The comprehensive index can be improved by adjusting the parameter weights. The specific tool state is determined by the parameter sample, and the threshold of the comprehensive wear coefficient can also be adjusted through parameters, depending on the usage target and application scenario. In some specific implementation cases, the normal state is below 0.4, the initial wear fluctuates within a large range, the accelerated wear gradually approaches 1, and exceeding 1 indicates failure. In this embodiment,

[0067] The degree of tool wear can be inferred from the surface roughness of the machined workpiece. Wear 1 indicates initial wear, with burrs beginning to appear on the sample edges, but the degree is relatively small. Wear 2 indicates advanced wear or even failure, with a significant increase in burrs in the sample grooves.

[0068] (2) CNC machine tool wear monitoring method based on acoustic vibration sensing technology:

[0069] As shown in Figures 2, 3, and 4, in the signal preprocessing, the acquired real-time vibration signal is low-pass filtered and the sampling rate is reduced to 5kHz.

[0070] In the test metric algorithm, the time-domain tolerance t s Set to 3, amplitude sensitivity to 100, and spectral tolerance t. f The threshold coefficient T is set to 100. The working area in the time domain features is constructed using 15 initial samples, while the frequency domain features mainly analyze the 1kHz frequency range.

[0071] Figure 8 shows the quantitative results of ASR. Notably, in the original sample, the ASR value was almost zero. However, after artificial wear of the cutting tool, due to the unevenness of the cutting force and the instability of the vibration response, the ASR value exhibited significant irregular oscillations. The extreme ASR value during the fourth machining operation reflected the peak of the time response signal, indicating that chipping occurred during this machining process. After high wear, the cutting force decreased significantly, the vibration response weakened, and fine cracks and chipping continued to appear. Therefore, the ASR value oscillated repeatedly within a certain range. The experimental results show that the ASR index is highly effective in monitoring tool chipping and fracture.

[0072] 2. Figure 9 shows the quantification results of POC. The initial tool POC value fluctuated consistently below 0.4. Using 0.4 as the baseline, in wear 1, the POC value partially exceeded the threshold. In wear 2, the POC value became even more unstable, with POC values ​​exceeding the baseline in multiple datasets. This result indicates that POC is more sensitive to later tool wear.

[0073] 3. Figure 10 shows the quantification results of SNR. For the initial sample, the SNR value remained below 0.4, even lower, with 0.4 as the baseline. In wear 1, the SNR value increased slightly, mostly exceeding the baseline value. In wear 2, the SNR value deviated significantly from the baseline, showing an overall gradually increasing trend. Therefore, SNR is more sensitive to the later stages of tool wear.

[0074] Given the quantitative characteristics of POC and SNR, both of these metrics can clearly provide more significant feedback on the later stages of tool wear.

[0075] 4. Figure 11 shows the quantification results of WEC. In the initial sample, the WEC values ​​fluctuated very little, remaining below 0.2. In wear 1, the WEC values ​​increased significantly, all exceeding 0.6. As tool wear intensifies, the lack of cutting force and the dispersion of energy lead to an increase in the number of interference peaks and a decrease in their respective energy ratios. Therefore, a slight decrease in WEC values ​​was observed. The WEC index is significantly more sensitive to early tool wear.

[0076] Initially, the tool's CWI value remained stable below 0.4 with minimal fluctuation. The initial wear of the tool exhibited distinct stratification, fluctuating between 0.4 and 0.8, deviating from its original state. Under severe tool wear, the CWI index continuously increased, rapidly exceeding the wear range and eventually surpassing 1. The results effectively distinguished three stages of tool wear: initial wear, uniform wear, and accelerated wear. With this weighted allocation, the CWI index successfully achieved the function of monitoring tool damage at all stages.

[0077] Finally, it is worth noting that the weighting in this method is based on statistical principles. However, the specific weighting of indicators should be fine-tuned using machine learning and other parameter adjustment methods based on the real-time status of the tool, thereby achieving high-precision tool wear identification.

[0078] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. A method for monitoring CNC machine tool tool wear based on acoustic vibration sensing technology, characterized in that... include: Obtain the tool number and machining status from the machine tool; The cutting vibration signal during machine tool processing is collected by an acoustic vibration sensor; the cutting vibration signal is preprocessed to extract time-domain and frequency-domain characteristic indicators, and a comprehensive wear coefficient is calculated; the tool condition is determined based on the range of the comprehensive wear coefficient; if the tool reaches the failure level, a warning signal is issued. The time-domain characteristic indicators include the amplitude overflow ratio indicator. ； Frequency domain characteristic indicators include peak offset coefficient indicators Spectrum noise ratio Wear energy coefficient index Overall wear coefficient The calculation expression is: ;in: These are the amplitude spillover ratio indicators. Peak offset coefficient index Spectrum noise ratio Wear energy coefficient index The corresponding weighting coefficients; used in calculating the amplitude spillover ratio index During the process, a tolerable working range is constructed based on sample data of the cutting vibration signal from a normal tool. Its expression is: ;in, Indicates the tolerable working range of the cutting tool. This represents the average value of the time signal envelope of the cutting vibration signal in the sample data. The standard deviation of the time signal envelope of the cutting vibration signal in the sample data. Indicates time-domain tolerance; amplitude spillover ratio index The expression is: Where γ represents amplitude sensitivity, It is the envelope range of the vibration signal of the tool to be monitored; It is the tolerable operating range of the tool vibration signal envelope; peak offset coefficient index. In the calculation process, its expression is: ;in, 、 and They represent the first The characteristic values, peak amplitude, and peak frequency of the cutting vibration signal during each process. and This represents the amplitude and frequency of the average spectrum of the dominant frequency in the sample data; This represents the average value of the peak offset coefficient (POC) in the sample data. The standard deviation of the peak offset coefficient (POC) in the sample data. Spectrum tolerance represents the acceptable degree of tool wear; it is used in calculating the spectrum noise ratio. During the process, its expression is: ;in, Indicates the first The characteristic values ​​of the process, Indicates the indicator function, i.e., if The value is 1 if it is 1, otherwise it is 0. Represents the first frequency component corresponding to each frequency component. The amplitude of the cutting vibration signal spectrum during the process; The threshold coefficient, i.e. Indicates the threshold; It is the average value of the spectral noise ratio (SNR) in the sample data; It is the standard deviation of the spectral noise ratio (SNR) in the sample data; in calculating the wear energy coefficient index. During the process, its expression is: ;in, Indicates the first The characteristic values ​​of the process, This indicates the amplitude of the interference peak that exceeds the threshold amplitude; Indicates the threshold; It is the average value of the wear energy coefficient (WEC) in the sample data. It is the standard deviation of the wear energy coefficient index WEC in the sample data.

2. The CNC machine tool tool wear monitoring method based on acoustic vibration sensing technology according to claim 1, characterized in that: Weighting coefficients Parameters can be determined through weight allocation, parameter optimization, or deep learning methods.

3. The CNC machine tool tool wear monitoring method based on acoustic vibration sensing technology according to claim 1, characterized in that: When a tool reaches a failure level, the warning signals issued may include: illuminating the system alarm light, stopping the machine tool, and displaying the tool number of the failed tool through the user interface.

4. A CNC machine tool tool wear monitoring system based on acoustic vibration sensing technology, used to execute the CNC machine tool tool wear monitoring method based on acoustic vibration sensing technology as described in any one of claims 1 to 3, characterized in that, The system includes: an acoustic vibration sensor, fixedly mounted on the machine tool, for detecting cutting vibration signals; an analog signal acquisition card, electrically connected to the acoustic vibration sensor, for receiving cutting vibration signals, preprocessing the cutting vibration signals and converting them into digital signals; and a signal processing module, communicatively connected to both the machine tool and the analog signal acquisition card, for obtaining tool number and machining status from the machine tool and for calculating the comprehensive wear coefficient.

5. A CNC machine tool tool wear monitoring system based on acoustic vibration sensing technology according to claim 4, characterized in that: The acoustic vibration sensor is an encapsulated acoustic vibration sensor, which is fitted and mounted on the spindle or spindle mount of the machine tool.

Images