A machine tool tool wear detection method and system

Abstract

The application relates to the technical field of wear detection, in particular to a machine tool cutter wear detection method and system; the method comprises the following steps: acquiring a cutting force signal and a vibration signal in a machine tool machining process, and dividing a time period with a preset time length as a time window; taking the vibration signal on the single axis with the highest energy in each time window as a main vibration signal; calculating an environmental interference index of each time window; calculating a corrected grinding energy of each time window; based on the corrected grinding energy and the environmental interference index, calculating a comprehensive wear evaluation index of each time window; the comprehensive wear evaluation index is positively correlated with the corrected grinding energy and is negatively correlated with the environmental interference index; based on the comprehensive wear evaluation index, the cutter wear is detected; the false alarm problem caused by environmental interference in the machine tool cutter wear detection process is solved.

Description

Technical Field

[0001] This application relates to the field of wear detection technology, and in particular to a method and system for detecting wear of machine tool cutting tools. Background Technology

[0002] In the field of high-performance precision machining, the machining accuracy and long-term stability of machine tools are crucial to the consistency of product dimensions and surface quality. As core consumable parts directly involved in cutting, the wear state of machine tool cutting tools directly affects cutting force fluctuations, vibration energy distribution, machining temperature, and surface roughness. If tool wear is not identified in time, it can lead to workpiece dimensional deviations, surface burns, and even chipping, and may also cause abnormal vibrations in the spindle system, resulting in equipment damage and downtime. Therefore, accurate and real-time detection of tool wear is a crucial prerequisite for intelligent manufacturing systems to achieve self-sensing and self-decision-making. Currently, tool wear monitoring technologies mainly include direct and indirect methods. Direct methods typically rely on microscopes, cameras, or laser displacement sensors to photograph or scan the tool edge morphology, judging the degree of wear through contour changes. This method offers high accuracy but suffers from expensive equipment, weak resistance to oil contamination, and sensitivity to ambient light, making it unsuitable for online applications. Indirect methods indirectly reflect the state variables of tool wear by measuring vibration signals, cutting force signals, acoustic emission signals, and current signals generated during the cutting process, and then using statistical features or machine learning models for judgment. This method has a simple structure and can collect data in real time, making it a commonly used monitoring approach in current intelligent manufacturing production lines.

[0003] Traditional tool wear monitoring relies heavily on operator sensory experience or single signal thresholds. The former is highly subjective and unsuitable for automated production. The latter only monitors the amplitude or power spectrum energy of cutting force and vibration signals. The core problem is that it cannot distinguish between actual wear and signal changes caused by environmental interference such as sudden changes in material hardness, load fluctuations, impacts from hard particles, or loose clamping. It is easy to misjudge interference as wear, resulting in a high false alarm rate. This seriously weakens the reliability of the system and the trust of operators, making it difficult for the system to play its role in actual production or even rendering it idle. It cannot solve the problem of balancing production cycle and tool life. Summary of the Invention

[0004] To address the problem of false alarms caused by environmental interference during machine tool wear detection, this application provides a machine tool wear detection method and system.

[0005] In a first aspect, this application provides a method for detecting machine tool tool wear, employing the following techniques:

[0006] The cutting force and vibration signals during the machine tool machining process are acquired, and time windows of preset duration are defined. The vibration signal with the highest energy on a single axis within each time window is taken as the main vibration signal. The environmental interference index for each time window is calculated. The environmental interference index is positively correlated with the amplitude span of the cutting force signal within the corresponding time window, negatively correlated with the product of the length of the time window and the preset machining movement speed, and positively correlated with the concentration of energy distribution in the low-frequency band of the power spectrum of the main vibration signal.

[0007] The corrected wear energy for each time window is calculated. This corrected wear energy is positively correlated with the sum of energy densities higher than the reference power spectrum in the high-frequency band of the power spectrum, and negatively correlated with the absolute value of the difference between the sharpness of the main vibration signal and the sharpness of the reference signal. Based on the corrected wear energy and the environmental interference index, a comprehensive wear assessment index for each time window is calculated. This comprehensive wear assessment index is positively correlated with the corrected wear energy and negatively correlated with the environmental interference index. Tool wear is detected based on the comprehensive wear assessment index.

[0008] The beneficial effects are as follows: By simultaneously acquiring cutting force and vibration signals during machine tool processing, and dividing the sampled data into preset time windows, the axial signal with the highest energy within each time window is selected as the main vibration signal, thereby constructing two core quantitative indicators: environmental interference index and corrected wear energy. This distinguishes environmental factors from the actual tool wear signal, and achieves wear trend judgment through comprehensive wear assessment indicators. The technical effect is to significantly reduce false wear judgments caused by environmental interference, reduce false alarm rates, improve the stability and reliability of wear detection, and make tool life assessment more accurate, thereby optimizing tool changing strategies and improving equipment utilization.

[0009] Furthermore, the method for dividing the low-frequency band and the high-frequency band is as follows: the energy density of the power spectrum is accumulated from low to high along the frequency axis, and the frequency corresponding to when the energy density accumulation distribution reaches a first preset value is taken as the upper limit of the low-frequency band; the frequency starting point corresponding to when the energy density accumulation distribution exceeds a second preset value is taken as the lower limit of the high-frequency band.

[0010] The beneficial effects are as follows: by performing cumulative distribution analysis of the power spectrum of the vibration signal along the frequency axis, the boundaries between the low-frequency band and the high-frequency band are defined by the frequencies corresponding to the first preset value and the second preset value, thereby improving the rationality of the signal spectrum segmentation, making the low-frequency characteristics more focused on the working condition disturbance, and the high-frequency characteristics more sensitive to tool wear, effectively improving the accuracy of subsequent energy analysis.

[0011] Furthermore, the formula for calculating the environmental disturbance index is as follows: In the formula, The environmental disturbance index for the m-th time window; For the first Cutting force signal within a time window The maximum value of the amplitude; The cutting force signal within the m-th time window The minimum value of the amplitude; The length of the time window; Preset processing movement speed; Let be the concentration coefficient for the m-th time window. This represents the normalization function.

[0012] The beneficial effects are as follows: an environmental interference index is constructed, which jointly models the amplitude span of the cutting force signal, the length of the time window, the preset machining movement speed and the concentration coefficient, and uses a normalization function to eliminate dimensional differences. The intensity of non-wear disturbances caused by factors such as load change and material switching during machining is dynamically quantified. The environmental interference index is used to respond sensitively to external disturbances, thereby achieving signal suppression under environmental disturbances and providing a reference benchmark for subsequent wear energy correction, thus improving the system's adaptability to complex working conditions.

[0013] Furthermore, the concentration coefficient is obtained by performing a Fourier transform on the main vibration signal to obtain the power spectrum; calculating the arithmetic mean and geometric mean of the energy density of the power spectrum of all frequencies in the low-frequency band of the power spectrum; and using the ratio of the arithmetic mean and geometric mean as the concentration coefficient.

[0014] The beneficial effects are as follows: a concentration coefficient is constructed based on the ratio of the arithmetic mean to the geometric mean of the low-frequency power spectral density of the main vibration signal. The arithmetic mean reflects the total energy, while the geometric mean reflects the uniformity of energy distribution. The higher the ratio, the more concentrated the energy. The concentration coefficient can quantitatively characterize the degree of low-frequency energy accumulation, identify environmental vibration characteristics, provide accurate quantitative support for the environmental interference index, and improve the overall system's anti-interference capability.

[0015] Furthermore, the method for obtaining the signal sharpness is as follows: for the main vibration signal of each time window, calculate the difference between the amplitude of the main vibration signal at each sampling point and the mean of the amplitudes of all sampling points, calculate the ratio of the difference to the standard deviation of the amplitudes of all sampling points within the time window, and take the sum of all the ratios within the time window as the signal sharpness of the main vibration signal of each time window.

[0016] The beneficial effects are as follows: the signal sharpness is constructed by calculating the sum of the difference between the amplitude and the mean of the main vibration signal sampling point and the standard deviation, which reflects the transient change of the signal; the peak characteristics of the cutting process are highlighted and captured, the intensity of signal waveform change is reflected in real time, and the early wear or abnormal impact of the tool is sensitively identified, thereby improving the detection response speed and system real-time performance.

[0017] Furthermore, the method for obtaining the corrected wear energy is as follows:

[0018] Obtain the reference power spectrum and the sharpness of the reference signal;

[0019] For each time window, calculate the difference between the energy density of each frequency in the power spectrum of the main vibration signal and the energy density of the corresponding frequency in the reference power spectrum, use a positive value function to process the difference to obtain a first positive value, and calculate the normalized value of the sum of all the first positive values ​​in the high-frequency band of the power spectrum of the main vibration signal.

[0020] The absolute value of the difference between the sharpness of the main vibration signal and the sharpness of the reference signal within the calculation time window is calculated. The sum of the absolute value of the difference and the value 1 is calculated. The ratio of the normalized value to the sum is used as the corrected wear energy for each time window.

[0021] Furthermore, the method for obtaining the reference power spectrum and the reference signal sharpness is as follows: a predetermined number of cutting tests are performed on a qualified tool, and the vibration signal is recorded; the power spectrum of the vibration signal is used as the reference power spectrum, and the average value of the signal sharpness during the stable cutting phase of the vibration signal is used as the reference signal sharpness.

[0022] Furthermore, the method for obtaining the comprehensive wear assessment index is as follows: for each time window, calculate the sum of the numerical value 1 and the environmental disturbance index, and use the ratio of the corrected wear energy to the sum as the comprehensive wear assessment index for each time window.

[0023] Furthermore, the detection of tool wear based on the comprehensive wear assessment index includes: when the comprehensive wear assessment index for a consecutive preset number of time windows is greater than a preset wear threshold, the tool is determined to be in a severely worn state; when the comprehensive wear assessment index for a consecutive preset number of time windows is between a preset safety threshold and a preset wear threshold, the tool is determined to be in a wear warning state.

[0024] Secondly, this application provides a machine tool tool wear detection system, which employs the following technology:

[0025] A machine tool wear detection system includes a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a machine tool wear detection method according to the above is implemented.

[0026] The above-mentioned machine tool wear detection method is used to generate a computer program, which is then stored in a memory for loading and execution by a processor. This allows for the creation of a system based on the memory and processor, making it convenient to use.

[0027] This application has the following technical effects:

[0028] This application addresses the problems of large environmental interference, high misjudgment rate, fixed threshold and inability to dynamically reflect the true wear trend in existing machine tool wear detection. It proposes a multi-layer fusion detection method and system based on environmental interference index, modified wear energy and comprehensive wear assessment index.

[0029] This application simultaneously acquires cutting force and vibration signals during machine tool processing, dividing the signals into multiple time windows and selecting the vibration signal with the highest energy axial direction as the main vibration signal, thereby ensuring the representativeness and stability of the analyzed signals. By constructing an environmental interference index, the amplitude span of the cutting force signal, the length of the time window, and the preset processing speed are used to comprehensively reflect the degree of fluctuation in the processing environment. The stability of the machine tool's stress state is characterized by combining the concentration of energy distribution in the low-frequency power spectrum, enabling the system to adaptively identify non-wear factors such as clamping looseness and changes in material hardness. Furthermore, by constructing a corrected wear energy index, the difference between the current power spectrum and the reference power spectrum is calculated, retaining only the portion with energy higher than the reference spectrum. The reciprocal of the difference in signal sharpness is introduced as an adjustment factor to effectively suppress high-frequency spurious energy caused by transient impacts, making the changes in wear energy more realistically reflect the tool wear process. Subsequently, this application further combines the modified wear energy with the environmental disturbance index to propose a comprehensive wear assessment index. When environmental disturbance intensifies, the weight automatically decreases, thereby achieving adaptive adjustment under different working conditions. The index change trend is used to determine the status. Through a two-layer comparison of the safety threshold and the wear threshold, the tool is classified into "normal - warning - severe wear" levels. This avoids false alarms caused by single-point fluctuations and achieves smooth tracking of wear trends and early warning. It solves the problem of false alarms caused by environmental interference in the machine tool wear detection process. While ensuring detection accuracy, it significantly reduces the probability of misjudgment, extends the tool life, and improves the safety and intelligence level of machine tool operation, providing a highly reliable wear monitoring method for intelligent manufacturing. Attached Figure Description

[0030] Figure 1 This is a flowchart of a machine tool tool wear detection method according to this application. Detailed Implementation

[0031] This application discloses a method for detecting machine tool wear. It acquires cutting force and vibration signals during machine tool processing, and divides time periods into preset durations as time windows. The vibration signal with the highest energy on a single axis within each time window is taken as the main vibration signal. The environmental interference index for each time window is calculated. The corrected wear energy for each time window is calculated. Based on the corrected wear energy and the environmental interference index, a comprehensive wear assessment index for each time window is calculated. The comprehensive wear assessment index is positively correlated with the corrected wear energy and negatively correlated with the environmental interference index. Tool wear is detected based on the comprehensive wear assessment index. This method solves the problem of false alarms caused by environmental interference during machine tool wear detection.

[0032] Reference Figure 1 A method for detecting wear of machine tool cutting tools includes steps S1-S4.

[0033] Step S1: Acquire the cutting force signal and vibration signal during the machine tool processing, and divide the time period into preset durations as time windows; take the vibration signal on the single axis with the highest energy in each time window as the main vibration signal.

[0034] This application installs a three-axis accelerometer on the machine tool spindle and a dynamic cutting force sensor on the worktable. Vibration signals and cutting force signals during the tool cutting process are acquired through the three-axis accelerometer and the dynamic cutting force sensor. The vibration signals on the three axes are respectively denoted as... , , The cutting force signal is denoted as .

[0035] Furthermore, the vibration signal and cutting force signal are subjected to DC offset removal and amplitude normalization processing. The DC offset removal and amplitude normalization processing are well-known techniques and will not be elaborated upon in this application. A time window with a preset duration is set. In one embodiment of this application, the preset duration is 10 seconds, but the implementer can select other values ​​based on actual conditions. For the vibration signals on the three axes, the energy of the vibration signals on the three axes is calculated within each time window, and the vibration signal on the single axis with the highest energy is taken as the main vibration signal of the time window. .

[0036] Step S2: Calculate the environmental interference index for each time window. The environmental interference index is positively correlated with the amplitude span of the cutting force signal within the corresponding time window, negatively correlated with the product of the length of the time window and the preset machining movement speed, and positively correlated with the concentration of energy distribution in the low-frequency band of the power spectrum of the main vibration signal.

[0037] Furthermore, during the wear process of machine tool cutting tools, material switching or sudden load changes can cause sudden changes in cutting force and concentration of low-frequency vibration energy, which can easily be misjudged as wear signals.

[0038] Based on the above analysis, for the main vibration signal in each time window, a Fourier transform is performed on the main vibration signal to obtain its power spectrum. The power spectrum is divided into low-frequency and high-frequency bands. The arithmetic mean and geometric mean of the energy density of all frequencies in the low-frequency band of the main vibration signal's power spectrum are calculated. The ratio of the arithmetic mean to the geometric mean is used as the concentration coefficient for each time window. The method for dividing the low-frequency and high-frequency bands is as follows: This application analyzes the energy density curve of the vibration signal power spectrum and determines the frequency corresponding to when the cumulative energy density distribution reaches approximately 85% as the upper limit of the low frequency. When the cumulative power exceeds 90%, the starting point of the frequency is used as the lower limit of the high frequency, forming an adaptive frequency band boundary. This ensures that the low-frequency band reflects macroscopic mechanical vibration, while the high-frequency band highlights microscopic wear characteristics.

[0039] Furthermore, based on the concentration coefficient and the intensity of dynamic changes in cutting force signals within the time window, an environmental interference index for each time window is constructed, calculated using the following formula:

[0040] ;

[0041] In the formula, The environmental disturbance index for the m-th time window; For the first Cutting force signal within a time window The maximum value of the amplitude; The cutting force signal within the m-th time window The minimum value of the amplitude; The length of the time window; It is the preset machining movement speed set by the machine tool control system before machining, which is set by the operator and indicates how fast the tool travels on the workpiece; Let be the concentration coefficient for the m-th time window. The function represents the normalization function. In one embodiment of this application, the normalization method selected is maximum-minimum normalization. Implementers may select other normalization methods based on actual circumstances.

[0042] It should be noted that, Indicates the first The normalized span rate of the cutting force signal within a time window reflects the intensity of dynamic changes in cutting force per unit time; when materials are switched or the load changes abruptly, the cutting force fluctuates drastically. The normalized span rate increases significantly, resulting in a higher final environmental disturbance index. The cutting force increases; conversely, when the machining process is stable and the material is uniform, the cutting force changes gradually. The value is relatively small, indicating an environmental disturbance index. Lower; further, Refers to low frequency band The ratio of the arithmetic mean to the geometric mean of the energy density of the internal power spectrum reflects the concentration of energy distribution in the low-frequency band. When material switching or sudden load changes occur during machining, the stress state of the machine tool system changes, and vibration energy concentrates in the low-frequency band, increasing the concentration coefficient and the obtained environmental interference index. The concentration coefficient increases; conversely, when the vibration energy distribution is uniform and the processing environment is stable, the concentration coefficient is close to 1. At this time, the obtained environmental disturbance index... Lower.

[0043] Step S3: Calculate the corrected wear energy for each time window. The corrected wear energy is positively correlated with the sum of the energy densities in the high-frequency band of the power spectrum that are higher than the reference power spectrum, and negatively correlated with the absolute value of the difference between the sharpness of the main vibration signal and the sharpness of the reference signal.

[0044] Furthermore, based on the above steps, the environmental interference index is obtained. The environmental interference index can effectively reflect changes in working conditions, such as material switching or sudden load changes, but it cannot distinguish interference caused by transient impacts, such as loose clamping or impacts from hard points, which can lead to misjudgments in the detection of machine tool wear.

[0045] Based on the above analysis, and considering the changes in the amplitude of the main vibration signal, the signal sharpness of each time window is constructed, and the calculation formula is as follows:

[0046] ;

[0047] In the formula, The sharpness of the main vibration signal in each time window; The total number of sampling points within each time window. In one embodiment of this application, the total number of sampling points within a time window is 100. Implementers may select other values ​​based on actual circumstances. Let be the amplitude of the main vibration signal at the i-th sampling point in each time window. This represents the average amplitude of all sampling points of the main vibration signal within each time window. is the standard deviation of the amplitude of all sampling points of the main vibration signal in each time window.

[0048] It should be noted that in the above formula, signal sharpness is used to measure the strength of instantaneous impact or abrupt changes in the vibration signal. This application reflects the signal sharpness by comprehensively measuring the concentration and extreme value distribution of the signal waveform. This construction method can highlight the transient peak characteristics caused by tool wear, notches, or hard spots while maintaining the overall energy characteristics of the signal, thereby distinguishing between smooth cutting and abnormal impact states. The advantage of this index is that it can achieve real-time discrimination without relying on complex models. When the sharpness increases abnormally, it can indicate that the tool has entered an unsteady cutting stage or that there is a slight risk of chipping, improving the system's sensitivity and reliability to early wear signals.

[0049] Furthermore, for the main vibration signal within each time window Based on the magnitude of the energy density of the power spectrum of the main vibration signal, the corrected wear energy for each time window is constructed. The calculation formula is:

[0050] ;

[0051] In the formula, Let be the corrected wear energy for the m-th time window; It is the high-frequency band of the power spectrum of the main vibration signal in the m-th time window, and the vibration energy in this band gradually increases with tool wear; The frequency in the power spectrum of the main vibration signal in the m-th time window The corresponding energy density; The frequency in the reference power spectrum of the main vibration signal in the m-th time window The corresponding energy density, where the reference power spectrum is the power spectrum of the main vibration signal of a qualified cutting tool that is manually judged to be qualified under the same working condition; To obtain a positive function, the part of the power spectrum that is higher than the reference power spectrum is retained, while the part of the power spectrum that is lower than the reference power spectrum is set to 0. Only the energy increase is accumulated, so as to truly reflect the wear trend. For the first Main vibration signal within a time window The sharpness of the signal; The sharpness of the reference signal refers to the average sharpness during the stable cutting phase of a qualified cutting tool. value; The function represents the normalization function. In one embodiment of this application, the normalization method selected is the maximum-minimum value normalization method. Implementers may select other normalization methods based on actual circumstances.

[0052] The method for obtaining the stable cutting stage of a qualified cutting tool is as follows: After the tool is installed, a preset number of cutting tests are conducted on the qualified tool. Vibration signals and cutting force signals are recorded, and the fluctuation rates of the vibration and cutting force signals are analyzed. The fluctuation rate is the ratio of the signal variation amplitude to the signal mean. The stage with the smallest fluctuation and the longest duration in each test is obtained as the stable cutting stage. In one embodiment of this application, the preset number of tests is 10, and the implementer can select other values ​​based on the actual situation.

[0053] It should be noted that, Indicates the wear-sensitive high-frequency range Within this range, the net energy increment between the current power spectrum and the reference power spectrum shows a significant increase in high-frequency vibration energy when actual tool wear occurs. With numerous frequency points and large differences, the net energy increment increases, ultimately leading to higher corrected wear energy. The power spectrum increases; conversely, when the tool is in good condition and the vibration energy is stable, the obtained power spectrum is close to the reference power spectrum, and the summation term approaches 0, thus correcting for wear energy. reduce; To adjust the weight of the net energy increment, when there is clamping looseness or transient interference from hard point impacts during the cutting process, the main vibration signal will show a spike, and the sharpness of the signal will be... Significantly higher than the sharpness of the reference signal ,at this time, The value is much less than 1, thus significantly attenuating the net energy increment and preventing high-frequency energy surges from being misjudged as wear; conversely, when the cutting process is smooth and without abnormal impact, the signal sharpness is... near ,at this time, The value approaches 1, which does not inhibit the net energy increment, allowing it to truly reflect the tool condition.

[0054] Step S4: Based on the corrected wear energy and environmental interference index, calculate the comprehensive wear assessment index for each time window; the comprehensive wear assessment index is positively correlated with the corrected wear energy and negatively correlated with the environmental interference index; detect tool wear based on the comprehensive wear assessment index.

[0055] Furthermore, based on the above steps, the obtained corrected wear energy can reflect the wear trend, but it may still be artificially high when the operating conditions change, such as material switching or sudden load changes; while the environmental disturbance index can reflect the degree of instability of the current operating conditions.

[0056] Based on the above analysis, in order to effectively transfer wear energy when the environment is stable and suppress it when the environment is disturbed, this application constructs a comprehensive wear assessment index for each time window based on the environmental disturbance index and the corrected wear energy. The calculation formula is:

[0057] ;

[0058] In the formula, This is the comprehensive wear assessment index for the m-th time window; Let be the corrected wear energy for the m-th time window; Let m be the environmental disturbance index for the m-th time window.

[0059] It should be noted that, The reciprocal of environmental disturbance is used as the weight for correcting wear energy; the environmental disturbance index... The larger the value, the more unstable the current working condition, which may lead to material switching or sudden load changes, and the weight decreases; conversely, when the processing is stable, the weight approaches 1. This means that the wear energy is corrected by multiplying it by a weight. When the tool is actually worn and the operating conditions are stable, the wear energy is corrected. rise, The value approaches 1, therefore, the comprehensive wear assessment index It can fully reflect the degree of wear; when there are non-wear interferences such as material inhomogeneity or loose clamping, it can correct the wear energy. The index may be artificially inflated due to the increase in high-frequency energy, but the environmental interference index... Simultaneous rise, The value decreases, at which point the obtained comprehensive wear assessment index... The value is relatively small; when there is no wear and no interference, the wear energy is corrected. and environmental disturbance index The values ​​are all relatively small, at which point the obtained comprehensive wear assessment index is... The value is relatively stable.

[0060] In the above formula, this application uses the reciprocal of the environmental interference index as an adjustment weight to dynamically adjust the corrected wear energy. When the working conditions are stable, the weight is close to 1, and the corrected wear energy is fully retained. When the environmental interference is severe, the weight becomes smaller, the corrected wear energy is weakened, and the final comprehensive wear assessment index will be reduced. This can effectively prevent false alarms caused by non-wear factors such as uneven material and loose clamping, and make the comprehensive wear assessment index more reliably reflect the actual tool condition.

[0061] Furthermore, in the above steps, comprehensive wear assessment indicators for each time window were obtained. Since these indicators are affected by instantaneous fluctuations, this application analyzes their changing trends by measuring the indicators across multiple consecutive time windows. When the indicators for multiple consecutive windows exceed a preset safety threshold, a wear warning is triggered. When the indicators continue to rise and reach the preset wear threshold, it is determined to be severe wear. Ultimately, this achieves an effective distinction between instantaneous interference and actual wear, enabling tool condition assessment.

[0062] Specifically, a preset safety threshold is set based on the statistical calibration results of a large amount of experimental data. and preset wear threshold Specifically, experiments were conducted on the collected vibration and cutting force signals. Based on the entire process of the tool from healthy to slightly worn to severely worn during the experiment, the numerical distribution of the comprehensive wear assessment index at each stage was statistically analyzed. The 95th percentile of the healthy sample was used as the preset safety threshold, and the 5th percentile of the severely worn sample was used as the preset wear threshold.

[0063] In one embodiment of this application, the preset safety threshold is set to 0.4, and the preset wear threshold is set to 0.8. The implementer can select other values ​​based on actual circumstances to adjust the comprehensive wear assessment index of the window. It is compared with a threshold, specifically:

[0064] When the comprehensive wear assessment index of a consecutive preset number of time windows If the condition is deemed normal, processing can continue.

[0065] When the comprehensive wear assessment index for a consecutive preset number of time windows is When this occurs, it is determined to be a wear warning state, and it is recommended to strengthen monitoring.

[0066] When the comprehensive wear assessment index of a consecutive preset number of time windows If the wear condition is deemed severe, the system will issue a tool replacement prompt or a shutdown signal.

[0067] This application also discloses a machine tool wear detection system, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a machine tool wear detection method according to this application is implemented.

[0068] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

[0069] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for detecting wear of machine tool cutting tools, characterized in that, The process includes the following steps: acquiring cutting force and vibration signals during machine tool processing, dividing the time into preset durations as time windows; identifying the vibration signal with the highest energy on a single axis within each time window as the main vibration signal; calculating the environmental interference index for each time window, wherein the environmental interference index is positively correlated with the amplitude span of the cutting force signal within the corresponding time window, negatively correlated with the product of the length of the time window and the preset processing speed, and positively correlated with the concentration of energy distribution in the low-frequency band of the power spectrum of the main vibration signal; The corrected wear energy for each time window is calculated. This corrected wear energy is positively correlated with the sum of energy densities higher than the reference power spectrum in the high-frequency band of the power spectrum, and negatively correlated with the absolute value of the difference between the sharpness of the main vibration signal and the sharpness of the reference signal. Based on the corrected wear energy and the environmental interference index, a comprehensive wear assessment index for each time window is calculated. This comprehensive wear assessment index is positively correlated with the corrected wear energy and negatively correlated with the environmental interference index. Tool wear is detected based on the comprehensive wear assessment index.

2. The method for detecting machine tool wear according to claim 1, characterized in that, The method for dividing the low-frequency band and the high-frequency band is as follows: the energy density of the power spectrum is accumulated from low to high along the frequency axis, and the frequency corresponding to when the energy density accumulation distribution reaches a first preset value is taken as the upper limit of the low-frequency band. The frequency starting point corresponding to the cumulative energy density distribution exceeding the second preset value is taken as the lower limit of the high-frequency band.

3. The method for detecting machine tool wear according to claim 1, characterized in that, The formula for calculating the environmental disturbance index is as follows: In the formula, The environmental disturbance index for the m-th time window; For the first Cutting force signal within a time window The maximum value of the amplitude; The cutting force signal within the m-th time window The minimum value of the amplitude; The length of the time window; Preset processing movement speed; Let be the concentration coefficient for the m-th time window. This represents the normalization function.

4. The method for detecting machine tool wear according to claim 3, characterized in that, The concentration coefficient is obtained by performing a Fourier transform on the main vibration signal to obtain the power spectrum; calculating the arithmetic mean and geometric mean of the energy density of the power spectrum of all frequencies in the low-frequency band of the power spectrum; and using the ratio of the arithmetic mean and geometric mean as the concentration coefficient.

5. The method for detecting machine tool wear according to claim 1, characterized in that, The method for obtaining the signal sharpness is as follows: for the main vibration signal of each time window, calculate the difference between the amplitude of the main vibration signal at each sampling point and the mean of the amplitudes of all sampling points, calculate the ratio of the difference to the standard deviation of the amplitudes of all sampling points within the time window, and take the sum of all the ratios within the time window as the signal sharpness of the main vibration signal of each time window.

6. The method for detecting machine tool wear according to claim 1, characterized in that, The method for obtaining the corrected wear energy is as follows: Obtain the reference power spectrum and the sharpness of the reference signal; For each time window, calculate the difference between the energy density of each frequency in the power spectrum of the main vibration signal and the energy density of the corresponding frequency in the reference power spectrum, use a positive value function to process the difference to obtain a first positive value, and calculate the normalized value of the sum of all the first positive values ​​in the high-frequency band of the power spectrum of the main vibration signal. The absolute value of the difference between the sharpness of the main vibration signal and the sharpness of the reference signal within the calculation time window is calculated. The sum of the absolute value of the difference and the value 1 is calculated. The ratio of the normalized value to the sum is used as the corrected wear energy for each time window.

7. The method for detecting machine tool wear according to claim 6, characterized in that, The method for obtaining the reference power spectrum and the reference signal sharpness is as follows: a predetermined number of cutting tests are performed on a qualified tool, and the vibration signal is recorded; the power spectrum of the vibration signal is used as the reference power spectrum, and the average value of the signal sharpness in the stable cutting stage of the vibration signal is used as the reference signal sharpness.

8. The method for detecting machine tool wear according to claim 1, characterized in that, The method for obtaining the comprehensive wear assessment index is as follows: for each time window, calculate the sum of the numerical value 1 and the environmental disturbance index, and use the ratio of the corrected wear energy to the sum as the comprehensive wear assessment index for each time window.

9. The method for detecting machine tool wear according to claim 1, characterized in that, The method of detecting tool wear based on a comprehensive wear assessment index includes: determining that the tool is in a severely worn state when the comprehensive wear assessment index for a consecutive preset number of time windows is greater than a preset wear threshold; and determining that the tool is in a wear warning state when the comprehensive wear assessment index for a consecutive preset number of time windows is between a preset safety threshold and a preset wear threshold.

10. A machine tool tool wear detection system, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a machine tool tool wear detection method according to any one of claims 1-9.

Images